Coping with Syncope: Tilt Towards Pacingindianheartjournal.com/pdf/July-Aug2001.pdf · Coping with Syncope: Tilt Towards Pacing VK Bahl, Gautam Sharma Department of Cardiology, All

Indian Heart J 2001; 53: 415–422 Bahl et al. Management of Vasovagal Syncope 415

Coping with Syncope: Tilt Towards Pacing

VK Bahl, Gautam SharmaDepartment of Cardiology, All India Institute of Medical Sciences, New Delhi

Syncope, defined as a transient loss of consciousnessand postural tone, is one of the oldest recorded medical

problems. The term syncope is derived from the Greekword “synkoptein”, meaning “to cut short”. One of the mostfrequently encountered forms of syncope, the vasovagalor the “common faint” is caused by inappropriatereflex vasodilatation and bradycardia and is rarelylife threatening.1,2 Even in patients, with the so-called“malignant” form of the disease, who have profoundcardioinhibition culminating in asystole during tilt tests,no death has been reported on long-term follow-upstudies.3,4 However, patients with vasovagal syncope are atan increased risk of a fall, which can often result in seriousinjuries. Those who work as pilots, drivers, etc. can endangerothers as well. Patients who develop syncope withoutenough warning or prodromal symptoms are unable tobrace themselves for the usually inevitable “fall”. Thus,patients with syncope have a level of impairment similar tothat seen in patients with severe rheumatoid arthritis andpsychiatric illness.5,6 The quality of life deteriorates as thefrequency of syncopal spells increases. Therefore, vasovagalsyncope is not “truly benign” and these patients need to betreated. At present, there is no consensus on themanagement of patients with vasovagal syncope. Ifsyncopal episodes are infrequent, simple nonpharmacologicmeasures usually suffice. In those with severe symptoms,tilt test-guided therapy or empiric pharmacotherapy isrecommended. If drugs fail, permanent pacemakers areadvocated. However, this treatment algorithm has to beindividualized, taking into consideration age, occupation,symptomatic status and response to head-up tilt test(HUTT).

Non pharmacological Methods

Salt supplementation and increased fluid intake, byincreasing the extracellular fluid volume, may reduce theimpact of orthostatic blood pooling. These simple measuresare commonly recommended as initial therapy. Althoughlarge controlled trials are lacking, they have been shown tobe of some benefit in small studies.7,8

Indian Heart J 2001; 53: 415–422 Editorial

It may be of help to teach the patient about thepathophysiology of vasovagal syncope and ways to avoidknown situational triggers. Medications such asperipherally acting alpha-antagonists and nitrates shouldbe stopped. Several studies have demonstrated that theprobability of having a syncopal recurrence is significantlyreduced after a tilt test. This could be because of thecounseling and education that patients receive after the firsttest. The benefit of “tilt training” or orthostatic training hasbeen reported in a recent study.9 Forty-two patients wereinstructed to continue a program of daily tilt training athome: two 30-min sessions of upright standing against avertical wall. No medication was prescribed. At follow-upof 15.1+7.8 months, 36 patients remained completely freeof syncope.

Pharmacotherapy

Several pharmacologic agents targeting the afferent limb(fludrocortisone), efferent limb (beta-blockers) or both(midodrine) of this reflex-mediated syndrome have beenadvocated to manage these cases. Many small,nonrandomized trials have been reported using beta-blockers, clonidine, disopyrimide, angiotensin-convertingenzyme inhibitors, dextroamphetamine, etilefrine,fludrocortisone, methylphenidate, midodrine,phenylephrine, propan-theline, scopolamine, serotonine-reuptake inhibitors and theophylline. However, as thenatural history of this disease entity can be variable, criticalassessment of the efficacy of any pharmacologic agentrequires a randomized clinical trial.

Beta-blockers : These are the most widely prescribed drugsin clinically encountered vasovagal syncope, often asempiric therapy. They presumably act by inhibiting theactivation of ventricular mechanoreceptors and by blockingthe effects of surging catecholamines in the circulation.

Despite extensive clinical experience with the drug, thereis a dearth of controlled studies. Several nonrandomizedstudies10–19 have demonstrated its clinical efficacy withrecurrence rate reduced to 6%–28% with treatment. In alarge study by Cox et al.,15 296 consecutive patients wereput on tilt-guided or empiric beta-blocker therapy. Duringa follow-up of 28±11 months, 90% of patients on tilt-

Correspondence: Dr VK Bahl, Professor of Cardiology, All India Instituteof Medical Sciences, New Delhi 110029. e-mail: [email protected]

IHJ-Editorial.p65 10/17/01, 3:55 PM415

416 Bahl et al. Management of Vasovagal Syncope Indian Heart J 2001; 53: 415–422

guided beta-blocker therapy remained symptom-freecompared with 77% on empiric therapy. Similar results werereported in another large nonrandomized study by Nataleet al.19 They demonstrated a 6% recurrence rate ofsymptoms on tilt-guided beta-blocker therapy among 210patients, while the recurrence rate was 36% when empiricbeta-blocker therapy was given.

Only two randomized, placebo-controlled trials20,21 havebeen published. In the first double-blind, randomized,placebo-controlled study by Mahanonda et al.,20 42 patientswith a positive tilt test were randomized to atenolol orplacebo. They were subjected to a repeat tilt test after onemonth. The response rates (negative HUTT) were 62% v.5% (p=0.0004) for the atenolol and control groups,respectively. Moreover, 71% patients reported a subjectivesense of well-being in the atenolol group compared with29% in the placebo group (p=0.02). The frequency ofsyncopal attacks reduced from 6.0+9.4 per week to 0.6+1.6per week in the atenolol group (p=0.025). This was the firstcontrolled trial demonstrating the benefit with beta-blockers. However, many a question regarding the long-term efficacy of the drug, its tolerability and side-effectsremained unanswered due to the short follow-up of onlyone month.

The second study was published 6 years after the earliertrial. In this prospective, double-blind, randomized andplacebo-controlled study,21 50 patients with syncope, ofwhom 20 (40%) had an abnormal tilt test, were randomizedto atenolol or a placebo. During the one year of follow-up,16 patients on atenolol and 11 patients on placebo hadrecurrence of syncope (61% v. 45%, p=0.09). The authorsconcluded that beta-blockers were no more beneficial thanplacebo. However, only 20 patients (40%) in this study hadan abnormal response to tilt testing at baseline. Previousstudies, including the favourable study by Mahanonda etal.,20 included only patients with a positive tilt test in thetrial. The difference in the inclusion criteria in the twostudies may have had a significant impact on the resultsobtained. It is indeed surprising that the investigatorsassigned even patients with a negative tilt test (steep angleof 80° for 45 min) to the trial. There are no prior studies inthe literature showing the efficacy of elective beta-blockertherapy in patients with presumed neurocardiogenicsyncope. The fact that only 5 patients included in this studyhad responded favorably to intravenous atenolol may alsobe responsible for the reported high incidence of recurrence.Previous studies15,17,22 have demonstrated that patients whorespond to intravenous beta-blockers during tilt testing, alsohave a favorable response to oral beta-blockers on long-termfollow-up. In the study by Sra et al.,17 all patients with a

negative tilt test during esmolol infusion also had a negativetilt response with oral metoprolol. Of the patients with apositive tilt test during esmolol infusion, 90% continued tohave a similar response with oral therapy. Because only 5patients (25%) had responded favorably to intravenousatenolol in the study by Madrid et al.21 it is not surprisingthat the response to oral atenolol was correspondingly poor.Although the authors had questioned the efficacy of beta-blockers in vasovagal syncope, the conclusions of the studyneed to be re-examined and interpreted critically.

It appears that beta-blockers are ef fective in thetreatment of patients with vasovagal syncope; however,further scrutiny is needed in large randomized trials withappropriate numbers and extended follow-up. For now, beta-blockers retain their hallowed place as the first line oftreatment for vasovagal syncope. Other medications shouldbe used if the patient does not respond or does not toleratebeta-blockers because of side-effects such as bradycardia,impotence, bronchospasm or AV blocks.

Fludrocortisone: There is a wealth of clinical experiencewith fludrocortisone. Though controlled studies are lacking,it is commonly used to treat vasovagal syncope, especiallyin children. Fludrocortisone is a mineralocorticoid that actsby increasing the renal absorption of sodium, therebyexpanding the blood volume. It may also affect thebaroreceptor sensitivity and enhance the ef fects ofcirculating catecholamines by increasing the sensitivity ofblood vessels to their vasconstrictive effects.23

In a study by Scott et al.,12 58 pediatric patients wererandomized to receive atenolol or fludrocortisone acetate.At the end of 6 months, 48 patients (83%) showed clinicalimprovement (no recurrence or significant decrease infrequency/severity of episodes). Both drugs were equallyeffective. Neither the response to intravenous beta-blockers,nor the presence of tachycardia was predictive of long-termresponse to oral beta-blockers. This is contrary to earlierstudies in which a positive response to intravenous beta-blockers predicted a favorable long-term response to oralbeta-blockers.15,17,22 Similarly, the presence of tachycardiaduring the tilt test had predicted a favorable response to oralbeta-blocker therapy in an earlier study by Leor et al.24 In astudy18 involving 21 children who were treated withfludrocortisone, 20 experienced no recurrences of syncopeon follow-up for 20 months. Although small non-randomized studies have demonstrated the usefulness ofthis drug, no randomized, placebo-controlled trial has beenpublished. Despite this, fludrocortisone continues to be apopular choice amongst clinicians because of its patient-friendly profile and minimal side-effects (hypertension,edema, acne, depression and rarely, hypokalemia).



Alpha-adrenergic agonists: These agents increase thevenous tone, thereby reducing the venous pooling and thuspossibly the paradoxical activation of cardioreceptors. Theincrease they cause in arteriolar tone may counterbalancethe reflex vasodilatation and hypotension that culminatein overt syncope.

Though the initial trials showed promise, they were open-label studies with a small number of patients and useddifferent drugs such as phenylephrine,25 methylphenidate26

and dextroamphetamine.27 Only midodrine has proved tobe of use in a randomized, placebo-controlled trial.28 Wardet al.,28 randomized 16 patients to midodrine or placebo andreported significant improvement in the frequency andrecurrence of symptoms. Also, the quality of life improvedsignificantly in the patients on midodrine. In another studywith midodrine,29 11 patients were followed up for a meanof 4.3 months. There was a significant clinical improvementin 9 patients with complete resolution of symptoms in 5 ofthem.

Small, nonrandomized trials had also found etilefrine,another alpha-adrenergic agonist, to be beneficial. However,recently a double-blind, randomized, placebo-controlledtrial with etilefrine failed to confirm its efficacy.30 Both thetreatment and placebo arms (63 patients in each) had thesame recurrence rate of syncope (24%). The median timeto the first syncope (106 v. 112 days) and the incidence andnumber of presyncopal episodes were also not significantlydifferent. This well-crafted VASIS study30 has put a questionmark on the usefulness of alpha-adrenergic agonists invasovagal syncope. Despite the initial promise, largerandomized trials are needed to further the claims ofmidodrine as an agent that can be relied upon in thiscondition.

Serotonin-reuptake inhibitors: Serotonin is aneurotransmitter which mediates the development ofhypotension and bradycardia, both of which are essentialcomponents of vasovagal syncope. Therefore, serotonin-reuptake inhibitors such as paroxetine, fluoxetine andsertraline could be of value in the treatment of this entity.Grubb et al.31 anecdotally observed that treating patientswho suffered from endogenous depression as well asneurocardiogenic syncope with fluoxetine resulted in theresolution of their clinical syncope. This observation pavedthe way for clinical trials with this class of drugs. In theirstudy, 16 patients with resistant vasovagal syncope and apositive tilt test were treated with fluoxetine and restudied5–6 weeks later. Of the 13 patients (3 could not tolerate thedrug) 7 (53%) were rendered tilt test negative and remainedasymptomatic over a mean follow-up of 19±9 months.

Di Girolamo et al.32 studied the effects of paroxetinehydrochloride on vasovagal syncope in a randomized,double-blind, placebo-controlled trial. The response rates(negative HUTT after 1 month) were 61.8% and 38.2% inthe paroxetine and placebo groups, respectively. During thefollow-up of 25.4±8 months, recurrence of syncope wasseen in 17.6% in the paroxetine group as compared to52.9% in the placebo group (p<0.0001). Only 1 patient(2.9%) discontinued the drug due to severe headache. Thisis a significant study; it being a placebo-controlled,randomized trial showing evidence of improvement.

Sertraline has also been found to be effective in childrenwith vasovagal syncope in small nonrandomized trials.33,34

Incapacitating side-effects such as anxiety, insomnia,drowsiness, anorexia and fatigue restrict the use ofserotonin receptor antagonists.

Disopyramide: This class IA antiarrhythmic agent has anegative inotropic, anticholinergic and a direct peripheralvasoconstrictive effect which theoretically makes it an idealdrug to combat syncope. Small, nonrandomized studieshave shown disopyramide to be beneficial. In a study byMilstein et al.,35 a repeat tilt test after oral disopyramidetherapy for 3 days was performed in 10 patients. All of themhad a negative test and during a subsequent follow-up after20±5 months, all but one patient remained asymptomatic.In another study,24 of the 6 nonresponders to beta-blockertherapy, subsequently treated with disopyramide, only 1 hadrecurrence of syncope in the follow-up period of 9±3months. In a double-blind, randomized trial, Morillo et al.36

studied 22 patients who were serially tilt tested withintravenous and then oral disopyramide. Neither form ofdisopyramide was found to be superior to placebo inpreventing tilt-induced or spontaneous syncope. Duringfollow-up of 29±8 months, syncope recurred in 27% ofpatients on drug therapy compared to 30% on placebo.Because of the potential for pro-arrhythmias and thepresence of significant anticholinergic effects, disopyramideis not considered as first-line therapy for vasovagal syncope.

Other pharmacologic agents: Anticholinergic agentsmight be effective by reducing the high vagal tone thatoccurs during vasovagal syncope. Small numbers ofpatients were successfully treated with transdermalscopolamine in a few nonrandomized studies.19,37 However,Lee et al.38 found no benefit of the drug over placebo in arandomized trial of 60 patients.

Theophylline blocks the action of adenosine which is amediator of hypotension and bradycardia. However, twosmall, nonrandomized studies39,40 have demonstratedconflicting results with the drug.



Oral enalapril, an angiotensin-converting enzymeinhibitor, has also shown promise in 30 patients withvasovagal syncope in a randomized, double-blind, placebo-controlled trial.41 All the 15 patients given enalapril had anegative repeat tilt test. During a follow-up of 13.4+2.1months, 14/15 (93%) patients on enalapril remainedsymptom-free compared with 3/15 in the placebo group.Further studies are required to assess the full potential ofthis class of drugs in treating patients with vasovagalsyncope.

Role of tilt testing in guiding pharmacotherapy: Tilttest-guided therapy has generally been found to be moreeffective than empiric treatment.15,17,22 In a study of 303patients,19 44 received empiric therapy, 210 were treatedwith tilt test-guided medications, and 49 refused treatment.Of the 210 patients on tilt test-guided therapy, 130 wereon beta-blockers, 35 on theophylline, 10 on ephedrine, 31on disopyramide and 4 on combination therapy. Empirictherapy consisted of beta-blockers in most of the patients.Recurrence of symptoms was seen in only 6% of the patientson tilt test-guided therapy compared with 36% on empirictherapy and 67% in no treatment group.

Conclusions: Although several pharmacologic agentshave been used for the treatment of vasovagal syncope,there is a remarkable absence of scientific proof of theirefficacy from large, prospective, randomized clinical trials.In fact, only three agents have shown to be of benefit inrandomized clinical trials—atenolol,20 midodrine27 andparoxitene.32 Beta-blockers are still the old favorite as theother two have significant side-effects. However, the onlytwo randomized trials available on their efficacy haveyielded conflicting results. Of late, clinicians have realizedthe need of looking beyond beta-blockers, and have startedconsidering other treatment options.

Pacing for Vasovagal Syncope

The need for a therapy beyond drugs was felt when drug-resistant fainters with “malignant syncope” wereencountered. These patients form a high risk for “falls withthe faint” and have a severely restricted lifestyle withsignificant physical and psychosocial handicap. Thebradycardia or asystole component of the vasovagalsyncope which along with hypotension culminates in theclinical cascade, theoretically makes pacing support duringthe episode seem the final answer to this vexing problem.

In earlier studies, vasovagal syncope induced by tilttesting was used as a model for clinical syncope, todetermine whether pacing could be helpful. In these studies,

patients who had a positive tilt test with marked bradycardiaunderwent a second tilt test with temporary pacing at ratesof 85–100 beats/min. Fitzpatrick et al.42 using temporarydual-chamber pacing with rate hysteresis could preventsyncope in 5 out of 7 patients. In these 5 cases, syncopewas prevented despite the onset of a vasovagal reactionattested to by fall in blood pressure and the development ofsufficient bradycardia to initiate pacing. Similarly, Samoilet al.43 reported prevention of syncope in 3 out of 6 patientswith pacing. In the study by Sra et al.,44 22 patients withbradycardia or asystole along with hypotension during tilttesting underwent a second tilt-test with dual-chambertemporary pacing at a rate of 20 beats/min above theirresting rate. They found that 15 patients still had presyncopeand 5 patients went on to have frank syncope during pacing.Although the authors felt that pacing was not useful invasovagal syncope, a shift from syncope at baseline in 18patients to syncope in just 5 patients during pacingrepresents a significant clinical improvement as far as franksyncope is concerned. The results were similar to the earlierstudy by Fitzpatrick et al.,42 in which the authors hadconcluded that modification of the vasovagal episode duringtemporary pacing could translate into a clinical advantagewith permanent pacing. Overall, temporary pacing wasshown to prevent syncope in more than half the patients.

The initial studies with permanent pacing werehistorically controlled. Peterson et al.45 reported theirexperience of 37 patients with vasovagal syncope who wereimplanted dual-chamber pacemakers with rate hysteresis.These pulse generators had sensors that detected thefall in heart rate to 40–50 beats/min and responded withpacing rates of 80–90 beats/min. During the follow-up of39±19 months, 89% of patients reported symptomaticimprovement and 62% had no recurrence of syncope.

A similar benefit was reported using a permanent dual-chamber rate-drop response pacing algorithm by Benditt etal.46 With this feature, the pacemaker can be programmedto detect a small rapid drop of heart rate and then pace at arelatively high rate to provide chronotropic support duringthe time of presumed vasodilatation. All the parameters,i.e. the lowest rate, top rate and the time period areprogrammable. During an average follow-up of 6 months,syncope was prevented in 78% of cases.

A third type of pulse generator with automatic rate-dropsensing mode was tested by Sheldon et al.47 In this, the dual-chamber pacemaker paces if the sensed heart rate drops by>15 beats/min/beat. The initial pacing rate is 15 beats/minbelow the rate preceding the rate decrease, and then slowlydecreases at 0.5 beats/min/beat until the intrinsic heart rateexceeds it. With this “rate smoothening” pacing algorithm,



the frequency of syncope was reduced by 93% and 50% hadno recurrence. The median time to syncope recurrencebefore and after pacing was 7 days and 5.3 months,respectively. There was also a marked improvement in thequality of life.

Thus, irrespective of the algorithm used, benefit wasdemonstrated in these nonrandomized studies withpermanent dual-chamber pacing, and the results werecomparable. A recent study has now confirmed the benefitof dual-chamber pacing over single-chamber ventricularpacing.48 However, as these trials were historicallycontrolled, the next step was to demonstrate the benefit incontrolled studies.

Randomized pacemaker trials: There are only a fewcontrolled trials available to study the efficacy of pacing invasovagal syncope.

The North American Vasovagal Pacemaker Study (VPS-I)49 was the first randomized trial to show benefit withpermanent pacing. In this trial, 54 patients with more thansix lifetime episodes of syncope and a positive HUTT (withsyncope or presyncope and relative bradycardia), wererandomized to dual-chamber pacing with rate-dropresponse and no pacing. Patients in both the groups werepermitted to receive medical treatment according to thejudgment of treating physician. It was found that syncoperecurred in 19/27 (70%) patients who did not receivepacemaker compared to 6/27 (22%) in the pacemakergroup, showing a 85.4% relative risk reduction. However,there was no difference in the occurrence of presyncope,with at least one episode of presyncope reported by 74% ofthe no pacemaker group and 63% of the pacemaker group.The mean time from randomization to syncope alsoincreased in the pacemaker group (112 days) compared tono pacemaker group (54 days) . However, it was an open-label study, therefore a placebo-type effect or psychologicalbenefit from receiving a pacemaker cannot be excluded.Also, there was no standardization of medical therapy inthe study. At the end of the study, it was not clear whetherthe conventional bradycardia support benefited the patientsor the rate-drop response alogrithm. This particular aspectwould become clear after publication of the results of theVPS-II trial, comparing the effectiveness of DDD pacing withand without the rate-drop response, which is currently inprogress.

The Vasovagal Syncope International Study (VASIS)investigators50 showed that plain DDI pacing with ratehysteresis reduced the likelihood of syncope in a select high-risk group. They included 42 patients who had at least threesyncopal episodes in the preceding two years with a positivecardioinhibitory response to HUTT. Ninteen patients

received DDI pacemakers programmed to 80 beats/min withrate hysteresis of 45 beats/min and 23 received nopacemaker. No drugs were prescribed in either group. Therewas recurrent syncope in 14/23 (61%) patients after amedian time of 5 months in the no pacemaker group versus1/19 (5%) in the pacemaker group (after 15 months)(p <0.0006). This benefit was maintained over a long period(mean follow-up 3.7±2.2 years). The Kaplan–Meieractuarial estimates of the first recurrence of syncope after1, 3 and 5 years were 0%, 6% and 6% in the pacemakergroup and 39%, 50% and 75% in no pacemaker group,respectively. There was recurrence in only 5%, a rate muchlower than previous studies, including VPS-I. In this trial,the highest-risk category patients with a severecardioinhibitory response to the tilt test were selected – 79%of pacemaker-treated patients and 91% of controls had anasystolic response during HUTT. Thus, in this study, patientswith a more severe cardioinhibition were included,compared with the VPS-I study, in which only one-thirdof patients developed a heart rate of less than 40 beats/min at baseline tilt test. Like the previous VPS trial, theplacebo effect cannot be excluded as the control group didnot include implantation of a device. Recurrences ofpresyncope and dizziness were not studied.

Ammirati et al.51 recently published the first randomized,controlled trial comparing the effect of medical treatmentwith pacemaker therapy (DDD with rate-drop response) inpatients with recurrent vasovagal syncope. The inclusioncriteria were similar to those of the VASIS study.45 Of the93 patients randomized, 46 received a DDD pacemaker withrate-drop response. The other 47 patients were givenatenolol 100 mg/day. There was a recurrence in 2/46(4.3%) patients in the pacemaker group after a median timeof 390 days as compared with 12/47 (25.5%) patients onatenolol therapy after a median of 135 days (p=0.004).Pacemaker therapy was associated with 93% actuarialprobability of remaining free of syncopal recurrence after3 years, whereas in the medically treated group, theprobability was reduced to only less than 67% for the sameperiod of time. This study was not blinded and there was abias towards giving the pacemaker option to older, highlysymptomatic patients. Data on presyncope were notcollected. Although 26% of patients in the atenolol groupreported side-effects, only one patient discontinued therapy.

In a small study of 12 severely symptomatic children,McLeod et al.48 performed a three-way, double-blind,randomized, crossover study in which the pacemakers wereprogrammed to sensing only (no pacing), single-chamberventricular pacing with rate hysteresis or dual-chamberpacing with rate-drop response for 4-month periods, with



each patient randomized to one of the 6 possible orders ofthe 3 pacing modes. They found that both VVI and DDDpacing significantly reduced the episodes of syncopecompared with no pacing (p=0.0078). This proved thebeneficial role of pacing over and above any possible placeboeffect. VVI and DDD pacing were comparable in terms ofreducing syncope but the dual-pacing mode was superiorwhen presyncope was also considered (p=0.039).

These trials included only a highly selected group of“fainters”—high frequency of syncopal attacks, severecardioinhibition with asystole or syncope during HUTT.Clinical improvement has been demonstrated in all therandomized studies. However, how much of this benefit isbecause of the psychological impact of a surgical procedureimparting a placebo effect has to be further looked into. Inthe VPS-I study, there was an absence of any reduction inthe frequency of presyncopal attacks in the paced patients.The investigators used this negative point to eliminate anyconcerns regarding placebo effect, arguing that if placeboeffect could decrease the incidence of syncope, it should alsoreduce the frequency of episodes of presyncope or light-headedness. The study by McLeod et al.48 is the only one sofar that has demonstrated the superiority of pacing over nopacing. However, this was only a small study involving 12children with documented asystole of >4 s.

Selection of patients: Almost all trials have used tilttesting to identify patients who would benefit frompermanent pacing. But the hypothesis on which thisselection is based, i.e. the greater the cardioinhibition, themore the likelihood of pacing therapy being effective, hasyet to be confirmed. Follow-up studies of asystolicresponders on tilt testing did not reveal a more malignantoutcome.3,4 In the study by Peterson et al.45 on pacing,patients with an asystolic response during tilt testing faredno better than those with less severe cardioinhibition. It isalso uncertain whether bradycardia during tilt testingautomatically translates into bradycardia during clinicalsyncope. Whether a negative test on temporary pacing canpredict a long-term favorable ef fect on permanentpacemaker is also not proved. In the VASIS study,50 the long-term efficacy of a permanent pacemaker was clearlydemonstrated despite repeat tilt testing at 15 days showingno benefit of pacing over placebo (the positive response ontilt test being 59% and 61%, respectively). Similarly, in thestudy by Sheldon et al.,47 the outcome of the tilt test did nothave any bearing on whether patients had recurrence ofsyncope after pacemaker insertion. One of the 2 patientswith negative tilt test results had a recurrence, comparedto 5 of 10 patients with positive test results.

The higher frequency of syncopal episodes has also been

shown to be predictive of a greater probability of syncopalrelapse.52,53 However, in all the randomized trials, there wasa low recurrence rate despite the predicted high risk ofrelapse in the highly symptomatic study population. Forexample, in the control arm of the VASIS study, there was arecurrence rate of 0.44% per year during a follow-up ofmore than 3 years. Because this was lower than in the 2years preceding enrollment, there was an obviousspontaneous decrease in the syncopal episodes even in theabsence of any active or placebo treatment. The cycliccourse of vasovagal syncope, the possible therapeutic effectof tilt training and patient education are confoundingfactors that may be responsible for the observation. Patientswho are older and have a history of traumatic falls are oftenconsidered likely candidates for permanent pacing.However, logical as it seems, there is no data to substantiatethis. At present, tilt testing and clinical presentation arerelied upon to select patients for pacing; the future may liein implantable loop recorders that would correctly identifythe cardiac event responsible.

Type of pacing: The pacemakers used to treat vasovagalsyncope detect decrease in heart rate as the sensed eventin 1 of the 3 options. All the three sensing modes—rate-drop response,46,49,51 rate smoothening47 and ratehysteresis42,45,50—have demonstrated comparable benefitwith both temporary and permanent pacing. No large studycomparing these different modes is available. Future trialssuch as VPS-II might show the superiority of one specializedpacing algorithm over the other.

Ideal end-point: One of the important issues is to establisha consensus for assessing the efficacy of treatment invasovagal syncope. Syncope is known to occur in clusterswith long, symptom-free intervals. Therefore, recurrenceof syncope is not an ideal end-point to establish efficacy.Other end-points such as time to first recurrence,asymptomatic intervals and overall syncopal burden overa period of time should also be considered.

Conclusions: Does the demonstrable benefit in threesuccessive randomized trials mean lowering of the thresholdfor permanent pacing in vasovagal syncope? At present,cautious optimism would be advisable. A careful assessmentof the clinical profile, the physical, psychological and socialhandicap, and other factors such as occupational necessityis required. Tilt testing remains the only test that one turnsto, despite its limitations. In future, implantable looprecorders would probably prove to be more useful. Larger,multicenter studies with longer follow-up are needed toresolve many unanswered questions. But for now, it seemsprudent to limit pacemaker use to a few select, severely



symptomatic patients who are resistant to drugs and whoare particularly prone to injuries or accidents.

References

1. Kapoor WN, Karft M, Wieand S, Peterson JR, Levy GS. A prospectiveevaluation and follow-up of patients with syncope. N Engl J Med 1983;309: 197–204

2. Kushner JA, Kou WH, Kadish AH, Morady F. Natural history ofpatients with unexplained syncope and a non-diagnosticelectrophysiologic study. J Am Coll Cardiol 1989; 14: 391–396

3. Dhala A, Natale A, Sra J, Deshpande S, Blanck Z, Jazayeri MR, et al.Relevance of asystole during head-up tilt testing. Am J Cardiol 1995;75: 251–254

4. Brignole M, Menozzi C, Gianfranchi L, Bottoni N, Lolli G. The clinicaland prognostic significance of the asystolic response during the head-up tilt test. Eur J Cardiac Pacing Electrophysiol 1992; 2: 109–113

5. Linzer M, Pontinen M, Gold DT, Divine GW, Felder A, Brooks WB.Impairment of physical and psychosocial function in recurrentsyncope. J Clin Epidemiol 1991; 44: 1037–1043

6. Rose S, Koshman M, McDonald S, Sheldon R. Health related qualityof life in patients with neuromediated syncope [Abstr]. Can J Cardiol1996; 12: 131E

7. El-sayed H, Hainsworth R. Salt supplement increases plasma volumeand orthostatic tolerance in patients with unexplained syncope. Heart1996; 75: 134–140

8. Younoszai AK, Franklin WH, Chan DP, Cassidy SC, Allen HD. Oralfluid therapy: a promising treatment for vasodepressor syncope. ArchPediatr Adolesc Med 1998; 152: 165–168

9. Reybronck T, Heidbuchee H, Van de Werf F, Ector H. Tilt training: anew treatment for recurrent neurocardiogenic syncope and severeorthostatic intolerance. Pacing Clin Electrophysiol 2000; 23: 493–498

10. Sheldon R, Rose S, Flanagan P, Koshman ML, Killam S. Effect of betablockers on the time to first syncope recurrence in patients after apositive isoproterenol tilt-table test. Am J Cardiol 1996; 78: 536–539

11. Slotwiner DJ, Stein KM, Lippman N, Markowitz SM, Lerman BB.Response of neurocardiac syncope to beta blocker therapy:interaction between age and parasympathetic tone. Pacing ClinElectrophysiol 1997; 20: 810–814

12. Scott WA, Pongiglione G, Bromberg BI, Schaffer MS, Deal BJ, Fish FA,et al. Randomized comparison of atenolol and fludrocortisone acetatein the treatment of pediatric neurally mediated syncope. Am J Cardiol1995; 76: 400–402

13. Biffi M, Boriani G, Sabbatani P, Bronzetti G, Frabetti L, Zannoli R, etal. Malignant vasovagal syncope: a randomized trial of metoprololand clonidine. Heart 1997; 77: 268–272

14. Klingenheben T, Kalusche D, Li YG, Schopperl M, Hohnloser SH.Changes in plasma epinephrine concentration and in heart rateduring head-up tilt testing in patients with neurocardiogenic syncope:correlation with successful therapy with beta-receptor antagonists.J Cardiovasc Electrophysiol 1996; 7: 802–808

15. Cox MM, Perlman BA, Mayor MR, Silberstein TA, Levin E, Pringle L,et al. Acute and long-term beta adrenergic blockade for patients withneurocardiogenic syncope. J Am Coll Cardiol 1995; 26: 1293–1298

16. Cohen MB, Snow JS, Grasso V, Lehnert L, Goldner BG, Jadonath RL,et al. Efficacy of pindolol for treatment of vasovagal syncope. Am HeartJ 1995; 130: 786–790

17. Sra JS, Murthy VS, Jazayeri MR, Shen YH, Troup PJ, Avitall B, et al.Use of intravenous esmolol to predict efficacy of oral beta-adrenergicblocker therapy in patients with neurocardiogenic syncope. J Am CollCardiol 1992; 19: 402–408

18. Grubb BP, Temesy-Armos P, Moore J, Wolfe D, Hahn H, Elliott L. Theuse of head-upright tilt table testing in the evaluation and

management of syncope in children and adolescents. Pacing ClinElectrophysiol 1992; 15: 742–748

19. Natale A, Sra J, Dhala A, Wase A, Jazayeri M, Deshpande S, et al.Efficacy of different treatment strategies for neurocardiogenicsyncope. Pacing Clin Electrophysiol 1995; 18: 655–662

20. Mahanonda N, Bhuripanyo K, Kangkagate C, Wansanit K, KulchotB, Nademanee K, et al. Randomized double-blind, placebo-controlledtrial of oral atenolol in patients with unexplained syncope and positiveupright tilt table test results. Am Heart J 1995; 130: 1250–1253

21. Madrid AH, Ortega J, Rebollo JG, Manzano JG, Segovia JG, SanchezA, et al. Lack of efficacy of atenolol for the prevention of neurallymediated syncope in a highly symptomatic population: a prospective,double-blind, randomized and placebo-controlled study. J Am CollCardiol 2001; 37: 554–559

22. Natale A, Newby KH, Dhala A, Akhtar M, Sra J. Response to betablockers in patients with neurocardiogenic syncope: how to predictbeneficial results. J Cardiovasc Electrophysiol 1996; 7: 1154–1158

23. Mion D Jr, Rea RF, Anderson EA, Kahn D, Sinkey CA, Mark AL. Effectsof fludrocortisone on sympathetic nerve activity in humans.Hypertension 1994; 23: 123–130

24. Leor J, Rotstein Z. Vered Z, Kaplinsky E, Truman S, Eldar M. Absenceof tachycardia during tilt-test predicts failure of beta-blocker therapyin patients with neurocardiogenic syncope. Am Heart J 1994; 127:1539–1543

25. Strieper MJ, Campbell RM. Efficacy of alpha-adrenergic agonisttherapy for prevention of pediatric neurocardiogenic syncope. J AmColl Cardiol 1993; 22: 594–597

26. Grubb BP, Kosinski D, Mouhaffel A, Pothoulakis A. The use ofmethylphenidate in the treatment of refractory neurocardiogenicsyncope. Pacing Clin Electrophysiol 1996; 19: 836–840

27. Susmano A, Volgman AS, Buckingham TA. Beneficial effects ofdextroamphetamine in the treatment of vasodepressor syncope.Pacing Clin Electrophysiol 1993; 16: 1235–1239

28. Ward CR, Gray JC, Gilroy JJ, Kenny RA. Midodrine: a role in themanagement of neurocardiogenic syncope. Heart 1998; 79: 45–49

29. Sra J, Maglio C, Biehl M, Dhala A, Blanck Z, Deshpande S, et al. Efficacyof midodrine hydrochloride in neurocardiogenic syncope refractoryto standard therapy. J Cardiovasc Electrophysiol 1997; 8: 42–46

30. Raviele A, Brignole M, Sutton R, Alboni P, Giani P, Menozzi C, et al.for the Vasovagal Syncope International Study (VASIS) Investigators.Effect of etilefrine in preventing syncopal recurrence in patients withvasovagal syncope: a double-blind, randomized placebo-controlledtrial. Circulation 1999; 99: 1452–1457

31. Grubb BP, Wolfe DA, Samoil D, Temesy-Armos P, Hahn H, Elliott L.Usefulness of fluoxetine hydrochloride for prevention of resistantupright tilt induced syncope. Pacing Clin Electrophysiol 1993; 16: 458–464

32. Di Girolamo ED, Di Iorio CD, Sabatini P, Leonzio L, Barbone C, BarsottiA. Effects of paroxetine hydrochloride, a selective serotonin reuptakeinhibitor on refractory vasovagal syncope: a randomized double-blindplacebo-controlled study. J Am Coll Cardiol 1999; 33: 1227–1230

33. Grubb BP, Samoil D, Kosinski D, Kip K, Brewster P. Use of sertralinehydrochloride in treatment of refractory neurocardiogenic syncopein children and adolescents. J Am Coll Cardiol 1994; 24: 490–494

34. Lenk M, Alehan D, Ozme S, Celiker A, Ozer S. The role of serotoninereuptake inhibitors in preventing recurrent unexplained childhoodsyncope. A preliminary report. Eur J Pediatr 1997; 156: 747–750

35. Milstein S, Buetikofer J, Dunnigan A, Benditt DG, Gornick C. ReyesWJ. Usefulness of disopyramide for prevention of upright tilt-inducedhypotension and bradycardia. Am J Cardiol 1990; 65: 1339–1344

36. Morillo CA, Leitch JW, Yee R, Klein GJ. A placebo controlled trial ofintravenous and oral disopyramide for prevention of neurallymediated syncope induced by head-up tilt. J Am Coll Cardiol 1993;22: 1843–1848

37. Grubb BP, Temesy-Armos P, Hahn H, Elliott L. Utility of upright tilt



testing in the evaluation and management of syncope of unknownorigin. Am J Med 1991; 90: 6–10

38. Lee TM, Su SF, Chen MF, Liau CS, Lee YT. Usefulness of transdermalscopolamine for vasovagal syncope. Am J Cardiol 1996; 78: 480–482

39. Nelson SD, Stanley M, Love CJ, Coyne KS, Schaal SF. The autonomicand haemodynamic effects of oral theophylline in patients withvasodepressor syncope. Arch Intern Med 1991; 151: 2425–2429

40. Lee TM, Chen MF, Su SF, Chao CL, Liau CS, Lee YT. Excessivemyocardial contraction in vasovagal syncope demonstrated byechocardiography during head-up tilt test. Clin Cardiol 1996; 19:137–140

41. Zeng C, Zhu Z, Liu G, Hu W, Wang X, Yang C, et al. Randomized, doubleblind, placebo-controlled trial of oral enalapril in patients withneurally mediated syncope. Am Heart J 1998; 136: 852–858

42. Fitzpatrick A, Theodorakis G, Ahmed R, Williams T, Sutton R. Dualchamber pacing aborts vasovagal syncope induced by head-up 60°tilt. Pacing Clin Electrophysiol 1991; 4: 13–19

43. Samoil D, Gruff BP, Brewster P, Moore J, Temesy Armos P. Comparisonof single and dual chamber pacing techniques in prevention ofupright tilt induced vasovagal syncope. Eur J Cardiac PacingElectrophysiol 1993; 1: 36–41

44. Sra JS, Jazayeri MR, Avitall B, Dhala A, Deshpande S, Blanck Z, et al.Comparison of cardiac pacing with drug therapy in the treatment ofneurocardiogenic (vasovagal) syncope with bradycardia or asytole.N Engl J Med 1993; 328: 1085–1090

45. Peterson ME, Chamberlain-Webber R, Fitzpatrick AP, Ingram A,Williams T, Sutton R. Permanent pacing for cardioinhibitorymalignant vasovagal syndrome. Br Heart J 1994; 71: 274–281

46. Benditt DG, Sutton R, Gammage MJ, Markowitz T, Gorski J, NygaardGA, et al. Clinical experience with Thera DR rate-drop response pacingalgorithm in carotid sinus syndrome and vasovagal syncope. PacingClin Electrophysiol 1997; 20: 832–839

47. Sheldon RS, Koshman ML, Wilson W, Kieser T, Rose S. Effect of dualchamber pacing with automatic rate drop sensing on recurrentneurally mediated syncope. Am J Cardiol 1998; 81: 158–162

48. McLeod K, Wilson N, Hewitt J, Stephenson J. Double-blind trial ofcardiac pacing for severe childhood neurally mediated syncope withreflex anoxic seizures [Abstr]. Pacing Clin Electrophysiol 1999; 22: 807

49. Connolly SJ, Sheldon R, Roberts RS, Gent M. The North AmericanVasovagal Pacemaker Study (VPS). A randomized trial of permanentcardiac pacing for the prevention of vasovagal syncope. J Am CollCardiol 1999; 33: 16–20

50. Sutton R, Brignole M, Menozzi C, Raviele A, Alboni P, Giani P, et al.Dual-chamber pacing in the treatment of neurally mediated tilt-positive cardioinhibitory syncope. The Vasovagal SyncopeInternational Study (VASIS). Circulation 2000; 102: 294–299

51. Ammirati F, Colivicchi F, Santini M. Permanent cardiac pacing versusmedical treatment for the prevention of recurrent vasovagal syncope.Circulation 2001; 104: 52–57

52. Sheldon R, Rose S, Flanagan P, Koshman ML, Killam S. Risk factorsfor syncope recurrence after a positive tilt-table test in patients withsyncope. Circulation 1996; 93: 973–981

53. Grimm W, Degenhardt M, Hoffmann J, Menz V, Wirths A, Maisch B.Syncope recurrence can better be predicted by history than by head-up tilt testing in untreated patients with suspected neurally mediatedsyncope. Eur Heart J 1997; 18: 1465–1469


Indian Heart J 2001; 53: 423–444 Sra et al. New Techniques for Mapping Arrhythmias 423

New Techniques for Mapping Cardiac Arrhythmias

Jasbir Sra, Joy M ThomasUniversity of Wisconsin Medical School, Milwaukee Clinical Campus, Electrophysiology Laboratories of St Luke’s

and Sinai Samaritan Medical Centers, Milwaukee, Wisconsin, USA

Essential to the effective management of any cardiacarrhythmia is a thorough understanding of the

mechanisms of its initiation and sustenance.1–13

Conventionally, this has been achieved by careful study ofthe surface electrocardiogram (ECG) and correlation of thechanges therein with data from intracavitary electrogramspicked up by catheters which were maneuvered to lie incontact with the endocardium in various key locationswithin the cardiac chambers. A record of theseelectrograms documenting multiple sites simultaneously isstudied to deduce the mechanisms of an arrhythmic event.However, these methods do not cover a vast area of theendocardial surface. Painstaking spot-by-spot maneuveringof the catheter is required to trace the origin of anarrhythmic event and its activation sequence in theneighboring areas. To obtain a more accurate picture of thespread of activation in all directions from the origin of thearrhythmia would require covering every possible area ofthe endocardium with an electrogram-recording catheter.

Conventional mapping techniques thus have significantlimitations. Using current techniques it is difficult toconceive the three-dimensional (3-D) orientation of cardiacstructures as these use a limited number of recordingelectrodes guided by fluoroscopy. Although catheters usingmultiple electrodes to acquire data points are available, theexact location of an acquired unit of electrophysiologicaldata is difficult to ascertain due to inaccurate delineationof the location of anatomic structures. The laboriousprocess of precise mapping with conventional techniquesexposes the electrophysiologist, staff and patient toundesirable levels of radiation from the extendedfluoroscopy time. Another drawback of conventionaltechniques is the inability to identify several sites and thenreturn to the most appropriate or optimal site. This inabilityto identify, for example, the site of a previous ablationincreases the risk of repeated ablation of areas already dealtwith and the likelihood that new sites may be missed. Theability to identify the position of the ablation catheter withina 3-D model where the sites are tagged would thus help toavoid ablation of unwanted areas and minimize the use of

fluoroscopy. Although several mapping techniques havebeen developed in the last few years, electroanatomic(commonly called “CARTO”) and noncontact mapping aremost commonly used. The purpose of this report is to give acomprehensive description of these techniques and theiruse in the treatment of some key arrhythmias.

Electroanatomic Mapping

Electroanatomic or CARTO mapping is a nonfluoroscopicmapping system which uses a special catheter to generate3-D electroanatomic maps of the heart chambers.13–39 Thissystem uses magnetic technology to accurately determinethe location and orientation of the mapping and ablationcatheter while simultaneously recording local electrogramsfrom the catheter tip. By sampling electrical and spatialinformation from different endocardial sites, the 3-Dgeometry of the mapped chamber is reconstructed in real-time and analyzed to assess the mechanism of arrhythmiaand the site appropriate for ablation.

Fundamentals of electroanatomic mapping and themapping system: Electroanatomic mapping is based onthe premise that a metal coil generates an electrical currentwhen placed in a magnetic field. The magnitude of thecurrent depends on the strength of the magnetic field andthe orientation of the coil in it. The CARTO mapping systemconsists of an ultralow magnetic field emitter, a locationsensor inside the mapping and ablation catheter tip, a dataprocessing unit and a graphical display unit to generate theelectroanatomic model of the chamber being mapped. Themagnetic field emitter, mounted under the operating table,consists of three coils which generate a low-intensitymagnetic field, around 0.05–0.2 Gauss. In comparison themagnetic field inside a magnetic resonance imaging (MRI)machine is approximately 150–250 000 Gauss. Themapping and ablation catheter is similar to the routineablation catheter. It is a 7 F quadripolar, deflectable,steerable catheter and can be easily manipulated inside theheart.

The data of the amplitude, frequency and phase of themagnetic field are gathered and analyzed by the processingunit and displayed on the display unit. The CARTO mappingsystem uses a triangulation algorithm similar to the Global

Review Article

Correspondence: Professor Jasbir Sra, 2801 W Kinnickinnic River Pkwy#777, Milwaukee, WI 53215, USA. e-mail: [email protected]

IHJ-009-01.p65 10/17/01, 12:06 PM423

424 Sra et al. New Techniques for Mapping Arrhythmias Indian Heart J 2001; 53: 423–444

Positioning System (GPS). The sensor in the catheter tipdetects the intensity of magnetic field generated by eachcoil, allowing for determination of its distance from eachcoil. These distances determine the area of theoreticalspheres around each coil (Fig. 1) and the intersectionbetween these three spheres determines the location of thetip of the catheter. The accuracy of determination of thelocation is highest in the center of the magnetic field;therefore, it is important to position the location pad underthe patient’s chest.

In addition to the x, y and z coordinates of the cathetertip, the CARTO mapping system can determine threeorientation determinants—roll, yaw and pitch (Fig. 1). Theposition and orientation of the catheter tip can be seen onthe screen and monitored in real-time as it moves withinthe electroanatomic model of the chamber mapped. Thecatheter icon has four color bars (green, red, yellow andblue) enabling the operator to view the catheter as it turnsclockwise or counterclockwise (Fig. 2). In addition, sincethe catheter always deflects in the same direction, eachcatheter will always deflect towards a single color. Hence,to deflect the catheter to a specific wall, the operator shouldfirst turn the catheter so that this color faces the desiredwall.

Important aspects of mapping: Several critical elementsneed to be defined before appropriate arrhythmia mappingcan be performed (Fig. 3). As CARTO mapping is not animaging technique, f luoroscopy is initially used fororientation by taking known anatomic locations in the heartas references to create the model of the mapped chamber.

Fig. 2. The catheter always deflects towards a single color. Hence, to deflect thecatheter to a specific wall, the operator can turn the catheter so that the samecolor faces the desired wall, following which the catheter is deflected.

When mapping the heart the system may deal with fourtypes of motion artifacts:(i) cardiac motion—the heart is in constant motion; thus

the location of the mapping catheter changesthroughout the cardiac cycle;

(ii) patient motion;(iii) respiratory motion—intrathoracic change in the

position of the heart during the respiratory cycle; and(iv) system motion.

Several steps are taken by the CARTO mapping systemto compensate for these possible motion artifacts and toensure that the initial map coordinates are appropriate.

First, a reference signal called the reference electrogram isselected. This is the fiducial marker on which the entiremapping procedure is based. The location of the catheter isgated to a fiducial point in the cardiac cycle which isdetermined only once per cardiac cycle. In building a 3-Dmap, both the anatomic location and the electrogram atthe anatomic point are acquired by the mapping catheter,which is moving from point to point. The timing of thefiducial point is used to determine the activation timing inthe mapping catheter in relation to the acquired points andto ensure collection of data during the same part of thecardiac cycle and is, therefore, vital to the performance ofthe system. The fiducial point is defined by the user byassigning a reference channel and an annotation criterion.

The timing of all electrophysiologic information displayedon the completed 3-D map is relative to the fiducial point. Asdepicted in Fig. 3 during an accessory pathway mappingsurface ECG lead is selected with maximum dV/dt being thefiducial point. All of the local activation timing informationrecorded by the mapping catheter at different anatomiclocations during mapping will be relative to this fiducial

Fig. 1. Depending upon the distance of the mapping and ablation catheter fromthe magnetic coils, theoretical spheres are created around each coil (A and B).The intersection between these three spheres determines the location of the tipof the catheter. In addition to the x, y and z coordinates, the system alsodetermines the roll, yaw and pitch orientation of the catheter (C). Adaptedfrom: The CARTO Physician’s manual. Cordis-Webster, Johnson and Johnson.Reproduced by permission.

IHJ-009-01.p65 10/17/01, 12:07 PM424


point, with the acquisition being gated so that each point isacquired during the same part of the cardiac electricalsignal. It is important that the rhythm being mapped ismonomorphic and the fiducial point is reproducible at eachsampled site. The system has a great deal of flexibility interms of choosing the reference electrogram, fiducial pointand gating locations. Any ECG lead or intracardiacelectrogram in bipolar or unipolar mode may serve as areference electrogram. For the purpose of stability whenintracardiac electrograms are selected, coronary sinuselectrograms are usually chosen. The fiducial point of thereference electrogram may be the maximum or minimumslope.

The second important concept is that of an anatomicreference. Once the mapping catheter is placed inside theheart, its location in relation to the fixed magnetic fieldsensors placed under the patient can be determined. Thus,if the mapping catheter moves from one location to another,

The final important concept in CARTO mapping is thedetermination of local activation time or LAT. Once thereference electrograms, anatomic reference and WOI have

Fig. 3. Two sinus beats from a patient with a right anterior accessory pathwayand manifest pre-excitation. The surface lead II has been selected as the referenceelectrogram and is displayed on the top panel, the unipolar intracardiacelectrogram (IC3) near the ventricular insertion site of the accessory pathwayis displayed on the bottom panel, and the system is set up to use the maximumdV/dt of the reference electrogram (lead II) as the fiducial point. A vertical purpleline intersects the fiducial point, and is drawn down vertically to intersect theintracardiac electrogram. A second vertical purple line has been drawn tointersect the onset of an accessory pathway potential in IC3. The difference intime between the fiducial point and the local activation point on IC3 is termedthe local activation time (LAT), i.e. the time that activation occurs at theanatomic point where the mapping catheter is located relative to the fiducialpoint. The vertical green bars reflect the window of interest (WOI) that hasbeen preset by the user. The CARTO system will automatically measure timingintervals inside the WOI and ignore events occurring outside it. This allows theuser to include or exclude electrical signals, and facilitates the acquisition ofdetailed, meaningful maps. From: Gepstein et al. Electroanatomical mappingof the heart: basic concepts and implications for the treatment of cardiacarrhythmias. Pacing Clin Electrophysiol 1998; 21: 1271. Reproduced bypermission.

the CARTO mapping system tracks the location. However,several of the factors mentioned earlier, including a changein the patient’s position during the procedure, can result inloss of orientation of the structures. To overcome the effectof motion artifacts, a reference catheter with a sensorsimilar to the mapping catheter is used. This referencecatheter is fixed in its location inside the heart or on thebody surface. When the patient moves, this anatomicreference sensor moves with the patient. The CARTOmapping system continuously calculates the position of themapping catheter in relation to the anatomic reference, thussolving the problem of any possible motion artifacts.

The intracardiac reference catheter has the advantageof moving with the patient’s body and with the heart duringthe phases of respiration. However, the intracardiacreference catheter may change its position during the courseof the procedure, especially during manipulation of theother catheters. It is therefore better to use an externallypositioned reference catheter strapped to the back of thepatient’s chest in the interscapular area. The movement ofthe ablation catheter is then tracked relative to the positionof this reference catheter.

The next most important aspect in ensuring the accuracyof the initial map coordinates is to define a window of interest(WOI). This is defined as the time interval relative to thefiducial point, during which the local activation time (LAT)is determined. The total length of the WOI should not exceedthe cycle length of the tachycardia and is usually 90% ofthe cycle length of the tachycardia. The boundaries are setrelative to the reference electrogram. Thus, as depicted inFigs 3 and 4, the window is defined by two intervals, oneextending before the reference potential and the other afterit.

Fig. 4. Selecting the window of interest for atrial flutter. In this example, thewindow is divided equally into two 90 ms intervals: one before the timing of thereference atrial potential (fiducial point) and one after. Reference electrogram isthe proximal coronary sinus (CS). Cycle length of atrial flutter is 200 ms.MC: mapping catheter; II: surface ECG lead II. From: Sra et al. Electroanatomicmapping to identify breakthrough sites in recurrent typical human flutter. PacingClin Electrophysiol 2000; 23: 1482. Reproduced by permission.

IHJ-009-01.p65 10/17/01, 12:07 PM425


been chosen, the mapping catheter is moved from point topoint. These points can be acquired in a unipolar or bipolarmodel. Unipolar recordings have the advantage of providinga more precise measurement of LATs because the steepestnegative intrinsic deflection (minimum dV/dt, minimumslope) correlates well with maximal Na+ conductance.However, in areas of scar tissue, unipolar recordings maysuffer from poor signal-to-noise ratio. In this case, it maybe preferable to use the bipolar signal as it may reduce thefar-field effect, although the local activation is less preciselydefined.

The LAT at each sampled site is calculated as the timeinterval between the fiducial point on the referenceelectrogram and the corresponding local activationdetermined from the unipolar or bipolar local electrogramrecorded from that site.

Electroanatomic mapping procedure: Followingselection of the reference electrogram, positioning of theanatomic reference and determination of the WOI, themapping catheter is positioned in the mapping chamberunder fluoroscopic guidance. The catheter is initiallypositioned at known anatomic points that serve aslandmarks for the electroanatomic map. For example, tomap the right atrium, points such as the superior vena cava,inferior vena cava (IVC), bundle of His and the ostium ofthe coronary sinus are marked. The catheter tip is thenadvanced slowly along the different walls. Analysis of theanatomic position of these points and the electrogramsobtained from them generate real-time 3-D models on amonitor display. The selected points are connected by linesto form several adjoining triangles (Fig. 5) in a global modelof the chamber that can be easily rotated to any degree,including standard views. Next, gated electrograms are usedto create an activation map which is superimposed on theanatomic model.

To create the activation map, points are color-coded (redfor the earliest electrical activation areas; orange, yellow,green, blue and purple for progressively delayed activationareas). Between these points, colors are interpolated andthe adjoining triangles are colored with these interpolatedvalues. However, if the points are widely apart, nointerpolation is done. The degree to which the system willinterpolate activation times is programmable (as thetriangle-fill threshold) and can be modified if necessary. Theactivation sequence in the mapped chamber can also berepresented as the propagation map, in which the wholechamber is blue and electrical activation waves are seen inred, spreading throughout the chamber as a continuousanimated loop.

In addition, a voltage map, also color-coded, can becreated and superimposed on the anatomic model to showthe amplitudes of all selected points with red as the lowestamplitude and orange, yellow, green, blue and purpleindicating progressively higher amplitudes. Myocardialscars are seen as low voltage, and their delineation may helpin understanding the location of the arrhythmia, e.g.ventricular tachycardia (VT). Points can also be marked ortagged as “location only”, enabling one to return preciselyto the region of interest after several points have beeninvestigated or after an impedance rise is noted during theablation procedure.

The electroanatomic model, which can be seen in a singleview or in multiple views simultaneously and freely rotatedin any direction, forms a reliable road map for navigationof the ablation catheter. Any portion of the chamber canbe seen in relation to the catheter tip in real-time, and pointsof interest can easily be reached even without fluoroscopy.Most important, areas of previous ablations or pointsselected previously for ablation can be returned to easily.Potential ablation targets can be studied, compared andtreated, if desired.

Clinical applications: The capability of the CARTOmapping system to associate relevant electrophysiologicinformation with the appropriate spatial location in theheart and to accurately determine the 3-D location andorientation of the ablation catheter is of great value inablation procedures.19–39

The technology enables the physician to both performthe mapping procedure and define the mechanismsunderlying the arrhythmia. An ablation strategy can bemore precisely designed and the system allows for a moreaccurate return to the locations of interest.

Over the last four years, over 2000 patients have beentreated for various cardiac arrhythmias using the CARTOmapping system at our institution; Figs 6 to 18 are

Fig. 5. The selected points are connected by lines to form several adjoiningtriangles. These subsequently create a global model of the chamber-gatedelectrograms and are used to create an activation map that can be superimposedon the anatomic model. The points are color-coded and red represents the earliestelectrical activation with orange, yellow, green, blue and purple representingprogressively delayed activation areas.

IHJ-009-01.p65 10/17/01, 12:07 PM426


representative example of some of the arrhythmias mappedusing this system in our institution. As shown in Figs 6 to8, which depict atrial flutter, and Figs 12 to 15, which depictfocal atrial tachycardia, differentiating between re-entrantand focal arrhythmia is made quite easy. Followingidentification of the mechanism involved in the genesis ofthe arrhythmia, the electroanatomic map generated can beused to design an appropriate ablation strategy. Definingthe target sites for ablation is usually based upon acombination of anatomical and electrical criteria.

CARTO strategy for specific arrhythmias:Atrial flutter: If the entire circuit is mapped, a re-entranttachycardia such as atrial flutter is characterized by a rangeof activation times that will equal the cycle length of thetachycardia and also by close spatial orientation betweenarbitrary early and late sites (Figs 6 and 7). After taking theusual anatomic landmarks mentioned before, our approach

Fig. 6. Three-dimensional anteroposterior view of a right atrial activation mapduring typical atrial flutter (upper panel). On the color scale, red represents thearea of earliest endocardial activation. Propagation map showing four clockwisestages of right atrial flutter; propagation of the electrical impulse (clockwise)through the right atrium is represented by red in each successive stage (lowerpanel). From: Sra et al. Electroanatomic mapping to identify breakthrough sitesin recurrent typical human flutter. Pacing Clin Electrophysiol 2000; 23: 1483.Reproduced by permission.

Fig. 7. Coronary sinus (CS) pacing during sinus rhythm in a patient withrecurrent atrial flutter. In the upper panel, the right atrial activation pattern isshown in the left anterior view. Red represents the area of earliest activation;orange, yellow, green, blue and purple indicate progressively delayed activationin relation to reference atrial potential. In the lower 6 panels, propagation mapsshow electrical wavefronts travelling in opposite directions, i.e. through theisthmus and the septum and high right atrium, and then colliding in the lateralright atrium, thus confirming the activation pattern seen in the upper panel.Slightly delayed conduction across the isthmus is typical of slower conductiondue to previous ablation. IVS: inferior vena cava; RA: right atrium; SVC: superiorvena cava. From: Sra et al. Electroanatomic mapping to identify breakthroughsites in recurrent typical human flutter. Pacing Clin Electrophysiol 2000; 23:1486. Reproduced by permission.

is to do a detailed tricuspid annulus/IVC isthmus mappingby withdrawing the catheter at 2–3 mm intervals andtaking several points along the line. Voltage, activation andpropagation maps are then created to look for any high-voltage areas which are better avoided in first-time ablationsor to look for high-voltage and breakthrough sites duringcases of recurrent atrial flutter, as these sites should betargeted under these circumstances. Bipolar electrogramsrecorded from the mapping catheter are filtered at 30–400 Hz.Areas showing <1 mV are considered to represent scartissue. Our strategy is to ablate during atrial flutter; however,the ablation can be performed during sinus rhythm as well.

Ablation is usually performed at 50–60 °C for 30–60 s at each site. At the end of lesion delivery, if significant

IHJ-009-01.p65 10/17/01, 12:07 PM427


atrial activity is still visible, it is important to give morelesions or go around that site if it is identified as a high-voltage area.

Ablation success is defined by: (i) termination of atrialflutter during application of radiofrequency (RF) current(Fig. 8); (ii) inability to induce atrial flutter duringprogrammed atrial stimulation at baseline and duringisoproterenol infusion; and (iii) demonstration duringpacing of a line of bi-directional conduction block betweentwo points, i.e. from the coronary sinus ostium to the rightatrium adjacent to the tricuspid annulus posterior andlateral to the ablation line (Figs 9 and 10). Immediately afterablation, a limited assessment is performed using10–15 points during coronary sinus (CS) and low atrialpacing on either side of the ablation line. After waiting for30–45 min, detailed mapping and programmed stimulationis repeated.

Fig. 8. Termination of atrial flutter during ablation. Shown from top to bottomare surface ECG leads I, II, and V1, intracardiac recordings for the high rightatrium (HRA), proximal coronary sinus (CSp), distal coronary sinus (CSd),distal ablation (RFd) and time line (T). Counterclockwise atrial flutter whichterminates during application of radiofrequency current is shown on early CSprecordings.

Fig. 9. Demonstration of conduction block following ablation of atrial flutter.Upper panel: endocardial activation sequence in the right atrium during coronarysinus pacing after successful ablation of a breakthrough site of recurrent flutteris shown. Red represents the area of earliest activation and purple the last areaactivated. An area of conduction block seen in the isthmus as the lower lateralportion of the right atrium is the last area activated. Propagation map showingfour counterclockwise stages of right atrial flutter during coronary sinus pacingis shown in the lower panel. Propagation of electrical impulse (counterclockwise)through the right atrium is represented by red in each successive stage. There isno longer conduction across the isthmus. Reversal of the activation patternsuggests that ablation has created a complete line of block across the isthmus.From: Sra et al. Electroanatomic mapping to identify breakthrough sites inrecurrent typical human fluter. Pacing Clin Electrophysiol 2000; 23: 1485.Reproduced by permission.

The main advantage of using the CARTO mappingsystem for atrial flutter is the visibility of an ablation lineas it is performed so that no area is left out or repeatedlyablated. In patients with recurrent atrial flutter followingprevious ablation attempts, CARTO mapping is particularlyhelpful. As mentioned before, a detailed map of the isthmusis constructed. Since the system records both the activationtime as well as the electrogram amplitude from each pointselected, a voltage map which reveals the unablated areacan be prepared. The RF energy can then be delivered tothe site with more precision. The activation wave frontduring atrial flutter halts at the previously completed linebut continues through the breakthrough site. As shown inFig. 11, the CARTO mapping system is also useful forablation of atrial flutter if recurrence occurs immediatelyfollowing ablation.

Focal atrial tachycardia: In contrast to re-entrantarrhythmias such as atrial flutter, focal atrial tachycardiais characterized by a total range of activation times thatare usually shorter than the cycle length of the tachycardia,and a very well-defined early activation site surrounded bylater activation sites. An area of interest showing earlyactivation is identified followed by detailed mapping of thatarea (Figs 12–15). As two simultaneous maps can bedisplayed on the CARTO mapping system, a precise site canbe targeted. Furthermore, if unipolar electrograms at earlyright atrial septal sites show initial positive deflection, thelocation of tachycardia is most likely left atrial.

IHJ-009-01.p65 10/17/01, 12:08 PM428


Simultaneous maps of the two atria can then be drawn (Fig.14) to localize the earliest activation site.

Ventricular tachycardia: It has been shown that the chancesof terminating VT are better if (i) RF current is delivered atsites where there may be entrainment with concealedfusion; (ii) the post-pacing interval approximates thetachycardia cycle length (<10 ms difference in cycle length);(iii) the stimulus to QRS duration is >60 ms or <70% oftachycardia cycle length; and (iv) there are mid-diastolicand pre-systolic potentials.33,34,39 Attempted RF ablation ofVT, however, is quite complicated in patients with priormyocardial infarction (MI) because of broad re-entrant

Fig. 10. Demonstration of bi-directional line of conduction block to confirmsuccessful ablation of atrial flutter. The same color scale indicates progressionmarked by as described before. After atrial flutter ablation, pacing is performedfrom the coronary sinus ostium and lower right atrium to demonstrate bi-directional conduction block. There is no conduction across the isthmus;conduction is markedly delayed in the posterior right atrium (left panel) withearliest activation now in the septal region. Pacing of the low right atrium (rightpanel) confirms no conduction across the isthmus as well. The presence of earlyand late adjacent points and progressively delayed activation (counterclockwiseduring coronary sinus pacing and clockwise during low right atrial pacing) isimportant to demonstrate conduction block. RA: right atrium; SVC: superiorvena cava. From: Sra et al. Electroanatomic mapping to identify breakthroughsites in recurrent typical human flutter. Pacing Clin Electrophysiol 2000; 23:1488. Reproduced by permission.

Fig.11. Ablation of breakthrough site of atrial flutter (left anterior obliqueprojection). Brown circles represent ablation sites. Pink circles represent thetricuspid valve annulus (TVA)/inferior vena cava (IVC) regions. Despite thisapparently complete line of ablation created across the isthmus, atrial flutterrecurs after ablation. Propagation map (middle panel) shows area ofbreakthrough near the TA. After re-ablation of this breakthrough site (greencircles), the gaps are eradicated and no further breakthrough is seen duringcoronary sinus pacing. Note: the middle panel has been rotated slightly to showbreakthrough sites more clearly. CS: coronary sinus. From: Sra et al.Electroanatomic mapping to identify breakthrough sites in recurrent typicalhuman flutter. Pacing Clin Electrophysiol 2000; 23: 1490. Reproduced bypermission.

Fig. 12. Electroanatomic mapping of the right atrium in a patient with a historyof atrial septal defect repair and supraventricular tachycardia. The rhythm wasmisdiagnosed as atrial flutter during conventional mapping. Duringelectroanatomic mapping, the tachycardia was correctly diagnosed to be a focalatrial tachycardia. The front end of the atrium is opened up to show the earliestsites of activation as shown by the green circle. The brown circles represent theablation sites.

Fig. 13. Propagation map of the patient shown in Fig. 12 illustrating the siteof early activation.

IHJ-009-01.p65 10/17/01, 12:08 PM429


circuits for VT. The prospect of success is even poorer whenVT has multiple morphologies because of extensivemyocardial scarring, or it causes rapid hemodynamicdeterioration due to rapid heart rate and depressed leftventricular function. These features are often acontraindication to catheter ablation, as it is difficult to keeppatients in tachycardia for a long time.

If it were feasible to interrupt broad re-entrant circuitsby using lesions with RF ablation, catheter ablation couldbe successful but this requires capabilities beyond what canbe achieved using conventional mapping techniques toidentify the critical regions of slow conduction: (i) toprecisely identify the margins of the scar; (ii) to identify andthen return precisely to areas of interest; (iii) to visualizelines being created; and (iv) to perform ablation during sinusrhythm in unstable patients.

Such capabilities may now be feasible with the CARTOmapping system, which can be used to create 3-D maps ofcardiac chambers, including those of the left ventricle.

In our laboratory, CARTO mapping is performed using aNavistar Cordis-Webster mapping and ablation catheter.The catheter is introduced and positioned in the left ventricleusing a trans-septal or retrograde transaortic approach.During stable sinus rhythm, the margins of the left ventricleare defined by placing the catheter at the apex, sweeping it

Fig. 14. Propagation map showing the spread of activation to the rest of theatrium.

Fig. 15. Side-by-side right and left atrial electroanatomic maps in a patientwith left atrial tachycardia. Various anatomical locations including patentforamen ovale (PFO) have been tagged and labeled. During right atrial mapping,the earliest activation was thought to occur from a site just below the bundle ofHis. However, during left atrial mapping with the catheter positioned across thePFO, the earliest activation was found to be the left side of the septum. Ablationof this site led to successful termination of the tachycardia.

Fig. 16. Electroanatomic map of the left ventricle. A voltage map is createdand the various locations are labeled. The red color represents scar tissue. Thebrown tags represent the location of two ventricular tachycardia morphologieswhich were successfully ablated.

IHJ-009-01.p65 10/17/01, 12:09 PM430


back to the base and margin under the mitral valve annulusand over other areas, such as the interventricular septumand the lateral walls. We prefer the trans-septal approachto prevent the repeated introduction of the mapping andablation catheter through the aortic approach in case of arise in impedance. The trans-septal approach also makes iteasier to maneuver the catheter for mapping and ablationin the left ventricle.

After generation of the voltage map, pace mapping (attwice the diastolic threshold) is performed at the border ofthe scar tissue to replicate clinical and induced VT. Pointsthat most closely matched these previously documentedpatterns of VT are tagged for ablation (Fig. 18). For a site tobe considered appropriate, pace mapping has to mimic VTin at least 11 out of 12 leads. In cases of incessant, buttolerated VT, mapping can be performed during the VT. IfVT occurs during pacing, the software allows switchingbetween maps acquired during sinus rhythm and thoseacquired during VT. During VT, points are taken rapidly ifthe patient is hemodynamically stable, then accepted orrejected based on morphology, i.e. whether it was a QRScomplex acquired during VT or an ectopic or fusion beat.Activation mapping is performed to assess the site of earliestbreakthrough. Multiple points are taken and tagged, andthese early points are then used for subsequent ablation.

For the purpose of dealineating scar tissue, the infarctedmyocardium is distinguished from healthy myocardium by

a reduction in electrical voltage (bipolar voltage <1.0 mV).After appropriate sites are identified by means of either

pace mapping or activation mapping, the catheter ispositioned at each site that had been tagged, and RF energyimpulses (50–60 V for up to 60 s) are delivered to createlesions. If the electrograms at the catheter tip showapproximately <50% reduction in amplitude, ablation isrepeated in 30-s increments until the desired effect isobtained. Depending on the pace mapping technique,lesions are created at sites where 12-lead ECGs obtainedduring pacing most closely match the targeted VT. Next, aseries of lesions are created, transecting the edges of the

Fig. 17. Voltage maps and sites of ablation in four different patients withventricular tachycardias. A different view is shown in each patient.

Fig. 18. Detailed voltage map of a patient with extensive myocardial infarctionshows extensive scar tissue. Red represents an area of dense scarring or of lowestvoltage (amplitude ≤1.0 mV). Orange, yellow, green, blue and purple indicateprogressively increased voltage and healthier tissue. The pink circle representsthe site of the pace map which mimicked the ventricular tachycardia morphologymost closely. Brown circles represent the lesion sites transecting the edges ofthe scar. The lesion sites are encircled with black to delineate the spots moreclearly.

IHJ-009-01.p65 10/17/01, 12:09 PM431


scar (Fig. 18). Where available, data from activations mapsare used to identify appropriate ablation sites. Duringactivation mapping, the earliest identified site is targeted.Following this, linear lesions are created to transect themargins of the scar.

Successful ablation is defined as: (i) termination ofongoing VT during application of RF current; and (ii)inability to induce the targeted VT during programmedventricular stimulation, both at baseline and duringisoproterenol infusion, using similar pacing protocols eachtime as at the baseline study.

Used alone, pace mapping during sinus rhythm is notspecific enough to target an appropriate site for ablation.However, in one study, combining this technique with longstimulus to the QRS interval identified the site in asignificant proportion of cases. Identification capability isenhanced even further when pace mapping is combinedwith substrate mapping. Because conventional techniquesmake it difficult to precisely localize the zone of slowconduction. With pace mapping during sinus rhythm, it isdifficult to map several areas and then go back to the mostappropriate or optimal site. Hence, focal ablation of re-entrant circuits may be insufficient to interrupt these broadre-entrant circuits. The risk of hemodynamic collapse inpatients who are unable to tolerate VT also frequentlyprecludes successful mapping and ablation; such patients,including those in our study, have significant left ventriculardysfunction and large areas of scarring. Furthermore,should an impedance rise be noted during ablation usingthe fluoroscopic technique, it is difficult to return preciselyto the original ablation site.

To delineate myocardial scarring, the novel method ofcreating voltage maps of the left ventricle by means ofCARTO mapping provides promising advantages, one beingthat maps can be created during sinus rhythm at leisure.Since, as mentioned earlier, the lower voltage of amyocardial scar accurately distinguishes it from healthymyocardium, in conjunction with the pacing technique thevoltage map can clearly identify areas of interest. Eventhough pace mapping may fail to precisely identify a zoneof slow conduction, ablation guided by CARTO mapping cancreate linear lesions across the edges of a scar which mayencompass all re-entrant zones, i.e. the outer loop, zones ofslow conduction and the inner loop. This could also explainwhy ablation at one region sometimes terminates VT of twodifferent morphologies. In the event of a hemodynamicallystable VT, activation maps can be created and one mayswitch rapidly between activation and pace mapping shouldhemodynamic instability necessitate termination oftachycardia. Furthermore, with any rise in impedance

during the ablation, one can return to the exact site ofprevious ablation at any time because all ablation pointshave been tagged. Thus, by facilitating the creation ofcontinuous lesions, CARTO mapping can benefit evenpatients who have refractory and hemodynamicallyunstable VT. This was clearly evidenced in our study39 by astatistically significant reduction in the number of shocksdelivered by the implantable cardioverter–defibrillator (ICD)and antiarrhythmic medications in patients with verysymptomatic and refractory VT.

Clinical implications: Electroanatomic mapping can thusgreatly benefit patients with symptomatic, refractorytachycardia. Despite some limitations mentioned below, thecapability of this technique to create several potential pointsof interest and return to them provides significantadvantages over conventional techniques. Atrial flutter,especially recurrent flutter, can be ablated more precisely.The conjunctive use of voltage mapping and pace mappingduring sinus rhythm can help in ablating even multiplehemodynamically unstable VTs. As for patients withhemodynamically tolerated VT, further studies are neededto assess whether this technique will obviate the need foran ICD or antiarrhythmic drug therapy.

Limitations of CARTO mapping: Creation of anelectroanatomic map may be a lengthy process of taggingmany points, depending on the spatial details needed toanalyze a given arrhythmia. Since these points do notprovide real-time, constantly updated information, moretime may be needed for making new maps to see a currentendocardial activation sequence, to detect a change inarrhythmia, or to fully visualize multiple VTs. To createoptimum maps, sustained arrhythmia is usually needed totake enough data points.

Noncontact Mapping

The inability to accurately relate electrophysiologicinformation to a specific spatial location in the heart limitsconventional techniques for RF catheter ablation of complexcardiac arrhythmias. Recent advances such as CARTOmapping offer advantages over conventional mapping byreconstructing a 3-D image of the cardiac chamber.However, in CARTO mapping, the degree of resolution islimited by the time available to acquire data points and, inmany instances, use of the technique is restricted tohemodynamically stable and sustained arrhythmias.Arrhythmia mapping would be significantly advanced ifclinicians had the ability to collect thousands of data pointsin a single beat without the need for point-by-point contact

IHJ-009-01.p65 10/17/01, 12:09 PM432


electrograms. A high-resolution noncontact mappingsystem capable of single-beat mapping has been validatedin humans40–52 and has been used at our institution overthe past 2 years in over 200 patients with complex cardiacarrhythmias. These included patients with atrial flutter, VT,atrial tachycardia and atrial fibrillation (AF).

Fundamentals of the noncontact mapping system:The mapping technique using the EnSite 3000™(Endocardial Solutions, St Paul, MN) mapping system hasbeen studied extensively. This technique is based on thepremise that endocardial activation creates a chambervoltage field which obeys LaPlace’s equation (Fig. 19).Noncontact intracavitary electrodes can be used to detectthe potential field on the endocardial surface. The potentialsin this field are generally lower in amplitude and frequencythan the source potential of the endocardium itself, whichlimits their utility in the raw form. To improve accuracy andstability in reconstructions, an inverse LaPlace’s equationis used. Once the potential field has been established, over3000 activation points can be displayed as computedelectrograms or as isopotential maps. A three-step processof establishing geometry, identifying the area of interest andnavigating the ablation catheter in this area is used to mapand treat arrhythmias.

The balloon or multielectrode array (MEA) comprises abraid of 64 polyamide-insulated, 0.003" diameter wires(Fig. 20). For electrophysiologic studies, any mapping

catheter can be used. The catheter location system uses alow-level, 5.68 kHz current emitted by the distal electrodewhich returns to each of the two intra-chamber ringelectrodes on the MEA (Fig. 21). Since the positions of boththe array electrodes and the current-sink electrodes areknown, a custom algorithm is generated to determine theposition of the roving catheter by demodulating the 5.68 kHzpotentials on the MEA. The locator system serves severalpurposes:(1) Three-dimensional modeling: As a conventional cathetermoves around the cardiac chamber under fluoroscopic andelectrographic guidance, a series of coordinates are builtwithin the endocardium from which a contoured model is

Fig. 19. The potential distribution on the multielectrode array (MEA) or balloon(solid line) created by the endocardial activation is described by LaPlace’sequation. The highest chamber voltage is at the site of origin of the electricalimpulse. Although the electrode closest to the origin of impulse is influencedthe most, all the electrodes are influenced, with the degree of influencediminishing with the distance between the electrode and each endocardial point.The system reconstructs the voltage field (dotted lines) by solving LaPlace’sequation using the boundary element method. (See text for details).

Fig. 20. The multielectrode array (MEA) once left deployed in a chamber isinflated into a balloon with 64 electrodes. The magnified version of one of theseelectrodes is shown on the right.

Fig. 21. The EnGuide™ location signal indicates the position of the mappingand ablation catheter. A 5 kHz signal is emitted from the tip of the contactcatheter and sensed on the proximal and distal ring electrodes of the EnSite3000™ catheter. The signals received establish an EnGuide™ field, thusestablishing the location of the catheter.

IHJ-009-01.p65 10/17/01, 12:10 PM433


generated (Fig. 22). To accomplish this, the systemautomatically stores only the most distant points visited bythe catheter, then uses a bicubic spline-smoothingalgorithm to create a high-resolution contoured model ofthe chamber. By gating each sample to a surface-lead Rwave, a contoured wire frame model with end-diastolicdimensions is created (Fig. 22).

(2) Catheter guidance and location: The locator signal can alsobe used to display and log a catheter’s position within theendocardial model and to guide it to the region of interestfor ablation.

The MEA is deployed in the cardiac chamber of interestand the geometry created. Once the chamber geometry hasbeen delineated, tachycardia is induced and the tachycardiasite mapped using the EnSite 3000™ system. As theelectrical activity within the endocardium is sensed by theMEA’s microelectrodes, the array potentials are sampled at1200 Hz and passed to a higher-order boundary elementcomputation. The boundary element uses an inverseformation to solve LaPlace’s equation for each sample,yielding endocardial potentials. Using this method, morethan 3000 isopotential points are generated and may bedisplayed as computed (virtual) electrograms and asisopotential maps.

The EnSite 3000™ system reliably replicates the voltageelectrogram signal displayed by the contact catheterelectrodes. In a study that compared contact recordingswith computed electrograms during sinus rhythm andpacing, a mean correlation of 0.96 was demonstrated inreproducing morphology. When pacing techniques were

Fig. 22. The mapping catheter, as it moves around the chamber, establishes aconvex hull model of the chamber anatomy. Using a bicubic spline-smoothingalgorithm, the convex model is converted into a high-resolution contoured modelof the chamber.

used in this study, the distance from the RF lesion to thepacing plunge electrode was approximately 4.0±3.3 mm(mean 3 mm).44

Principle of inverse solutions and filter settings: Theelectrical activity detected by the electrodes on the surfaceof the MEA is generated by the potential field on theendocardial surface of the chamber of interest. Asdiscussed, the potential distribution on the MEA created bythe endocardial activation is described by LaPlace’sequation, s2V=0. The highest chamber voltage is at thesite of origin of the electrical impulse. Although theelectrode closest to the origin of the impulse is influencedthe most, all the electrodes on the MEA are influenced, thedegree of influence diminishing with the distance betweenthe electrode and each endocardial point.

The electrograms detected by the noncontact MEA areof lower amplitude and frequency than those detected atthe source by contact electrograms which, in turn, limitsthe utility of contact electrograms in raw form.49–56 Atechnique called the boundary element method (BEM),based upon the inverse solution to LaPlace’s equation isused. The potential field created on the MEA surface dependsupon the geometry of both the MEA and the endocardiumand their relationship in space. The MEA geometry isdetermined at the time the catheter is manufactured whilethe endocardial geometry is obtained at the time ofgeometric reconstruction as described above. With thisinformation, it is then possible to compute endocardialelectrograms from the MEA potentials by inverse solutionof LaPlace’s equation.

The regions or space with complex electrical phenomenaare computed into smaller regions so that behavior withinthe element can be accurately determined using the FiniteElement Method (FEM). The properties of electricalconduction between such different elements are equated,resulting in a linear system of many equations. The BEMworks similar to the FEM but only the two-dimensional(2-D) surface (i.e. the boundary) is modeled rather than theentire 3-D domain, resulting in far fewer elements andequations to deal with.

However, small noise in the measured potentials in thecavity could potentially lead to large errors in thereconstructed endocardial electrograms if careful attentionis not paid to the application of this method. The techniqueof regularization is used to overcome this problem.55,56

Several sources of noise can affect the recording andinclude:

(1) electrical interference from the surroundingenvironment;

IHJ-009-01.p65 10/17/01, 12:10 PM434


(2) electrochemical fluctuations at the surface of therecording electrodes; and

(3) noise from the amplifier circuitry due to electricalfluctuations.

Environmental noise comes primarily from power lines andmanifests as common-mode noise across the body.Electrical artifacts fall into three categories:

(1) Main power supply noise, i.e. 50–60 Hz. This noise is acharacteristic of the local electrical power source. Itis a medium-frequency noise which is usually steadyin amplitude.

(2) High-frequency noise is a random noise which is usuallythe result of using many electrical devices on the samepower source. This noise does not always have a steadyamplitude and usually manifests as a thickening ofthe signal recording.

(3) Low-frequency noise is a random noise which makesthe ECG signals appear as if they “float” on waves.

Common-mode noise is rejected by connecting a unipolarreference to the bipotential amplifiers in the standarddifferential configuration. Noise due to electrochemicalfluctuations at the surface of the electrodes is primarily lowfrequency in nature and is reduced by coupling theelectrodes to the amplifiers through a high-pass filter whichattenuates frequencies below 0.1 Hz.

The small noise fluctuations in amplifier circuitry aregenerally broadband in frequency and are uncorrelatedbetween channels. Regularization is required to attenuatethese random fluctuations and any residual noise thatremains after reduction of the main power supply, high- andlow-frequency noise (described earlier). Randomfluctuations between amplifier channels translate intoerrors in measurement at the recording sites which arerandomly associated with these sites. These errors can bethought of as small and rapid fluctuation steps of potentialacross the surface of the MEA. Regularization attenuatesthese rapid fluctuations across space at any given instantin time much as a low-pass filter would attenuate rapidfluctuations across time at any given point in space.

The accuracy of the noncontact mapping system thusdepends on inverse solutions and a methodology which, inturn, will depend upon accurate and detailed geometriccreation. In addition, the accuracy will depend upon theapplied regularization technique. Several techniques havebeen used to improve the accuracy of the process ofregularization. The geometry and voltage are modeled usingbicubic spline surfaces instead of the standard linearelements. This allows for a more physiologically realisticmodel of the endocardium. Small errors in the model may

still occur, however, due to undersampling of theendocardial points or due to complexity of the chambergeometry.

Reconstruction of the map and mapping protocol:The MEA is deployed in the cardiac chamber of interest andthe mapping catheter is then moved to known anatomicallocations which are tagged. A detailed geometry of thechamber is then reconstructed by moving the mappingcatheter around the chamber. Unlike the CARTO mappingsystem, the noncontact system allows for the patient to bein sinus rhythm or tachycardia during creation of thegeometry. Activated clotting time (ACT) is kept at 250 s and300 s for right-sided and left-sided mapping, respectively.

The noncontact mapping system is capable ofreconstructing and interpolating more than 3000 unipolarelectrograms over the endocardium during mapping and,in turn, isopotential or isochronal maps can bereconstructed from these electrograms. The color rangerepresents voltage or timing of onset. In addition,electrograms (virtual electrograms) may be selected usingthe mouse from any part of the created geometry anddisplayed as waveforms.

Because BEM is used for inverse solutions, the 3-Dmyocardium is treated as a 2-D endocardial surface. Thereconstructed electrograms are subject to the sameelectrical principles as contact catheter electrograms, asthey contain far-field electrical information from thesurrounding endocardium as well as the underlyingmyocardium signal vector, and distance from measurementcan affect the contribution to the electrogram. Althoughstructures such as the papillary muscles may contribute tothe reconstructed electrograms, their contribution is notsignificant relative to the much larger area of thesurrounding endocardium.

Validation of reconstructed electrograms: Noncontactmapping has been validated in several studies. In the studyby Gornick et al.44 three methods of validation were used:(i) driven electrodes in an in vitro tank were located; (ii)waveforms generated from an array catheter were comparedwith catheter contact waveforms in canine left ventricles;and (iii) sites of local left ventricular endocardial activationwere located and marked with RF lesions. Tank testinglocated a driven electrode to within 2.33±0.44 mm.Correlation of timing and morphology of computed versuscontact electrograms was 0.96. Radiofrequency lesionsmarked 17 endocardial pacing sites to within 4.0±3.2 mm.However, the accuracy of the location changes the furtherthe distance from the center of the balloon.

Although electrograms recorded by contact and

IHJ-009-01.p65 10/17/01, 12:10 PM435


reconstructed noncontact electrograms are similar in bothnear- and far-field components of electric potential, theremay be variations. In several studies,41 results from 34 leftventricular electrograms were compared to contactelectrograms, demonstrating a mean timing difference of0.9 ms and a mean morphology cross-correlation of 0.83over a population of 31 000 recorded cycles. One-half of thetime differences were within 2 ms and 77% were within10 ms.

Clinical applications: Over the past 2 years, noncontactmapping has been used in our laboratory to treat over 200complex cardiac arrhythmias. Some representativeexamples are depicted in Figs 23 to 30.

Atrial flutter: The patient mapped in Fig. 23 had atrialflutter which was ablated at the time of mitral valve surgeryusing a malleable EP Technologies™ RF probe (BostonScientific, San Jose, CA). However, 3 weeks after surgery,the patient was admitted with dyspnea and palpitations. Thesurface ECG of the tachycardia was again suggestive of atrialflutter with typical negative “saw tooth” waves on theinferior ECG leads.

Using the EnSite 3000™ noncontact mapping system,the activation sequence of the re-entrant circuit in the rightatrium was revealed to be counterclockwise atrial flutter.Detailed mapping identified the breakthrough site in the Fig. 23 (b). Pre-ablation color-coded isopotential maps during sinus rhythm

of the tricuspid valve (TV), inferior vena cava (IVC), isthmus (in caudal view)and surface ECG leads 1 and 2, and reconstructed electrograms (A to E) duringseptal pacing (top panel) from the coronary sinus ostium (CSO) and low lateral(LAT) right atrial pacing (bottom panel). Reconstructed electrogram A–Erepresent the direction of spread of the waveforms during septal pacing and lowlateral right atrial pacing. During conduction over the isthmus, the wavefronttraverses the gap while stopping at the ablation line at other sites. Time ofendocardial activation displayed on the color-coded map is indicated by a verticalyellow cursor on the reconstructed electrograms. Stimulus to atrial activationtime is different in the two figures and probably represents different conductioncharacteristics, depending upon how the waveform is traveling. S: stimulusartifact. From: Sra et al. Noncontact mapping for radiofrequency ablation ofcomplex cardiac arrhythmias. J Interven Card Electrophysiol 2000; 5: 325.Reproduced by permission

isthmus (Fig. 23a). A computed electrogram showed doublepotentials at the sites of block and continuous conductionthrough the site of “gap”. Following overdrive pacing of theatrial flutter, bi-directional conduction over the gap wasconfirmed during pacing from the low lateral right atriumand the CSO (Fig. 23b). Three lesions were delivered at the“gap” site. Radiofrequency ablation performed at the “gap”following induction of atrial f lutter terminated thetachycardia.

To confirm the absence of conduction across the isthmusfollowing ablation, atrial pacing was done from the proximal

Fig. 23 (a). Identification of breakthrough sites in typical atrial flutter. Oncolor-coded isopotential maps of the right atrium ECG lead II reconstructedelectrograms (3 to 7) during atrial flutter, the site of “gap” at the tricuspidannulus (TA) and inferior vena cava (IVC) isthmus is clearly seen. The “gap”represents the reconstructed electrogram(8). Reconstructed electrograms ofthe isthmus show double potentials at the area ablated earlier where noconduction is seen and continuous activity at the “gap” area where conductionoccurs during atrial flutter. CSO: coronary sinus ostium; His: bundle of His;SVC: superior vena cava. From: Sra et al. Noncontact mapping for radiofrequencyablation of complex cardiac arrhythmias. J Interven Card Electrophysiol 2001;5: 325. Reproduced by permission.

IHJ-009-01.p65 10/17/01, 12:11 PM436


that the leading edge of a wavefront of atrial depolarizationspread out from that site. Following RF ablation at this site,the tachycardia was noninducible despite an aggressivestimulation protocol, including programmed stimulationand burst atrial pacing at baseline and during isoproterenolinfusion. The time to prepare the balloon catheter andreconstruct right atrial geometry was 5 min and 19 min,respectively. During follow-up of 11 weeks, the patient hashad no recurrence of arrhythmia and is off all medications.

Atypical atrial flutter and atrial fibrillation (AF)originating from the right atrium: In the patientdepicted in Fig. 25, several documented prior tracings weresuggestive of paroxysmal atrial fibrillation (AF), typicalatrial flutter with “saw tooth” flutter waves in the inferiorleads and, possibly, atrial tachycardia. Based upon cardiacevaluation prior to the referral, including cardiaccatheterization, there was no evidence of coronary arterydisease. Because flecainide and amiodarone did not controlhis symptoms adequately, the patient underwentelectrophysiological evaluation and RF catheter ablation.Right atrial geometry was created using the EnSite 3000™noncontact mapping system. During atrial pacing, typicalatrial flutter with counterclockwise rotation was easilyinduced, mimicking one of three documented arrhythmias.

Fig. 23 (c). Bi-directional block post-ablation. Color-coded sequentialisopotential maps modeled to the right atrium of the patient, surface ECG lead1 and reconstructed electrograms (A to E) during coronary sinus (CS) pacing(top panel), and low lateral atrial pacing (bottom panel) are shown. The frontend of the map is opened to show the multielectrode array, represented by theyellow ellipsoid. During pacing from the CS, the depolarizing wavefront is blockedat the tricuspid valve (TV)/ inferior vena cava (IVC)/isthmus region, as evidencedby markedly delayed activation at the low lateral atrium. Similarly, during lowerlateral atrial pacing, bi-directional block is evidenced by markedly delayedactivation of coronary sinus ostium (CSO). Reconstructed electrograms A to Eshow double potentials (DP) suggesting completed ablation line. His: bundle ofHis; S: stimulus artifact; SVC: superior vena cava. From: Sra et al. Noncontactmapping for radiofrequency ablation of complex cardiac arrhythmias. J IntervenCard Electrophysiol 2001; 5: 325. Reproduced by permission.

Fig. 24. Focal atrial tachycardia. Isopotential map of the right atrium andreconstructed electrograms (5 to 9) during a nonsustained episode of atrialtachycardia. Earliest activation of the tachycardia is shown (right panel). Deepnegative initial deflection in the reconstructed electrogram also suggests thatthe tachycardia originates close to the wavefront's origin on the isopotentialmap. CSO: coronary sinus ostium; His: His bundle region; IVC: inferior venacava; RAA: right atrial appendage; SVC: superior vena cava. From: Sra et al.Noncontact mapping for radiofrequency ablation of complex cardiacarrhythmias. J Interven Card Electrophysiol 2001; 5: 326. Reproduced bypermission.

coronary sinus. No conduction was observed across theisthmus and the lower lateral atrial wall showed verydelayed conduction (Fig. 23c). Next, to confirm bi-directional block across the isthmus, pacing was performedin the lower lateral part of the right atrium. Mappingshowed very delayed activation in the lower interatrialseptum near the coronary sinus ostium, suggesting a lineof block in the isthmus in this direction as well. The timetaken to prepare the balloon catheter and to reconstructright atrial geometry were 6 min and 27 min, respectively.During a follow-up of 10 months, the patient has had norecurrence of atrial flutter and is off all antiarrhythmicmedications.

Atrial tachycardia: Figure 24 shows a patient in atrialtachycardia. This patient presented with frequent runs ofnonsustained atrial tachycardia. Routine atrial stimulationprotocol failed to induce any arrhythmia. However, duringisoproterenol infusion, a short burst of atrial pacing wasable to induce runs of atrial tachycardia (less than 10 beats)but these runs could not be sustained.

Using the noncontact mapping system, detailed mappingidentified the earliest activation of the atrial tachycardia.On the computed electrogram, a negative initial slope wasconsistent with the site of earliest activation, which showed

IHJ-009-01.p65 10/17/01, 12:11 PM437


Atrial fibrillation originating from the pulmonaryveins: Figure 26 depicts a 63-year-old man with a historyof symptomatic persistent AF and hypertension but nocoronary artery disease. He had an estimated leftventricular ejection fraction of 40% and an enlarged leftatrium (4.8 cm). While on sotalol, the patient hadpremature atrial contractions (PACs) which reinitiated AFfollowing two external cardioversion attempts. Followingthis, the patient was placed on amiodarone for 4 weeks, then

Fig. 25 (a). Focal atrial fibrillation. Sequential isopotential maps of the rightatrium modeled to the patient’s chamber dimensions. The image is oriented tohighlight the crista terminalis region. Surface ECG leads and reconstructedelectrograms revealing premature atrial beat degenerating into atrial fibrillationare shown at the bottom. Timing on maps is indicated by a yellow cursor, shownpropagating through the premature atrial beat on reconstructed electrogramsnext to each isopotential map. The ectopic activation wavefront spreads out fromthe mid crista terminalis region. An area of nonconduction is seen just belowthe site of focal activation. CSO: coronary sinus ostium; TV: tricuspid valve;His: bundle of His; RAA: right atrial appendage. From: Sra et al. Noncontactmapping for radiofrequency ablation of complex cardiac arrhythmias. J IntervenCard Electrophysiol 2001; 5: 327. Repoduced by permission

Fig. 25 (b). Atypical nonisthmus-dependent atrial flutter. Sequentialisopotential maps and reconstructed electrograms, as in Fig.23, are shown. Theactivation wavefront travels clockwise through the same region (D to F) as seenin Fig. 23. Six RF lesions that transected the region of wavefront in Map Dsuccessfully terminated the atrial flutter. Abbreviation as in Fig. 23. From: Sraet al. Noncontact mapping for radiofrequency ablation of complex cardiacarrhythmias. J Interven Card Electrophysiol 2001; 5: 327. Reproduced bypermission.

Catheter ablation of the triscupid–IVC isthmus led totermination of the atrial flutter. Postablation, bi-directionalblock across the isthmus was demonstrated, as in the patientdescribed earlier, during pacing from the coronary sinus andlow lateral right atrium.

Following ablation of typical atrial flutter, the patienthad a spontaneous episode of AF similar to one of hisdocumented arrhythmias. Evaluation of the reconstructedelectrograms showed a premature atrial beat degeneratinginto AF. Sequential isopotential maps (Fig. 25a) showed theectopic activation waveform spreading out from the regionof the mid crista terminalis. Following ablation at this siteand during programmed atrial stimulation, a thirdtachycardia was induced. A positive P wave morphology wasseen on the inferior ECG. This third tachycardia was thoughtto be similar to one of his clinical tachycardias as well;however, noncontact mapping using isopotential mapsshowed an atypical nonisthmus-dependent atrial flutter(Fig. 25b). Radiofrequency lesions transecting the narrowestconduction zone successfully terminated the tachycardia.Time to prepare the balloon catheter and to reconstruct theright atrial chamber were 4.5 min and 16 min, respectively.During a follow-up of 4 months, the patient has stayed freeof arrhythmias and is off all medications.

Fig. 26. Left atrial mapping of focal atrial fibrillation. Isopotential map of theleft atrium with the front end opened up to show the MEA. Surface ECG leadsand reconstructed electrograms (6–10) show a sinus beat and a prematureatrial contraction (PAC) originating from the right superior pulmonary vein;this location was confirmed angiographically and echocardiographically.Ablation at this target site led to clinical cure of atrial fibrillation.

IHJ-009-01.p65 10/17/01, 12:12 PM438


externally cardioverted again. Two days later, the patientwas readmitted with dyspnea and palpitations and foundto be back in AF with a rapid ventricular response.Amiodarone was discontinued and the ventricular ratecontrolled with beta-blockers and calcium-channelblockers. Three weeks later, he was brought in to thelaboratory for mapping and ablation of AF.

The MEA and mapping catheter were placed in the leftatrium using a trans-septal approach and 3-Dreconstruction was performed. Transesophagealechocardiography, which had also been used to perform thetrans-septal catheterization, confirmed the location of thepulmonary veins, and these sites were tagged. Followingexternal cardioversion, the patient had frequent PACs, butno AF was reinitiated. Mapping of the left atrium showedthe origin of the PACs to be the right superior pulmonaryvein.

Following ablation at this site, during a 20-week follow-up period, the patient remained free of AF on half the dosageof amiodarone (200 mg/day) which had failed to maintainsinus rhythm prior to ablation. At the last follow-up visit,the patient had remained free of AF for 12 weeks and wasoff amiodarone entirely. Time to prepare the balloon andreconstruct the left atrium were 5 min and 29 min,respectively, and total procedure time was 219 min.

Myocardial ventricular tachycardia: Figure 27 depictsa 69-year-old man with a history of ischemiccardiomyopathy, a left ventricular ejection fraction of 19%and VT for which he had received an ICD. The patient hadreceived over 50 shocks from the ICD during a period of 3months despite a trial of antiarrhythmic medicationsincluding sotalol, procainamide, mexiletine andamiodarone. During programmed ventricular stimulation,sustained monomorphic VT similar to his clinical VT waseasily induced. Because of hemodynamic instability, thetachycardia needed to be immediately cardioverted.

Following trans-septal catheterization, 3-D mapping wasperformed by positioning the MEA in the left ventricle andinducing VT which was immediately cardioverted to sinusrhythm. By using a short segment of the tachycardia, a zoneof slow conduction in the left ventricle was identified duringthe mapping of this segment. After this, 5 RF lesions weredelivered, transecting this site. Repeated programmedventricular stimulation failed to induce any tachycardia.

During a 3-month follow-up period, the patient had oneepisode of VT while on amiodarone (200 mg/day). Time toprepare the balloon and reconstruct the left atrium were 4min and 23 min, respectively, and the total procedure timewas 167 min.

Idiopathic left ventricular tachycardia: Ventriculartachycardia with QRS morphology of right bundle branchblock (RBBB) and left-axis deviation in young patients whohave no structural disease has been reported. Endocardialactivation mapping during VT can identify the earliest siteof activation in the inferoposterior left ventricular septum.High-frequency potentials 30–40 ms before the VT–QRScomplex, thought to represent Purkinje potentials, havebeen identified in some patients and these data have beenused for VT ablation. In this example, single-beat mappingusing data derived from a noncontact mapping techniquewas used to delineate a part of the tachycardia circuit in ayoung patient with recurrent VT (Fig. 28) and identify mid-diastolic potentials, thus helping to define the mechanismof the tachycardia (Figs 29 and 30). The information wasthen used to identify the appropriate site for successfulablation. The MEA was deployed in the left ventricle via aretrograde transaortic approach and a 7 F mapping andablation EP Technologies ™catheter (Boston Scientific, SanJose, CA) was deployed via a trans-septal puncture. Onceleft ventricular geometry was delineated, VT was inducedand the site of tachycardia mapped using the EnSite 3000™system. Mid-diastolic potentials were identified oncomputed electrograms (Fig. 29) and confirmed to be partof the tachycardia circuit when a premature ventricularcontraction was found to be resetting the VT withoutaltering the duration of QRS. Activation from the potentialinscribing the QRS on the surface electrogram was notedto spread over the Purkinje fascicle network before exiting,

Fig. 27. Ventricular tachycardia circuit mapping. Sequential isopotential mapsof the left ventricle surface ECG leads and reconstructed electrogram (1–5)during ventricular tachycardia are shown. At the end of ventriculardepolarization, as shown by the timing of the cursor on the reconstructedelectrograms, a zone with no depolarization is seen on the isopotential map.Timing of the endocardial activation is represented by the yellow cursor; end ofdepolarization (Frame A); spread of activation over the diastolic pathway(Frames B to C); and exit site of the circuit (Frames E to F); which is confirmedby the presence of the cursor at the inscription of the QRS complex on surfaceECG. Thus, a zone of slow conduction during diastole was identified and targetedfor ablation. From: Sra et al. Noncontact mapping for radiofrequency ablationof complex cardiac arrhythmias. J Interven Card Electrophysiol 2001; 5: 329.Reproduced by permission.

IHJ-009-01.p65 10/17/01, 12:12 PM439


thus identifying part of the circuit. Ablation at the site ofthe mid-diastolic potential successfully ablated VT with nofurther initiation of the tachycardia. During a 4-monthfollow-up, this patient has been free of symptoms and hasnot needed medication.

Clinical implications: Noncontact mapping, whosehallmark is single-beat mapping, is thus an importantdevelopment in the management of cardiac arrhythmias.Noncontact mapping may be particularly helpful in patientswho have recurrent atrial flutter following a previousablation because it can identify the “gap” or breakthroughsite, which can then be targeted precisely. As any numberof maps can be superimposed on the initial geometry, bi-directional block at the ablation site can be rapidly identified

Fig. 28. Twelve-lead electrocardiogram of ventricular tachycardia inducedduring atrial pacing. Idiopathic left ventricular tachycardia is identified by rightbundle branch block morphology and left-axis deviation. From: Sra et al.Endocardial noncontact activation mapping of idiopathic left ventriculartachycardia. J Cardiovasc Electrophysiol 2000; 11: 1410. Reproduced bypermission.

Fig. 29. Mid-diastolic potential during ventricular tachycardia (VT). Top tobottom, surface electrocardiographic leads II and V1, and computed electrogramsfrom the selected site are shown. Time from mid-diastolic potential to onset ofQRS is 79 ms and tachycardia cycle length is 370 ms. Following a spontaneouspremature ventricular beat, VT is reset, as evidenced by shortening of cycle lengthto 200 ms. However, potential to QRS duration of 79 ms is unaltered, suggestingthat this potential is part of the tachycarida circuit. From: Sra et al. Endocardialnoncontact activation mapping of idiopathic left ventricular tachycardia. JCardiovasc Electrophysiol 2000; 11: 1411. Reproduced by permission.

Fig. 30. Mapping of the tachycardia circuit. Isopotential maps of the leftventricle, surface electrocardiographic leads, and computed electrograms areshown. Early activation (EA) (top right) and timing of wavefronts representedby a white cursor (bottom left) suggests fascicular activation followed bybreakthrough at the inferoseptal area (bottom right). The resultant wavefrontsubsequently depolarized the septum and the rest of the ventricle (not shown inthe figure). From: Sra et al. Endocardial noncontact activation mapping ofidiopathic left ventricular tachycardia. J Cardiovasc Electrophysiol 2000; 11:1411. Reproduced by permission.

during pacing following ablation. Tagging ablation areasduring delivery of each RF impulse and a constantly visibleablation line offer another advantage; they ensure that noarea is overlooked or ablated repeatedly.

Atypical atrial flutters are characterized by macro re-entrant tachycardias that do not use the tricuspid–IVC–Eustachian ridge isthmus and do not exhibit re-entryaround atrial incisions.57,58 Because this arrhythmia iscomplex and conventional mapping is inadequate, dataabout various forms of atypical flutter in humans arecurrently inadequate and, in most cases, treatment has beenrestricted to pharmacological management. As shown inthis report, noncontact mapping may be particularly helpfulin delineating macro re-entrant circuits and, moreimportantly, identifying the precise location where ablationmay terminate this type of tachycardia.

In patients who have nonsustained ectopic atrialtachycardia, noncontact mapping may allow rapidlocalization of the arrhythmic focus by simply selecting anarea of interest that shows the earliest endocardialactivation on the initial map and creating a detailed mapof this area using a single-beat mapping technique. Ourapproach is to start with a right atrial map, but the leftatrium is mapped if the point of earliest activation is in theseptal region and has an initial positive deflection onreconstructed electrograms or if the surface ECG issuggestive of left atrial tachycardia.

Identifying trigger sites of pulmonary vein ectopy isanother potential advantage of the noncontact mappingsystem. As recent studies have shown, pulmonary vein

IHJ-009-01.p65 10/17/01, 12:12 PM440


ectopy initiates AF in many patients.59–63 The presence ofectopy occuring spontaneously during variouselectrophysiological maneuvers as well as that initiating AFfollowing external cardioversion can all be used to identifyablation sites. None of these are readily identified bysequential mapping using conventional techniques. Thetask is further complicated by unpredictable andinconsistently inducible PACs. However, high-resolution3-D reconstruction of the left atrium (over 3000reconstructed electrograms) coupled with single-beatmapping makes it possible to precisely map these targets inpatients with infrequent PACs from just a single beat.

Despite ICD implantation, recurrent VT can causemortality and significant morbidity. For catheter ablationto succeed, it is critical to identify its re-entrant circuits,including zones of slow conduction and the exit site, whichis suggested by inscription of the QRS complex on thesurface electrocardiogram as the wavefront exits from suchregions. Using conventional techniques, it has been shownthat the chances of terminating VT are better if RF currentis delivered at sites where: (i) there may be entrainment withconcealed fusion; (ii) the post-pacing interval approximatesthe tachycardia cycle length; (iii) the stimulus to QRSduration is >60 ms; and (iv) there are mid-diastolic and pre-systolic potentials.64–66 However, with conventionalmapping in patients who have rapid hemodynamicdeterioration due to rapid rate and depressed ventricularfunction, the prospects of success are poor because it is notpossible to keep these patients in tachycardia duringmapping. A preferable alternative in such unstablepatients—now feasible with noncontact mapping—is tocreate linear lesions that transect critical regions, i.e. areasof slow conduction, and then return precisely to areas ofinterest, visualizing lines of ablation as they are beingcreated and performing the ablation during sinus rhythm(Fig. 31). As stated earlier, comparing reconstructed and

contact electrograms recorded from the same site duringVT has shown an excellent cross-correlation of 0.86±0.16and a mean time error of 1.67±10.06 ms, validating itsaccuracy.

Zipes et al.67 identified the following triad of diagnosticcharacteristics for this VT: (i) induction by atrial pacing;(ii) a configuration of RBBB with left-axis deviation; and(iii) no structural heart disease. Subsequently, Belhassen etal.68 demonstrated the sensitivity of this VT to verapamil.Debate has also centered on whether this tachycardia is re-entrant or triggered. It has also been proposed that thetachycardia circuit may be focal as suggestions have beenmade that ventricular extrastimuli can capture the ventriclewithout resetting the VT.69 Termination of the tachycardiawith programmed stimulation, an inverse relationshipbetween the extrastimulus and the first VT beat,entrainment criteria, and mid-diastolic potentials aresuggestive of re-entry with an excitable gap as a mechanismof VT.70,71 Okumura et al.70 were able to make severalobservations regarding the entrainment of the tachycardia:(i) there was pacing with constant fusion with the lastcaptured electrogram being entrained but not fused; and(ii) with increased pacing rate, there was progressive butconstant fusion and the activation site was not capturedantidromically (direct capture rather than orthodromiccapture). It has also been suggested that, due to closenessof the site to the posterior fascicles, entrainment is morefrequently achieved when pacing is accomplished from theright ventricular outflow tract rather than from the rightventricular apex. Successful ablation at sites demonstratingpotentials originating in the Purkinje fibers suggest thatthese may be part of the tachycardia circuit. In a study byNakagawa et al.,72 distinct Purkinje or p potentials wereindentified preceding the QRS during tachycardia by 15 to42 ms. In some instances, mid-diastolic potentials have alsobeen described.73 Due to ablation at different sites in the leftventricle, it has also been suggested that this arrhythmiaentity may not represent a homogenous group.74,75

Although pace mapping has been used during VT ablation,the data derived are often insufficient.

In the example described in this manuscript, noncontactmapping technology, by acquiring extensive data to createan activation sequence simultaneously during a single beatof tachycardia, was able to delineate some of thephysiological aspects of this intriguing arrhythmia. Unlikeobservations in prior studies, the precise delineation of partof the tachycardia circuit does not suggest that this is focal.Furthermore, the re-entrant mechanism of the tachycardiais validated. These findings are further corroborated by thediscovery that a single premature ventricular beat was able

Fig. 31. Electroanatomic map (left) of the left ventricle, as seen in Fig. 18, ina patient with extensive scar and myocardial ventricular tachycardia.Noncontact mapping (right) in the same patient showed a zone of slowconduction along the margins of the scar.

IHJ-009-01.p65 10/17/01, 12:13 PM441


to reset the tachycardia without altering the relationshipbetween the potential and the onset of the QRS. A 79 mspotential-to-QRS distance suggests that this was not aPurkinje potential as described in some earlier studies. Norwas this seen during sinus rhythm. There are insufficientdata to indicate any significant conduction system diseasein these patients.

The zone of slow conduction has been thought to bedependent upon slow inward calcium current.74,75 Also, therole of verapamil in some of these tachycardias confirmsthis hypothesis. The tachycardia circuit involved thePurkinje fascicle network, which may have beensignificantly diseased as suggested by prolonged conductiontime from mid-diastolic potential to onset of QRS on thesurface electrogram. These findings suggest that somedisease in the Purkinje system, by enabling a slowing ofconduction, may create the conditions for a re-entranttachycardia to occur.

Limitations: Although noncontact mapping is new andexciting, there are limitations to this technique. Caution willbe needed to interpret the great amount of new data notpreviously available by any prior technology. The overallaccuracy of the reconstructed electrograms may decreasewith distance of the area mapped from the MEA, thuscreating problems in mapping large cardiac chambers.Although the risk of complications is low, aggressiveanticoagulation measures because of MEA deployment inthe cardiac chamber expose patients to potential bleedingcomplications. The ACT needs to be kept around 250 s forright-sided and 300 s for left-sided procedures. It is alsopossible that some of the arrhythmias described, such asatrial tachycardia, atrial flutter and VT, could have beenablated with other techniques such as the CARTO mappingsystem or conventional mapping techniques. However,single-beat mapping has a clear advantage over othermapping techniques when dealing with complexarrhythmias and arrhythmias which are hemodynamicallyunstable.

Acknowledgment

The authors gratefully acknowledge the help of BarbaraDanek in the editorial preparation of this manuscript, andof Brian Miller and Brian Schuerrer in creating the graphics.

References

1. Akhtar M, Gilbert C, Wolf F, Schmidt DH. Reentry within the His–Purkinje system: elucidation of reentrant circuit utilizing right bundleand His bundle recordings. Circulation 1978; 58: 295–304

2. Denker SJ, Mahmud R, Tchou P, Jazayeri M, Al-Bitar I, Akhtar M.

Demonstration of catheter ablative technique for control ofventricular tachycardia due to macro-reentry within the His–Purkinje system [Abstr]. J Am Coll Cardiol 1986; 7: 243A

3. Rosen KM, Mehta A, Miller RA. Demonstration of dualatrioventricular nodal pathways in man. Am J Cardiol 1974; 33: 291–294

4. Ross DL, Johnson DC, Denniss AR, Cooper MJ, Richards DA, Uther JB.Curative surgery for atrioventricular junctional (“AV-nodal”)reentrant tachycardia. J Am Coll Cardiol 1985; 6: 1383–1392

5. Sra J, Jazayeri M, Avitall B, Dhala A, Deshpande S, Blanck Z, et al.Dual versus multiple atrioventricular nodal pathways in patientsundergoing selective fast or slow pathway ablation [Abstr]. J Am CollCardiol 1992; 19: 271A

6. Dhala A, Sra J, Blanck Z, Gonzalez-Zuelgaray J, Dwyer A. Markedlydifferent sequence of atrial activation during retrograde conductionover “fast” and “slow” pathways in patients with co-existent commonand uncommon atrioventricular nodal reentry: isn’t the atrium anecessary part of reentrant circuit? [Abstr]. Circulation 1991; 84:235

7. Akhtar M, Tchou PJ, Jazayeri M. Mechanisms of clinical tachycardias.Am J Cardiol 1988; 61: 9A–19A

8. Akhtar M, Mahmud R, Tchou P, Denker S, Gilbert CJ. Normalelectrophysiologic responses of the human heart. Cardiol Clin 1986;4: 365–386

9. Josephson ME, Horowitz LN, Spielman SR, Waxman HL, GreenspanAM. Role of catheter mapping in the preoperative evaluation ofventricular tachycardia. Am J Cardiol 1982; 49: 207–220

10. Saoudi N, Atallah G, Kirkorian G, Touboul P. Catheter ablation of theatrial myocardium in human type I atrial flutter. Circulation 1990;81: 762–771

11. Schalij MJ, van Rugge FP, Siezenga M, van der Velde ET. Endocardialactivation mapping of ventricular tachycardia in patients: firstapplication of a 32-site bipolar mapping electrode catheter. Circulation1998; 98: 2168–2179

12. Thomas JM, Novis M, Batra R, Mohan JC, Kaul UA, Arora R. Atrialelectrogram patterns in human chronic atrial fibrillation—an in vivomultisite mapping study [Abstr]. Indian Heart J 1998; 50: 660

13. Ben-Haim SA, Osadchy D, Schuster I, Gepstein L, Hayam G, JosephsonME. Nonfluoroscopic, in vivo navigation and mapping technology. NatMed 1996; 2: 1393–1395

14. Gepstein L, Hayam G, Ben-Haim SA. A novel method fornonfluoroscopic catheter-based electroanatomical mapping of theheart: in vitro and in vivo accuracy results. Circulation 1997; 95:1611–1622

15. Shpun S, Gepstein L, Hayam G, Ben-Haim SA. Guidance ofradiofrequency endocardial ablation with real-time three-dimensional magnetic navigation system. Circulation 1997; 96:2016–2021

16. Smeets JL, Ben-Haim SA, Rodriguez LM, Timmermans C, Wellens HJ.New method for nonfluoroscopic endocardial mapping in humans:accuracy assessment and first clinical results. Circulation 1998; 97:2426–2432

17. Gepstein L, Evans SJ. Electroanatomical mapping of the heart: basicconcepts and implications for treatment of cardiac arrhythmias.Pacing Clin Electrophysiol 1998; 21: 1268–1278

18. The CARTO physician’s manual. Cordis Webster, Johnson and Johnson.19. Kottkamp H, Hindricks G, Breithardt G, Borggrefe M. Three-

dimensional electromagnetic catheter technology: electroanatomicalmapping of the right atrium and ablation of ectopic atrial tachycardia.J Cardiovasc Electrophysiol 1997; 8: 1332–1337

20. Saxena A, Dhala AA, Akhtar M, Jazayeri M, Blanck Z, Deshpande S,et al. Electroanatomical mapping of cardiac arrhythmias. A newapproach to catheter ablation. Indian Heart J 1998; 50: 557–564

21. Nakagawa H, Lazzara R, Khastgir T, Beckman KJ, McClelland JH, ImaiS, et al. Role of the triscuspid annulus and the eustachian valve/ridge

IHJ-009-01.p65 10/17/01, 12:13 PM442


on atrial flutter: relevance to catheter ablation of the septal isthmusand a new technique for rapid identification of ablation success.Circulation 1996; 94: 407–424

22. Poty H, Saoudi N, Abdel Aziz A, Nair M, Letac B. Radiofrequencycatheter ablation of type I atrial flutter: prediction of late success byelectrophysiological criteria. Circulation 1995; 92: 1389–1392

23. Shah DC, Jais P, Haissaguerre M, Chouairi S, Takahashi A, Hocini M,et al. Three-dimensional mapping of the common atrial flutter circuitin the right atrium. Circulation 1997; 96: 3904–3912

24. Sra J, Bhatia A, Dhala A, Blanck Z, Rathod S, Boveja B, et al.Electroanatomic mapping to identify breakthrough sites in recurrenttypical human flutter. Pacing Clin Electrophysiol 2000; 23: 1479–1492

25. Willems S, Weiss C, Ventura R, Ruppel R, Risius T, Hoffmann M, et al.Catheter ablation of atrial flutter guided by electroanatomic mapping(CARTO): a randomized comparison to the conventional approach. JCardiovasc Electrophysiol 2000; 11: 1223–1230

26. Iesaka Y, Takahashi A, Goya M, Fujiwara H, Nitta J, Nogami A, et al.Nonlinear ablation targeting an isthmus of critically slow conductiondetected by high density electroanatomical mapping for atypical atrialflutter. Pacing Clin Electrophysiol 2000; 23: 1911–1915

27. Marchlinski F, Callans D, Gottlieb C, Rodriguez E, Coyne R, KleinmanD. Magnetic electroanatomical mapping for ablation of focal atrialtachycardias. Pacing Clin Electrophysiol 1998; 21: 1621–1635

28. Dorostkar PC, Cheng J, Scheinman MM. Electroanatomical mappingand ablation of the substrate supporting intraatrial reentranttachycardia after palliation for complex congenital heart disease.Pacing Clin Electrophysiol 1998; 21: 1810–1819

29. Hebe J, Antz M, Ouyang F, Volkmer M. Mapping and ablation ofincisional atrial reentry tachycardias in patients after surgery forcongenital heart disease using a 3D electroanatomical mappingsystem [Abstr]. Circulation 1998; 97 (Suppl): 1–616

30. Nakagawa H, Jackman WM. Use of a 3-dimensionalelectroanatomical mapping system for catheter ablation ofmacroreentrant right atrial tachycardia following atriotomy. JElectrocardiol 1999; 32: 16–21

31. Natale A, Breeding L, Tomassoni G, Rajkovich K, Richey M, BeheiryS, et al. Ablation of right and left ectopic atrial tachycardias using athree-dimensional nonfluoroscopic mapping system. Am J Cardiol1998; 82: 989–992

32. Reithmann C, Hoffmann E, Dorwarth U, Remp T, Steinbeck G.Electroanatomical mapping for visualization of atrial activation inpatients with incisional atrial tachycardias. Eur Heart J 2001; 22:237–246

33. Nademanee K, Kosar EM. A nonfluoroscopic catheter-based mappingtechnique to ablate focal ventricular tachycardia. Pacing ClinElectrophysiol 1998; 21: 1442–1447

34. Stevenson WG, Delacretaz E, Friedman PL, Ellison KE. Identificationand ablation of macroreentrant ventricular tachycardia with CARTOelectroanatomical mapping system. Pacing Clin Electrophysiol 1998;21: 1448–1456

35. Tomassoni G, Stanton M, Richey M, Leonelli FM, Beheiry S, NataleA. Epicardial mapping and radiofrequency catheter ablation ofischemic ventricular tachycardia using a three-dimensionalnonfluoroscopic mapping system. J Cardiovasc Electrophysiol 1999;10: 1643–1648

36. Cooke PA, Wilber DJ. Radiofrequency catheter ablation ofatrioventricular nodal reentry tachycardia utilizing nonfluoroscopicelectroanatomical mapping. Pacing Clin Electrophysiol 1998; 21:1802–1809

37. Leonelli F, Richey M, Beheiry S, Rajkovich K, Natale A. Tridimensionalmapping-guided modification of the sinus node. J CardiovascElectrophysiol 1998; 9: 1214–1217

38. Worley SJ. Use of a real-time three-dimensional magnetic navigationsystem for radiofrequency ablation of accessory pathways. Pacing Clin

Electrophysiol 1998; 21: 1636–164539. Sra J, Bhatia A, Dhala A, Blanck Z, Deshpande S, Cooley R, et al.

Electroanatomical guided catheter ablation of ventriculartachycardias causing multiple defibrillator shocks. Pacing ClinElectrophysiol 2001 (In Press)

40. Taccardi B, Arisi G, Macchi E, Baruffi S, Spaggiari S. A newintracavitary proble for detecting the site of origin of ectopicventricular beats during one cardiac cycle. Circulation 1987; 75: 272–281

41. Schilling RJ, Peters NS, Davies DW. Simultaneous endocardialmapping in the human left ventricle using a noncontact catheter:comparison of contact and reconstructed electrograms during sinusrhythm. Circulation 1998; 98: 887–898

42. Schilling RJ, Friedman PA, Stanton MS, Beatty G, Voth E, Hauck J.Mathematical reconstruction of endocardial potentials with non-contact multi-electrode array. Field Clinical Training Manual.Endocardial Solutions, Inc.

43. Schilling RJ, Peters NS, Davies DW. Feasibility of a noncontact catheterfor endocardial mapping of human ventricular tachycardia.Circulation 1999; 99: 2543–2552

44. Gornick CC, Adler SW, Pederson B, Hauck J, Budd J, Schweitzer J.Validation of a new noncontact catheter system for electroanatomicalmapping of left ventricular endocardium. Circulation 1999; 99: 829–835

45. Strickberger SA, Knight BP, Michaud GF, Pelosi F, Morady F. Mappingand ablation of ventricular tachycardia guided by virtualelectrograms using a noncontact, computerized mapping system. JAm Coll Cardiol 2000; 35: 414–421

46. Kadish A, Hauck J, Pederson B, Beatty G, Gornick C. Mapping of atrialactivation with a noncontact, multielectrode catheter in dogs.Circulation 1999; 99: 1906–1913

47. Sra J, Bhatia A, Krum D, Akhtar M. Endocardial noncontact activationmapping of idiopathic left ventricular tachycardia. J CardiovascElectrophysiol 2000; 11: 1409–1412

48. Sra J, Bhatia A, Krum D, Akhtar M. Noncontact mapping forradiofrequency ablation of complex cardiac arrhythmias. J IntervenCardiac Electrophysiol 2001; 5: 323–331

49. Khoury DS, Rudy Y. Reconstruction of endocardial potentials fromintracavitary probe potentials: a model study. Proc Comput CardiolDurham, NC, Oct 11–14, 1992; 9–12

50. Beatty GE, Remole SC, Hansen R, Johnston MD, Holte JE, Benditt DG.Noncontact electrical extrapolation technique to reconstructendocardial potentials [Abstr]. Pacing Clin Electrophysiol 1994; 17:765

51. Derfus DL, Pilkington TC. Assessing the effect of uncertainty inintracavitary electrode position on endocardial potential estimates.IEEE Trans Biomed Eng 1992; 39: 676–681

52. Peters NS, Jackman WM, Schilling RJ, Beatty G, Davies DW. Humanleft ventricular endocardial activation mapping using a novel non-contact catheter. Circulation 1997; 95: 1658–1660

53. Khoury DS, Taccardi B, Lux RL, Ershler PR, Rudy Y. Reconstructionof endocardial potentials and activation sequences from intracavitaryprobe measurements. Localization of pacing sites and effects ofmyocardial structure. Circulation 1995; 91: 845–863

54. Khoury DS, Rudy Y. A model study of volume conductor effects onendocardial and intracavitary potential. Circ Res 1992; 71: 511–525

55. Messinger-Rapport BJ, Rudy Y. Regularization of the inverse problemin electrocardiography: a model study. Math Biosci 1988; 89: 79–118

56. Tikhonov AN, Arsenin VY. Solutions of II-Posed Problems. WashingtonDC: VH Winston and Sons; 1997. pp. 27–94

57. Kalman JM, Olgin JE, Saxon LA, Lee RJ, Scheinman MM, Lesh MD.Electrocardiographic and electrophysiologic characterization ofatypical atrial flutter in man: use of activation and entrainmentmapping and implications for catheter ablation. J CardiovascElectrophysiol 1997; 8: 121–144

IHJ-009-01.p65 10/17/01, 12:13 PM443


58. Kall JG, Rubenstein DS, Kopp DE, Burke MC, Verdino RJ, Lin AC, et al.Atypical atrial flutter originating in the right atrial free wall.Circulation 2000; 101: 270–279

59. Zipes DP, Knope RF. Electrical properties of the thoracic veins. Am JCardiol 1972; 29: 372–376

60. Haïssaguerre M, Jaïs P, Shah DC, Takahashi A, Hocini M, Quiniou G,et al. Spontaneous initiation of atrial fibrillation by ectopic beatsoriginating in the pulmonary veins. N Engl J Med 1998; 339: 659–666

61. Lin WS, Prakash VS, Tai CT, Hsieh MH, Tsai CF, Yu WC, et al.Pulmonary vein morphology in patients with paroxysmal atrialfibrillation initiated by ectopic beats originating from the pulmonaryveins: implications for catheter ablation. Circulation 2000; 101:1274–1281

62. Haïssaguerre M, Jaïs P, Shah DC, Garrigue S, Takahashi A, LavergneT, et al. Electrophysiological endpoint for catheter ablation of atrialfibrillation initiated from multiple pulmonary venous foci. Circulation2000; 101: 1409–1417

63. Natale A, Breeding L, Tomassoni G, Rajkovich K, Richey M, BeheiryS, et al. Ablation of right and left ectopic atrial tachycardias using athree-dimensional nonflouroscopic mapping system. Am Heart J1998; 82: 989–992

64. Stevenson WG, Khan H, Sager P, Saxon LA, Middlekauff HR,Natterson PD, et al. Identification of reentry circuit sites duringcatheter mapping and radiofrequency ablation of ventriculartachycardia late after myocardial infarction. Circulation 1993; 88:1674–1670

65. El- Shalakany A, Hadjis T, Papageorgiou P, Monahan K, Epstein L,Josephson ME. Entrainment/mapping criteria for the prediction oftermination of ventricular tachycardia by single radiofrequencylesion in patients with coronary artery disease. Circulation 1999; 99:2283–2289

66. Stevenson WG, Sager PT, Natterson PD, Saxon LA, Middlekauff HR,Wiener I. Relation of pace mapping QRS configuration andconduction delay due to ventricular tachycardia reentry circuits in

human infarct scars. J Am Coll Cardiol 1995; 26: 481–48867. Zipes DP, Foster PR, Troup PJ, Pedersen DH. Atrial induction of

ventricular tachycardia. Reentry versus triggered automaticity. Am JCardiol 1979; 44: 1–8

68. Belhassen B, Rotmensch HH, Laniado S. Response of recurrentsustained ventricular tachycardia to verapamil. Br Heart J 1981; 46:679–682

69. Lin FC, Finley CD, Rahimtoola SH, Wu D. Idiopathic paroxysmalventricular tachycardia with QRS pattern of right bundle branchblock and left axis deviation. A unique clinical entity with specificproperties. Am J Cardiol 1983; 52: 95–100

70. Okumura K, Matsuyama K, Miyagi H, Tsuchiya T, Yasue H.Entrainment of idiopathic ventricular tachycardia of left ventricularorigin with evidence for reentry with an area of slow conduction andeffect of verapamil. Am J Cardiol 1998; 62: 727–732

71. Lerman BB, Stein KM, Markowitz SM, Mittal S, Slotwiner DJ.Ventricular tachycardia in patients with structurally normal hearts.In: Zipes DP, Jalife J (eds). Cardiac electrophysiology from cell to bedside.3rd ed. Philadelphia: W.B. Saunders Co.; 2000. pp.640–656

72. Nakagawa H, Beckman KJ, McClelland JH, Wang X, Arruda M,Santoro I, et al. Radiofrequency catheter ablation of idiopathic leftventricular tachycardia guided by a Purkinje potential. Circulation1993; 88: 2607–2611

73. Kottkamp H, Chen X, Hindricks G, Willems S, Haverkamp W, WichterT, et al. Idiopathic left ventricular tachycardia: new insights intoelectrophysiological characteristics and radiofrequency catheterablation. Pacing Clin Electrophysiol 1995; 18: 1285–1297

74. Okumura K, Yamabe H, Tsuchiya TS, Tabuchi T, Iwasa A, Yasue H.Characteristic of slow conduction zone demonstrated duringentrainment of idiopathic ventricular tachycardia of left ventricularorigin. Am J Cardiol 1996; 77: 379–383

75. Ohe T, Shimomura K, Aihara N, Kamakura S, Matsuhisa M, Sato I,et al. Idiopathic sustained left ventricular tachycardia: clinical andelectrophysiologic characteristics. Circulation 1988; 77: 560–568

IHJ-009-01.p65 10/17/01, 12:13 PM444

Indian Heart J 2001; 53: 445–450 Al-Mubarak et al. Carotid Artery Stenting 445

Carotid Artery Stenting: Current Status andFuture Prospects

Nadim Al-Mubarak, Gary S Roubin, Jiri J Vitek, Gishel New, Sriram S IyerThe Lenox Hill Heart and Vascular Institute of New York, New York, USA

Carotid artery stenting (CAS) has emerged as anendovascular treatment alternative to carotid

endarterectomy (CEA) for the management ofatherosclerotic obstructive extracranial carotid arterydisease. Evidence is accumulating in support of its efficacyin preventing stroke.1–3 This treatment modality is beingincreasingly utilized to treat the high-risk surgicalpopulation and randomized trials have been undertaken tocompare its efficacy with that of CEA.4–5 More suitableequipment and distal protection devices currently underevaluation are expected to enhance the procedural safetyand clinical outcomes of this treatment. This articlediscusses the current indications, patient selection andfuture prospects of CAS.

Surgical Background

The clinical significance of an endovascular approach toocclusive disease of the carotid artery is best understood inthe context of its surgical counterpart, CEA. Based on threelandmark randomized trials comparing CEA with medicaltherapy, the former is considered to be the standardtreatment for both symptomatic and asymptomaticextracranial carotid artery stenoses.6–9 However, theappropriate clinical indication for CEA requires carefulanalysis of the population studied in these trials and theresults obtained. The North American Symptomatic CarotidEndarterectomy Trial (NASCET) yielded the largest andmost rigorously collected data with a neurological overviewand provides a unique insight into the outcome of CEAwhen performed in centers of excellence.6 In this study, CEAcombined with the best medical therapy was clearly superiorto medical therapy alone in reducing the risk of stroke in aselect group of patients with symptomatic extracranialcarotid artery stenoses of >50%. This benefit was directlyrelated to the severity of stenosis and was greatest forpatients with stenoses of >90%.9 The overall combined30-day rate of stroke and death in the CEA group was 6.7%.

Indian Heart J 2001; 53: 445–450 Review Article

The second important study that examined the benefitof CEA in symptomatic patients was the European CarotidSurgery Trial (ECST)8 which also demonstrated a clearbenefit of surgery over medical therapy in reducing the riskof future stroke. The overall 30-day incidence of stroke anddeath in the surgical arm was similar to that seen in NASCETbut was higher in women (10.6%), in patients with a systolicblood pressure >180 mmHg (12.3%) and in the presenceof peripheral vascular disease (12.3%). The AsymptomaticCarotid Atherosclerosis Study (ACAS)7 also showed a clearbenefit of CEA plus medical therapy in patients withasymptomatic extracranial carotid artery stenosis of >60%.In this study, the estimated 30-day perioperative risk ofstroke or death was 2.7%. The population in this study wasat low risk for CEA, as patients with important commonco-morbid conditions were excluded. In all these trials, CEAwas complicated by cranial nerve palsies in approximately7% of the patients, and in 13% there were a variety of localwound problems or medical complications includingcongestive heart failure, myocardial infarction and cardiacarrhythmias. These operative complications are importantsince they are negligible in patients undergoing CAS.

Despite the benefit observed in these trials, their resultscannot be extrapolated to the general population, largelybecause the risk of CEA is likely to be significantly higher inthe excluded population. These include patients withcommon conditions such as prior ipsilateral CEA, significantcoexisting coronary artery disease, as well as those withsignificant renal, hepatic or pulmonary co-morbidities.Numerous observational studies have reported increasedrates of perioperative stroke and death in these groups.10–14

In addition, patients who underwent CEA in the presenceof an occluded contralateral carotid artery had highperioperative combined stroke and death rates (14%).6

Therefore, the perceived benefit of CEA in these patients maynot be the same as that seen in the randomized population.

Furthermore, these trials were conducted by pre-selectedhigh-volume surgeons who qualified for participation onlyif they demonstrated low perioperative stroke or deathrates.6 Recent data, however, have shown definitive evidencethat the actual incidence of stroke and death from CEA in

Correspondence: Dr Gary S Roubin, 130 East 77th Street, 9th floor BlackHall, New York, NY 10021, USA. e-mail: [email protected]

Revie Article-01.p65 10/17/01, 4:01 PM445

446 Al-Mubarak et al. Carotid Artery Stenting Indian Heart J 2001; 53: 445–450

the community is much higher than that reported inrandomized trials.15–17 Data from Medicare mortalitystatistics also showed that in-hospital mortality (not themore rigorous 30-day neurological assessment) wassignificantly higher in low-volume CEA centers thanin high-volume ones, and was higher in the latterthan in NASCET/ACAS centers (2.5%, 1.9% and 1.4%,respectively).18 Approximately 80% of CEAs in the UnitedStates are performed in low-volume hospitals by operatorscarrying out less than 20 procedures annually.15 Moreover,published CEA outcomes are greatly influenced by themanner in which they are reported; specifically, stroke ratesreported by surgeons were lower (2.2%) than those reportedwhen neurologists were also involved in the authorship(7.7%).19–20

Accepting these limitations, the American HeartAssociation has set guidelines for the performance of CEA.According to these, CEA should be performed only if thecombined stroke and death rates can be kept at <6% insymptomatic and <3% in asymptomatic patients withsevere extracranial carotid artery stenoses.21

Rationale for an Endovascular Approach

The documentation of a significant incidence of bothneurological and non-neurological complicationsassociated with CEA in landmark studies emphasizes theneed to pursue an alternative and safer method for treatingcarotid bifurcation disease, especially in a subset of patientsthought to be at high risk from CEA.

Carotid artery stenting is an endovascular, less invasivetreatment approach which offers several advantages. Themajority of CEAs are still carried out under generalanesthesia with recognized attendant risks, especially fromfrequently coexisting coronary artery disease or heartfailure. In addition, neurological complications are onlyapparent following recovery from anesthesia. In contrast,stenting is performed in the nonsedated conscious patient;under these conditions, neurological changes areimmediately recognized and diagnostic intracranialangiography and appropriate therapeutic measures can bepromptly initiated.

Carotid endarterectomy is difficult in patients with highcarotid artery stenoses or bifurcations, those with proximal/ostial common carotid artery (CCA) lesions and short, obesenecks. In these situations, extensive exposure of the carotidartery is required, which adds considerable risk to theprocedure. Patients with prior irradiation to the neck forhead and neck cancers and prior radical neck dissection alsopresent a challenge for the surgeon due to the unusual

location of the lesions and extensive fibrosis in and aroundthe arterial wall. Similarly, patients with restenosis afterprior CEA are sometimes at higher risk for repeat CEA dueto scarring. These conditions usually require generalanesthesia and some may require mobilization of themandible. It is also more difficult for the surgeon to exposethe artery for shunt placement. Fibrosis and scarringaround the artery require extensive dissection, increasingthe risk for cranial nerve injuries and lead to delayed woundhealing. In some patients, CEA cannot be performed andan interposition graft might be required. Generally, noneof the above conditions are a problem for the endovascularapproach and, accordingly, CAS may clearly be preferablein these patients.22–24 In addition, stenting offers thepotential advantages of a shorter hospital stay, shorterrecovery period and reduced cost.

Patient Selection

The current indications for CAS are predominantlydetermined by the operator’s experience and results. Ingeneral, there should be no contraindication to proceed withstenting, provided the procedure can be performed with acombined stroke and death rate of <6% in symptomatic and<3% in asymptomatic patients.25 Similarly, based on thecurrent knowledge of surgical outcome of carotid arterydisease, stenting should only be carried out in lesions inwhich stenosis is >50% in symptomatic and >70% inasymptomatic patients (using NASCET angiographicmeasurements).25 The risk/benefit ratio of the procedurefor an individual patient should be established prior to theintervention. Based on our learning curve experience andcurrent technology, several factors associated withincreased or decreased risk of procedural events have beenidentified (Table 1). However, as technology improves,particularly with the application of cerebral protectiondevices, these factors may need to be revised.

During the initial learning curve of the operator, casesassociated with higher periprocedural risk should be avoidedand those with low procedural risk selected. In thesepatients, especially those below the age of 80 years, stentingis associated with a low rate of periprocedural events(Table 2). In higher-risk subsets, much greater experienceis required to achieve similar results. Our current practiceis to treat high-risk patients, especially the elderly, with distalprotection. Although at the beginning of the learningexperience there is a tendency to accept high-risk patientsfor CEA, this temptation should be avoided if the patienthas one or more of the higher-risk descriptors listed inTable 1. A number of situations that pose a high risk for



CEA may, however, be at low risk for stenting and representideal indications, such as restenosis after prior CEA orstenoses in those with prior neck irradiation and/or radicalneck dissection.22–24 Patients with discrete proximal or ostialCCA lesions, discrete lesions in the distal internal carotidartery (ICA) or lesions involving high bifurcations are alsoconsidered ideal candidates for stenting.

It must be emphasized that unfavorable lesioncharacteristics may only be apparent following initialangiography. The operator should be prepared to abandonthe procedure at this point and consider elective CEA ordecide whether to continue medical management. Thejudgement of the operator is critical in achieving a lowcomplication rate. Since there are usually reasonablealternative therapies, it must be emphasized that failure tocomplete the procedure is acceptable but an avoidablecomplication is not.

Clinical Results

Over the past 5 years, we challenged this new technique bytreating a skewed, high surgical-risk population which hadbeen excluded from prior randomized CEA trials.1,26 A largenumber of these patients were referred by physicians,including vascular surgeons, because of a variety ofconditions that put them at higher risk for CEA. Theseconditions included: prior ipsilateral CEA, prior ipsilateralneck irradiation, contralateral occlusion and severecoexisting coronary artery disease requiring staged/simultaneous carotid and coronary artery intervention. Inaddition, a variety of distally located lesions, highbifurcations and proximal CCA disease were treated. Allvarieties of lesions were attempted, including severelyulcerated or calcified lesions, long severe lesions, “string signlesions” and completely occluded vessels. Similarly, all typesof severely diseased aortic arch vessels, including tortuous,calcified, atherosclerotic and stenotic CCAs were accessedto complete the procedure. In general, our patients belongedto an older age group (mean age: 69±10 years) who sufferedfrom a number of co-morbidities including coronary arterydisease, hypertension and diabetes mellitus.

From September 1994 through May 2001 we completedstenting of 935 hemispheres in 847 consecutive patients.The overall 30-day all-stroke and death rate was 7%. The30-day mortality was 1.3%, and major and minor disablingstroke rates were 0.9% and 4.7%, respectively. Thetechnique has gained wide acceptance at present becausemajor adverse events have been uncommon and minorstrokes (usually noted only during detailed examination bythe neurologist) have not been of functional significance tothe patient. Neurological events continue to be mostlynondisabling strokes. Deaths and disabling strokes havebeen uncommon, especially in patients <80 years. Whilethe rate of embolic events in patients below the age of 80years has been very low, higher rates of neurologicalcomplications were seen in patients >80 years. Over our5-year experience, we have modified the eligibility criteria.

Table 1. Risk status associated with carotid stenting

Patients at increased risk for neurological complications

ClinicalAdvanced age (≥80 years)Prior major disabling stroke, or cerebral atrophy/dementiaUnstable neurological symptoms (recent TIA or stroke)

AnatomicalSeverely tortuous, calcified and atherosclerotic aortic arch/

arch vesselsSevere tortuousity just distal to the bifurcationCoexisting proximal common carotid artery lesionTotal occlusion or long sub-total occlusions—“string sign

lesions”Severe concentric calcificationAngiographic evidence of a large thrombus

Patients at lower risk for embolic events

ClinicalAge ≤80 yearsLess severe stenosis

AnatomicalStraight, noncalcified, “smooth arch vessels”Nontortuous bifurcationAbsence of common carotid artery disease (except at

adjacent bifurcation)Absence of thrombusAbsence of kinks, loops, bend-points at lesion siteShort lesionsPrior CEA

CEA: carotid endarterectomy; TIA: transient ischemic attack

Table 2. Lenox Hill experience (November 1997–May 2001)

<80 years >80 years

Asymptomatic patientsPatients/hemispheres 305/326 59/64

Minor stroke 8 (2.5) 4 (6.2)Major stroke 1 (0.3) 2 (3.1)Deaths 1 (0.4) 0

Symptomatic patientsPatients/hemispheres 149/160 21/22

Minor stroke 5 (3.1) 2 (9.1)Major strokes 0 1 (4.5)Death 1 (0.6) 0

Values in parentheses are percentages



We now strongly believe that given the availability ofcurrent technology and taking into account the experienceof the operator, careful patient selection is critical formaintaining low complication rates. Our recent experienceat the Lenox Hill Hospital is summarized in Table 2.

Our results have been reproduced by several groups.1–4

Wholey et al.,2 in a global registry that includes theoutcomes of 5210 cases of CAS from 24 centers aroundthe world, reported a technical success rate of 98.4%. Theoverall 30-day event rate (all-strokes/deaths) was 5.07%.The majority of these patients were treated without theadvantage of distal protection.

Stent restenosis: Of the first 225 successfully stentedpatients, follow-up carotid angiography was performed in121, and carotid ultrasound in 29 patients. Of these,restenosis (defined as 50% diameter narrowing) developedin 8 (5.3%) patients. Only 4 of these patients (2.7%) hadrestenosis which warranted repeat intervention.1 Restenosisof stents in the carotid artery can be easily treated withrepeat balloon dilatation. Similar rates for restenosis(6-month restenosis rate of 2%) were reported by Wholeyet al.2 in the global registry. Furthermore, the late (>1 year)luminal loss following CAS has been shown to be favorable.27

Late outcome: Systematic follow-up of 528 patients(99.6%) with 604 successfully stented vessels at a mean of17±12 months (range 12–55 months) was obtained.26

There was an incremental 3.2% incidence of fatal andnonfatal stroke beyond 30 days. Freedom from fatal andnonfatal stroke at 3 years was 88%±2% and freedom fromipsilateral nonfatal stroke and all-fatal stroke was 92%±1%.These results compare very favorably with those seen in theNASCET trial.

Future Directions

It is now evident that experienced operators can performCAS with low complication rates.1–4 As with CEA, pooroutcomes from CAS may result from poor technical skillsand experience. In our view, the most successful results canbe attained through collegial collaboration between thevarious vascular disciplines. There is an urgent need to trainendovascular specialists who wish to perform thisprocedure. The coming years should be focused oneducation and technical training through symposia,workshops and live demonstration courses. There is also anongoing need to improve the equipment available for CAS.Often, technical problems are attributed to the inadequacyof currently available devices. There is a need for lower-profile stent delivery systems, better access sheaths and

specially designed guidewires and balloons. It is likely thata variety of different stent designs will be required foroptimal treatment of anatomical variations of thebifurcation of the carotid artery.

Despite optimal antiplatelet therapy, some patientsdevelop neurological events. These invariably occur duringthe procedure, occasionally within 1–2 hours and veryrarely within the 4 weeks following the procedure.Obstructive carotid artery lesions are known to containfriable thrombotic and atherosclerotic components whichcan embolize during intervention and are responsible forthe majority of the neurological events during CAS. Thishas been demonstrated in an ex vivo human carotid arterymodel by Ohki et al.28 as well as by several transcranialDoppler studies during CAS.29 Embolization can occur fromaggressive guidewire manipulation, balloon dilatation(particularly larger peripheral balloons), during stentdeployment and post-stent dilatation.30 The majorchallenge from a technical perspective has been thedevelopment of “distal protection devices” (Fig. 1) whichreliably prevent any particulate matter released during CASfrom embolizing to the brain. A number of “distal protection”strategies are currently being evaluated for their efficacy inminimizing the risk of embolic neurological events.31 Threeapproaches are under investigation, all of which providedistal protection during the three critical stages of the carotidintervention: predilation, stent deployment andpostdilatation. Each of these strategies has its inherentadvantages and limitations. One approach, first proposed byVitek et al.32 and later pioneered by Jacques Théron,33–34

involves the use of a distal occlusion balloon which interruptsflow during critical maneuvers likely to release emboli. Thecolumn of blood containing embolic material is thenaspirated prior to deflating the balloon. Théron reported animpressive reduction of embolic complications from 28% to2% using this technique.33–34 Théron’s concept has now beenrefined with more advanced technology and similar favorableoutcomes have been reported.35 The second approachinvolves reversing the flow within the ICA. This is achievedby either occluding the CCA and diverting the blood into theexternal carotid artery (ECA)—Kachel technique—or bysimultaneously occluding the ipsilateral CCA and the ECA,diverting the blood into the guiding catheter which isexternally connected to the contralateral femoral vein (Paroditechnique).36 While these two approaches, especially thelatter, are advantageous in tortuous arteries, they may notbe suitable for the 5%–10% of patients with poorcontralateral circulation such as incomplete circle of Willis,who do not tolerate prolonged carotid artery occlusion orflow reversal. The third approach involves the deployment



of an atraumatic embolic filter which is placed prior todefinitive lesion dilatation and removed after completion ofstent positioning and expansion. The filter has the advantageof providing constant cerebral perfusion, thereby allowingmore time for careful and precise intervention of the lesion.Currently available devices are deployed and retrieved on a0.014" or 0.018" shaft which serves as the guidewire for the

balloons and stent delivery system. This approach is limitedby the high profile of the currently available filter systemswhich are difficult to use if the vessel is tortuous. Preliminarydata on filter protection are also encouraging,36–38 and largeclinical trials are currently in progress. Cerebral protectiondevices have the potential to greatly enhance the safety ofCAS and, when available, they will extend this specialized,less invasive approach to the treatment of carotid arterystenoses to community-based interventionalists.

For CAS to attain acceptance as a less traumatic, safeand effective alternative to CEA, it must be validated in arandomized controlled trial. One such trial—the Carotidand Vertebral Artery Transluminal Angioplasty Study(CAVATAS)—has been completed and a second larger NIH-sponsored study is planned—Carotid RevascularizationEndarterectomy Versus Stent Trial (CREST).5,39 TheCAVATAS, a prospective randomized controlled trial, wasconducted in Great Britain through a collaborative effortbetween neurologists, radiologists and vascular surgeons.39

In general, the trial included a high-risk, symptomaticpopulation of patients with high-grade carotid arterystenosis. Inclusion criteria were much broader than thoseof the NASCET trial and, accordingly, this study representsthe first prospective evaluation of CEA in higher-riskpatients with independent neurological assessment. Thestudy was undertaken at large regional centers in Britainby experienced vascular surgeons. In contrast, theradiologists—stent operators—involved in the trial wereoperating within their learning curves for carotidintervention. Approximately 20% of the patients actuallyreceived stents (usually for “bail-out” application). Inaddition, the stenting approach in this trial was suboptimal,inferior stents and large-profile peripheral balloons wereused, and the technical approach was employed without aguiding sheath using 0.035" wires. Despite these differencesin operator experience and the relative maturity of thetechniques, both early and late outcomes were similar. The30-day combined rates of major stroke and death wereapproximately 6% in both groups and the 2-yearneurological outcomes were identical.

The CREST, a randomized controlled trial of CEA versusCAS, plans to recruit 2500 patients with symptomaticstenosis (>50% diameter narrowing, NASCET angiographiccriteria). Primary end-points will be the incidence of death,any stroke and myocardial infarction at 30 days, and theincidence of ipsilateral stroke at 4 years after the procedure.The credentials of both surgeons and endovascularinterventionists are being rigorously evaluated to ensurethat the trial is conducted according to the higheststandards.40

Fig. 1. Distal protection devices: (A) the distal-balloon protection: GuardWire®

(Percusurge Inc.); (B) the NeuroShield filter protection system (MedNova Ltd.);(C) the ACCUNET filter protection system (Guidant Inc.).



References

1. Yadav JS, Roubin GS, Iyer S, Vitek J, King P, Jordan WD, et al. Electivestenting of the extracranial carotid arteries. Circulation 1997; 95:376–381

2. Wholey MH, Wholey M, Mathias K, Roubin GS, Diethrich EB, HenryM, et al. Global experience in cervical carotid artery stent placement.Catheter Cardiovasc Interv 2000; 50: 160–167

3. Shawl F, Kadro W, Domanski MJ, Lapetina FL, Iqbal AA, DoughertyKG, et al. Safety and efficacy of elective carotid artery stenting in high-risk patients. J Am Coll Cardiol 2000; 35: 1721–1728

4. Shawl FA. Carotid stenting in patients with symptomatic coronaryartery disease: a preferred approach. J Invasive Cardiol 1998; 10: 432–442

5. Hobson RW. CREST (Carotid Revascularization Endarterectomyversus Stent Trial): background, design and current status. SeminVasc Surg 2000; 13: 139–143

6. North American Symptomatic Carotid Endarterectomy Trial.Methods, patient characteristics, and progress. Stroke 1991; 22: 711–720

7. Executive Committee for the Asymptomatic Carotid AtherosclerosisStudy. Endarterectomy for asymptomatic carotid artery stenosis.JAMA 1995; 273: 1421–1428

8. European Carotid Surgery Trialists’ Collaborative Group. MRCEuropean Carotid Surgery Trial: interim results for symptomaticpatients with severe (70–99%) or with mild (0–29%) carotid stenosis.Lancet 1991; 337: 1235–1243

9. North American Symptomatic Carotid Endarterectomy TrialCollaborators. Beneficial effect of carotid endarterectomy insymptomatic patients with high-grade carotid stenosis. N Engl J Med1991; 325: 445–453

10. Akins CW. Combined carotid endarterectomy and coronaryrevascularization operation. Ann Thorac Surg 1998; 66: 1483–1484

11. Ballotta E, Da Giau G, Guerra M. Carotid endarterectomy andcontralateral internal carotid artery occlusion: perioperative risks andlong-term stroke and survival rates. Surgery 1998; 123: 234–240

12. Bartlett FF, Rapp JH, Goldstone J, Ehrenfeld WK, Stoney RJ. Recurrentcarotid stenosis: operative strategy and late results. J Vasc Surg 1987;5: 452–456

13. Bass A, Krupski WC, Dilley RB, Bernstein EF. Combined carotidendarterectomy and coronary artery revascularization: a soberingreview. Isr J Med Sci 1992; 28: 27–32

14. Borger MA, Fremes SE, Weisel RD, Cohen G, Rao V, Lindsay TF, et al.Coronary bypass and carotid endarterectomy: does a combinedapproach increase risk? A meta-analysis. Ann Thorac Surg 1999; 68:14–20

15. Mayo Asymptomatic Carotid Endarterectomy Study Group. Resultsof a randomized controlled trial of carotid endarterectomy forasymptomatic carotid stenosis. Mayo Clin Proc 1992; 67: 513–518

16. Brott T, Thalinger K. The practice of carotid endarterectomy in a largemetropolitan area. Stroke 1984; 15: 950–955

17. Cebul RD, Snow RJ, Pine R, Hertzer NR, Norris DG. Indications,outcomes, and provider volumes for carotid endarterectomy. JAMA1998; 279: 1282–1287

18. Wennberg DE, Lucas FL, Birkmeyer JD, Bredenberg CE, Fisher ES.Variation in carotid endarterectomy mortality in the Medicarepopulation: trial hospitals, volume, and patient characteristics. JAMA1998; 279: 1278–1281

19. Rothwell PM, Robertson G. Meta-analyses of randomised controlledtrials. Lancet 1997; 350: 1181–1182

20. Rothwell PM, Slattery J, Warlow CP. Clinical and angiographicpredictors of stroke and death from carotid endarterectomy:systematic review. BMJ 1997; 315: 1571–1577

21. Morey SS. AHA updates guidelines for carotid endarterectomy. AmFam Physician 1998; 58: 1898, 1903–1904

22. Al-Mubarak N, Roubin GS, Gomez CR, Liu MW, Iyer SS, Vitek JJ.Carotid stenting for severe radiation-induced extacranial carotidartery occlusive disease. J Endovasc Ther 2000; 7: 36–40

23. Al-Mubarak N, Roubin GS, Liu MW, Dean LS, Gomez CR, Iyer SS, etal. Early results of percutaneous intervention for severe coexistingcarotid and coronary artery disease. Am J Cardiol 1999; 84: 600–602

24. Al-Mubarak N, Roubin GS, Vitek JJ, Gomez CR. Simultaneous bilateralcarotid stenting for restenosis after endarterectomy. Cathet CardiovascDiagn 1998; 45: 11–15

25. American Heart Association. 1999 Heart and Stroke Statistical Update.Dallas, Texas: American Heart Association, 1998

26. Roubin GS, New G, Iyer SS, Vitek JJ, Al-Mubarak N, Liu MW, et al.Immediate and late clinical outcomes of carotid artery stenting insymptomatic and asymptomatic carotid stenosis: a 5–year prospectiveanalysis. Circulation 2001; 103: 532–537

27. Al-Mubarak N, Roubin GS, Liu MW, Iyer SS, Dean LS, Vitek JJ. Lateluminal loss following carotid artery stenting [Abstr]. CathetCardiovasc Interv 1999; 120: 45

28. Ohki T, Marin ML, Lyon RT, Berdejo GL, Soundararajan K, Ohki M, etal. Ex vivo human carotid artery bifurcation stenting: correlation oflesion characteristics with embolic potential. J Vasc Surg 1998; 27:463–471

29. Markus HS, Clifton A, Buckenham T, Browo MM. Carotid angioplasty:detection of embolic signals during and after the procedure. Stroke1994; 25: 2403–2406

30. Al-Mubarak N, Gary SR, Vitek JJ, New G, Iyer SS. Does the distal-occlusion balloon reduce microembolization during carotid stenting?[Abstr]. Circulation 2000; 102: II-475

31. Ohki T, Veith F. Carotid stenting with and without protection devices:should protection be used in all patients? Semin Vasc Surg 2000; 13:144–152

32. Vitek JJ, Raymon BC, Oh SJ. Innominate artery angioplasty. Am JNeuroradiol 1984; 5: 113–114

33. Theron JG. In: Protected angioplasty and stenting of atheroscleroticstenosis at the carotid artery bifurcation. Philadelphia: W. B. SaundersCo.; 1998. pp. 466–473

34. Theron J. Protected carotid angioplasty and carotid stents. J Mal Vasc1996; 21 (Suppl A): 113–122

35. Henry M, Amor M, Klonaris C, Henry I, Masson I, Chati Z, et al.Angioplasty and stenting of extracranial carotid arteries. Tex HeartInst J 2000; 27: 150–158

36. Parodi JC, La Mura R, Ferreira LM, Mendez MV, Cersosimo H,Schoenholtz C, et al. Initial evaluation of carotid angioplasty andstenting using three different cerebral protection devices. J Vasc Surg2000; 32: 1127–1136

37. Reimer B, Corvaja N, Moshiri S, Sacca S, Albeiro R, Di Mario C, et al.Cerebral protection with filter devices during carotid artery stenting.Circulation 2001: 104; 12–15

38. Al-Mubarak N, Colombo A, Iyer SS, Corvaja N, Vitek JJ. Early resultsof carotid artery stenting with a filter protection system [Abstr].J Heart Disease 2001; 2: 55

39. Endovascular versus surgical treatment in patients with carotidstenosis in the Carotid and Vertebral Artery TransluminalAngioplasty Study (CAVATAS): a randomized trial. Lancet 2001; 357:1729–1737

40. Al-Mubarak N, Roubin GS, Hobson RW 2nd, Ferguson R, Brott T,Moore W. Credentialing of stent operators for the CarotidRevascularization Endarterectomy vs. Stenting Trial “CREST”[Abstr]. Stroke 2000; 73: 97


Indian Heart J 2001; 53: 451–457 Kumar et al. Thrombolytic Therapy for Stuck Prosthetic Valves 451

Role of Thrombolytic Therapy for Stuck Prosthetic Valves:A Serial Echocardiographic Study

Sudeep Kumar, Naveen Garg, Satyendra Tewari, Aditya Kapoor, PK Goel, Nakul SinhaDepartment of Cardiology, Sanjay Gandhi PGIMS, Lucknow

Prosthetic valve replacement is commonly performed totreat patients with significant valvular lesions. Most

often, a mechanical prosthesis is used; this is inherentlyassociated with an increased risk of thromboembolism andprosthetic valve thrombosis (PVT). The incidence of PVTand thromboembolism depends on the location and type ofprosthesis used and averages 0.2% and 1.8% per patient-year, respectively.1 The incidence of PVT may be as high as13% in the first year or 20% overall for tricuspid mechanical

prostheses,2 while it is 0.2%–6% per patient-year for left-sided prostheses.3 Redo valve replacement/debridement hasbeen the conventionally advocated treatment, but it isassociated with significant mortality and morbidity, withmortality ranging from 37.5% to 54.5% for emergencycases and 8% to 20.3% for urgent cases.4,5 Thrombolytictherapy (TLT) is an alternative to surgery and is consideredto be the treatment of choice for tricuspid PVT. Recently,TLT has emerged as an alternative to left-sided valvethrombosis also.6–9 In this study, we evaluated the efficacyand safety of TLT in left-sided PVT using serial Dopplerechocardiography.

Background: Thrombotic occlusion of a prosthetic valve continues to be an uncommon but serious complication.Intravenous thrombolytic therapy has been proposed as an alternative to surgical treatment, but only in criticallyill patients.Methods and Results: Forty-one consecutive patients presenting with 48 episodes of prosthetic valve thrombosis(44 mitral and 4 aortic) were treated with thrombolytic therapy under serial echocardiographic guidance. Therewere 14 male and 27 female patients. The anticoagulation status was inadequate in 89.6% of episodes. Atrialfibrillation was present in 47.9% of episodes. The prostheses involved in these episodes were tilting disc in 45,bileaflet in 2, and ball and cage type in 1. The Sorin prosthetic valve was the most commonly involved. The timeinterval between valve replacement and thrombosis ranged from 1 month to 108 months (mean 20.4±20.6months). Patients were in New York Heart Association functional class III in 47.9% and in class II in 43.9% ofepisodes. Thrombolytic agents used were streptokinase and urokinase in 44 and 4 episodes, respectively. Themean duration of thrombolytic therapy was 27.9±15.0 hours and the overall success rate was 87.5%. Patientsdeveloped peripheral embolism with almost complete recovery in 5 episodes while significant bleeding thatrequired termination of thrombolytic therapy was observed in 2 episodes. Redo valve replacement was done in 3episodes because these patients did not improve on thrombolytic therapy (all 3 cases were of recurrent prostheticvalve thrombosis and were found to have pannus peroperatively). Three patients died during thrombolytic therapybecause of persistent heart failure. Six patients experienced a total of 13 epidoses of recurrent prosthetic valvethrombosis including index episodes (rethrombosis in 5, re-rethrombosis in 1). They were treated with repeatedthrombolysis with a success rate of 76.92%. The mean duration of thrombolytic therapy in these episodes was36.1±14.0 hours.Conclusions: In patients with prosthetic valve thrombosis, intravenous thrombolysis guided byechocardiography is a safe and effective method that may expand the indications for nonsurgical treatment ofprosthetic valve thrombosis. By using serial echocardiography, the duration of thrombolytic therapy can betailored to the patient’s requirement for normalization of valve hemodynamics. (Indian Heart J 2001; 53:451–457)

Key Words: Thrombolysis, Prosthetic valve thrombosis, Echocardiography

Original Article

Correspondence: Professor Nakul Sinha, Department of Cardiology,Sanjay Gandhi PGIMS, Lucknow, UP 226014. e-mail: [email protected]

IHJ-915-00.p65 10/17/01, 2:03 PM451

452 Kumar et al. Thrombolytic Therapy for Stuck Prosthetic Valves Indian Heart J 2001; 53: 451–457

Methods

Patient population: Forty-one consecutive patientsadmitted to the intensive care unit of our institute betweenJuly 1994 and June 2000 with PVT were included. All thepatients included in the study had clinical andechocardiographic evidence of prosthetic valve dysfunction(thrombotic) and no contraindication for TLT. The clinicalcriteria of prosthetic valve obstruction were taken as recentonset significant dyspnea, orthopnea and/or paroxysmalnocturnal dyspnea, and embolic episode, associated withdiminished or absent prosthetic valve clicks with or withoutaudible stenotic or regurgitant murmurs across theprosthetic valve.10 The echocardiographic criteria ofprosthetic valve obstruction in the mitral position weretaken as a calculated mitral valve area of less than 1.6 cm2

with an associated end-diastolic gradient >10 mmHg.11 Inthe aortic position, a peak systolic gradient >50 mmHg wasconsidered abnormal for valves ≥20 mm in size.12

Clinical evaluation and basal investigations: Allpatients underwent detailed clinical evaluation with specialemphasis on diminished prosthetic valve sounds, murmurof recent onset, any evidence of pulmonary venoushypertension, pulmonary arterial hypertension andsystemic embolism. Basal routine investigations, whichincluded hemogram, biochemical tests, prothrombintime with International Normalized Ratio (INR),electrocardiography and chest X-ray were also carried out.Adequacy of anticoagulation was defined according to theAmerican Heart Association guidelines.13

Echocardiographic evaluation: Detailed echocardio-graphic evaluation including transesophagealechocardiography was done using ATL Ultramark®–9ultrasound system (Advanced Technology Laboratories Inc.Bothell, WA, USA) and HP sonos 5500 ultrasound system(Hewlett Packard, Massachusetts, USA). Each patient wascarefully interrogated echocardiographically for thepresence of any clot and/or pannus.14 Two-dimensionalechocardiography was done with special emphasis onlooking for opening excursion and completeness of closureof the prosthesis and its beat-to-beat variability, if any. Thesharpness of opening and closing angles was evaluated byM-mode echocardiography. Doppler echocardiography wasdone to measure the pressure gradients across theprosthesis.15 The prosthetic valve area was calculated usingpressure half-time16 and/or continuity equation, and rightventricular systolic pressure was calculated using tricuspidregurgitation velocity. Cardiac Doppler measurementsincluded an average of three cardiac cycles in patients with

sinus rhythm and an average of five cardiac cycles in thosewith atrial fibrillation. Color flow imaging was done to lookfor the presence of any regurgitation across the prosthesis.Regurgitation was graded according to previously definedcriteria.17 All echocardiographic recordings were taken bytwo independent observers, and a third one was taken incase of any discrepancy.

Thrombolysis: Thrombolytic therapy was considered forpatients fulfilling the criteria of PVT with nocontraindication for TLT. Informed consent was taken ineach case. Streptokinase (Streptase; Hoechst MarionRoussel, Frankfurt, Germany) or urokinase (Urokinase;TTK Pharmaceutical, Chennai, India) were used as agentsfor thrombolysis. Streptokinase was given in a loadingdose of 2 50 000 units over 30 minutes, followed by aninfusion of 1 00 000 units/hour. Hydrocortisone wasadministered (100 mg intravenously) before the initiationof TLT with streptokinase. An infusion of urokinasewas given at a dose of 4400 IU/kg/hour. In addition, allpatients received the conventional symptomatictherapy, which included diuretics. The patients weremonitored 2-hourly for symptomatic improvement, signsof peripheral embolization and audibility as well ascharacter of prosthetic valve sounds. Serial transthoracicechocardiographic examination was done at 12-hourintervals after the start of TLT. The end-point oftherapy was taken as near-normalization of Dopplerechocardiographic transvalvular gradients according topreviously defined normal values,17 along with clinicalimprovement. Death of the patient, clinical deteriorationduring thrombolytic infusion, or lack of improvement inobjective measurements even after 72 hours ofthrombolytic infusion were regarded as failure of therapy.All patients received heparin infusion at a rate of 1000units/hour after TLT for a variable period of time untiladequate anticoagulation was achieved with nicoumaloneor warfarin sodium. In addition, all patients received aspirin150 mg per day from the day of admission.

Statistical analysis: Continuous variables are presentedas mean±standard deviation and compared using pairedStudent’s t test. Categorical variables are expressedas percentages and analyzed by the Z test. Differenceswere considered statistically significant if the p value was<0.05.

Results

Demographic profile: Over a period of 72 months (July1994 to June 2000), a total of 1720 valve replacements

IHJ-915-00.p65 10/17/01, 2:03 PM452


were done in 1566 patients (707 male, 859 female) in ourcenter [Sorin 946 (55%), Medtronic Hall 564 (32.7%),Starr–Edward 164 (9.53%) and St Jude 46 (2.67%)]. Duringthis period, 41 patients with mechanical prosthetic valvereplacement presented with 48 episodes of PVT in ourhospital. The rate of PVT was highest for Sorin (4.5%),followed by St Jude (2.17%), Medtronic Hall (0.35%), andStarr–Edward (0.06%) prostheses (Table 1). There wasfemale preponderance with a male/female ratio of 14/27.The age of the patients ranged from 13 to 54 years (meanage 33.8±9.3 years). Mitral valve prostheses were involvedin 44 episodes and aortic valve prostheses in only 4 episodesof PVT. Both in the mitral as well as in aortic position, Sorinvalve prothesis was the most commonly involved. Therewere only 2 patients with double valve replacement (DVR)presenting with PVT; mitral prosthesis was involved in oneand aortic in the other.

Clinical profile: The interval from surgery to time ofpresentation varied from 1 month to 108 months (meanduration 20.4±20.6 months). Most of these episodes(n=35, 72.91%) occurred within two years of operation(within 12 months: 45.83%; 13–24 months: 27.08%; after24 months: 27.08%). The mean duration of symptoms was

10.85±12.57 days. Atrial fibrillation was present in23 (47.9%) episodes (all with mitral PVT). At the time ofdiagnosis of PVT, only 5 (10.41%) patients had adequateanticoagulation (Table 1).

The commonest clinical presentation (n=45, 93.75%)was subacute pulmonay edema. Isolated embolism was theclinical presentation in 2 (4.16%) episodes and cardiogenicshock in 1 (2.08%). All the patients of PVT, except thosepresenting with an embolic event, showed absence ormuffling of prosthetic valve sounds. The majority (n=43,89.6%) also had associated systolic or diastolic murmur ofrecent onset.

Echocardiographic profile: The diagnosis of PVT wasconfirmed by echocardiography including transesophagealechocardiography. Two-dimensional echocardiographyshowed diminished excursion of the disc in all cases oftilting disc or bileaflet disc prostheses, and decreasedmovement of the ball in cases of ball and cage typeprostheses. On M-mode examination, all the patients hadblunted opening and closing angles in cases of tilting discprostheses. Doppler examination revealed significantpressure gradients across the prostheses in all episodes. Inthe majority of episodes of mitral PVT (n=39, 88.6%), therewas associated mitral regurgitation at the time ofpresentation [mild (grades I and II) in 36, moderate (gradeIII) in 3], while in all the episodes of aortic PVT there wasassociated aortic regurgitation [moderate (grade III) in 3,severe (grade IV) in 1]. Pulmonary arterial hypertension(PAH) was present in all episodes of mitral PVT (mild in 1,moderate in 20, and severe in 23 patients). The averagesystolic pulmonary arterial pressure in patients with mitralPVT was 70.7±14.5 mmHg (range 48–120 mmHg). In allepisodes of mitral PVT, patients had severe stenosis of theprostheses at the time of presentation (mean valvular areawas 0.78±0.11 cm2) (Table 2).

Efficacy of thrombolysis: Streptokinase was thecommonly used agent (in 91.66% of episodes), whileurokinase was used in only 4 (8.34%) episodes, all inpatients with recurrent PVT in whom streptokinase hadbeen used previously (Tables 2 and 3). The mean durationof streptokinase infusion was 26.5±13.77 hours, rangingfrom 11 to 72 hours. The duration of infusion was longerin episodes of mitral PVT (29.0±15.1 hours) compared tothose in aortic PVT (16.0±5.23 hours). The mean durationof urokinase infusion was 34.0±25.6 hours, ranging from16 to 72 hours. Thrombolytic therapy was successful in atotal of 42 (87.5%) episodes. The success rate was slightlyhigher for episodes of mitral PVT (39/44, 88.6%) comparedto those of aortic PVT (3/4, 75%). After successful

Table 1. Demographic profile and clinical characteristicsof 48 episodes of prosthetic valve thrombosis in 41 patients

All episodes of Mitral PVT Aortic PVTPVT (n=248) (n=44) (n=4)

Age 33.8±9.29 33.8±9.6 33.0±4.55Sex (M/F) 14/27 12/25 2/2Prosthesis

Sorin 43 (89.58) 40 (90.9) 3 (75)Medtronic Hall 2 (4.16) 2 (4.54) –St Jude 1 (2.08) 1 (2.27) –Carbomedics 1 (2.08) 1 (2.27) –Starr–Edward 1 (2.08) – 1 (25)

Duration of symptoms (days) 10.85±12.57 11.11±13.04 8.0±5.03Duration from date of surgery 20.4±20.6 19.1±16.6 34.5±49.2

(months)Atrial fibrillation 23 (47.9) 23 (52.27) –Anticoagulation status

Inadequate 41 (85.41) 38 (86.36) 3 (75)Adequate 5 (10.41) 4 (9.09) 3 (25)Stopped 2 (4.16) 2 (4.54) –

Mode of presentationDyspnea 45 (93.75) 41 (93.2) 4 (100)NYHA class II 21 (43.75) 19 (43.18) 2 (50.0)NYHA class III 23 (47.9) 21 (47.72) 2 (50.0)NYHA class IV 1 (2.08) 1 (2.27) –Embolism 2 (4.16) 2 (4.54) –Shock+Dyspnea 1 (2.08) 1 (2.27) –

Decreased prosthetic valve sound 46 (95.83) 42 (95.45) 4 (100)Regurgitant murmur 43 (89.58) 39 (88.6) 4 (100)

PVT: prosthetic valve thrombosis; NYHA: New York Heart AssociationValues in parentheses are percentages

IHJ-915-00.p65 10/17/01, 2:03 PM453


thrombolysis, there was near normalization oftransvalvular gradients and pulmonary arterial pressureswith disappearance of regurgitation across the prostheses.

Systemic embolism was observed in 5 (10.41%)episodes—2 patients had cerebrovascular embolism withalmost full functional recovery, another 2 had peripheralarterial embolization to the right dorsalis pedis withspontaneous complete recovery; 1 patient had both strokeand peripheral embolism after 13 hours of TLT andrecovered spontaneously from both events. Only 2 (4.6%)patients developed significant bleeding—TLT was stoppedin both of them. One patient developed a large hematomain the left arm 11 hours after initiation of TLT so that it had

to be stopped. Echocardiography showed good openingexcursion of the prosthesis with normalization of gradients,and the hematoma was managed conservatively. Anotherpatient developed significant gingival bleeding after 16hours, forcing termination of TLT. This patient also had goodopening excursion of the prosthesis with nearnormalization of gradients at the time of stopping TLT.

Three (6.25%) patients required redo surgery (mitralvalve replacement in all), as they did not show anyimprovement with TLT even after 72 hours of infusion.Peroperatively, all these patients had pannus associated withthrombus on their prostheses (Fig. 1). Of these 3 patients,2 improved following redo valve replacement while the thirddied in the perioperative period because of intractable heartfailure.

Four (8.33%) patients died during the study. Threepatients died while on TLT (1 patient presented withcardiogenic shock and could receive TLT for only 2 hours; 2others died of persistent heart failure after 14 and 28 hoursof TLT, respectively), while the fourth patient died in theperioperative period.

Thrombolysis in patients with recurrent thrombosis:Six patients experienced a total of 13 episodes of recurrentPVT including index episodes (rethrombosis in 5 and re-rethrombosis in 1 patient). Four were females and 2 weremales. All these patients had Sorin prostheses in the mitralposition. The mean duration of presentation from the timeof surgery was 18.6±12.0 months (range 1–47 months).The mean interval between thrombotic episodeswas 8.57±4.35 months (range 1–13 months). Theanticoagulation status was inadequate in 9 (69.23%) suchepisodes. Atrial fibrillation was present in 11 (84.61%)episodes. The mode of presentation was subacutepulmonary edema in all these episodes with a meanduration of 6.8±7.2 days (range 2–20 days).

Urokinase was used in 4 and streptokinase in 9 episodes.The mean duration of TLT was 36.1±14.0 hours (range24–48 hours). The success rate of TLT was 76.92% (allepisodes treated with streptokinase were successful). In 3(23.07%) episodes, patients were taken up for redo surgery.During surgery, all 3 patients were found to be have pannusassociated with thrombus (Fig. 1). Two improved aftersurgery while 1 died in the perioperative period because ofpersistent heart failure.

Discussion

Prosthetic valve thrombosis is a life-threateningcomplication, requiring emergency management, usuallyin the form of thrombectomy or valve replacement.

Table 2. Effect of thrombolytic therapy on echocardio-graphic profile in patients with prosthetic valvethrombosis (n=48 episodes)

Pre-thrombolysis Post-thrombolysis p value

Mitral prosthesis (n=44)Gradients (mmHg)

Peak 34.23±8.5 13.0±6.8 <0.0001Mean 21.7±7.25 6.6±3.2 <0.0001EDG 9.14±4.22 2.3±1.2 <0.0001

RVSP (mmHg) 70.7±14.5 36.0±12 <0.0001Regurgitation 88.63% – <0.0001Valve area (cm2) 0.78±0.11 2.13±0.55 <0.0001

Aortic prosthesis (n=4)Gradients (mmHg)

Peak 88.0±32.5 20.5±14.2 <0.0001Mean 54.0±23 13.0± 9.6 <0.0001

Regurgitation 100% – <0.0001

EDG: end-diastolic gradient; RVSP: right ventricular systolic pressure

Table 3. Outcome of thrombolytic therapy in 48 episodesof prosthetic valve thrombosis

Total episodes Mitral PVT Aortic PVT(n=48) (n=44) (n=4)

Agents usedStreptokinase 44 (91.66) 40 (90.9) 4 (100)Urokinase 4 (8.33) 4 (9.1) –

Mean dosage (infusion in units)Streptokinase (units ×105) 28.0±13.8 29.25±13.9 16.5±5.2Urokinase (units ×103) 149.6±112.64 149.6±112.64 –

Duration of therapy (in hours) 27.9±15.0 29.0±15.1 16.0±5.23Success rate 42/48 (87.5) 39/44 (88.6) 3/4 (75)Complications

Embolism 5 (10.41) 5 (11.36) –CNS 3 3 –Limbs 3 3 –

Hemorrhage 2 (4.6) 2 (4.54) –Failure of therapy 6 (12.5) 5 (11.36) 1 (25)

Surgery 3 3 –Death (during TLT) 3 2 1

CNS: central nervous system; PVT: prosthetic valve thrombosis; TLT: thrombolytic therapyValues in parentheses are percentages

IHJ-915-00.p65 10/17/01, 2:03 PM454


Recently, TLT has been recommended as an alternativetherapeutic modality but only in critically ill patients.6,8,9

However, the majority of patients with prosthetic valvethrombosis do not fall into this category. The first series ofTLT for left-sided heart valves was reported by Witchitz in1980.7 In five larger series, success rates between 75% and100% have been reported.3,6,7,9,12 Our study is oneof its kind, wherein consecutive patients with PVT,irrespective of their presentation, were given TLT andfollowed up with a comprehensive assessment of valvefunction (clinically and echocardiographically). By usingserial echocardiography, the duration of TLT was tailoredto the patient’s requirement for normalization of valvehemodynamics. Our results are in agreement with previousstudies including Indian ones (Table 4).3,6,7,9,12,18,19,20 Theoverall success rate of TLT was 87.5%. It is interesting tonote that most patients (87.5%) required TLT for less than72 hours (an average of 27.9±15.0 hours of infusion beingrequired for the entire group). Thrombolytic therapy wasequally succesful in all patients irrespective of the New YorkHeart Association (NYHA) class. In a recently publishedlarge series of 110 patients from India, completehemodynamic response, unaffected by the NYHA class, wasseen in 81.8% of episodes.9

In our series, the total incidence of complications was27.08% (13/48 episodes, including an incidence of 10.41%for embolism). The fear of potential embolism during TLThas restricted the use of TLT for left-sided valve thrombosis.Gupta et al.9 had observed embolic episodes in19.1% duringtherapy, while Ledain et al.6 reported an incidence ofapproximately 18% of systemic embolism during TLT; theyattributed these episodes to breaking of the thrombuspresent on the prostheses. Roudaut et al.3 noted peripheralembolism in 22% of the cases (14/64), of which 4 had lethalcerebrovascular accident and an additional 6 had transientischemic attacks. The incidence of peripheral embolism isconsistently higher (18%–22%) in western series3,6,7

compared to Indian ones (2.8% in a series by Reddy et al.,19

12% in the series by Vasan et al.,12 and 10.41% in thepresent study). The possible reason for the difference mightbe that the study populations in the western series wereelderly (mean age around 50 years v. 33.8 years in thepresent study). Secondly, the majority of patients in thewestern series were in atrial fibrillation. The incidence ofatrial fibrillation was 82.14% in the series by Vitale et al.,5

63% in the study by Miller et al.,21 and 53% in a series byOzkan et al.22 Only a few western studies have reported anincidence less than ours.14 Hence the likelihood of a clot inthe left atrium is higher, with the added risk of embolizationonce the valve opens. Since transesophageal echocardio-graphy was not performed in these studies, this theorycannot be confirmed. Thus it is possible that the incidenceof systemic embolism in our study population is not as highas reported earlier.

Rethrombosis is the most important complication duringfollow up, reported to occur in 12%–24% of patients aftervariable follow-up intervals.3,8,9,23 In our series, 6 patientsexperienced a total of 13 episodes of recurrent PVTincluding the index episode (2.2% patients hadrethrombosis while 2.4% patients had re-rethrombosis).Repeat TLT was given on 7 occasions of which 4 weresuccessful. The efficacy of TLT was similar in previousreports.3,19,24 The proposed mechanisms of recurrentPVT are inadequate anticoagulation and pannusformation.19,25,26 In our study, anticoagulation wasinadequate in 9 (69.23%) episodes while pannus was foundin 3 (23.07%) episodes and confirmed during surgery in all3 cases.

Until recently, surgical intervention was the onlymodality of treatment for PVT. However, the reportedmortality is highly variable, ranging from 0% to 44%,25,26

with an average mortality of 8%–10%. This variation ispossibly related to the patient’s characteristics at inclusion.Patients in a relatively stable condition have lessermortality.25 The reported immediate mortality with TLT islower, ranging from 0% to 17%.3,8,12 In our study, theimmediate mortality on TLT was 7.46% (3/41).Rethrombosis remains the major problem, even withsurgery, with an incidence ranging from 1% to 5% withthrombectomy and debridement. 25,26,27 Another importantconsideration, often overlooked, is the financial burden ofa second operation, more so in developing countries. In fact,we initiated TLT as first-line therapy in PVT at our centerbecause most of the patients could not afford a repeatoperation, and found it to be cost-effective.

Conclusions: In this series, successful intravenousthrombolysis for PVT was achieved with a low risk of

Fig. 1. Sorin mitral prosthesis, removed during surgery from a patient withrethrombosis, showing pannus on both sides of the prosthesis (a) on the leftatrial side, and (b) on the left ventricular side.

a b

IHJ-915-00.p65 10/17/01, 2:03 PM455


Table 4. Indian studies on thrombolytic therapy in prosthetic valve thrombosis

Vasan Rajashekhar Reddy Agarwal Gupta Presentet al.12 et al.18 et al.19 et al.20 et al.9 study(1992) (1994) (1994) (1997) (2000)

Patients/episodes 16/16 12/13 38/44 42/42 110* 41/48

Mean age (years) 40.8±13.6 32.5±12.9 32.0±12.0 NA 35.4±10.8 33.8±9.29

Sex (M/F) 6/10 5/7 20/18 18/24 52/58 14/27

Prosthesis –

Bjork–Shiley 16 8 2 44 –Medtronic Hall – 3 – – 2Carbomedics – – 35 29 1St Jude – 1 1 37 1Sorin – – – 36Starr–Edwards – – – 1

Duration of symptoms (days) 37.6±51.8 24.7±31.6 9.6** NA*** 14**** 10.8±12.5

NYHA class II/III/IV 4/4/8 5/4/4 0/11/33 8/10/24 25/51/34 21/23/2

Inadequate ATC 31.05% 34.46% 70.0% 62.5% 41.5% 89.59%

AF 25.0% 41.67% 13.15% 38.0% 31.8% 47.9%

TLT agents STK STK STK/UK STK STK/UK STK/UK

Infusion (hours) 42.5±22.7 20.5±11.2 NA NA 42.8±20.4 27.9±15.0

Success 100.0% 92.0% 88.6% 88.0% 91.8% 87.5%

Complications

Bleeding

Minor 2 2 1 1 10 2

Major 1 – 3 2 9 –

Embolism 2 2 1 1 21 5

CVA 1 – – 2 11 3

Surgery – – 1 3 NA 3

Death – 1 – 5 8 3

NA: not available; NYHA: New York Heart Association; ATC: anticoagulation; AF: atrial fibrillation; TLT: thrombolytic therapy;STK: streptokinase; UK: urokinase; CVA: cerebrovascular accident

*Details of episodes of recurrent thrombosis not available; **Standard deviation not mentioned; ***Mean duration of symptoms not mentioned; ****Medianduration

complications and a high rate of success. Thrombolytictreatment was equally effective in PVT of both the mitraland aortic valves and in all types of prostheses. We believethat intravenous infusion of thrombolytics, monitoredthereafter by echocardiography, is a safe and effectivemethod of nonsurgical treatment of PVT. Serialechocardiography helps in tailoring the duration of TLTrequired for normalization of valve hemodynamics.

References

1. Cannegieter SC, Rosendaal FR, Briet E. Thrombotic and bleedingcomplications in patients with mechanical heart valve prosthesis.Circulation 1994; 89: 635–641

2. Thorburn CW, Morgan JJ, Shanaban MX, Chang VP. Long term results

of tricuspid valve replacement and the problem of prosthetic valvethrombosis. Am J Cardiol 1983; 51: 1128–1132

3. Roudaut R, Labbi T, Lorient-Roudaut MF, Gosse P, Baudet E, FonatanF, et al. Mechanical cardiac valve thrombosis: is fibrinolytic justified?Circulation 1992; 86 (Suppl II):8–15

4. Husebye DG, Pluth JR, Piehler JM, Schaff HV, Orszulak TA, Puga FG,et al. Reoperation on prosthetic heart valves: an analysis of risk factorsin 552 patients. J Thorac Cardovasc Surg 1983; 86: 543–552

5. Vitale N, Renzulli A, Cerasuolo F, Caruso AF, Fista M, de Luca L, et al.Prosthetic valve obstruction: thrombolysis versus operation. AnnThorac Surg 1994; 57: 365–370

6. Ledain LD, Ohayon JP, Colle JP, Lorient-Roudant FM, Roudant RP, BissePM, et al. Acute thrombotic obstruction with disc valve prosthesis:diagnostic consideration and thrombolytic treatment. J Am CollCardiol 1986; 7: 743–751

7. Witchitz S, Veyrat C, Moisson P, Scheinman N, Rozenstain L.Fibrinolytic treatment of thrombus on prosthetic valves. Br Heart J1980; 44: 545–554

IHJ-915-00.p65 10/17/01, 2:03 PM456


8. Graver LM, Gelber PM, Tyras DH. The risks and benefits ofthrombolytic therapy in acute aortic and mitral prosthetic valvedysfunction: report of a case and review of literature. Ann Thorac Surg1988; 46: 85–88

9. Gupta D, Kothari SS, Bahl VK, Goswami KC, Talwar KK, ManchandaSC, et al. Thrombolytic therapy for the prosthetic valve thrombosis:short and long term results. Am Heart J 2000; 140: 906–916

10. Balram A, Kaul U, Ram Rao BV, Iyer KS, Rajani M, Rao IM, et al.Thrombotic obstruction of Bjork–Shiley valves: diagnostic andsurgical considerations. Int J Cardiol 1984; 6: 61–73

11. Radhakrishnan S, Bahl VK, Bajaj R, Bhatia ML, Srivastava S. Dopplerechocardiographic evaluation of normal and thrombosed Bjork–Shiley mitral prosthetic valves. Int J Cardiol 1988; 20: 387–393

12. Vasan RS, Kaul U, Sanghavi S, Kamalakar T, Negi PC, Srivastava S, etal. Thrombolytic therapy for prosthetic valve thrombosis: a studybased on serial Doppler echocardiographic evaluation. Am Heart J1992; 123: 1575–1580

13. Robert OB, Carbello B, De Leon A, Edmunds LH. ACC/AHA Guidelinesfor the Management of Patients with Valvular Heart Disease: a Reportof the American College of Cardiology/American Heart AssociationTask Force on Practice Guidelines (Committee on management ofpatients with valvular heart disease). J Am Coll Cardiol 1998; 32:1486–1588

14. Barbeteseas J, Nagueh SF, Pitsavos C, Toutiouzas PK, Quinoes MA,Zoghibi WA. Differentiating thrombus from pannus formation inobstructed mechanical prosthetic valves: an evaluation of clinical,transthoracic and transesophageal echocardiographic parameters.J Am Coll Cardiol 1998; 32: 1410–1417

15. Nanda NC, Cooper JW, Mahan EF III, Fan P. Echocardiographicassessement of prosthetic valves. Circulation 1991; 84 (Suppl 3):1228–1239

16. Burstow DJ, Nishimura RA, Bailey KR, Reeder GS, Holmes DR Jr,Seward JB, et al. Continuous wave Doppler echocardiographicmeasurement of prosthetic valve gradients: a simultaneous Dopplercatheter correlative study. Circulation 1989; 80: 504–514

17. Feigenbaum H. Acquired valvular heart disease. In: Feigenbaum H(ed). Echocardiography. 5th ed. Philadelphia, USA: Lea & Febiger Co.;1994. pp. 239–349

18. Rajashekhar D, Balakrishnan KG, Venkitachalam CG, Tharakan JA,Titus T, Pillai VR, et al. Thrombolytic therapy for prosthetic cardiacvalve thrombosis. Indian Heart J 1994; 46: 101–105

19. Reddy NK, Padmanabhan TN, Singh S, Kumar DN, Raju PR,Satyanarayana PV, et al. Thrombolysis in left sided prosthetic valveocclusion: immediate and follow-up results. Ann Thorac Surg 1994;58: 462–470

20. Agarwal D, Dubey S, Saket B, Bhargava M, Mehta N, Lohchab SS, etal. Thrombolytic therapy for prosthetic valve thrombosis in ThirdWorld countries. Indian Heart J 1997; 49: 383–386

21. Miller DC, Oyer PE, Stinson EB, Reitz BA, Jamieson SW, BaumgartnerWA, et al. Ten to fifteen years reassessment of the performancecharacteristics of the Starr Edward model 6120 mitral valveprosthesis. J Thorac Cardiovasc Surg 1983; 85: 1–20

22. Ozkan M, Kaymaz C, Kirma C, Sonmez K, Ozdemir N, Balkanay M, etal. Intravenous thrombolytic treatment of mechanical prostheticvalve thrombosis: a study using serial transesophagealechocardiography. J Am Coll Cardiol 2000; 35: 1881–1889

23. Kontos CJ, Schaff HV, Orszulak TA, Puga FJ, Pluth JR, Danielson GK.Thrombotic obstruction of disc valves: clinical recognition andsurgical management. Ann Thorac Surg 1989; 48: 60–65

24. Lengyel M, Fuster V, Keltai M, Roudaut R, Schulte HD, Seward JB, etal. Guidelines for management of left sided prosthetic valvethrombosis: a role for thrombolytic therapy, Consensus Conferenceon Prosthetic Valve Thrombosis. J Am Coll Cardiol 1997; 30: 1521–1526

25. Daviri E, Sareli P, Wisenbaugh T, Cronje SL. Obstruction ofmechanical heart valve prosthesis: clinical aspects and surgicalmanagement. J Am Coll Cardiol 1991; 17: 646–650

26. Martinell J, Fraile J, Artiz V, Cortina J, Fresneda P, Rabago G.Reoperation for left sided low profile mechanical prosthesisobstructions. Ann Thorac Surg 1987; 43: 172–175

27. Munclinger MJ, Patel JJ, Mitha AS. Thrombolysis of thrombosed StJude Medical prosthetic valves: rethrombosis—a sign of tissueingrowth. J Thorac Cardiovasc Surg 1998; 115: 248–249

IHJ-915-00.p65 10/17/01, 2:03 PM457

458 Jadhav et al. Carotid Intima–Media Thickness Indian Heart J 2001; 53: 458–462

Carotid Intima–Media Thickness as an IndependentPredictor of Coronary Artery Disease

Uday M Jadhav, NN KadamDepartment of Non-Invasive Cardiology, MGM New Bombay Hospital, New Mumbai

Recently, considerable attention has been directed at thewall thickness of the carotid arteries as an early marker

of atherosclerotic disease and as a means of showing theeffectiveness of medical therapies in treating atherosclerosis.1

Noninvasive techniques such as B-mode ultrasound candirectly assess the intima–media thickness (IMT), whichcorresponds to the thickness of the histologic intima andmedia.2,3 Intima–media thickness is defined as the distancefrom the leading edge of the lumen–intima interface of thefar wall to the leading edge of the media–adventitia interfaceof the far wall.4,5 The definition of IMT of the commoncarotid artery (CCA) is not uniform and it includes twodifferent pathological changes: a general intima–media

thickening and a local atherosclerotic change (plaqueformation).6 Increased IMT is likely to be a physiologicaleffect of aging that corresponds to diffuse intimal thickening,especially in elderly persons and IMT is distinct frompathological plaque formation.7

The rate of change of IMT with age in cross-sectionalstudies has been calculated to be about 0.01 mm/year inthe general population and 0.03–0.06 mm/year in patientswith coronary artery disease (CAD).8 The role played bycarotid artery IMT in predicting clinical coronary eventshas generated considerable interest in studies done in thewestern population. The increasing incidence ofcardiovascular disease in the Indian society and the possibleutility of the simple technique of IMT measurementmotivated us to undertake this study in order to assess itsrole in predicting CAD.

Correspondence: Dr Uday M Jadhav, MGM New Bombay Hospital,Sector 3, Vashi, New Mumbai 400703. e-mail: [email protected]

Original Article

Background: A noninvasive technique of measuring carotid artery intima–media thickness has recentlygenerated considerable interest as a marker of atherosclerosis and in the prediction of clinical coronary eventsand coronary artery disease. The present study evaluated the association of carotid artery intima–media thicknessin the prediction of coronary artery disease in a western Indian population.Methods and Results: Carotid artery intima–media thickness was measured with a B-mode scan in an ongoingstudy of 266 patients, who were further subdivided into 4 subgroups: those with non-insulin dependent diabetesmellitus; hypertension; diabetes mellitus with hypertension; and those without diabetes or hypertension (labeledas controls). The maximal intima–media thickness greater than 0.8 mm at the far wall of the common carotidartery, excluding raised lesions and plaques, was selected as the highest value for comparison. The subgroupswere further divided into those with and without apparent coronary artery disease. A statistically significantintima–media thickness greater than 0.8 mm was observed in 59.2% of the subjects with coronary arterydisease as against 40.8% in those without the disease on univariate analysis. A higher incidence of intima–media thickness of more than 0.8 mm was observed in all subgroups with coronary artery disease as againstthose without the disease, which was most marked in the hypertensive group (22.2% v. 3.6%) and contributedto the increased arterial thickness in diabetics with concomitant hypertension. Multivariate regression analysisrevealed carotid artery intima–media thickness to be associated with coronary artery disease with an odds ratioof 2.40.Conclusions: Carotid artery intima–media thickness is a simple, noninvasive and reproducible clinical tool toevaluate atherosclerosis and predict coronary artery disease in Indian subjects. Prospective studies in a largernumber of subjects, particularly in those undergoing coronary angiography, will help in establishing the role ofthis technique. (Indian Heart J 2001; 53: 458–462)

Key Words: Carotid arteries, Coronary disease, Ultrasonography

IHJ-160-01.p65 10/17/01, 1:30 PM458

Indian Heart J 2001; 53: 458–462 Jadhav et al. Carotid Intima–Media Thickness 459

Methods

A total of 266 randomly selected patients, both hospitalizedas well as outpatient, were studied. Informed consent wasobtained from the subjects (166 males and 100 females).Subjects were divided into two groups: with CAD (99subjects) and without CAD (167 subjects). They were furthersubdivided into 3 subgroups in context to risk stratificationfor CAD: non-insulin dependent diabetes mellitus (78subjects); hypertension (74 subjects); and diabetes withconcomitant hypertension (63 subjects). Subjects notbelonging to these categories were termed as controls. The51 subjects labeled as controls were randomly selected andincluded those with risk factors such as dyslipidemia,smoking and obesity. The ages of the subjects ranged from36 to 64 years and their mean age was 50.5 years. Baselinecharacteristics of the subjects in the groups with andwithout CAD are shown in Table 1 and the baseline

characteristics of the various subgroups in context to thecontrol cases are shown in Table 2.

Ultrasonographic scanning of the carotid arteries wasperformed in the supine position with the neck extended,using a high-frequency imaging probe (7.5 mHz) with aHewlett-Packard scanner, at a depth of 2 cm, as the carotidvessels are relatively superficial. The carotid vessels werefollowed from the clavicular head cephalad to theirbifurcation and 3–4 cm of the proximal internal andexternal carotid arteries were studied. The IMT wasmeasured at end-diastole at 4 different points in the far wallof both the CCAs. The maximum IMT and not the meanwas taken into consideration for calculating the results.Raised lesions and carotid plaques were excluded whilecalculating the maximum IMT, as the study was specificallyaimed at correlating the IMT at sites free of plaques althoughconventionally, plaques have been included in most studies.Age-adjusted IMT values >0.8 mm were labeled as thehigher quartile for correlating the association between IMTand the prevalence of CAD. Interobserver variability was0.034 mm and intraobserver variability 0.029 mm. Filmswere recorded in all subjects for documentation.

Coronary artery disease was diagnosed by clinicalpresentation, with electrocardiography, symptom-limitedexercise stress test, Doppler echocardiography and coronaryangiography documentation, when feasible. Of the 99subjects with CAD, 47 had evidence of myocardial infarctionor acute coronary syndrome, 22 had evidence of CAD oncoronary angiography, and 30 had ECG, stress test andechocardiographic manifestations suggestive of CAD.

Clinical examination included blood pressuremeasurement, cardiovascular examination, anthro-pometric measurements and body mass index. Biochemicalassessment included fasting and postprandial blood sugarlevels, glycated hemoglobin, urine for microalbumin, and

Table 2. Baseline characteristics of the study subgroups

Parameters Controls Diabetes Hypertension Diabetes and p valuehypertension

Age (years) 47.1±9.9 50.4±8.7 48.9±8.9 52.8±7.5 <0.159Male (%) 34 (66.7) 47 (60.3) 52 (70.3) 33 (52.4) <0.152Smoking (%) 11 (21.6) 9 (11.5) 15 (20.3) 11 (17.5) <0.406Body mass index (kg/m2) 25.02±2.73 24.97±2.60 26.75±3.34 26.63±4.21 <0.002Fasting plasma glucose (mg/dl) 85.62±14.4 151.77±65.3 87.17±15.7 137.33±43.1 <0.001Total cholesterol (mg/dl) 197.38±34.4 215.35±50.7 194.47±48.3 215.68±38.9 <0.006LDL-cholesterol (mg/dl) 120.1±36.3 126.1±47.6 115.06±43.4 135.3±36.2 <0.040HDL-cholesterol (mg/dl) 41.33±7.4 43.6±7.7 41.9±7.8 42.4±7.0 <0.354Triglycerides (mg/dl) 159.2±87 162.4±102.6 162.5±82.3 176±107.2 <0.794Cholesterol–HDL ratio 4.88±1.1 5.09±1.3 4.69±1.0 5.05±1.0 <0.109Apolipoprotein A1 (mg/dl) 1.32±0.31 1.27±0.37 1.37±0.43 1.35±0.35 <0.387Apoliprotein B (mg/dl) 1.25±0.36 1.40±0.52 1.21±0.41 1.43±0.47 <0.012


Table 1. Baseline characteristics of the study group

Parameters CAD (n=99) Non-CAD (n=167) p value

Age (years) 52.8±8.7 48.2±8.5 <0.001Male (%) 65 (65.7) 101(60.5) <0.400Smoking (%) 31 (31.3) 15 (9) <0.001Hypertension (%) 18 (18.2) 56 (33.5) <0.002Diabetes mellitus (%) 55 (51.5) 86 (55.5) <0.002Body mass index (kg/m2) 25.55±3.37 26.05±3.36 <0.245Fasting plasma glucose (mg/dl) 124.3±56.7 114.5±48.7 <0.140Total cholesterol (mg/dl) 208.4±43.4 20.5±46.8 <0.567LDL-cholesterol (mg/dl) 126.9±39.2 122.5±44 <0.414HDL-cholesterol (mg/dl) 41.7±7.7 42.9±7.4 <0.211Triglycerides (mg/dl) 165.6±80.1 167.1±103.6 <0.899Cholesterol–HDL ratio 5.07±1.1 4.85±1.1 <0.130Apolipoprotein A1 (mg/dl) 1.27±0.34 1.36±0.39 <0.064Apoliprotein B (mg/dl) 1.41±0.40 1.27±0.48 <0.016


IHJ-160-01.p65 10/17/01, 1:30 PM459


comprehensive lipid profile inclusive of apolipoproteins A1and B. Results of the biochemical parameters in all thesubgroups have been shown in Tables 1 and 2.

Statistical analysis: Chi-square and odds ratio withconfidence interval (CI) and multivariate regression analysiswere computed with the SPSS statistical software package(Version 10); p values <0.05 were considered significant.

Results

The relationship between CAD and IMT is shown in Table 3and Fig. 1. Among subjects with evidence of CAD, 59.2%had IMT values exceeding 0.8 mm as compared to 40.8%in those without obvious evidence of CAD, which was highlysignificant (p<0.001) on univariate analysis. The odds ratiofor an arterial thickness exceeding 0.8 mm was 3.04 inthose with CAD as against those without obvious CAD,making it an important marker for preclinical and clinicalCAD.

The importance of increased arterial thickening in high-risk subgroups is shown in Table 4, with an additive adverseeffect of concomitant hypertension in subjects with diabetesmellitus with or without CAD, as reflected by a higher

incidence of IMT exceeding 0.8 mm. On subgroup analysis(Table 5), IMT exceeding 0.8 mm had the strongestcorrelation for CAD in subjects with hypertension (with anincidence of 22.2%) as against those without CAD (only3.6%), although all high-risk subgroups showed a positivecorrelation as illustrated in Fig. 2. Carotid artery IMT valuesremain in the higher quartile in subjects with hypertension

Fig. 1. Distribution of patients with CAD and without CAD by intima–mediathickness.

Table 3. CAD–IMT crosstabulation

Number of Number ofpatients patients

without CAD with CAD Total

IMT <0.8 mm 147 (67.4) 70 (32.6) 217 (100)

IMT >0.8 mm 20 (40.8) 29 (59.2) 49 (100)

Pearson Chi-square:12.4, p<0.001; Odds ratio: IMT <0.8 mm orIMT >0.8 mm for CAD=3.0 (95% CI 1.6–5.7)IMT: Intima–media thickness; CAD: coronary artery diseaseValues in parentheses are percentages

Table 5. IMT: without CAD and with CAD for individualsubgroups

Number of Number of Number of Number ofcontrol patients with patients with patients with

patients NIDDM hypertension NIDDM+HT

Without CAD

IMT <0.8 mm 24 (96.0) 46 (85.2) 54 (96.4) 23 (71.9)IMT >0.8 mm 1 (4.0) 8 (14.8) 2 (3.6) 9 (38.1)

With CAD

IMT <0.8 mm 23 (88.5) 18 (75.0) 14 (77.8) 15 (48.4)IMT >0.8 mm 3 (11.5) 6 (25.0) 4 (22.2) 16 (51.6)

Pearson Chi-square:13.5, p<0.004 for subgroup without CAD; PearsonChi-square:12.0, p<0.007 for subgroup with CADNIDDM: non-insulin dependent diabetes mellitus; HT: hypertensionValues in parentheses are percentages

Table 4. IMT subgroup crosstabulation

Number of Number of Number of Number ofcontrol patients with patients with patients with

patients NIDDM hypertension NIDDM+HT

IMT <0.8 mm 47 (92.2) 64 (82.1) 68 (91.9) 38 (60.3)IMT >0.8 mm 4 (7.8) 14 (17.9) 6 (8.1) 25 (39.7)

Pearson Chi-square: 27.9, p<0.001;NIDDM: non-insulin dependent diabetes mellitus; IMT: intima–media thickness;HT: hypertensionValues in parentheses are percentages

Fig. 2. Percentage of patients with intima–media thickness above 0.8 mm bygroups and subgroups.

IHJ-160-01.p65 10/17/01, 1:30 PM460

Indian Heart J 2001; 53: 458–462 Jadhav et al. Carotid Intima–Media Thickness 461

and diabetes mellitus, but the incidence increasessignificantly in the presence of CAD.

Multivariate regression analysis using age-adjusted IMTas a dependent variable and diagnosis (in the context ofsubgroups), sex, smoking, LDL-cholesterol, HDL-cholesterol,total cholesterol to HDL-cholesterol ratio, apolipoprotein A1to B ratio and microalbuminuria as independent variablesshowed a strong association with CAD (p< 0.019, oddsratio: 2.40) (Table 6).

Discussion

Ultrasonography is a reliable and accurate technique todetermine IMT in the superficial arteries. Reproducibility ofIMT determination is best in the CCA of healthy subjectsand in patients with advanced atherosclerosis.9–12 B-modeultrasound scan including carotid IMT measurement andperipheral plaque detection may be of clinical value in thescreening of patients with CAD.13

The Rotterdam study,14 which involved a 4-year follow-up of almost 8000 patients, showed that there was, in fact,a significantly higher incidence of stroke and myocardialinfarction in patients who had increased IMT. Bots et al.15

in the Rotterdam study showed that the 10-year absoluterisk of coronary heart disease rose from 13% to 23.4%,whereas the risk of death within 11.5 years rose from 15%in the lowest quintile to 46% in the highest quintile.

Hodis et al.,16 in a long-term follow-up study, showedthat for each 0.03 mm increase per year in carotid arteryIMT, the relative risk for nonfatal myocardial infarction orcoronary death was 2.2 and the relative risk for coronaryevents was 3.1. Absolute thickness and progress in thicknesspredicted risk for a coronary event beyond that predicted

by coronary arterial measures of atherosclerosis and serumlipid levels. The B-mode score is at least as useful as otherwell-known risk factors for identifying patients with CAD.17

The Cardiovascular Health Study Collaborative ResearchGroup18 has recently shown in 4476 subjects withoutclinical cardiovascular disease followed up over 6 years,that the relative risk for myocardial infarction or stroke forthe quintile with the highest IMT as compared with thelowest was 3.87. The present study has also shown a riskestimate of 3.04 for CAD in subjects with an increased IMT.It is a strong independent predictor for new cardiovascularevents, even after statistical adjustment for other traditionalrisk factors. This finding has clinical significance, particularlywith regard to the elderly.

This study also shows a significant increase in IMT inthe hypertensive population which is consistent withreported studies of greater IMT of the CCA in sedentaryand active hypertensives than normotensives.19,20 This isof therapeutic importance in diabetics with concomitanthypertension, as shown in an earlier study.21 Mohan etal.22 demonstrated an increased IMT in 243 south Indiandiabetic and non-diabetic subjects in the Chennai UrbanPopulation Study, and the findings are further confirmed inthis study from the western part of the country. Besides theusual conventional risk factors, a significant positivecorrelation has been shown between plasma homocysteinelevels and carotid artery IMT in the context of CAD in theNHLBI Family Heart Study,23 which can have futureimplications for Indian subjects. Carotid artery IMT exhibitsless variability, is associated with cardiovascular risk factors,and increased levels can predict myocardial infarction andstroke. Aggressive management of risk factors can decreaseIMT.24

Conclusions: The present study is an example of evidence-based medicine in correlating two vascular beds, coronaryand carotid, which share the same atherosclerotic riskfactors. Carotid artery IMT—a simple, safe, noninvasive andreproducible method—can be used as a surrogate markerin the prediction of atherosclerosis and CAD, as has beendemonstrated. The results can be improved upon and mademore precise by increasing the sample size. The presentstudy has some limitations in that it is not a prospectivecohort study and a long-term follow-up of subjects withoutCAD is required to demonstrate the utility of IMT inpredicting CAD. Measurement of the mean of maximal IMTin both the CCA and carotid bulb and inclusion of plaqueswill contribute to a higher IMT and strengthen the resultsfurther. The obvious limitation of the study, however, is thelack of angiographic documentation in both the groups

Table 6. Multivariate regression analysis for CAD

Parameter β SE p value Odds ratio

Sex 0.321 0.318 >0.05 1.4Smoking 1.487 0.407 <0.001 4.4Subgroup –0.083 0.138 >0.05 0.9LDL-cholesterol –0.316 0.431 >0.05 0.7HDL-cholesterol –0.157 0.389 >0.05 0.8Total cholesterol

to HDL-C ratio –0.003 0.320 >0.05 1.0Apolipoprotein A1

to B ratio 0.916 0.297 <0.003 2.5Microalbuminuria 0.498 0.332 >0.05 1.6Carotid IMT 0.887 0.376 <0.018 2.4

SE: standard error; LDL-cholesterol: low-denisty lipoproteincholesterol; HDL-cholesterol: high-density lipoprotein cholesterol;IMT: intima–media thickness

IHJ-160-01.p65 10/17/01, 1:30 PM461


for the presence of CAD, apart from 69 patients in the groupwith CAD. This could not be performed because of logistic,ethical and socio-economic reasons. IMT measurementshould be included as a diagnostic tool, given the paucityof facilities for invasive techniques in India. This will help inearly identification of clinical and preclinical CAD. Finally,long-term follow-up studies from India are required toaddress the utility of this technique for therapeuticintervention.

Acknowledgments

We gratefully acknowledge the help provided by theDepartments of Pathology and Ultrasonography, MGM NewBombay Hospital and Dr DP Singh for the statistical analysis.

References

1. Bluth EI. Evaluation and characterisation of carotid plaque. SeminUltrasound CT MR 1997; 18: 57–65

2. Murthy SV, Venkatramana K. Extracerebral carotid atherosclerosis—Doppler evaluation. Indian J Radiol Imag 1998; 8: 19–25

3. Jadhav UM. Non-invasive early prediction of atherosclerosis bycarotid intima–media thickness. Asian J Clin Cardiol 2001; 4: 24–28

4. Pignoli P, Ultrasound B-mode imaging for arterial wall thicknessmeasurement. Atherosclerosis Rev 1984; 12: 177–184

5. Pignoli P, Tremoli E, Poli A, Oreste P, Paoletti R. Intimal plus medialthickness of the arterial wall: a direct measurement with ultrasoundimaging. Circulation 1986; 74: 1399–1406

6. Ando F, Takekuma K, Niino N, Shimokata H. Ultrasonic evaluationof common carotid intima–media thickness (IMT)—influence oflocal plaque on the relationship between IMT and age. J Epidemiol2000; 10: S10–S17

7. Homma S, Hirose N, Ishida H, Ishii T, Araki G. Carotid plaque andintima–media thickness assessed by B-mode ultrasonography insubjects ranging from young adults to centenarians. Stroke 2001;32: 830–835

8. Mohan V, Premlatha G, Ravikumar R. Intima-medial thickness ofthe carotid arteries—an excellent measurement of atherosclerosis.Asian J Diabetol 1999; 1: 43–46

9. Kanters SD, Algra A, van Leeuwan MS, Banga JD. Reproducibilityof in vivo carotid intima media thickness measurements: a review.Stroke 1997; 28: 665–671

10. Rosfors S, Hallerstam S, Jensen-Urstad K, Zetterling M, Carlstrom C.Relationship between intima–media thickness in the commoncarotid artery and atherosclerosis in the carotid bifurcation. Stroke1998; 29: 1378–1382

11. Kanters SD, Elgersma OE, Banga JD, van Leeuwan MS, Algra A.

Reproducibility of measurements of intima–media thickness anddistensibility in the common carotid artery. Eur J Vasc Endovasc Surg1998; 16: 28–35

12. Sramek A, Bosch JG, Reiber JH, Van Oostayen JA, Rosendaal FR.Ultrasound assessment of atherosclerotic vessel wall changes:reproducibility of intima–media thickness measurements in carotidand femoral arteries. Invest Radiol 2000; 35: 699–706

13. Balbarini A, Buttitta F, Limbruno U, Petronio AS, Baglini R, StrataG, et al. Usefulness of carotid intima–media thickness measurementand peripheral B-mode ultrasound scan in the clinical screening ofpatients with coronary artery disease. Angiology 2000; 51: 269–279

14. Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Commoncarotid intima–media thickness and risk of stroke and myocardialinfarction: the Rotterdam Study. Circulation 1997; 96: 1432–1437

15. Bots ML, Hoes AW, Hofman A, Witteman JC, Grobbee DE. Cross-sectionally assessed carotid intima–media thickness relates to longterm risk of stroke, coronary heart disease and death as estimatedby available risk functions. J Intern Med 1999; 245: 269–276

16. Hodis HN, Mack WJ, LaBree L, Selzer RH, Liu CR, Liu CH, et al. Therole of carotid arterial intima–media thickness in predicting clinicalcoronary events. Ann Intern Med 1998; 128: 262–269

17. Craven TE, Ryu JE, Espeland M, Kahl FR, Mckinney WM, Toole JF, etal. Evaluation of associations between carotid arteryatherosclerosis and coronary artery stenosis. Circulation 1990; 82:1230–1242

18. O’Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL,Wolfson SK, et al. for the Cardiovascular Health StudyCollaborative Research Group. Carotid artery intima and mediathickness as a risk factor for myocardial infarction and stroke inolder adults. N Engl J Med 1999; 340: 14–22

19. Bigazzi R, Bianchi S, Nenci R, Baldari D, Baldari G, Campese VM.Increased thickness of the carotid artery in patients with essentialhypertension and microalbuminuria. J Hum Hypertens 1995; 9:827–833

20. Casiglia E, Palatini P, Da Ros S, Pagliara N, Putao M, Dorigatti F,Pauletto P. Effect of blood pressure and physical activity on carotidartery intima–media thickness in stage 1 hypertensives andcontrols. Am J Hypertens 2000; 13: 1256–1262

21. Jadhav UM, Kadam NN. Proceedings of the 9th Annual Conferenceof the Hypertension Society of India: 14–15 October, 2000,Ahmedabad, India

22. Mohan V, Ravikumar R, Shanthi Rani S, Deepa R. Intimal medialthickness of the carotid artery in South Indian diabetic and non-diabetic subjects: the Chennai Urban Population Study (CUPS).Diabetologia 2000; 43: 494–499

23. Tsai MY, Arnett DK, Eckfeldt JH, Williams RR, Ellison RC. Plasmahomocysteine and its association with carotid intima–medial wallthickness and prevalent coronary heart disease: NHLBI FamilyHeart Study. Atherosclerosis 2000; 151: 519–524

24. Fathi R, Marwick TH. Noninvasive tests of vascular function andstructure: why and how to perform them. Am Heart J 2001; 141:694–703

IHJ-160-01.p65 10/17/01, 1:30 PM462

Indian Heart J 2001; 53: 463–466 Isser et al. Lp(a) in Young Patients with MI 463

Lipoprotein (a) and Lipid Levels in Young Patients withMyocardial Infarction and Their First-Degree Relatives

HS Isser, VK Puri, VS Narain, RK Saran, SK Dwivedi, S SinghDepartment of Cardiology, King George’s Medical College, Lucknow

Coronary artery disease (CAD) is now a majorpublic health problem in India and is emerging as a

major killer.1 Of all ethnic groups, people of Indian originhave one of the highest incidences of CAD,2,3 and diffuseand severe CAD frequently occurs in Indians at an earlyage.4 The prevalence of premature CAD in Indians is up to3 times higher when compared with people of a similar agegroup in the western world.5

One of the most striking characteristics of prematureCAD in Indians is a relatively low prevalence of traditionalcoronary risk factors.6 Little is known about thepathogenesis of atherosclerosis and CAD in Indians who,

while belonging to different religious groups with varyingcultural backgrounds and living in different geographicalregions, share the common problem of high mortality fromthe disease.7 Available data suggest the presence of apowerful risk factor unaffected by even maximalmodification of lifesyle.8 Familial aggregation of a numberof cardiovascular diseases has long pointed to the possiblerole of primary genetic factors in their pathogenesis. Manystudies in the past have demonstrated that lipoprotein (a)[Lp(a)], a genetically determined lipoprotein, is one of themost powerful8,9 and most prevalent10 independent risk,factors for premature CAD.11

Because the disease affects Indians in the prime of theirlives and careers and has significant socio-economicconsequences, there is an urgent need to define thepathophysiological process involved and to look for familial

Background: Studies among emigrant Indians have stressed the role of a powerful genetic factor, lipoprotein (a),in the causation of premature coronary artery disease. This study was carried out to assess lipoprotein (a) andlipid levels in 50 consecutive young north Indian patients (age less than 45 years, mean age 39+3.7 years) withmyocardial infarction, their first-degree relatives (n=125, mean age 36+16 years), and age- and sex-matchedcontrols (n=50, mean age 34+6.9 years).Methods and Results: Blood samples for lipid estimation were taken within 24 hours of myocardial infarctionand after overnight fasting for twelve hours. Lipoprotein (a) levels were estimated by the ELISA technique usingpreformed antibodies while lipid levels were estimated by kits using the colorimetric method. All were malepatients. The mean lipoprotein (a) level was 22.28+5.4 mg/dl in patients, 13.88+5.19 mg/dl in their first-degree relatives and 9.28+22.59 mg/dl in controls. In addition, it was significantly higher in young patientswith myocardial infarction and their relatives as compared to controls (p<0.001 for patients v. controls andp<0.05 for relatives v. controls). There was no significant difference in the levels of total cholesterol and low-density lipoprotein cholesterol among the three groups. High-density lipoprotein cholesterol was significantlylower in young patients with myocardial infarction (30.16+9.45 mg/dl) and their first-degree relatives(33.28+8.45 mg/dl) as compared to controls (46.8+8.04 mg/dl) (p<0.001 for patients v. controls and p<0.01for relatives v. controls). Triglyceride levels were significantly higher in patients as compared to controls (202+76mg/dl v. 149+ 82.99 mg/dl, p<0.05). Smoking was more prevalent in young patients with myocardial infarctionas compared to controls (44% v. 36%, p<0.05).Conclusions: Smoking, high lipoprotein (a) and triglyceride levels and low high-density lipoprotein levels maybe important risk factors for coronary artery disease in the younger population; also, there is familial clusteringof high lipoprotein (a) levels in first-degree relatives of young patients with myocardial infarction.(Indian HeartJ 2001; 53: 463–466)

Key Words: Lipoproteins, Myocardial infarction, Coronary disease

Original Article

Correspondence: Dr HS Isser, Consultant Cardiologist, Batra Hospitaland Medical Research Centre, 1, Tughalakabad Institutional Area,New Delhi 110062

IHJ-867-00.p65 10/17/01, 1:59 PM463

464 Isser et al. Lp(a) in Young Patients with MI Indian Heart J 2001; 53: 463–466

clustering, so that an intensive program of primaryprevention and alteration of modifiable risk factors can bedrawn.

Lipoprotein (a) is fully expressed in the first year of lifeand high levels have almost the same predictive value as afamily history of premature CAD.12,13 Therefore, “tracking”Lp(a) levels from childhood may be a better option thandetecting other dyslipidemias which are not fully expresseduntil middle age. Keeping the aforementioned objectives inmind, Lp(a) and lipid levels were determined in youngpatients (<45 years) with myocardial infarction (MI) andtheir first-degree relatives, and were compared with thoseof a control population.

Methods

Subjects included in this study were consecutive patientswith MI, less than 45 years of age admitted in our hospitalduring the study period, and their first-degree relatives. Allyoung patients with MI were male. Informed consent of thepatients, their relatives and controls was taken beforeenrollment in the study.

Patients included in the study were less than 45 years ofage and admitted with acute myocardial infarction (AMI)which was diagnosed by the presence of at least 2 of 3criteria, i.e. anginal chest pain of more than 30 minutes’duration; ST segment elevation more than 1 mm abovebaseline in at least two leads of a standard 12-leadelectrocardiogram (ECG); and creatinine kinase andcreatinine kinase-MB elevation to at least twice the upperlimit of the normal range.

First-degree relatives of these patients, i.e. their brothers,sisters, children and parents were included in the study. Thefollowing patients were excluded from the study: patientsmore than 45 years of age; patients with chronic renalparenchymal disease or nephrotic syndrome; patients withconcomitant liver disease; and patients with any disablingterminal disorder. Age- and sex-matched controls with nofamily history of premature CAD were taken from amongthe medical and paramedical staff. They had no evidenceof manifest CAD and their ECGs were normal. Patients, theirfirst-degree relatives and controls were subjected to detailedclinical examination with special reference to theconcomitant risk factors, i.e. smoking, diabetes,hypertension, obesity and family history of premature CAD.

Blood samples of the patients were taken within 24hours of MI and after 12-hour overnight fasting. Bloodsamples of the relatives and controls were also taken after12-hour overnight fasting. For isolation of plasma, a 10 mlblood sample was centrifuged at 2000 rpm at 4°C for20 min and aspirated into tubes preflushed with nitrogen.Lipoprotein (a) was isolated by density gradient ultra-

centrifugation. Isolated Lp(a) was purified by lysinesepharose chromatography and the purity established bygel electrophoresis. Lipoprotein (a) level was estimated bythe ELISA method, using preformed antibodies.

Serum concentrations of total cholesterol (TC) andtriglyceride (TG) were estimated by the colorimetric methodand high-density lipoprotein (HDL) cholesterol wasmeasured by the CHOD–PAP method using BoehringerMannheim kits. Low density lipoprotein (LDL) cholesterolwas calculated using the Friedewald formula, i.e.LDL-cholesterol

= Total cholesterol – [(Triglycerides/5) + HDL]

Statistical analysis: All values are expressed as themean+standard deviation. To see the inter-relationshipbetween two variables, the Pearson Product MovementCorrelation Coefficient was used while to test thesignificance of the correlation, the Student’s t test was used.

Results

The subjects were divided into 3 groups, young patients withMI, first-degree relatives of patients, and age- and sex-matched controls. Lipoprotein (a) level and lipid profile wereestimated in 50 patients with MI, 125 first-degree relativesof these patients and in 50 age- and sex-matched controls.Baseline characteristics of the patients, their relatives andcontrols are shown in Table 1. All patients with MI andcontrols were male. Smoking was the most prevalent riskfactor, present in 44% of young patients with MI ascompared to 36% of controls (p<0.05).

Lipoprotein (a) and lipid levels estimated in these groupsare shown in Table 2. Distribution of Lp(a) in the 3 groups(Table 3) shows that 30/50 (60%) young patients with MIhad Lp(a) levels between 20 and 30 mg/dl, while 101/125(82.5%) of their first-degree relatives had lower levels(between 10–20 mg/dl). By comparison, 34/50 (68%)

Table 1. Demographic profile of the patients, theirrelatives and controls

Parameters Young patients First-degree Controlswith MI relatives

Number 50 125) 50Age (years) 39+3.71 36+16) 34+6.9)Male 50 87 ) 50Female 0 38) 0Hypertension 10 (20). 6 (5) 8 (16)Diabetes 7 (14) 7 (5.6)0 6 (12)Smoking 22 (44)0 34 (27)0 18 (36)0Truncal obesity 10 (20)0 12 (9.6)0 6 (12)Family history of CAD 12 (24)0 125) Nil)

MI: myocardial infarction; CAD : coronary artery diseaseValues in parentheses are percentages

IHJ-867-00.p65 10/17/01, 1:59 PM464

Indian Heart J 2001; 53: 463–466 Isser et al. Lp(a) in Young Patients with MI 465

controls had still lower Lp(a) levels (below 10 mg/dl). Nopatient with MI had Lp(a) levels less than 10 mg/dl. Veryhigh levels of Lp(a) (>30 mg/dl) were found in 5/50 (10%)young patients with MI, 4/125 (3.2%) of their first-degreerelatives and none of the controls.

Statistical comparison of Lp(a) and lipid levels amongthe 3 groups shows that TC and LDL-cholesterol levels didnot dif fer significantly in the 3 groups (p value notsignificant). High-density lipoprotein cholesterol wassignificantly lower in young patients with MI and their first-degree relatives as compared to controls (p<0.001 forpatients v. controls and p<0.01 for relatives v. controls).Triglyceride levels were also significantly higher in patientsas compared to controls (p<0.05).

Lipoprotein (a) level in patients with MI and their first-degree relatives was significantly higher as compared tocontrols (p<0.001 and p< 0.05 for patients v. controls andrelatives v. controls, respectively).

Discussion

As compared to other ethnic groups, prevalence of CAD inthe young is much higher in Indians.14 Manifestation ofCAD in a more extensive form at a younger age despite arelatively low longitudinal burden of conventional riskfactors points to some other risk factors, presumablygenetically determined, which predispose our population

Table 3. Lp(a) levels in the study and control populations

Lp(a) Young patients First-degree Controlswith MI relatives

n=50 n=125 n=50

<10 mg/dl Nil 12(9.4) 34(68)10–20 mg/dl 15(30) 101(82.5) 16(32)20–30 mg/dl 30(60) 8(5.43) Nil>30 mg/dl 5(10) 4(2.67) Nil

Lp(a): lipoprotein (a); MI: myocardial infarction; mg/dl: milligram per deciliterValues in parentheses are percentages

to an increased risk at a much younger age. If these factorscan be identified, preventive measures can be instituted intime.

Prevalence of hypertension (20%) and diabetes mellitus(14%) was not very high in our study population and wassimilar to the incidence reported in previous studies.15

Smoking was the most prevalent risk factor in youngpatients with MI. Previous studies have shown familialclustering of CAD and its risk factors.16 However, we do notknow whether data from studies done on overseas Indianscan be extrapolated to those living in India and Indian dataregarding Lp(a) and lipid levels in young patients with MIand their relatives is lacking. In our study, we found thatthe difference in total- and LDL-cholesterol levels in youngpatients with MI and controls was statistically insignificant.High-density lipoprotein levels in patients and their first-degree relatives were significantly lower while TG levels weresignificantly higher in patients as compared to controls(p<0.05). Low HDL-cholesterol and high TG levels inIndians have been reported earlier.17

Lipoprotein (a) is considered to be an independent riskfactor for premature18 and multivessel CAD.19 Lipoprotein(a) consists of 2 different components, apolipoprotein (B)-100 which binds to LDL receptors and acts as an atherogenicprotein and apolipoprotein (A) which resemblesplasminogen and competes with the latter for binding tofibrinogen and fibrin monomer, thus acting as aprothrombotic agent. Thus, Lp(a) functions as a dualpathogen which is highly atherogenic and is alsoprothrombotic. The mean Lp(a) level in our study was22.24+5.4 mg/dl in patients, 13.88+5.19 mg/dl in theirfirst-degree relatives and 9.28+2.59 mg/dl in controls. TheCADI (Coronary Artery Disease in Asian Indians) studyshowed that the Lp(a) level in Asian Indians is high.Bhatnagar et al.16 also reported high mean Lp(a) levelsamong Asian Indians living in the UK and their siblings inIndia. Shaukat et al.20 found that mean Lp(a) levels werenearly double in sons of Asian Indians (age range 15–30years) as compared to sons of White parents of similar age19 mg/dl v. 10 mg/dl). A recent study by Gupta et al.21

showed that TC, LDL-cholesterol, HDL-cholesterol and TGlevels were not significantly different in cases of CAD andcontrols but Lp(a) levels were significantly higher in cases(p <0.04). A study by Gambhir et al.22 reported that lowHDL-cholesterol and high Lp(a) levels were independent riskfactors for premature CAD. These studies and our datasuggest the genetic nature of the Lp(a) trait and clusteringof high Lp(a) levels in families of patients with prematureCAD.

Conclusions: Our study suggests that smoking, high Lp(a),high TG, and low HDL levels are the most important risk

Table 2. Lp(a) and lipid levels in the study and controlpopulations

Parameters Young patients First-degree Controlswith MI relatives

Number 50 125 50Lp(a) (mg/dl) 22.24+5.4 13.88+5.19 9.28+2.59Total cholesterol (mg/dl) 191.43+44.07 178.93+34.65 173.45+52.21Triglycerides (mg/dl) 202.00+76.10 153.74+82.92 149.60+82.99LDL-cholesterol (mg/dl) 118.01+35.31 113.11+32.81 99.05+38.92HDL-cholesterol (mg/dl) 30.16+9.45 33.28+8.45 46.81+8.04

Lp(a) : lipoprotein (a); LDL: low-density lipoprotein; HDL : high-densitylipoprotein; mg/dl: milligram per deciliter

IHJ-867-00.p65 10/17/01, 1:59 PM465

466 Isser et al. Lp(a) in Young Patients with MI Indian Heart J 2001; 53: 463–466

factors for CAD in the young. First-degree relatives of youngpatients with MI also show high Lp(a), high TG and lowHDL levels. These findings have important preventive andtherapeutic implications. Extensive lifestyle modificationand therapeutic interventions should begin early and atlower levels of TC and LDL-cholesterol, particularly inpersons with a family history of premature CAD and inpersons with high Lp(a) levels.

References

1. Reddy KS. Cardiovascular disease in India. World Health Stat Q 1993;46: 101–107

2. Enas EA, Yusuf S, Mehta JL. Prevalence of coronary artery disease inAsian Indians. Am J Cardiol 1992; 70: 945–949

3. Chadha SL, Radhakrishnan S, Ramachandran K, Kaul U, GopinathN. Epidemiological study of coronary heart disease in urbanpopulation of Delhi. Indian J Med Res 1990; 92: 424–430

4. Jalowiec DA, Hill JA. Myocardial infarction in the young and inwomen. Cardiovasc Clin 1989; 20: 197–206

5. Lowry PJ, Lamb P, Watson RD, Ellis KE, Singh SP. Littler WA, et al.Influence of racial origin on admission rates of patients withsuspected myocardial infarction in Birmingham. Br Heart J 1991;66: 29–35

6. Krishnaswami S, Prasad NK, Jose VJ. A study of lipid levels in Indianpatients with coronary arterial disease. Int J Cardiol 1989; 24: 337–345

7. Enas EA, Garg A, Davidson MA, Nair VM, Huet BA, Yusuf S. Coronaryheart disease and its risk factors in the first-generation immigrantAsian Indians to the United States of America. Indian Heart J 1996;48: 343–353

8. Enas EA. Avoiding premature coronary artery disease deaths inAsians in Britain. Guidelines for pharmacological intervention areneeded. BMJ 1996; 312: 376

9. Dahlen CG. Lp(a) lipoprotein in cardiovascular disease. Atherosclerosis1994; 108: 111–126

10. Geneset J, McNamara JR, Ordovas JM, Jenner JL, Silberman SR,Anderson KM, et al. Lipoprotein cholesterol, apolipoprotein A1 andB and Lp(a) abnormalities in men with premature coronary arterydisease. J Am Coll Cardiol 1992; 19: 792–802

11. Rhoads GG, Dahlen G, Berg K, Morton NE, Dannenberg AL.Lipoprotein (a) as a risk factor for myocardial infarction. JAMA 1986;256: 2540–2544

12. Marquez A, Mendoza S, Hamer T, Glucek GJ. High Lp(a) in childrenfrom kindreds with parental premature myocardial infarction.Circulation 1989; 80 (Suppl): 1–97

13. Rifai N, Heiss G, Doetsch K. Lp(a) at birth in blacks and whites.Atherosclerosis 1992; 92: 123–129

14. Enas EA, Yusuf S, Garg A, Davidson L, Thomas T, Pearson TA.Lipoprotein (a) levels in Indian physicians: comparison with whiteand black physicians in the USA. Indian Heart J 1994; 46: 1

15. Pahlajani DB, Chawla MH, Kapashi KA. Coronary artery diseasepattern in the young. J Assoc Physicians India 1989; 37: 312–314

16. Bhatnagar D, Anand IS, Durrington PN, Patel DJ, Wander GS,Mackness MI, et al. Coronary risk factors in people from the Indiansubcontinent in west London and their siblings in India. Lancet 1995;345: 405–409

17. Hughes LO, Wojciechowski AP, Raftery EB. Relationship betweenplasma cholesterol and coronary artery disease in Asians.Atherosclerosis 1990; 83: 15–20

18. Scanu AM. Lipoprotein (a): a genetic risk factor for prematurecoronary heart disease. JAMA 1992; 267: 3326–3329

19. Wang XL, Tam C, McCredie RM, Wilcken DE. Determinants of severityof coronary artery disease in Australian men and women. Circulation1994; 89: 1974–1981

20. Shaukat N, Douglas JT, de Bono DP. Can physical activity explain thedifferences in insulin levels and fibrinolytic activity between youngIndo-origin and European relatives of patients with coronary arterydisease? Fibrinolysis 1995; 9: 55–63

21. Gupta R, Kastia S, Rastogi S, Kaul V, Nagar R, Enas EA. Lipoprotein(a) in coronary heart disease: a case control study. Indian Heart J 2000;52: 407–410

22. Gambhir JK, Kaur H, Gambhir DS, Prabhu KM. Lipoprotein (a) as anindependent risk factor for coronary artery disease in patients below40 years of age. Indian Heart J 2000; 52: 411–415.

IHJ-867-00.p65 10/17/01, 1:59 PM466

Indian Heart J 2001; 53: 467–476 Verma et al. Experience with AutoCapture™ Pacing 467

A Cardiac Evoked Response Algorithm ProvidingAutomatic Threshold Tracking for Continuous Capture

Verification: A Single-Center Prospective Study

Puneet K Verma, Jai K Sharma, Idris A Khan, Harinder K Bali,Jagmohan S Varma, Mandeep Bhargava, Yash P Sharma, Anil Grover

Department of Cardiology, Postgraduate Institute of Medical Education and Research, Chandigarh

Present-day multiparameter pacemaker programmingis sophisticated and time consuming. The industry has

provided solutions, incorporating automatic functions intothe latest pacemakers that may help to speed up patientfollow-up and optimize individual parameter adjustment.The development of an automatic pacemaker algorithmthat can independently verify pacemaker capture anddetermine pacing threshold is likely to provide an addedmargin of safety and help to prolong longevity of thepacemaker. One such algorithm is the ‘AutoCapture’ which

requires detection of an evoked response (ER) following thepacemaker depolarizing stimulus. It is configured to controlthe effectiveness of each pacing pulse, deliver a back-uppacing pulse, increase the output in case of ineffectivepacing and adjust the pulse amplitude every 8 hours to theactual pacing threshold.1

The recognition of a cardiac ER by the pacemaker ishampered by the polarization voltage at the electrode–tissueinterface.2 Typically, the peak amplitude of a paced ER isbetween 5 and 30 mV. The polarization is actually theafterpotential of the pacing pulse, which is in the order ofhundreds of millivolts and lasts for tens of milliseconds afterthe end of a pacing pulse.3 It often encroaches into the

Correspondence: Dr Puneet K Verma, 5177/3, MHC, Manimajra,Chandigarh 160101. e-mail: [email protected]

Original Article

Background: The AutoCapture™ algorithm as implemented in Regency® and Microny® pacemakers (PacesetterInc., Sylmar, CA, USA) provides beat-by-beat monitoring of capture based on proper detection of the evokedresponse, provides high output back-up pulse when loss of capture occurs, performs periodic threshold evaluationsand acquires the capture threshold data in a time-based event counter for later retrieval. The safety and efficacyof this algorithm was prospectively evaluated at a tertiary care hospital of north India.Methods and Results: Fifty-four patients (38 males, mean age 66±13 years) received a ventricular pacemakermodel Regency® SC+ with low polarization bipolar lead for high-grade atrioventricular block (n=42) and sicksinus syndrome (n=12). Evoked response and polarization signal were assessed initially at 24 hours postimplant,and follow-up measurements were systematically conducted at week 1 and months 1,3 and 6. Further evaluationof eligible patients was performed at 6-monthly intervals. Lead implantation parameters were optimum in allpatients. At 6 months, the algorithm was functional in 51 patients. The pacing threshold increased to 0.89±0.36 V(p<0.001) in the first month and stabilized thereafter. Significant saving of energy was accomplished by aconstant output safety margin of 0.3 V instead of the traditional 100%. While the evoked response signalremained stable throughout the study period, the potential signal increased significantly from 0.6±0.7 mV to1.0±0.6 mV (p<0.001) in the first month and remained steady subsequently. Back-up pacing in the event of exitblock was confirmed in all 25 patients who underwent a 24-hour Holter test. Based on the suggested sensemargins, ventricular undersensing was observed in 7 (28%) patients, the majority of whom had competitivecardiac rhythms. An elderly patient with pneumonic illness succumbed to pulmonary embolism at 6 months.Conclusions: This large single-center experience on AutoCapture demonstrates the success of this algorithm inlow-energy ventricular pacing without compromising the patient’s safety. (Indian Heart J 2001; 53:467–476)

Key Words: Pacemaker, AutoCapture, Heart block

IHJ-179-01.p65 10/17/01, 1:38 PM467

468 Verma et al. Experience with AutoCapture™ Pacing Indian Heart J 2001; 53: 467–476

native ER so that in order to reliably sense the ER, thepolarization artifact has to be minimized or abolished.4 Thisfeature labeled AutoCapture (Pacesetter Inc., St. JudeMedical Co., Sylmar, CA, USA) has been incorporated in theMicrony® and Regency® (SC+ and SR+) models, of whichthe latter are available for clinical use in India. In a single-center study, its effectiveness was evaluated prospectively inconjunction with a low threshold, low polarization bipolarlead.

The objectives of this study were: (i) to evaluate thepacing threshold characteristics of the Membrane E leadsover a 6-month follow-up utilizing the automatic capturethreshold capability and diagnostics of Regency® SC+ pulsegenerator; (ii) to assess the clinical performance of theAutoCapture™ algorithm as implemented in the ventriculardemand (VVI) pacing systems with low polarization leads;and (iii) to study the safety and efficacy of the AutoCapturefunction.

Methods

The study included 54 patients (36 males, 18 females)undergoing permanent pacemaker implantation forsymptomatic bradycardia due to a variety of causesincluding sick sinus syndrome (SSS) and high-gradeatrioventricular block (Table 1). The mean age of thepatients was 66 years. Clinical and 2-D transthoracicechocardiographic evaluation revealed an absence ofstructural heart disease in 25 patients (group I) while theother 29 patients had organic heart disease of multifactorialetiology (group II).

All patients received the VVI pacemaker Regency®

SC+2402L with the ventricular pacing lead Membrane E1450T as recommended by the manufacturer (PacemakerInc.). In the majority of patients, the physician’s judgmentto use ventricular-based and fixed-rate pacing unitsdepended on the economic affordability. All enrolled patientssigned an informed consent prior to the implantation. Thefollowing data were collected from the patients’ history andexamination: age, gender, indications for pacing, type ofarrhythmia, preimplant symptoms, concomitant cardiacconditions, and drug therapy at the time of implant.

Pacemaker: The VVI pacemaker Regency® SC+ providesseveral automatic functions summarized by theAutoCapture™ algorithm.1 The idea of AutoCapture is todetect in a bipolar sensing configuration the presence ofthe ER signal between 15 and 62.5 ms after a unipolarpacing pulse. If the pacemaker detects no ER, a safety pacingpulse of 4.5 V amplitude and 0.49 ms duration isimmediately delivered (Fig. 1). Besides the beat-to-beat safety

control, the pacing pulse is automatically adjusted to thepatient’s pacing threshold. The threshold search isautomatically activated after two consecutive losses ofcapture, resulting in the delivery of a back-up pulse, orevery 8 hours, or with the application of a magnet. Duringa threshold search, the pacemaker reduces the pulseamplitude in steps of 0.3 V while holding the pulse durationconstant until the pacing pulse becomes ineffective. Thecapture threshold is defined as the lowest amplitude thatresults in two consecutive captures. When two successivepaced stimuli fail to capture, the pulse amplitude isautomatically increased in 0.3 V steps until the capture isrestored. When the threshold search is complete, the systemautomatically sets the output to 0.3 V above the capturethreshold value. The device can monitor the evolution ofstimulation threshold over time at different samplingintervals, ranging from 2 s to 24 hours. In the freeze mode,data collection stops when 256 samples have been acquiredbut these remain stored in the memory until retrieved bythe physician at a follow-up visit. In the continuous mode,once 256 samples have been acquired, the system will deletethe oldest data while storing the most recent measurements.

A potentially life-threatening condition may result if lead

Table 1. Preimplant data and indications for pacing

Variables Value

Mean age (range), years 66±13 (6–90)Number of patients (males/females) 54 (38/16)Side of the implant (left/right) L30; R24Follow-up (months) 20.3±14.1

Structurally normal heart (2-D echocardiogram) 25Hypertension 25Diabetes mellitus 17Dilated cardiomyopathy 4Restrictive cardiomyopathy 2Aortic stenosis 1Coronary artery disease 1Hypothyroidism 1

Complete AV block 42Left bundle branch block 10Sinus arrest 6First-degree AV block 5Tachy–brady syndrome 4Atrial flutter 4Right bundle branch block 3Second-degree AV block 2Sinus bradycardia 2Atrial fibrillation 1

Each patient may have more than one indicationAV: atrioventicular

IHJ-179-01.p65 10/17/01, 1:38 PM468


and the titanium nitride tip are tissue tolerant, implyingthat only a minimum of fibrous tissue build-up occurs.

Implantation procedure and follow-up: All patientsreceived a Regency® SC+ model 2402L (Pacesetter Inc.), amultiprogrammable, single-chamber, cardiac pulsegenerator with AutoCapture function programmed to VVImode. The pacing system was implanted using standardendocardial implant techniques. Membrane E 1450 T leadswith connector IS-1 BI (VS.12) were exclusively used in thisstudy. A stringent implant protocol was followed and thepacing threshold, impedance, and spontaneous R-waveamplitude were measured using a standard pacing systemanalyzer (PSA). The pacing thresholds were determined ata pulse width of 0.42 ms, and stimulation impedance at anoutput of 5.0 V/0.42 ms. The minimum acceptable R-waveamplitude (bipolar intracardiac signal) was specified to 5 mV.On the following day, measurements of standard pacingand sensing thresholds were obtained followed by the ERsignal, ER sensitivity and polarization signal (PS) using thePacesetter’s Analyser Programmer System (APSµ) with the3264.ENG software. Thereafter, the AutoCapture functionwas switched on in eligible patients. The minimumacceptable value for the ER signal was 4.0 mV. Themeasurements were repeated in all patients at 1 week (timeof discharge from hospital) and at follow-up visits at 1,3and 6 months. Further evaluation in eligible patients wasperformed at 6-monthly intervals. At one-month follow-up, 24-hour Holter monitor studies were performed inselected patients (n=25) who gave consent for theinvestigation. These Holters were automatically andmanually analyzed for exit blocks, pauses, back-up pulsesand pseudo (fusion) beats. When noncapture occurred, itwas recorded whether or not there was an appropriateback-up pulse.

During the ER/PS assessment, the programmer turnsthe AutoCapture “off”, increases the pulse amplitude to 4.5 Vfor a series of paired pulses, and increases the pacing rate to100 ppm. The first pulse in each pair captures the heartand produces an ER and PS. The second pulse, delivered110 ms after the first, occurs when the myocardium isphysiologically refractory, which allows for an independentmeasurement of lead polarization. In simpler terms, ERsignal is the measure of the amplitude of myocardialdepolarization resulting from the pacemaker stimulus whilethe PS is a measurement of the residual electrical chargefrom a pacing pulse. The pacemaker was programmed toAutoCapture ‘on’ with the sensitivity and ER setting assuggested by the programmer. The device determines anER sense margin as close to 100% as possible, but minimally

Fig. 1. Illustration of AutoCapture™ algorithm. (A) Detection window forevoked response and polarization signal. (B) Delivery of back-up safety pulsein the absence of ER detection. (C) Holter illustration of successful back-uppulse capture (arrows).

polarization or afterpotential is high enough to overlap withthe ER signal. In the worst scenario, the pacing pulse isineffective, but the pacemaker detects the polarizationpotential that looks like an ER and, therefore, no back-uppulse is delivered. To obviate this, AutoCapture requirespacing leads that have low polarization. Also, the ERsensitivity—the level beyond which the pacemaker detectsthe ER—has to be individually programmed according tothe measured ER signal and the polarization potential.

Pacing lead: The Membrane E 1450T is a tined, bipolar,steroid-eluting pacing lead with an electrode surface areaof 3.5 mm2. The electrode tip consists of titanium coveredby titanium nitride. This is coated with a semipermeablemembrane, which is also used as a steroid carrier (acopolymer with 13 µg of dexamethasone). The membrane

A

B

C

IHJ-179-01.p65 10/17/01, 1:38 PM469


80%. To accomplish this, the amplitude of the signalresulting from the first pulse is divided by two. Of the availableER settings, the closest setting to this value is then comparedto the PS. AutoCapture is not recommended if the amplitudeof PS exceeds 60% of the proposed ER sensitivity. The pulseduration was set to 0.37 ms, and the stimulation thresholdversus time counter was programmed in the ‘freeze’ modeat a sampling frequency of 1-hourly for 7 days, 6-hourlyfor 1 month, 12-hourly for 3 months, and 24-hourly every6 months. The programmed and measured data along withthe test results were obtained, recorded and printed at eachevaluation session. All measurements for this study wereperformed at least twice in the supine position, and themean of the two measurements is given. However, therewere only minor differences between the two measurements.

Statistical analysis: The descriptive statistics in the reportare expressed as mean±standard deviation (unless statedotherwise). The difference between the paired variables wascalculated using appropriate t-tests. A p value of <0.05was considered significant.

Results

The lead parameters at implantation are given in Table 2.In 19 patients, the R wave could not be assessed at the timeof implant. At follow-up, the lead impedance and sensingparameters remained steady while the pacing thresholds

increased significantly in the first week (0.5±0.17 V v.0.78±0.39 V; p<0.001), with a further increase at onemonth (0.78±0.39 V v. 0.89±0.36 V; p<0.01) beforestabilizing (Table 3). The results of the automatic capturefunction correlated well with the capture thresholdmeasured using the Vario technique (Fig. 2).

Table 2. Lead implantation parameters (n=54)

Parameters Value

Pacing threshold (V) at pulse width 0.42 ms 0.50±0.20

Lead impedance (ohm) 651.7±143.6

R wave (mV) (n=35) 12.1±5.40

Table 3. Pacing and sensing parameters at follow-up

Parameters Day 1 Week 1 1 month 3 months 6 months 1 year 2 years

Threshold (V) at PW 0.37 ms 0.50±0.17* 0.78±0.39** 0.89±0.36## 0.89±0.37 0.87±0.41. 0.90±0.46 0.81±0.25Vario threshold (V) at PW 0.37 ms 0.49±0.19* 0.73±0.34***.0.85±0.35##00 0.84±0.36 0.85±0.40. 0.87±0.40 0.78±0.26Impedance (ohm) (n=54) 687±118* 673±1130. 692±10400 722±114 725±116# 741±114 742±127.R-wave sensitivity (mV) 5.4±3.0* 5.4±3.00 5.9±3.100 5.5±2.9 5.5±3.0. 5.5±2.7 4.2±1.8.R-wave sensing margin (%) 159±42.*. 157±3400 161±41000 159±450 160±360. 165±410 163±370.ER signal (mV) 11.0±5.50. 11.1±4.900 11.1±5.0000 10.9±4.70 10.7±4.90. 11.1±4.70 10.5±5.30PS (mV) 0.6±0.7* 0.8±0.7#. 1.0±0.6**. 1.1±0.5 1.1±0.5. 1.2±0.4 1.2±0.5

PW: pulse width; ER: evoked response; PS: polarization signal* v. # p=NS; ** v. ## p<0.01* v. ** and # v. ** p<0.001

Fig. 2. Evolution of stimulation threshold during follow-up (AutoCapture™v. Vario test).

The ER signal remained stable with minimal variation(11.0±5.5 mV at implant, 10.7±4.9 mV at 6-month, and11.1±4.7 mV at one-year follow-up). Two patientsdeveloped drug-induced renal failure and showed a steepdecline in ER signal with concomitant rise in pacingthresholds in the first week, but surprisingly never regainedthe amplitude despite correction of uremia. One of themsuccumbed to pulmonary embolism during a prolongedpneumonic illness at 6 months, with the pacing thresholdsonce again rising progressively and the ER declining furtherprior to death (Fig. 3).

The PS showed a significant increase at 1 month(0.6±0.7 mV v. 1.0±0.6 mV; p<0.001) and stabilizedthereafter (Fig. 4). The ER sensitivity throughout the follow-

IHJ-179-01.p65 10/17/01, 1:38 PM470


up period varied between 3.8±1.5 and 5.6±2.3 mV (meansensing margin 96%–156%). In three patients, theAutoCapture algorithm could not be switched on at day 1because of low ER signal (ER 2.0–3.3 mV, n=2) and highPS (1.6 mV, n=1). The pacemaker recommended not toprogram the automatic threshold tracking ‘on’ in onepatient permanently and in two patients intermittently. Afourth patient lost AutoCapture in the first week, whichpersisted into the first month (ER 3.1–3.3 mV, PS 1.0–1.1mV), but subsequently regained the function as the PSstabilized at <1.0 mV despite a low ER (3.3–3.5 mV). Twopatients had a low-to-borderline ER signal (3.3–4.6 mV)throughout the follow-up period but continued to showgood AutoCapture function because of low PS (<1.0 mV).Hence the AutoCapture function could continuesuccessfully even in patients with low ER signal providedPS remained low (<1.0 mV). At 6 months, the algorithmwas consistently effective in 51 (94%) of 54 patients.

There was no difference in the ER signal according togender and side of implant (Tables 4 and 5). However, ingroup II, the ER signal was significantly greater than thatin group I (12.3±5.7 mV v. 9.1±3.4 mV; p<0.05). Similarly,the ER sensitivity was numerically greater in group II bothat the time of implant (4.2±1.6 mV v. 3.4±1.2 mV; p<0.05)and at 6 months (6.5±2.8 mV v. 4.7±1.1 mV; p<0.01). Acomparison of the clinical and pacing parameters of thetwo study groups is given in Table 6.

In eligible patients with the AutoCapture ‘on’ the energyconsumption was extremely low with the batteryparameters remaining stable over time (Table 7, Fig. 5).Battery impedance can be used indirectly as an indicatorfor total current consumption over time. In all patients, themeasured battery voltage remained constant at 2.78 V and

Fig. 3. Progressive increase in pacing threshold in a terminally ill patient,finally resulting in high voltage (4.5 V) stimulation.

Fig. 4. Evolution of sensing and pacing parameters in AutoCapture-eligiblepatients.

Table 4. Evoked response (ER) signal (mV) in mean and women


Men (n=38) 10.8±5.5 11.3±4.6 11.2±4.8 11.1±4.7 10.9±5.1 10.1±4.3 10.3±5.3Women (n=16) 11.7±5.6 10.5±5.5 10.7±5.7 10.2±4.6 10.5±4.7 12.9±5.2 15.6±3.0

N 54 51 51 51 51 38 18

p=NS for comparison at all times

Table 5. Evoked response (ER) signal (mV) in left and right implants


Left (n=38) 11.9±6.3 11.0±4.9 11.3±5.1 11.3±4.7 11.0±5.1 10.9±3.9 12.5±5.6Right (n=16) 9.9±4.1 11.1±4.9 10.9±5.0 10.3±4.6 10.4±4.9 10.9±5.1 10.7±5.4

N 54 51 51 51 51 38 18

p=NS for comparison at all times

IHJ-179-01.p65 10/17/01, 1:39 PM471


the impedance was in the preimplant range (<1000 ohm)at the last recorded follow-up visit. The term 2×T is used todenote a state had the AutoCapture function not beenswitched on and the pulse amplitude of the pacemakerbeen manually programmed to twice the pacing thresholdas is the conventional practice. Autobatch refers to a set ofautomatically programmed values that the pacemakerreverts to if the AutoCapture function is sporadically turnedoff. Factory settings yield those values where the pacemakerhas never been programmed after implantation, a relativelyrare condition in patients who follow-up regularly at tertiarycare hospitals having pacemaker clinics. A similar statisticaltrend emerged in the pacing parameters in all modes ingroups I and II. There was no significant difference in theparameters between the groups (Table 8).

Twenty-four Holter monitor recordings were obtainedin 25 patients (Table 9). In a total recorded time of 580hours, there were 200 episodes (0.4% of all paced events)of loss of capture by initial impulse, either spontaneous orduring threshold checks, all of which were backed up by ahigh voltage output (Fig. 1C). In 7 patients (28%), therewere 24 episodes of back-up pulse delivery during native orfusion beats (0.25%) of all paced events (Fig. 6). There wasno incidence where delivery of back-up pulses occurreddespite capture by the initial impulse, as indicative of ERundersensing. The likelihood of multiple occurrences ofpseudofusion beats had a correlation with the presence ofspontaneous cardiac activity.

Table 6. Clinical and pacing parameters in group I andgroup II

Parameters Group I (n=25) Group II (n=29)

Age (years) 64.1±6.50 67.8±9.4Gender 17 M; 8 F 21 M; 8 FDiagnosis AVB 19; SSS 6 AVB 23; SSS 6Sidedness (left/right) L12; R 13 L 18; R 11Implant pacing threshold (V) 0.46±0.19 0.48±0.20Implant impedance (ohm) 0.650±105.3 653.3±140.2Implant R wave (mV) 11.8±4.40 12.4±6.10Follow-up (months) 21.8±15.4 19.0±13.2Pacing threshold (V)

at 6 months 0.90±0.49 0.84±0.33Pulse amplitude (V)

at 6 months 1.22±0.49 1.15±0.33Pulse current (mA)

at 6 months 1.62±0.63 1.72±0.71Pulse charge (µC)

at 6 months 0.66±0.43 0.61±0.28Pulse energy (µJ)

at 6 months 0.85±0.73 0.78±0.58Battery current (µA)

at 6 months 3.21±0.44 3.17±0.31ER signal (mV) at 6 months 9.1±3.4* 12.3±5.7*PS (mV) at 6 months 1.2±0.4. 1.1±0.5R sensitivity (mV)

at last follow-up 4.90±3.4.. 6.1±2.9(mean sensing margin) (162%) (165%)

ER sensitivity (mV)(Implant – last follow-up) 3.4±1.2*–4.7±1.1** 4.2±1.6*–6.5±2.8**(mean sensing margin) (96%) (150%) (96%) (161%)

AVB: atrioventricular block; SSS: sick sinus syndrome*p<0.05; **p<0.01

Table 7. Energy parameters in AutoCapture on patients (n=51) at 6-month follow-up

Parameters AutoCapture (6 months) 2×T Autobatch Factory setting

Pulse amplitude (V) 1.19±0.42 1.91±0.89 2.82±0.71 3.98±0.16Pulse current (mA) 1.68±0.66 2.75±1.16 4.0±1.18 5.82±1.28Pulse charge (µC) 0.63±0.36 0.93±0.40 1.40±0.46 2.03±0.43Pulse energy (µJ) 0.82±0.66 2.18±2.26 4.05±2.51 7.68±1.46Battery current (µA) 3.18±0.37* 3.82±1.22* 5.0±2.51 7.65±1.62

*p<0.01 for AutoCapture v. 2×T; for all other parameters, p<0.001p<0.001 for 2×T v. Autobatch for all parameters

Table 8. Energy parameters in group I and group II patients at 6-month follow-up

Parameters AutoCapture (6 months) 2×T Autobatch

Group I Group II Group I Group II Group I Group II(n=25) (n=26) (n=25) (n=26) (n=25) (n=26)

Pulse amplitude (V) 1.22±0.49 1.15±0.33 1.94±1.01 1.86±0.71 2.90±0.77 2.73±0.62Pulse current (mA) 1.62±0.63 1.72±0.71 2.66±1.22 2.83±1.12 3.98±0.91 4.02±1.4Pulse charge (µC) 0.66±0.43 0.61±0.28 0.89±0.38 0.96±0.41 1.42±0.46 1.39±0.47Pulse energy (µJ) 0.85±0.73* 0.78±0.58 2.43±2.81* 1.90±1.58 4.39±2.78 3.71±2.14Battery current (µA) 3.21±0.44* 3.17±0.31* 3.97±1.55* 3.69±0.84 5.36±3.23* 4.66±1.51

*p<0.01 for AutoCapture v. 2×T; for all other parameters in groups I and II, p<0.001p=NS for 2×T v. Autobatch; for all other parameters in groups I and II, p<0.001

IHJ-179-01.p65 10/17/01, 1:39 PM472


Fig. 5. Energy consumption in AutoCapture-eligible patients. (A) Pulseparameters. (B) Battery parameters. At 6 months, an elderly female patient inflicted with a

pneumonic illness succumbed to pulmonary embolism(Fig. 3). In another patient, a microdisplacement of the leadwas suspected since a persistently high voltage stimulationwas observed despite normal pacing parameters during theentire follow-up (Fig. 7). The other adverse events aresummarized in Table 10.

Fig. 6. (A) Intermittent high-voltage stimulation secondary to R-waveundersensing. (B) Holter illustration of ventricular undersensing (arrows).

Table 9. Holter monitoring data (n=25)

Events Value

Total record time 580 hoursTotal QRS complexes recorded 2 520 000Total paced QRS complexes* 2 088 000 (83% of total QRS)Loss of capture (200 episodes) 8352 (0.4% of paced QRS)R-wave undersensing

(n=7, 24 episodes) 5200 (0.25% of paced QRS)ER undersensing Not recorded

*Excludes fusion, pseudofusion and native beats

Fig. 7. Lead microdisplacement: persistent high-voltage stimulation in apatient with normal AutoCapture™ parameters.

Table 10. Summary of adverse events

Adverse events Number

Death (fatal pulmonary embolism) 1Embolic stroke 1Lead displacement 1Pacemaker syndrome 3Pacemaker not needed 1Pocket infection 1

IHJ-179-01.p65 10/17/01, 1:39 PM473


Discussion

The concept of a threshold tracking pacemaker wasproposed by Preston and Bowers almost three decades ago.5

Ever since, various investigators have studied thedepolarization signal emanating from ventricular captureand tried to differentiate it from the polarization artifact.2,3,4,6

The first reported clinical experience with AutoCapture wasan Italian multicenter study (PACEMATE Study Group)which evaluated 54 patients implanted with the Microny®

SR+2425T SSIR pacemaker at 19 centers.7 The bipolarMembrane leads used in this study were either impregnatedor not with steroid. With both these leads, the evolution ofpacing thresholds was identical over a 6-week period.

The European Microny Study Group systematicallyfollowed up 113 patients from 16 centers for 1 year.1 Allpatients underwent a 24-hour Holter recording at 1-monthfollow-up that confirmed total reliability of this algorithmwithout any exit block. The acute and chronic thresholdsmeasured by Vario and AutoCapture tests correlated overthe study period, and the measured ER signal also remainedstable over time.

The present study confirms the reliability and effectivenessof the combination of Membrane leads and Regency® SC+pulse generators. Over a mean follow-up of 20 months,stable and adequate sensing and ER thresholds were foundalong with low capture thresholds, and the ability toautomatically program a low output, maximizing the systemlongevity without compromising patient safety. At 6 months,the algorithm was consistently effective in 51 of 54 patients.It is well known that myocardial capture and sensingthresholds evolve over time and may be affected by severalfactors.8,9 It is also clear that there is no correlation betweenacute and chronic stimulation thresholds, due to variousfactors that are involved in the maturation of the tissue–electrode interface.10 The increase in stimulation thresholdduring the first several weeks is related to the developmentof a conductive but nonexcitable fibrotic capsule thatencases the tip of the electrode. After 3–4 weeks, theinflammation subsides and the threshold may subsequentlydecline, remain stable, or increase depending on the chronicforeign body response at the electrode–tissue interface.11

In this study there was a statistically significant differencebetween the threshold at implant and week 1 (p<0.001)with a further significant rise at month 1 (p<0.01) afterwhich there were no significant threshold modifications.The mean Vario threshold was marginally lower than theAutoCapture threshold which is consistent with thepacemaker design, i.e. capacitor’s output is 0.1 V higherthan the marked value during the Vario test.12 A margin of

1.5–2 times with respect to the voltage threshold isrecommended to safeguard against the expected butunpredictable waxing and waning thresholds to minimizethe risk of exit block.13 This happens to be the standardpractice in pacemaker programming, but it producesexcessive battery current drain and rarely may result inloss of capture if the threshold exceeds the programmedoutput. The AutoCapture values of pulse and batteryparameters were much lower than the 2×T and Autobatchvalues (Tables 7 and 8) (p<0.01–0.001) which translatedinto lower pulse energies required to pace the heart alongwith lower battery current drain. This predicts extendedlongevity of the device. However, the current consumptionfor stimulation was measured at each visit; and thepercentage of pacing and back-up ratio were not takeninto account. Therefore, the current consumption forstimulation may vary between patients and the internalcurrent drain should also be added when estimating thetotal current consumption. The battery impedance can alsobe used as an indirect parameter of total currentconsumption over time. All our patients had the batteryimpedance in the preimplant range (<1000 ohm) at theirlast follow-up visit.

R-wave sensing remained stable in the patients over time,there being no significant difference between or within thegroups. Similarly, the ER signal remained steady on follow-up, but it was significantly greater in group II as was thenumerical value of ER sensitivity. Since group II comprisedpatients with hypertrophied ventricles (on 2-Dechocardiography), it is possible that the amplitude of ERsignal is related to the muscle mass. Interestingly, the ERdeclined steeply in the first week in two patients whodeveloped drug-induced renal failure and never reallyrecovered despite reversal of the uremic state. Similarly, theER showed a further progressive decline in one of thesepatients who developed a prolonged pneumonic illness andsuccumbed to pulmonary embolism at 6 months. Theseevents possibly indicate the adverse effects of uremia andpulmonary hypertension on right ventricular myocardialdepolarization. While the AutoCapture algorithm could notbe switched ‘on’ in 3 patients at day 1 because of low ERsignal and high PS, there were an equal number of patientswho had low ER but also low PS which allowed thealgorithm to function successfully. This is in agreement withthe concept that the main reason not to activateAutoCapture in these patients is primarily the increase inPS and not the decrease in ER amplitude.14,15 In a recentstudy involving 398 patients from 42 centers, the ERexhibited a small increase over a mean period of 1 year.16

IHJ-179-01.p65 10/17/01, 1:39 PM474


However, that did not result in an improvement of theAutoCapture function in patients with inadequate valuesat implant. Approximately 10% of patients could not haveAutoCapture enabled due to suboptimal ER/PS signaldespite satisfactory R wave and pacing thresholds.Unfortunately, patients with low ER signal cannot beidentified in advance by their electrical and clinical data,since at present, it cannot be determined by any PSA andthere is no known parameter that can help optimize theamplitude. As the ER derives from unipolar ventricularpacing, we assumed that the parameters that relate to thepacing pulse could be beneficial. However, right- and left-sided pacing pulses did not influence the ER amplitude andthe pacing threshold, and the pacing impedance did notcorrelate with the ER signal. Also, the ER was not affectedby gender. These observations are in agreement with arecent study.17

Our Holter assessment of the 25 select patients was alaborious exercise, since the Holter system lacked thesoftware for advanced pacemaker analysis, and the analysiswas carried out manually by two different investigators.The total number of paced QRS complexes formed morethan 80% of the total QRS complexes. All episodes of loss ofcapture were followed by high voltage stimulation (Fig. 1C),and ventricular undersensing was seen in 28% of thepatients which constituted 0.25% of all paced beats(Fig. 6B). This study did not find any Holter evidence of ERundersensing. It has been shown in earlier studies that morethan 20% of pseudofusion beats induce back-up pulsesand these pseudofusion beats account for the majority(>85%) of all back-up pulses.1 In a randomized study withRegency® SC+ and SR+ pacemakers, 162 patients werefollowed up at 27 Spanish centers for 6 months. About10% of the patients with the algorithm showed high voltagestimulation, half of whom showed a high number of fusionand pseudofusion beats.18 Similarly, in the North Americanmulticenter study, the Holter assessment in 51 consecutivepatients revealed high voltage back-up pacing in the eventof an exit block in all. The exit blocks constituted 0.4% of allpaced events. The incidence of R wave and ER undersensingwas 0.3% and 0.001% of all paced events, respectively.16

The other causes of high voltage stimulation include poorER signal, high capture threshold and microdisplacementof lead, which were also observed in this study.

We report our experience with the AutoCapture™algorithm in the Regency® SC+ pulse generator, thatprovided automatic confirmation of ventricular capture,adjustment of output and continuous threshold monitoring.It functioned in providing safe low-output pacing in themajority (>90%) of patients during a limited follow-up

period. The algorithm decreased the battery drain comparedto the AutoCapture ‘off ’ settings, thereby prolonging theservice life of the generator. The high energy back-up pulseduring exit block provided an additional safety feature overconventional devices. However, this algorithm could notdistinguish pseudofusion beats from the genuine loss ofcapture while delivering back-up pulses. The ER signalremained stable throughout the study period, while the PSincreased significantly in the first month and remainedsteady subsequently. Importantly, even though PSAs thatmeasure ER signal are yet to become available, a standardimplantation technique with the recommended leadprovided an acceptable rate of maintenance of theAutoCapture function. To the best of our knowledge, this isthe largest single center experience being reported.

References

1. Clarke M, Liu B, Schuller H, Binner L, Kennergren C, Guerola M, et al.Automatic adjustment of pacemaker stimulation output correlatedwith continuously monitored capture thresholds: a multicenterstudy. Pacing Clin Electrophysiol 1998; 21: 1567–1575

2. Curtis AB, Vance F, Miller-Shifrin K. Characteristic variation inevoked potential amplitude with changes in pacing stimulusstrength. Am J Cardiol 1990; 66: 416–422

3. Feld GK, Love CJ, Camerlo J, Marsella R. A new pacemaker algorithmfor continuous capture verification and automatic thresholddetermination: elimination of pacemaker afterpotential utilizinga triphasic charge balancing system. Pacing Clin Electrophysiol 1992;15: 171–178

4. Brouwer J, Nagelkerke D, De Jongste MJ, Boute W, Den Heijer P, LieKI. Analysis of the morphology of the unipolar endocardial-pacedevoked response. Pacing Clin Electrophysiol 1990; 13: 302–313

5. Preston TA, Bowers DL. Clinical applications of the thresholdtracking pacemaker. Am J Cardiol 1975; 36: 322–326

6. Baig MW, Cowan JC, Perrins EJ. Comparison of unipolar and bipolarventricular-paced evoked responses. Br Heart J 1992; 68: 398–402

7. Sermasi S, Marconi M, Libero L, Moracchini PV, Rusconi L, Mininno A,et al. Italian experience with Autocapture in conjunction with amembrane lead. Pacing Clin Electrophysiol 1996: 19: 1799–1804

8. Preston TA, Judge RD. Alteration of pacemaker threshold by drugand physiological factors. Ann N Y Acad Sci 1969; 167: 686–692

9. Sowton E, Barr I. Physiological changes in threshold. Ann N Y AcadSci 1969; 167: 679–685

10. Stokes K, Borzin G. The electrode–bio interface: stimulation. In:Barold SS (ed). Modern cardiac pacing. Mount Kisco, NY: FuturaPublishing Co; 1985. pp. 33–77

11. Bobyn JD, Wilson GJ, Mycyk TR, Klement P, Tait GA, Pillar RM, et al.Comparison of the porous-surfaced with a totally porous ventricularendocardial pacing electrode. Pacing Clin Electrophysiol 1981; 4:405–416

12. Lindberg J. Description of function Microny/Regency. PacesetterAB; 1997

13. Astrinsky EA, Furman S. Pacemaker output programming formaximum longevity. Pacing Clin Electrophysiol 1983; 16: 105–110

14. Kacet S, Jarwe M. Klug D. Evaluation of the compatibility of variousbipolar pacing leads with Autocapture [Abstr]. Pacing ClinElectrophysiol 1998; 21: 825

15. Lau C, Nishimura SC, Philippon F. Polarization signal and evoked

IHJ-179-01.p65 10/17/01, 1:39 PM475


response characteristics in current endocardial electrodes [Abstr].Pacing Clin Electrophysiol 1998; 21: 841

16. Lau C, Cameron DA, Nishimura SC, Ahern T, Freedman RA,Elbenbogen K, et al. A cardiac evoked response algorithm providingthreshold tracking: a North American multicenter study. Pacing ClinElectrophysiol 2000; 23: 953–959

17. Schuchert A, Ventura R, Meinertz T. Automatic threshold tracking

activation without the intraoperative evaluation of the evokedresponse amplitude. Pacing Clin Electrophysiol 2000; 23: 321–324

18. Madrid AH, Olague J, Cercas A, Ojo JLD, Munoz F, Moro C, et al. Aprospective multicenter study on the safety of a pacemaker withautomatic energy control: influence of the electrical factor onchronic stimulation threshold. Pacing Clin Electrophysiol 2000; 23:1359–1364

IHJ-179-01.p65 10/17/01, 1:39 PM476

Indian Heart J 2001; 53: 477–480 Kar et al. Single-Pass Physiological Pacing 477

Long-Term Performance of Single-PassPhysiological Pacing

AK Kar, Debabrata Roy, Manoranjan Mondal, Indranil DuttaDepartment of Cardiology, IPGMER, Kolkata

Atrioventricular (AV) synchronous pacing, with orwithout rate adaptivity, is currently favored in cases

with symptomatic AV nodal (AVN) disease. This mode ofpacing offers several advantages: (i) it improves cardiacoutput; (ii) decreases the chances of development ofpacemaker syndrome; and (iii) prevents atrial fibrillationand other atrial tachyarrhythmias. This last effect protectsagainst systemic embolization and thus prolongs lifeexpectancy as well as improves the quality of life.1–6 Thoughthe cost of implantation is considerable, the long-termtherapeutic costs may be reduced by AV synchronouspacing.7

Atrioventricular synchronous pacing can be done byeither a dual-lead DDD/DDDR or by single-lead VDD/DDD(in patients with sick sinus syndrome) mode. Double-leadsystems have many disadvantages—implantation is

technically difficult as dual venous access is required; thecost is high because two leads are used; and atrial leaddislodgment and wastage are frequent. Atrial lead wastageoccurs due to fracture of the wire retaining the J-shape ofthe lead. Single-lead pacing, therefore, has gainedpopularity.8

Atrial pacing by a single-pass lead using theconventional pulses, however, results in a high pacingthreshold, and can lead to diaphragmatic contraction dueto phrenic nerve stimulation in a significant number ofcases. To tackle this problem, two atrial rings are currentlyused in the single lead. The proximal ring emits a positivepolarity wave and, simultaneously, the distal ring emits anegative polarity wave of the same amplitude and width—these two waves overlap each other. These two separateimpulses stimulate the atrial myocardium concurrently,thereby decreasing the pacing threshold. This technique iscalled overlapping biphasic impulse (OLBI) stimulation.9

The current prospective study was undertaken toCorrespondence: Professor AK Kar, 82A, Chowringhee Road, Kolkata700020, West Bengal

Original Article

Background: Single-pass physiological pacing has several advantages over dual-lead physiological pacing. Thepresent study evaluated the long-term performance of single-pass pacing using the overlapping biphasic impulsestimulation technique.Methods and Results: A total of 30 patients with single-pass VDD pacing and 8 patients with single-pass DDDCpacing were followed up for 1 year by basal and magnet electrocardiograms and real-time telemetry. All thepatients showed satisfactory atrial sensing and pacing capture threshold. The atrial sensing thresholds at implantand at 1 month, 3 months, 6 months and 12 months of follow-up were 2.5±0.67 mV, 1.6±0.6 mV, 1.1±0.5 mV,1.0±0.5 mV and 1.0±0.04 mV, respectively. The corresponding values for atrial pacing threshold at a pulse waveof 0.5 ms were 2.5±1.0 V, 4.4±0.9 V, 3.8±1.2 V, 3.6±1.4 V and 3.8±1.4 V. Of the patients with DDDC pacing,88% showed stable pacing capture in the supine position, 75% in the upright position and 62% in both positions.Diaphragmatic contraction was seen in 25% of cases with DDDC pacing. No such event was seen in patients withVDD pacing.Conclusions: Single-pass pacing is safe, technically easy and cheap as compared to dual-lead systems. However,it would be prudent to recommend DDDC pacing in patients who require predominantly VDD pacing and onlyoccasionally atrial pacing, as the latter showed a low percentage of stable atrial pacing capture in both uprightand supine positions as well as a significant percentage of diaphragmatic contraction. (Indian Heart J 2001;53: 477–480)

Key Words: Pacemaker, Physiological pacing, Heart block

IHJ-159-01.p65 10/17/01, 1:26 PM477

478 Kar et al. Single-Pass Physiological Pacing Indian Heart J 2001; 53: 477–480

assess the long-term results of OLBI single-lead AVsynchronous pacing and single-lead VDD pacing.

Methods

A total of 145 patients (age range 26–70 years) withcomplete or intermittent AV block, who attended theoutpatient or emergency department of the cardiology unitof SSKM Hospital between January 1997 and December2000, were enrolled for the study. All the patients underwent24-hour Holter monitoring and exercise testing to detectany chronotropic incompetence. The exclusion criteriawere: (i) chronotropic incompetence (i.e. achieving <100beats/min with exercise); and (ii) persistent atrial fibrillation.

After the screening tests, 50 patients were excluded fromthe study as they could not fulfil the inclusion criteria andreceived VVI/VVIR pacing systems. Another 57 patients alsoreceived VVI/VVIR pacing systems as they could not affordthe cost of the single-pass system and were excluded fromthe study. Of the remaining 38 patients (30 males and 8females) finally included in the study, 30 patients had noevidence of sinus node disease and were taken up for VDDpacing with a single lead. The remaining 8 patients hadsinus node disease and therefore underwent DDD pacingwith an OLBI single lead (DDDC mode). The pacemakersystems used were—(A) for VDD: (i) Prodigy 8168(Medtronic) (n=12), (ii) Dromos (Biotronik) (n=4), and (iii)Advent 2060 LR (Pacesetter) (n=14); and (B) for DDDC:Eikos SLW (Biotronik) (n=8).

The single-pass lead was introduced through the rightcephalic vein in 29 patients. This procedure failed in theremaining 9 patients and the lead was introduced throughthe right subclavian vein using a percutaneous leadintroducer. The distal end of the single-pass lead waspositioned at the right ventricular apex so as to achieve apacing threshold of less than 1 V and an R wave of morethan 5 mV. The floating atrial rings were then positioned ateither high- or mid-right atrium, with an adequate loop sothat the “p” wave amplitude was more than 0.5 mV duringboth inspiration and expiration. The position with the lowestcapture threshold and highest “p” wave sensing wasselected. The lead was then fixed by a ligature on the sleeveand sutured. It was connected with the correspondingpacemaker (VDD or DDDC) and placed in the subcutaneouspocket. After implantation, all the patients were monitoredfor 24 hours to detect any sensing or pacing failure. Theywere discharged after removal of stitches on postoperativeday 7 and advised follow-up at the pacemaker clinic.

Follow-up: Each patient was followed up at 1 month,

3 months, 6 months and 12 months after pacemakerimplantation with basal and magnet electrocardiograms(ECGs) taken at each visit. In every patient, the followingdata were actively recruited by real-time telemetry:(i) ventricular sensing and pacing threshold; (ii) atrialsensing threshold; and (iii) atrial pacing threshold (only inpatients with DDDC pacing). These observations were takenin both the supine and sitting positions.

Statistical analysis: All the values were expressed asmean±SD. A p value <0.05 was taken as significant whenthe follow-up data were compared with the acute implantdata.

Results

All 38 patients were followed up postprocedure for 12months. The mean (±SD) values of the acute implantthreshold and real-time telemetry data at each follow-upare presented in Tables 1 and 2.

During follow-up, the atrial sensing threshold showed asignificant drop at 1 month, continued falling until a nadirwas reached at 3 months and thereafter stabilized at asignificantly lower value throughout the rest of the follow-up period.

Table 2. Real-time telemetry data on follow-up

Threshold Implant 1 month 3 months 6 months 12 months

AtrialAtrial sensing 2.5±0.67 1.6±0.6* 1.1±0.5* 1.0±0.5* 1.0±0.04*

(mV)Atrial pacing at 2.5±1.0 4.4±0.9* 3.8±1.2 3.6±1.4 3.8±1.4

PW 0.5 ms (V)Ventricular

Ventricular 8.0±2.3 8.0±2.4 7.8±2.3 7.9±2.1 7.8±2.3sensing (mV)

Ventricular 0.8±0.34 2.4±0.40 1.2±0.32 1.3±0.32 1.3±0.34pacing (V)

PW: pulse width*p<0.05

Table 1. Acute pacemaker implant thresholds

A. VentricularR wave 11.03±3.22 mVCapture threshold 0.63±0.13 VCurrent 1.06±0.37 mA

B. Atrial“p” wave 2.5±0.37 mVAtrial capture 2.61±1.1 V

IHJ-159-01.p65 10/17/01, 1:27 PM478

Indian Heart J 2001; 53: 477–480 Kar et al. Single-Pass Physiological Pacing 479

The atrial pacing capture threshold rose significantlyabove the acute threshold at 1 month, but thereafter stabilizedat a lower value from 3 months onwards. However, thislowering of the pacing threshold value was not significantwhen compared with the acute implant data (Table 2).

In 7 out of 8 patients (88%) with DDDC pacing, at least80% of the atrial spikes induced a “p” wave in the follow-upECGs, thereby satisfying the North American Society ofPacing and Electrophysiology (NASPE) criteria of stablepacing capture of the atrium. In 2 out of 8 patients (25%)with DDDC pacing, diaphragmatic contraction was notedwith atrial spikes. No such event was seen in patients withsingle-pass VDD pacing.

The effect of atrial capture in the supine and uprightpositions in patients with DDDC pacing is presented in Fig. 1.Seven out of 8 patients (88%) were responsive to atrialpacing in the supine position throughout the 1-year follow-up period. Five out of 8 patients (62%) were responsive toatrial pacing in the upright position in the immediatepostimplant stage and at 3 months, but from 6 monthsonwards, 6 patients (75%) became responsive to atrialpacing in this position. Five out of 8 patients (62%) wereresponsive to atrial pacing in both the supine and uprightpositions throughout the 1-year follow-up period.

Adequate ventricular pacing was seen in all patients inboth the VDD and DDDC modes (Table 1).

The conventional single-pass lead results in a high pacingthreshold and significant diaphragmatic contraction.11 Thisproblem is circumvented by using the OLBI principle.9

In the present study, though the acute atrial pacingthreshold was high in the DDDC mode, the chronic pacingthreshold was low. Adequate ventricular tracking of theatrial rate was achieved in all cases once the threshold valueswere established. However, diaphragmatic contraction wasnoted in 25% of cases in the DDDC mode. No such eventwas noted in patients with single-pass VDD pacing.

As the OLBI principle involves the use of floating atrialelectrodes, achieving stable atrial pacing in differentpositions is of primary concern during long-term follow-up. In the present study, 80% of the patients showed stableatrial pacing in the supine position and 75% in the uprightposition but only 62% showed stable atrial pacing in boththe supine and upright positions. Similar observations werereported by Bongiorni et al.9

Another drawback of using atrial dipoles in OLBI is thatthey may sense far-field cardiac events like R wave andskeletal myopotentials or afterpotentials.11 These far-fieldsensings obviously reduce AV synchrony. The problem isusually tackled by adjustment of the refractory periods. Inthe present study, no such oversensing problem was noted.

Longo et al.12 reported two cases with pacemaker-mediated tachycardia associated with retrogradeconduction in their follow-up of 36 patients with single-pass VDD pacing system. In the present study, no suchcomplication was noted.

Conclusions: We conclude that single-pass VDD and DDDCpacing using the OLBI principle are safe and technicallyeasy and do not have the problem of atrial lead dislodgmentor fracture. They are also economically cheaper than theconventional dual-lead DDD mode. As chronic atrial sensingwas satisfactory in our study, atrial tracking in both VDDand DDDC modes was adequate. However, since chronicstable atrial pacing was seen in only 62% of patients in allpositions and 25% of cases had diaphragmatic contraction,single-pass DDDC pacing with floating atrial rings using theOLBI principle should be used in patients who requirepredominantly VDD pacing and only occasionally atrialpacing. Thus, the single-pass lead system with floating atrialrings requires further refinement to ensure better long-termoutcome for atrial pacing.

References

1. Lames GA, Orav EJ, Stambler BS, Ellenbogen KA, Sgarbossa EB,Huang SK, et al. Quality of life and clinical outcomes in elderlypatients treated with ventricular pacing as compared with dual-

Discussion

The single-pass lead system, with its ease of introductionand lead positioning, has made the job of achieving AVsequential pacing much easier than the conventional dual-lead systems. As compared to dual-lead systems,10,11 it iscost-effective and minimizes the chances of dislodgment orfracture of the atrial lead.

62% 62% 62%

75% 75%

88% 88%

75%

88% 88%

0%

20%

40%

60%

80%

100%

0 Months 1 Months 3 Months 6 Months 12 Months

Res

pons

e pe

rcen

tage

Fig. 1. Effects of atrial capture on patients with DDDC pacemaker in the supineand upright positions.

� Supine� Upright

IHJ-159-01.p65 10/17/01, 1:27 PM479

480 Kar et al. Single-Pass Physiological Pacing Indian Heart J 2001; 53: 477–480

chamber pacing. Pacemaker Selection in the Elderly Investigators.N Engl J Med 1998; 338: 1097–1104

2. Alpert MA, Curtis JJ, Sanfelippo JF, Flaker GC, Walls JT, Mukerji V, etal. Comparative survival after permanent ventricular and dualchamber pacing for patients with chronic high degreeatrioventricular block with and without pre-existing congestiveheart failure. J Am Coll Cardiol 1986; 7: 925–982

3. Nowak B, Voigtlander T, Himmrich E, Liebrich A, Poschmann G,Epperlein S, et al. Cardiac output in single-lead VDD pacing versusrate-matched VVIR pacing. Am J Cardiol 1995; 75: 904–907

4. Linde-Edelstam C, Gullbery B, Nordlander R, Pehrsson SK,Rosenquist M, Ryden L, et al. Longevity in patients with high degreeatrioventricular block paced in the atrial synchronous or the fixedrate ventricular inhibition mode. Pacing Clin Electrophysiol 1992;15: 304–313

5. Recommendations for pacemaker prescription for symptomaticpatients. Report of Working Party of the British Pacing andElectrophysiology Group. Br Heart J 1991; 66: 185–191

6. Gregoratos G, Chetlin MD, Conill A, Epstein AE, Fellows C,Ferguson TB, et al. ACC/AHA guidelines for implantation of cardiac

pacemaker and anti-arrhythmic devices: a report of the ACC/AHATask Force of practice guidelines (Committee on PacemakerImplantation). J Am Coll Cardiol 1998; 31: 1175–1209

7. de Belder MA, Linker NJ, Jones S, Camm AJ, Ward DE. Costimplications of the British Pacing and Electrophysiology Group’srecommendations for pacing. BMJ 1992; 305: 861–865

8. Antoniolli GE. Single lead atrial synchronous ventricular pacing: adream come true. Pacing Clin Electrophysiol 1994; 17: 1531–1547

9. Bongiorni MG, Bedendi N. Atrial stimulation by means of floatingelectrodes. A multicentre experience. Pacing Clin Electrophysiol 1992;15: 1977–1981

10. Lau C, Nishimura SC, Harrison AW, Goldman BS. Maintainingphysiological pacing in AV block using a new generation single leadVDD pacemaker. Can J Cardiol 1996; 12: 579–583

11. Sethi KK, Bhargava M, Mohan JC, Pandit N, Khan JA, Gokhale CS,et al. Long-term performance of atrial leads. Indian Heart J 1994;46: 71–75

12. Longo E, Catrini V. Experience and implantation techniques with anew single pass lead VDD pacing system. Pacing Clin Electrophysiol1990; 13: 927–936

IHJ-159-01.p65 10/17/01, 1:27 PM480

Indian Heart J 2001; 53: 481–485 Sharma et al. Secondary Hypertension: Pheochromocytoma 481

Pheochromocytoma: A 10-Year Experience in aTertiary Care North Indian Hospital

N Sharma, S Kumari, S Jain, S VarmaDepartment of Internal Medicine, Postgraduate Institute of Medical Education and Research, Chandigarh

Secondary hypertension accounts for 5%–8% of allcases of hypertension.1 Among the different causes of

secondary hypertension, pheochromocytoma is responsiblefor 0.05%–0.3%.2–4 Accurate diagnosis of the tumorrequires a high index of suspicion coupled with laboratorydemonstration of excess catecholamine secretion. Thepresent study was carried out in a tertiary care hospital ofnorth India to study the profile of patients withpheochromocytoma.

Methods

From January 1989 to December 1996, data of patientswith pheochromocytoma were obtained from the recordsof patients attending the hypertension clinic of the NehruHospital which is attached to the Post Graduate Institute ofMedical Education and Research, Chandigarh, India. FromJanuary 1997 onwards, patients were studied prospectively.

A detailed medical history was taken and all the patientsunderwent a complete physical examination. Blood pressurewas recorded in all four limbs with the standard mercurysphygmomanometer in the supine and erect positions.Hypertension was defined according to the Joint NationalCommittee VI criteria5 and hypertensive retinopathyaccording to the Keith–Wagener–Barker’s classification.6 Inall young hypertensives (age <40 years), we routinelycarried out estimation of the 24-hour urinaryvanillylmandelic acid (VMA) excretion, an ultrasoundexamination of the abdomen and a captopril challengetechnitium99m DTPA renal scan.

Laboratory investigations carried out in all patients were:a hemogram including the erythrocyte sedimentation rate,blood biochemistry including blood urea, fasting andpostprandial blood sugar, serum creatinine, sodium,potassium, calcium, phosphorus and alkaline phosphatase,and urinalysis. The 24-hour urinary sample was analyzedfor VMA by spectrophotometry, and noradrenaline andadrenaline by flurimetry. Morning blood samples wereanalyzed for catecholamines. For localization of the tumor,an ultrasound examination of the abdomen followed by a

Background: The study was carried out to highlight the clinical and biochemical profile of patients withpheochromocytoma in a tertiary care center of North India.Methods and Results: Thirty consecutive cases of pheochromocytoma admitted over a period of 10 years toour Institute were analyzed. The chief clinical complaints of these 30 patients (17 males and 13 females, meanage 24±7 years) were palpitation (80%), headache (77%), sweating (60%), breathlessness (67%) and flushing(56%). The clinical triad of headache, flushing and sweating occurred in 26.7% of cases. On clinical examination,97% of the patients were hypertensive and 16.6% presented with malignant hypertension. Laboratorymeasurements showed that the levels of 24-hour urinary vanillylmandelic acid were elevated in 80% of cases.Levels of plasma adrenaline and noradrenaline were raised in 78% and 79% of cases, respectively. Anatomicallocalization of the tumor on computerized tomographic scan showed the presence of an adrenal tumor in 80%and extra-adrenal tumor in 20%. Surgical removal of the tumor could be carried out in 28 cases followingcontrol of the blood pressure with antihypertensive drugs including alpha and beta adrenoreceptor blockers.Conclusions: Pheochromocytoma should be suspected in all young hypertensive persons. The appropriatetherapy for this tumor is surgical removal preceded by adequate blood pressure control including the use ofalpha and beta adrenoreceptor antagonists. (Indian Heart J 2001; 53: 481–485)

Key Words: Pheochromocytoma, Hypertension, Tumor

Original Article

Correspondence: Dr Navneet Sharma, Assistant Professor, Departmentof Internal Medicine, Postgraduate Institute of Medical Education and Research,Chandigarh 160012. e-mail: [email protected]

IHJ-191-01.p65 10/17/01, 1:41 PM481

482 Sharma et al. Secondary Hypertension: Pheochromocytoma Indian Heart J 2001; 53: 481–485

contrast enhanced computerized tomographic (CT) scan ofthe chest and abdomen were carried out.

Results

Of the 30 patients with pheochromocytoma, there were 17males and 13 females. The main complaints of our patientsare given in Table 1. The clinical triad of headache, flushingand sweating was present in only 8 (26.7%) cases. A historyof hypertension of either persistent or intermittent naturecould be elicited in 29 cases. In one case, the tumor wasaccidentally discovered during ultrasonographic screeningfor gall stones. In this patient, there were neither any clinicalfeatures nor any abnormalities in laboratory estimationsof catecholamines to suggest the diagnosis ofpheochromocytoma.

The blood pressure data of our patients are presented inTable 2. A hypertensive crisis (malignant hypertension) wasseen in 5 cases (16.6%). Of the 30 cases, 6 did not have theclinical features, clinical triad, paroxysmal hypertension ora postural fall of the systolic blood pressure typical ofpheochromocytoma. Changes characteristic of grade Ihypertensive retinopathy were seen in 2 cases (6.6%), gradeII in 19 (63.3%), grade III in 2 (6.6%) and grade IV in 5(16.6%). One patient showed evidence of metastaticpheochromocytoma on fine needle aspiration cytology foringuinal lymphadenopathy. Associated findings presentwere hypertrophic cardiomyopathy in 1, annulo-aorticectasia with aortic regurgitation in 1, neurofibromatosis typeI in 2 and gall stones in 1.

Specific laboratory investigations for pheochromocytomarevealed that the 24-hour urinary excretion of VMA wasnormal in 20% (normal 8 mg/total urinary volume) but itwas elevated 1–1.5 times the normal limit in 20%, 1.5–2.0times in 16.7% and more than double the normal limit in43.3% of cases. Estimation of 24-hour urinary adrenalinecould be carried out in 13 cases only. Four cases had urinaryadrenaline excretion within the normal limit (normal24 µg/24 hours). In 7, it was raised 1–1.5 times the normallimit and in 1 case each was elevated 1.5–2 times and morethan double the normal limit, respectively. In 16 cases, the24-hour urinary noradrenaline excretion was measuredand was within 1–1.5 times the normal limit (normal66 µg/24 hours) in all. In 6 cases where the urinary VMAexcretion was within the normal limit, urinary estimationof noradrenaline was not carried out. Five of these 6 caseshad clinical features suggestive of pheochromocytoma. In23 patients, blood samples were taken for estimating theplasma adrenaline and noradrenaline levels. Plasmanoradrenaline was elevated 1–1.5 times, 1.5–2 times andmore than double the normal limit in 9, 5 and 4 cases,respectively (normal 1.6 ηg/ml). Plasma adrenaline waselevated 1–1.5 times, 1.5–2 times and more than doublethe normal limit in 13,1 and 4 cases (normal 0.4 ηg/ml),respectively. Tumor localization carried out by CT scanningdisclosed an adrenal tumor in 80% and an extra-adrenaltumor in 20% of cases (1 near the right renal hilum, 1 inthe bladder wall and 4 arising from the Organ ofZuckerkandl). In the single pregnant patient, an ultrasonicexamination of the abdomen showed a nonviable growth-retarded fetus. After control of the blood pressure withdrugs, she underwent a medical termination of pregnancyfollowed by surgical extirpation of the tumor. There were 2cases of malignant pheochromocytoma with a palpableabdominal mass. In 1, the confirmation of malignancy wasmade on fine needle aspiration cytology of the inguinal

Table 1. Clinical features of patients withpheochromocytoma

Parameter n=30

Age 24±7 years

Palpitation 80%Headache 77%Flushing 56%Sweating 60%Breathlessness 67%Anemia 2 (6.6)Cardiomegaly 9 (30)Left ventricular hypertrophy 13 (43.3)

Tumor localization (on computerized tomography)Adrenal

Left 8 (26.7)Right 12 (40)Bilateral 4 (13.3)

Extra-adrenal 6 (20)


Table 2. Blood pressure data of patients withpheochromocytoma at presentation

Hypertension SBP DBP(mmHg) n (mmHg) n

Stage 1 140–159: 3 (10) 90–99: 2 (66)Stage 2 160–179: 1 (3.3) 100–109: 6 (20)Stage 3 >180: 25 (83.3) >110: 21 (70)

Postural fallin SBP >20 15 (50)

SBP: systolic blood pressure; DBP: diastolic blood pressureValue in parentheses are percentages

IHJ-191-01.p65 10/17/01, 1:41 PM482


lymph nodes which showed metastasis and in the other,malignant spread to the liver and kidneys was demonstratedat surgery.

Therapeutically, the antihypertensive drugs used werephenoxybenzamine in 76.6%, enalapril in 16.6%,nifedipine in 73.3% and prazocin in 26.5% of cases. Beta-blockers were used in all the patients 3–6 days after the startof alpha-blocker therapy. Three or more antihypertensiveagents were required in 93% of our cases. Out of the 30cases, 28 were subjected to surgical removal of the tumor.Those with bilateral pheochromocytomas were subjectedto bilateral subtotal adrenalectomy followed by steroidreplacement therapy. Extra-adrenal pheochromocytomaswere surgically extirpated and adrenal exploration wascarried out. Two cases of malignant pheochromocytomaswere subjected to chemotherapy with cyclicalcyclophosphamide, vincristine and dacarbazine. Atdischarge, 83% of patients required only a singleantihypertensive agent (amlodipine) for control of bloodpressure. In 10%, two antihypertensive agents wererequired (amlodipine and prazocin) for blood pressurecontrol and in 7%, i.e. 2 cases of malignantpheochromocytoma, three antihypertensive agents(amlodipine, prazocin and atenolol) were used in maximumtolerated doses.

Discussion

Pheochromocytomas are tumors of neuroectodermalorigin, presenting clinically with the triad of headache,diaphoresis and palpitation. This clinical triad is suggestiveof pheochromocytoma and peculiar to the specific patientpopulation but occurs in only 11%–22% of cases.7–13 In ourstudy, it was seen in 26.7% of patients. Hypertension of apersistent or paroxysmal nature is one of the leading cluesfor the diagnosis of pheochromocytoma. Twenty-nineout of the 30 cases in our study were hypertensive andone was normotensive. A review of the National CancerRegistry in Sweden showed that of 439 cases ofpheochromocytoma, nearly 40% were diagnosed at autopsyand, of these, the findings were reported as incidental in14%.4 In another review from the Mayo Clinic of 54 casesof pheochromocytoma proven at autopsy, only 24% couldbe diagnosed during life. Of the remaining 76%, 54% hadbeen diagnosed to have hypertension during life and theremaining 22% had normal blood pressure.8 Hypertensionof a persistent nature without any paroxysms was evidentin 56.6% of cases and paroxysms of hypertension werenoted in the remaining 40%. Other studies have shown thepresence of hypertension in 55%–98% of all cases of

pheochromocytoma, paroxysmal elevation of bloodpressure in 24%–86% and persistent hypertension withoutany paroxysms in 14.6%–37%.7–13 A significant drop of thesystolic blood pressure occurred in the upright position in50% of our cases in contrast to 70% of cases in otherstudies.8–13 Postural fall of the systolic blood pressuredepends upon the status of the intravascular volume andadrenoreceptors. In our study, the intravascular volumestatus of all patients was normal as evidenced by a normalurea-to-creatinine ratio. We postulate that this posture-dependent decline in the systolic blood pressure could berelated to a downregulation of the adrenergic receptorswhich is known to occur in conditions where there is aconstant interplay between the receptor and the hormone.14

This adrenoreceptor downregulation is known to occur insituations where the adrenergic system is activated as acompensatory phenomenon, i.e. congestive heart failure.Of the other clinical associations seen withpheochromocytoma, neurofibromatosis type I occurs in1%–3.3% cases,10–13,15 whereas association withhypertrophic cardiomyopathy is confined to case reportsonly.16,17

The biochemical marker of pheochromocytoma is anelevated level of plasma or urinary catecholamines. Thescreening test used for the biochemical diagnosis ofpheochromocytoma in many series has been urinary VMAexcretion which has a sensitivity and specificity of 64%–90% and 87%–98%, respectively.18 In our study, 24-hoururinary VMA was elevated in 80% of the patients. Thediagnosis of pheochromocytoma in the remaining 20% wasbased upon a high index of clinical suspicion and a positiveabdominal imaging study. In other studies, elevated urinaryVMA excretion has been seen in 61%–94% of patients.7,9,18

The 24-hour excretion of urinary adrenaline was measuredin 13 patients and was raised in 9. In a review of 18 casesof pheochromocytoma, Loh et al.7 measured urinarycatecholamines (meta-adrenalines) in 6 cases; all 6 had anelevated level of urinary catecholamines. Modlin et al.9 ina study of 72 cases of pheochromocytoma showed elevatedlevels of urinary catecholamines in 93%. In our study, levelsof plasma noradrenaline and adrenaline were measured in23 out of 30 cases. The levels of plasma noradrenaline wereraised in 78.7% and of adrenaline in 78.3% of cases. Otherstudies have shown that in 90%–95% of patients with thistumor, levels of plasma adrenaline and noradrenaline aremarkedly elevated; the sensitivity and specificity of this testis 85% and 95%, respectively.7,9,18

Computerized tomography is the most reliableinvestigation to localize pheochromocytoma preoperatively.In 94%–95% of cases, CT has been shown to accurately

IHJ-191-01.p65 10/17/01, 1:41 PM483

484 Sharma et al. Secondary Hypertension: Pheochromocytoma Indian Heart J 2001; 53: 481–485

localize an abdominal tumor greater than 1 cm in size.19,20

For adrenal masses of less than 1 cm and for extra-adrenalmasses, this accuracy decreases. However, CT scanningcannot dif ferentiate between pheochromocytoma,adenoma and metastasis.18–20 In our study, CT could localizeabdominal pheochromocytoma in all the cases (sensitivityof 100%). There were 6 tumors (20%) located extra-adrenally. Extra-adrenal pheochromocytomas account forapproximately 11%–27%7,9,12,15 of all pheochromocytomasand less than 1% of these arise in the bladder wall from thesympathetic nervous system.21 Bladder tumors are usuallylocalized to the trigone and such patients give acharacteristic history of adrenosympathetic symptoms onmicturition, which was the case in our patient.

Occurrence of pheochromocytoma in pregnancy ushersin considerable risk to both the mother and fetus. A reviewof 89 pregnant patients with pheochromocytoma showeda maternal mortality rate of 48%.22 This study alsodemonstrated that, in the first or second trimester,presurgical therapy with alpha-blockers followed by surgicalresection 2 weeks later brought down the maternalmorbidity and mortality; however, fetal mortality remainedhigh (67%).22 Recently studies have shown that medicalmanagement with alpha-blockade and close blood pressuremonitoring with deferral of surgery till after delivery mayfurther reduce maternal and fetal mortality and morbidity.23

In our study, there was 1 pregnant patient withpheochromocytoma (late second trimester). The favorablematernal outcome in our case was related to a promptcontrol of blood pressure and medical termination ofpregnancy followed by quick surgical extirpation of thetumor.

Patients with malignant pheochromocytoma formanother group, the management of which is difficult.Therapy for malignant pheochromocytoma revolves aroundsurgical removal and chemotherapy with cycles ofalkylating agents, vinca alkaloids and dacarbazine. Thiscombination has shown a complete tumor response ofonly14%.24 In our study there were 2 cases of malignantpheochromocytoma. In one of them, despite chemotherapy,metastases persisted. Although a biochemical response wasachieved, control of blood pressure was difficult, requiringtherapy with multiple antihypertensive agents.

Conclusions: Our study of patients withpheochromocytoma has shown a higher incidence ofsymptoms such as headache, flushing and palpitation whencompared to studies from the West. The typical triad ofheadache, diaphoresis and flushing occurred in 26.7% ofcases only. Thus, investigations for pheochromocytoma

should be carried out in all young hypertensive patients. Inthese patients, with control of blood pressure usingadrenergic receptor blockade followed by surgical removal,a successful outcome and cure of hypertension can beachieved.

References

1. Kaplan NM. Systemic hypertension: mechanisms and diagnosis. In:Braunwald E (ed). Heart disease —a textbook of cardiovascular medicine.5th ed. New York: WB Saunders Co; 1997. pp. 811–812

2. Tucker RM, Labarthe DR. Frequency of surgical treatment forhypertension in adults at the Mayo Clinic from 1973 through 1975.Mayo Clin Proc 1977; 52: 549–555

3. Beard CM, Sheps SG. Kurland LT, Carney JA, Lie JT. Occurrence ofpheochromocytoma in Rochester, Minnesota 1950 through 1979.Mayo Clin Proc 1983; 58: 802–804

4. Stenstrom G, Svardsudd K. Pheochromocytoma in Sweden 1958–1981: an analysis of the National Cancer Registry data. Acta MedScand 1986; 220: 225–232

5. The Sixth Report of the Joint National Committee on prevention,detection, evaluation and treatment of high blood pressure. ArchIntern Med 1997; 157: 2413–2445

6. Swales JD. Essential hypertension. In: Weatherall DJ, Ledingham JGG(eds). Oxford textbook of medicine. 3rd ed. New York: Oxford UniversityPress; 1996. pp. 2536–2537

7. Loh KC, Schlossberg AH, Abbot EC, Salisbury SR, Tan MH.Pheochromocytoma: a ten year survey. Q J Med 1998; 90: 51–60

8. Sutton MG, Sheps SG, Lie JT. Prevalence of clinically unsuspectedpheochromocytoma: review of a 50-year autopsy series. Mayo ClinProc 1981; 56: 354–360

9. Modlin IM, Farndon JR, Shepherd A, Johnston ID, Kennedy TL,Montgomery DA, et al. Phaeochromocytoma in 72 patients: clinicaland diagnostic features, treatment and long-term results. Br J Surg1979; 66; 456–465

10. Lucon AM, Pereirra MA, Mendonca BB, Halpern A, Wajchenbeg BL,Arap S. Pheochromocytoma: study of 50 cases. J Urol 1997; 157:1208–1212

11. Stein PP, Black HR. A simplified approach to pheochromocytoma. Areivew of literature and report of one institution’s experience.Medicine (Baltimore) 1991; 70: 46–66

12. Melicow MM. One hundred cases of pheochromocytoma at theColumbia–Presbyterian Medical Center 1926–1976; aclinicopathological analysis. Cancer 1977; 40: 1987–2004

13. Januszewicz W, Wocial B. Clinical and biochemical aspects ofpheochromocytoma. Cardiology 1985; 72(Suppl 1): 131–136

14. Packer M. Pathophysiological mechanisms underlying the effects ofβ-adrenergic agonists and antagonists on the functional capacity andsurvival in chronic heart failure. Circulation 1990; 82 (Suppl 2)I77–I88

15. Mathew S, Perakath B, Nair A, Seshadiri MS, Shanthly N.Phaeochromocytoma: experience from a referral hospital in SouthernIndia. Aust N Z J Surg 1999; 69: 458–460

16. Shub C, Williamson MD, Tajik AJ, Eubanks DR. Dynamic leftventricular outflow tract obstruction associated withpheochromocytoma. Am Heart J 1981; 102: 286–290

17. Mardini MK. Echocardiographic findings in phaeochromocytoma.Chest 1982; 81: 394–395

18. Werbel SS, Ober KP. Pheochromocytoma: update on diagnosis,localization and management. Med Clin North Am 1995; 79:131–153

19. Stewart BH, Bravo EL, Haaga J, Meaney TF, Tarazi R. Localisation of

IHJ-191-01.p65 10/17/01, 1:41 PM484


pheochromocytoma by computed tomography. N Engl J Med 1978;299: 460–461

20. Sheedy PF II, Hattery PR, Stephens DH. Computerised tomographyof the adrenal gland. In: Haaga JR, Alfidi RJ (eds). Computerisedtomography of the whole body. St Louis: CV Mosby; 1983. p.681

21. Whalen RK, Althausen AF, Daniels GH. Extra-adrenalpheochromocytoma. J Urol 1992; 147: 1–10

22. Schenker JG, Cheoers I. Pheochromocytoma and pregnancy: review

of 89 cases. Obstet Gynecol Surg 1971; 26: 73923. Buergess CE. Alpha blockade and surgical intervention of

pheochromocytoma in pregnancy. Obstet Gynecol 1979; 53:266–270

24. Averbuch SD. Steakley CS, Young RC, Gelmann EP, Goldstein DS,Stull R, et al. Malignant pheochromocytoma: effective treatment withcombination of cyclophosphamide, vincristine and dacarbazine. AnnIntern Med 1988; 109: 267–273

IHJ-191-01.p65 10/17/01, 1:41 PM485

486 Radhakumary et al. Collagen Types in EMF Indian Heart J 2001; 53: 486–489

Endomyocardial Fibrosis is Associated with SelectiveDeposition of Type I Collagen

C Radhakumary, TV Kumari, Chandrasekharan C KarthaDivision of Cellular and Molecular Cardiology, and Biomedical Technology Wing,

Sree Chitra Tirunal Institute for Medical Sciences and Technology, Thiruvananthapuram

Endomyocardial fibrosis (EMF) is a distinct form of heartdisease leading to restrictive filling of the ventricles and

cardiac failure.1 The characteristic pathological feature ofEMF is the severe and massive focal or diffuse thickening ofthe endocardium in either one or both the ventricles.2

Ultrastructurally, the fibrous tissue is composed of collagenfibers which are compactly arranged with regularly orientedbundles of collagen fibrils, measuring from 400 to 800 Åin diameter, and with normal periodicity.3 A severe degreeof endocardial thickening and obliteration of theventricular cavity leads to restriction of diastolic filling inthe ventricle. However, systolic function is relatively wellpreserved till the late stages of the disease.

It is well known that enhanced collagen accumulationin pathological conditions of the heart is accompanied bya change in the relative amounts of fibrillar types I and III

collagens, the major collagen phenotypes in the heart. Thenature of the cardiac collagens and the relative proportionsof collagen types in EMF have not been documented in theliterature. In this communication, we report the collagencomposition in the cardiac tissues of patients with EMF.

Methods

Cardiac tissues from the ventricles of 13 patients with EMF(7 males and 6 females; age range 21–45 years) werecollected either at autopsy or during operation for valvereplacement. Control tissues were collected from 6 victims(4 males and 2 females; age range 24–49 years) of trafficaccidents or suicidal deaths who were subjected to medico-legal autopsies and whose hearts did not have any evidenceof cardiac disease.

Preparation of tissues: Extraction and phenotyping ofthe collagen were performed using a procedure describedby Miller and Rhodes.4 Collagen typing and quantification

Background: Endomyocardial fibrosis is a distinct form of heart disease leading to restrictive ventricular fillingand cardiac failure. The disease is characterized by a marked thickening of the endocardium due to the depositionof dense fibrous tissue composed of wavy bundles of collagen. Changes in collagen composition and an abnormalincrease in its concentration result in a stiffer myocardium and ventricular diastolic dysfunction. The nature ofcardiac collagens and the relative proportions of collagen types in endomyocardial fibrosis have not beendocumented in the literature.Methods and Results: This study analyzed collagen composition in the cardiac tissues of 13 patients withendomyocardial fibrosis and 6 individuals who were the victims of traffic accidents or suicidal deaths and didnot have any heart disease. We estimated the relative proportions of types I and III collagen after pepsin digestionof the tissue and separation of the emerging peptides by sodium dodecyl sulfate polyacrylamide gel electrophoresis.The mean type I:III collagen ratio was 0.51±0.06 in normal individuals, and 0.93±0.43 in patients withendomyocardial fibrosis (p<0.05). The alteration in the type I:III collagen ratio was due to a disproportionateincrease in type I collagen.Conclusions: The results indicate that a selective increase in type I collagen may contribute to the impaireddiastolic distension of the ventricles in patients with endomyocardial fibrosis. (Indian Heart J 2001; 53:486–489)

Key Words: Endomyocardial fibrosis, Cardiomyopathy, Collagen phenotype

Original Article

Correspondence: Professor CC Kartha, Division of Cellular and MolecularCardiology, Sree Chitra Tirunal Institute for Medical Sciences and Technology,Thiruvananthapuram 695011. e-mail: [email protected]

IHJ-198-01.p65 10/17/01, 1:50 PM486

Indian Heart J 2001; 53: 486–489 Radhakumary et al. Collagen Types in EMF 487

were done following the methods described by Mukherjeeand Sen.5 Specimens collected were labeled and stored at–20 °C till analysis. Before analysis, the tissues were mincedwith scissors, lyophilized and the dry weight taken. Thelyophilized material was then pulverized and samples wereinitially extracted for 24 hours at 4°C, with 1M NaCl in0.05 M tris buffer, pH 7.4, containing protease inhibitorssuch as ethylene dinitrilo tetra-acetic acid (EDTA) 20.0 mM,di-isopropyl flurophosphate (DFP) 1.0 mM, N-ethylmaleimide (NEM) 2.0 mM and pepstatin 1 µg/ml. Thesupernatant was retained along with all the supernatantsof subsequent 24-hour re-extractions with 0.05 M aceticacid at 4 °C. Samples were digested 3 times successively, eachtime for 24 hours’ duration with pepsin (1 mg/1ml) in0.5 M acetic acid. The extracts were pooled and collagenprecipitated by adding NaCl to a final concentration of2 M at 4 °C.

Biochemical analysis: The collagen precipitate collectedby centrifugation was re-dissolved in 0.5 M acetic acid anddialyzed against 0.02 M dibasic sodium phosphate. In thesesamples, collagen types were identified by sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE).Different lanes of gel were loaded with 100 µg of standardtypes I and III collagen and 100 µg of each of the extractedheart collagens. After electrophoresis, the gels were stainedwith Coomassie blue R250 and de-stained using a solutionof acetic acid and methanol. The stained gels were scannedat a wavelength of 530 nm using a soft laser scanningdensitometer gel scanner. Relative amounts of types I andIII collagen were estimated by determining the peak areasdistinctive for collagen types, and the relationship betweenthese areas and the amount of collagen applied to the gel.Concentrations of types I and III collagen in tissues werecalculated from the regression equation of the standards.Standard types I and III collagen were procured fromSIGMA.

Group comparison was done using the Student’s t test.

Results

Concentrations of types I and III collagen were calculatedafter quantification of the pepsin fragments representativeof each type. The total collagen concentration in the tissueextract was determined by the estimation of hydroxyprolinein 100 µg of tissue collagen from each sample loaded inSDS-PAGE gels.

Hearts with EMF had significantly increased type Icollagen concentration (68.64±24.19 µg) when comparedwith that of control samples (43.94±19.5 µg) (p<0.05),

while type III collagen concentration in hearts with EMFand control samples were not significantly different(79.42±22.70 µg and 87.6±38 µg, respectively). Thetype I:III collagen ratio was 0.51±0.06 in normal heartsand 0.93±0.43 in hearts with EMF (p<0.05). The resultsare summarized in Table 1.

Table 1. Type I:III collagen ratio in normal hearts andhearts with EMF together with mean values for types I andIII collagen in densitometric scans

Collagen Normal hearts Hearts with EMF(µg) (n=6) (n=13)

Type I 43.94±19.5 .68.64±24.19*Type III 087.6±38.0 79.42±22.70Type I:III ratio 00.51±0.06 00.93±0.43*

* p<0.05

To validate our procedure, we loaded 3 lanes of the gel:the first with 25 µg of collagen standard, the second with25 µg each of collagen standard and heart collagen, andthe third with 25 µg heart collagen, and estimated thepercentage of recovery. The results are shown in Table 2.

Table 2. Data on validation of collagen quantification

Area under peak (cm2)

Collagen 25 µg collagen 25 µg collagen 25 µg heart Predictedstandard standard+25 µg collagen correlation

heart collagen (%)

Type I 10.0 29.5 22.5 87Type III 20.5 29.0 11.0 77.2

The relationships between peak area and types I and IIIcollagen concentrations are shown in Fig. 1. Figure 2 is arepresentative densitometric scan of the gels. Smallquantities of types V and VIII collagen were also seen.However, their contribution to the total collagenconcentration was negligible. The lack of correlationbetween the age of the patients and percentage of type Icollagen in both the groups of patients is seen in Fig. 3.

Discussion

The present study aimed to examine the relative proportionof the major collagen types in the ventricular tissues ofpatients with EMF. Cardiac tissues obtained from patientswith EMF were compared with those obtained from normalsubjects. The ratio of different collagen types was estimatedby SDM-PAGE after solubilization of collagen molecules by

IHJ-198-01.p65 10/17/01, 1:50 PM487

488 Radhakumary et al. Collagen Types in EMF Indian Heart J 2001; 53: 486–489

limited proteolysis and quantification by densitometricanalysis.

Collagens constitute the major structural proteins of theinterstitium of the heart, which, in turn, forms a supportfor other myocardial structures. The two most abundantcollagen types in the human myocardium are types I andIII.6 Type I collagen fibers have a substantial tensile strength

Fig. 2(a). Densitometric scan of the control sample.

Fig. 3. Relationship between age in years and mean value of collagen types inpatients with EMF.

Fig. 2 (b). Densitometric scan of the EMF sample.

Fig. 1. Relationship between collagen concentration and peak area in cm2 indensitometric scans. whereas type III fibers possess a resistance that is ideal for

maintaining the structural integrity and distensibility ofthe network. The relative proportion of these collagens may,therefore, play an important role in determining thephysical properties of the extracellular matrix. Recently, ithas been recognized that while cardiac myocytes and thecoronary vasculature are central to the contractile functionand viability of the myocardium, so is the extracellularmatrix or cardiac interstitium and, in particular, theconstituent types I and III fibrillar collagen matrix. Abiochemical defect within this collagenous network or itsphysical interruption has the potential to adversely alterthe tensile strength and tethering support function, therebyaltering the architecture and mechanical properties of themyocardium. Collagen plays an important role not only inthe mechanical properties of the heart but also inmaintaining an even myocardial shape.7

In pathological conditions, remodeling of the collagenmatrix occurs both in terms of structure and biochemistry.Collagen phenotypes have been well characterized inpathological conditions such as cardiac hypertrophysecondary to systemic hypertension, dilated car-diomyopathy and ischemic cardiomyopathy.5,6,8–11 Theconcentration of myocardial collagen increases three- tosix-fold during hypertrophy secondary to valvular heartdisease and systemic hypertension. An increase in type I:IIIcollagen ratio has been reported in other myocardialdiseases and dilated cardiomyopathy.10,11 Such an increasein collagen and the associated structural remodelingresults in an altered myocardial function.12 Ischemiccardiomyopathy in humans is characterized by an increaseddeposition of type III collagen, without a change in type Icollagen concentration, resulting in a decrease in the typeI:III collagen ratio.15 Dilated cardiomyopathy is associatedwith an increase in the concentrations of both collagentypes, predominantly type I collagen, resulting in an

IHJ-198-01.p65 10/17/01, 1:51 PM488

Indian Heart J 2001; 53: 486–489 Radhakumary et al. Collagen Types in EMF 489

increase in the type I:III collagen ratio.11 Both these formsof cardiomyopathy are characterized by dilatation of theheart and poor systolic function.

The hallmark of EMF is contracted ventricles and alteredmyocardial diastolic function resulting from increaseddeposition of collagen in the endomyocardium. In thepresent study, we have shown that in patients withEMF, there is a quantitative increase in type I collagenconcentration, resulting in an elevated type I:III collagenratio. In an earlier study, Bolarin and Andy13 reported thatenzyme levels of collagen biosynthesis are increased inpatients with EMF, suggesting an enhanced collagensynthetic activity.

The abnormal increase in collagen concentrationcould account for the stiffer myocardium and ventriculardiastolic dysfunction seen in EMF. Our findings suggest thatEMF is not different from other myocardial diseases in thisrespect. Although the nature of the underlying changes incollagen metabolism in EMF remains unclear, alterationsin both the synthetic and the degradative processes may beimportant.

In summary, this study demonstrates that there is anincrease in the type I:III collagen ratio in patients with EMF.This alteration is similar to the observations made in othermyocardial diseases. This is the first demonstration that EMFis associated with a distinct alteration in the compositionof the interstitial collagens.

References

1. Kartha CC. Endomyocardial fibrosis: a case for the tropical doctor.Cardiovasc Res 1995; 30: 636–643

2. Kartha CC, Sandhyamani S. An autopsy study of tropicalendomyocardial fibrosis in Kerala. Indian J Med Res 1985; 82: 439–446

3. Kartha CC, Valiathan MS. Cardiac ultrastructure in tropicalendomyocardial fibrosis. Indian J Med Res 1988; 87: 275–282

4. Miller EJ, Rhodes RK. Preparation and characterization of differenttypes of collagen. Methods Enzymol 1982; 82: 33–64

5. Mukherjee D, Sen S. Alteration of collagen phenotypes in ischemiccardiomyopathy. J Clin Invest 1991; 88: 1141–1146

6. Weber KT. Cardiac interstitium in health and disease: the fibrillarcollagen network. J Am Coll Cardiol 1989; 13: 1637–1652

7. Bishop JE. Regulation of cardiovascular collagen distribution bymechanical forces. Mol Med Today 1998; 4: 69–75

8. Weber KT, Brilla CG. Pathological hypertrophy and cardiacinterstitium. Fibrosis and renin–angiotensin–aldosterone system.Circulation 1991; 83: 1849–1865

9. Caspari PG, Newcomb M, Gibson K, Harris P. Collagen in the normaland hypertrophied human ventricle. Cardiovasc Res 1977; 11: 554–558

10. Bishop JE, Greenbaum R, Gibson DG, Yacoub M, Laurent GJ. Enhanceddeposition of predominantly type I collagen in myocardial disease. JMol Cell Cardiol 1990; 22: 1157–1165

11. Marijianowski MM, Teeling P, Mann J, Becker AE. Dilatedcardiomyopathy is associated with an increase in the type I/type IIIcollagen ratio: a quantitative assessment. J Am Coll Cardiol 1995; 25:1263–1272

12. Caulfield JB, Janicki JS. Structure and function of myocardial fibrillarcollagen. Technol Health Care 1997; 5: 95–113

13. Bolarin DM, Andy JJ. Enzymes of collagen synthesis and type IIprocollagen amino-propeptide in serum from Nigerians with chronicendomyocardial fibrosis. Ann Clin Res 1983; 15: 123–127

IHJ-198-01.p65 10/17/01, 1:51 PM489

490 Haridas et al. Bilateral Ostial Aorto-Coronary Angioplasty Indian Heart J 2001; 53: 490–492

Percutaneous Transluminal Angioplasty with CuttingBalloon and Stenting for Isolated Bilateral Aorto-Coronary

Ostial Stenosis in a Young Female

KK Haridas, Viveka Kumar, T Rajesh, M Vijaya Kumar, Benny PannekalDepartment of Cardiology, Amrita Institute of Medical Sciences and Research Centre, Cochin

An isolated bilateral aorto-coronary ostial stenosis is arare occurrence.1 Generally, lesions located within

3 mm of the origin of a vessel are considered ostial, but someauthors consider lesions up to 5 mm from the origin asostial.2 Besides atherosclerosis, aorto-coronary ostiallesions may occur uncommonly in various other diseasestates such as syphilis, Takayasu’s arteritis, fibromusculardysplasia and postradiation fibrosis.3–7 Cannulation of thecoronary artery during aortic valve surgery, coronaryangiography or angioplasty can also potentially lead tocoronary ostial stenosis.8 A bilateral tight coronary ostiallesion is potentially lethal and can present as sudden cardiacdeath, myocardial infarction, angina or ischemiccardiomyopathy.6,7,9 An isolated ostial left main coronaryartery (LMCA) stenosis and bilateral coronary ostial diseasewere earlier managed by conventional coronary arterybypass surgery (CABG). Surgical aorto-ostioplasty has alsobeen successfully used for managing these conditions.9 Withimprovement in hardware and increasing experience,isolated ostial disease of the LMCA has been managed bystenting alone or debulking and stenting.10–12 A cuttingballoon has been used for ostial lesions with or withoutstenting.13 However, the use of cutting balloon angioplastyand stenting has not been described for the managementof isolated bilateral aorto-ostial coronary artery lesions. Wereport the case of a young female who had tight bilateral

aorto-ostial coronary stenosis presenting with non-Q-waveanterior wall myocardial infarction, managed by cuttingballoon angioplasty and stenting.

Case Report

A 23-year-old female, mother of a 3-year-old child, washospitalized with unstable angina. She had a history ofcrescendo angina for the past 6 months. Herelectrocardiogram (ECG) showed 3 mm ST segmentdepression in the anterior precordial leads during angina,the Trop-T test was positive and CPK-MB was mildly raised.Her hemogram, glucose tolerance test and renal profile werenormal, but the erythrocyte sedimentation rate (ESR) was76 mm and C-reactive protein (CRP) level was elevated.Serological tests for syphilis and autoimmune disease werenegative and the lipid profile was well below theatherosclerotic risk range. Her echocardiogram showedmild aortic regurgitation (AR) and hypokinesia of theanterolateral wall with normal left ventricular (LV) systolicfunction.The coronary angiogram showed 90% discreteostial stenosis of the right coronary artery (RCA) andLMCA. The RCA was collateralizing with the left anteriordescending (LAD) coronary artery (Fig. 1a and b). An LVangiogram exhibited mild hypokinesia of the anterolateralwall with normal LV systolic function and an aortogramrevealed mild AR. The aortic lumen as well as carotid,subclavian and renal arteries appeared smooth and diseasefree. In view of her young age and severe symptoms,angioplasty and stenting of both ostia were planned.

Coronary artery disease involving both coronary ostia (left main and right coronary) is extremely rare in apremenopausal female, without pre-existing coronary risk factors. We report a case of tight bilateral coronaryostial disease which presented in unusual clinical circumstances in a young female, which was successfullyrevascularized by single-stage aorto-ostial cutting balloon angioplasty and stenting. (Indian Heart J 2001;53: 490–492)

Key Words: Coronary ostial stenosis, Stenting, Cutting balloon

Brief Reports

Correspondence: Dr KK Haridas, Department of Cardiology, AmritaInstitute of Medical Sciences and Research Centre, Amrita Lane, Elamakkara,Cochin 682026. e-mail: [email protected]

IHJ-209-01.p65 10/17/01, 1:54 PM490

Indian Heart J 2001; 53: 490–492 Haridas et al. Bilateral Ostial Aorto-Coronary Angioplasty 491

She was pretreated with soluble aspirin and clopidogrel.The patient’s consent was obtained for percutaneoustransluminal coronary angioplasty (PTCA) and stenting ofthe lesions. The procedure was performed under narcolepticsedation and local anesthesia with monitored anestheticcare and intra-aortic balloon pump (IABP) standby.

Technique: A 7 F conventional arterial sheath was placedpercutaneously in the right femoral artery. The left coronaryostium was partially engaged with a 7 F JL 3.5 short-tipguiding catheter as there was a marked fall in blood pressureand angina with full engagement. As the patient hadsignificant angina, 3 mg intravenous (i.v.) metoprolol wasgiven. The patient was heparinized with 12 500 units ofheparin. A 0.014" floppy-tipped guidewire was used to crossthe lesion, keeping the wire tip in the distal LAD. The lesionwas predilated with a 3.5×10 mm cutting balloon at 6–8 atm for 10–15 s, and stented with a 10 mm Prolink stentmounted on a 3.5×10 mm balloon. The stent was placedvery carefully to ensure optimal ostial coverage withminimal protrusion of the stent into the aorta. The guidingcatheter was further disengaged by pushing the guidewiredeeper during positioning of the stent. A deploymentpressure of 12 atm was used. The stented segment waspostdilated with the guiding catheter disengaged using a4×9 mm balloon at 14 atm. There was no residual stenosisor dissection after postdilatation (Fig. 2a). The rightcoronary ostium was partially engaged with a 7 F JR 3.5guiding catheter. The lesion was crossed with a 0.014"floppy-tipped guidewire, keeping the wire tip in the distalposterior descending artery (PDA). The ostial lesion of theRCA was dilated with a 3.5×10 mm cutting balloon at 6–8 atm. The RCA ostium was then stented with a 10 mmstent mounted on a 3.5×10 mm balloon, ensuring optimalostial coverage with the stent. During placement of thestent, the guiding catheter was disengaged into the aortaby pushing the guidewire deeper. The end result was goodwith no residual dissection or stenosis (Fig. 2b). At 2-monthfollow-up, the patient was asymptomatic.

Fig. 1. Diagnostic coronary angiogram. (a) Ostial left main coronary arterystenosis. (b) Ostial right coronary artery stenosis.

Fig. 2. Post-PTCA stent implantation image. (a) stented left main coronaryartery ostium. (b) Stented right coronary artery ostium.

Discussion

Coronary ostial stenosis involving both the coronary ostiais a rare occurrence. There have been few reports of bilateralcoronary ostial stenosis in young subjects without knownconventional coronary risk factors. In the absence ofconventional coronary risk factors, coronary ostialnarrowing has been reported in fibromuscular dysplasia,syphilitic aortitis, postradiation and Takayasu’s arteritis orwithout any recognizable etiological background.3–7

Our patient was a menstruating young woman withoutany previous illness or conventional atherosclerotic riskfactors, who presented with progressive angina culminatingin an acute coronary syndrome. Serological tests for syphiliswere negative. The elevated ESR and CRP in the setting ofminimal myocardial necrosis did not point to a specificetiology. However, in the setting of a mild aorticregurgitation, the aforementioned condition could occur inTakayasu’s arteritis. Currently, the diagnosis of Takayasu’sarteritis is based on the typical distribution of vesselinvolvement, elevated surrogate inflammatory markers inthe serum and response to anti-inflammatory therapy. Themajor limitation in the diagnosis of Takayasu’s arteritis isthe absence of specific tests.

Tight bilateral ostial stenosis of the coronary arteriesposes an imminent risk of sudden death due to globalischemia or infarction.1 This is a clinical emergency fromthe therapeutic viewpoint. The choice of therapy in aorto-ostial stenosis of the coronary artery has conventionallybeen CABG or surgical patch aorto-coronary ostioplasty.3–9

Aizawa et al.3 reported a patient with syphilitic aortitis andbilateral coronary ostial stenosis managed by CABG.Borolotti et al.9 reported their experience with four patientsin whom surgical ostio-aortoplasty was carried out forisolated coronary ostial stenosis.

There were major concerns with a surgical approach inour patient. Apart from the young age of the patient, therisk of progressive involvement of the aorta and its branchesin aortoarteritis could compromise distal coronary

a b a b

IHJ-209-01.p65 10/17/01, 1:54 PM491

492 Haridas et al. Bilateral Ostial Aorto-Coronary Angioplasty Indian Heart J 2001; 53: 490–492

perfusion. Surgical aorto-coronary ostioplasty in aninflamed vessel also had the potential of restenosis in thesurgical sites.

Stenting the ostium of the LMCA with or withoutdebulking is an attractive and less invasive option with lowprocedural risk and acceptable intermediate- and long-termresults in atherosclerotic coronary artery disease.10–12

However, the role of stenting in nonatherosclerotic aorto-ostial stenosis finds little mention in the literature. Lee4

reported successful performance of PTCA of the LMCAostium in Takayasu’s arteritis. Park et al.10 and Silvestriet al.11 reported stenting of the LMCA with a high successrate. Debulking with directional coronary atherectomy andstenting of the LMCA has been described for a large LMCA(>3.5 mm). There is no significant difference in therestenosis rate between stenting with or without debulkingfor LMCA stenosis.12

Cutting balloon angioplasty works on the principle ofproducing minute longitudinal incisions in the wall of thevessel.13 Radial dilatation of the vessel after producing theselinear incisions has the potential of better plaquecompression without extensive plaque contusion ordissection.14 Vessel recoil is less, making this an ideal formof balloon angioplasty in ostial lesions.

Since restenosis following stenting for non-atherosclerotic coronary disease is not well understood, werecommend a longer and more frequent follow-up in thissituation.

For isolated bilateral coronary ostial lesions, PTCA usinga cutting balloon followed by stenting is technically feasible.Early experience in this rare clinical condition suggests thatthis technique is safe and effective. At present, there is apaucity of data to show the long-term viability of thistreatment method. Further studies are needed before itsadvantage over surgery—especially surgical ostioplasty—is established.

References

1. Innami R, Nagaoka H, Oonuki M, Manabe S. Diffuse myocardialinfarction caused by isolated bilateral coronary ostial stenosis in ayoung woman: report of a case. Surg Today 1999; 29: 1120–1124

2. Pritchard CL, Mudd JG, Barner HB. Coronary ostial stenosis.Circulation 1975; 52: 46–48

3. Aizawa H, Hasegawa A, Arai M, Naganuma F, Hatori M, Kanda T,et al. Bilateral coronary ostial stenosis and aortic regurgitation dueto syphilitic aortitis. Intern Med 1998; 37: 56–59

4. Lee HY. Percutaneous transluminal coronary angioplasty for ostialstenosis of the left coronary artery—a case report and literaturereview. Yonsei Med J 1995; 36: 462–465

5. Kawakami H, Matsuoka H, Koyama Y, Saeki H, Inoue K, Nishimura K,et al. Isolated left coronary ostial stenosis as a result of fibromusculardysplasia in a young man. Jpn Circ J 2000; 64: 988–999

6. Takewa Y, Kawata T, Yoshida Y, Kawachi K, Kitamura S. Radiationinduced coronary ostial stenosis: a case of redo coronary bypass forthe restenosis following patch angioplasty. Nippon Kyobu Geka GakkaiZasshi 1996; 44: 220–225

7. Aronow H, Kim M, Rubenfire M. Silent ischemic cardiomyopathy andleft coronary ostial stenosis secondary to radiation therapy. ClinCardiol 1996; 19: 260–262

8. Gambhir DS, Trehan V, Batra R, Singh S. Left anterior descendingcoronary ostial stenosis. Balloon angioplasty, atherectomy, stenting,device or bypass surgery. Indian Heart J 1998; 50 (Suppl 1): 18–26

9. Borolotti U, Milano A, Balbarini A, Tartarini G, Levantino M,Borzoni G, et al. Surgical angioplasty for isolated coronary ostialstenosis. Tex Heart Inst J 1997; 24: 366–371

10. Park SJ, Park SW, Hong MK, Cheong SS, Lee CW, Kim JJ, et al. Stentingof unprotected left main coronary artery stenosis: immediate andlate outcome. J Am Coll Cardiol 1998; 31: 37–42

11. Silvestri M, Barragan P, Sainsous J, Bayet G, Simeoni JB, RoquebartPO, et al. Unprotected left main coronary artery stenting—immediateand medium term outcomes of 140 elective procedures. J Am CollCardiol 2000; 35: 1543–1550

12. Black A, Cortina R, Bossi I, Choussat R, Fajadet J, Macro J. Unprotectedleft main coronary artery stenting: correlates of mid term survivaland impact of patient selection. J Am Coll Cardiol 2001; 37:382–388

13. Kondo T, Kawakuchi K, Awaji Y, Mochizuki N. Immediate and chronicresults of cutting balloon angioplasty: matched comparison withconventional angioplasty. Clin Cardiol 1997; 20: 459–463

14. Pompa JJ, Lansky AJ, Purkayastha D, Hall LR, Bonan R. Angiographicand clinical outcome after cutting balloon angioplasty. J InvasiveCardiol 1996; 8: 12A–19A

IHJ-209-01.p65 10/17/01, 1:54 PM492

Indian Heart J 2001; 53: 493–495 Subramanyan et al. Transcatheter Closure of Coronary Fistula 493

Transcatheter Closure of a Large Coronary Artery Fistulawith Amplatzer Duct Occluder: A New Approach

Raghavan Subramanyan, Ajit Agrawal, Atul AbhyankarRoyal Hospital and Sultan Qaboos University Hospital, Muscat, Oman

Coronary artery fistulae (CAF) are characterized byabnormal communication between a coronary artery

and a cardiac chamber or vessel. Often they are detected asan incidental finding in an asymptomatic patient with amurmur. When the flow through the fistula is large, CAFcan cause heart failure, coronary steal and myocardialischemia, progressive enlargement and rupture, arrhythmiaor infective endocarditis.1–3 Closure is indicated for large CAFto prevent complications. As transcatheter techniquescontinue to evolve, nonsurgical closure of unnecessaryvascular structures including CAF is becoming morecommon using Gianturco coils, interlocking detachablecoils, detachable balloons, polyvinyl alcohol foam orRashkind’s double umbrella.4–9 Recently, the Amplatzer ductoccluder (ADO) has been used to close CAF using a femoralartery-to-femoral vein wire loop for deploying the device inthe fistula.10–13 To our knowledge, the following caserepresents the first report of an ADO being deployed in aCAF by the retrograde arterial approach using a coronaryangioplasty-guiding catheter.

Case Report

SK, a 27-year-old male, was found to have a continuousmurmur when examined for atypical chest pain and fatigue.The murmur was best heard along the right sternal border.The chest X-ray showed mild cardiomegaly with prominentpulmonary vascular markings, but the electrocardiogram

(ECG) was normal. On two-dimensional echocardiography,the right atrium (RA) and right ventricle (RV) wereenlarged. Doppler echocardiography showed a continuousflow into the upper part of the RA. A right coronary artery(RCA)-to-RA fistula was suspected. Cardiac catheterizationdemonstrated a significant left-to-right shunt in the RA(Qp/Qs 1.7). The pressures in the pulmonary artery and leftventricle (LV) were normal. Aortogram and selectivecoronary angiograms were carried out. The proximal RCAwas dilated (9 mm). A large tortuous fistula was seen toarise from the RCA proximal to the ventricular branchesand run in a posterior direction to open high in the RA(Fig. 1). The mid-segment of the fistula showed a relativelynarrow neck with an aneurysmal pouch (10 mm indiameter) distal to the neck (Fig. 2). The flow through thefistula was large while the RCA branches distal to the fistulafilled very faintly. The left coronary artery and the LV werenormal on angiography.

Because of the patient’s symptoms, significant shuntdetected by oximetry and large size of the fistula, we decidedto close it by the transcatheter method. The RCA wasengaged with an 8 F multipurpose large-lumen coronaryangioplasty-guiding catheter. A 0.035" Terumo guidewirewas introduced through this catheter into the RCA andmanipulated through the fistula into the RA. Keeping agenerous loop of guidewire in the RA, the guiding catheterwas gently and carefully advanced on the wire until its tipcrossed the narrow neck of the CAF. The position andmanipulation of the catheter was guided by small handinjections of contrast. An ADO (10–8 mm; AGA MedicalCorp., Minnesota, USA) was loaded and introduced throughthe guiding catheter. At this stage, it was necessary to cut

We report a new retrograde approach for the successful closure of a large right coronary artery fistula in a27-year-old man using the Amplatzer duct occluder. The device was deployed through a coronary angioplasty-guiding catheter that had been advanced through the aorta and the dilated right coronary artery into the fistula.This method simplified the procedure by eliminating the need for making a femoral artery-to-femoral vein wireloop. (Indian Heart J 2001; 53: 493–495)

Key Words: Catheter intervention, Coronary anomaly, Amplatzer device

Brief Report

Correspondence: Dr Raghavan Subramanyan, Senior Consultant, RoyalHospital, PO Box 1331, PC 111, CPO Seeb, Sultanate of Omane-mail: [email protected]

IHJ-210-01.p65 10/17/01, 1:57 PM493

494 Subramanyan et al. Transcatheter Closure of Coronary Fistula Indian Heart J 2001; 53: 493–495

Fig. 1. Aortic root angiogram (right anterior oblique projection) showing themarkedly dilated proximal RCA and the large, densely opacified fistula connectedto the RA.

Fig. 2. Selective right coronary arteriography (right anterior obliqueprojection) shows the large fistula opening into the RA. The mid-segment showsa stenosis (arrow) with an aneurysmal pouch distal to it.

and shorten the guiding catheter so that the device couldbe delivered through it. The ADO was advanced into thefistula until its distal rim opened in the pouch beyond theneck. Initially the whole system and then the guidingcatheter alone were slowly drawn back to deploy theoccluder in a dumb-bell shape across the neck. Gentle pushand pull on the delivery wire confirmed that the occluderwas in a stable position. No symptoms, ECG orhemodynamic changes were noted and the murmurdisappeared. After waiting for 10 minutes, the device wasreleased. Repeat angiography showed complete closure ofthe CAF with improved flow and filling of the distal RCAand its branches (Figs. 3 and 4). Doppler echocardiographyon the following day showed that the device was stable and

Fig. 3. Aortic root angiogram (right anterior oblique projection) afterdeployment of the ADO (arrow) showing complete occlusion of the fistula.

Fig. 4. Aortic root angiogram (left anterior oblique projection) after deploymentof the ADO (arrow). The fistula is occluded and there is improved filling of thedistal branches of the RCA.

there was no residual flow. At follow-up after 6 months thepatient was well. His ECG was normal and Dopplerechocardiography showed no shunt.

Discussion

This case demonstrates a simple and quick method ofclosing a large CAF through the retrograde arterialapproach using an ADO. Earlier reports on the use of anADO in CAF have described deployment of the devicethrough the venous route.11–14 This necessitates a femoralartery-to-femoral vein guidewire loop for the introductionof a delivery sheath of appropriate size, which makes theprocedure longer and more difficult.

IHJ-210-01.p65 10/17/01, 1:57 PM494

Indian Heart J 2001; 53: 493–495 Subramanyan et al. Transcatheter Closure of Coronary Fistula 495

In our case, the fistula originated from the proximal partof the RCA. Thus, it was possible to cannulate the fistuladeeply without causing any major distortion or damage tothe vessel. We preferred to use a coronary angioplasty-guiding catheter for delivery of the ADO because its softatraumatic tip and relative resistance to kinking are definiteadvantages over the Amplatzer delivery sheath. Theanatomy of the CAF showed a favorable “neck”, whichallowed us to deploy the occluder in a stable position,although the wide (aortic) end of the ADO was directedtowards the distal part of the fistula. The large size of thefistula and the dilated pouch beyond the neck allowed theADO to expand without causing excessive distention of thevessel. This was important because CAF often have thinwalls that can rupture.

The course of the CAF suggested that it was in the atrialbranch of the RCA. Closure did not produce any untowardhemodynamic or ischemic changes, and the rhythmremained normal. In fact, filling of the distal RCA branchesimproved substantially once the steal through the CAF wasabolished. This method can be employed whenever thefistula is large and proximal in origin, allowing for safepassage of a relatively large delivery catheter or sheath. Weanticipate difficulty in using this approach in small childrenor in cases where the CAF arises in the distal part of thecoronary artery. Several occluding devices are available forclosure of CAF. Coils and microparticle embolization are notsuitable for large CAF due to the risk of incomplete closureand embolism.4,7–9,13 Detachable balloons and Rashkind’sdouble-umbrella device have been used in large fistulae butthey need a large delivery sheath which can be introducedonly over an arteriovenous wire loop.6–9 The ADO has severaladvantages such as a smaller delivery system, ability torecapture and reposition, and a high rate of completeclosure.14

Conclusions: In this report, we have shown that selectedCAF, especially if large and proximal in origin, can besuccessfully closed with an ADO through a simpleretrograde arterial approach alone, using a coronaryangioplasty-guiding catheter for delivery.

References

1. Kirklin JW, Barratt-Boyes BG. Congenital anomalies of the coronaryarteries. Cardiac surgery. New York: John Wiley and Sons; 1987.pp. 945–955

2. Liberthson RR, Sagar K, Berkoben JP, Weintraub RM, Levine FH.Congenital coronary arteriovenous fistula. Report of 13 patients,review of the literature and delineation of the management.Circulation 1979; 59: 849–854

3. Corvaja N, Moses JW, Vogel FE, Javit DJ, Ziolo G, Frumkin WJ, et al.Exercise-induced ventricular tachycardia associated with coronaryarteriovenous fistula and correction by transcatheter coilembolization. Catheter Cardiovasc Interv 1999; 46: 470–472

4. Strunk BL, Hieshima GB, Shafton EP. Treatment of congenitalcoronary arteriovenous malformations with micro-particleembolization. Cathet Cardiovasc Diagn 1991; 22: 133–136

5. Graeb DA, Morris DC, Ricci DR, Tyers GF. Balloon embolization ofiatrogenic aortocoronary arteriovenous fistula. Cathet CardiovascDiagn 1990; 20: 58–62

6. Krabill KA, Hunter DW. Transcatheter closure of congenital coronaryarterial fistula with a detachable balloon. Pediatr Cardiol 1993; 14:176–178

7. Dorros DG, Thota V, Ramireddy K, Joseph G. Catheter-basedtechniques for closure of coronary fistulae. Catheter Cardiovasc Interv1999; 46: 143–150

8. Reidy JF, Anjos RT, Qureshi SA, Baker EJ, Tynan MJ. Transcatheterembolization in the treatment of coronary artery fistulas. J Am CollCardiol 1991; 18: 187–192

9. Perry SB, Rome J, Keane JF, Baim DS, Lock JE. Transcatheter closureof coronary artery fistulas. J Am Coll Cardiol 1992; 20: 205–209

10. Hakim F, Madani A, Goussous Y, Cao QL, Hijazi ZM. Transcatheterclosure of a large coronary arteriovenous fistula using thenew Amplatzer duct occluder. Cathet Cardiovasc Diagn 1998; 45:155–157

11. Thomson L, Webster M, Wilson N. Transcatheter closure of a largecoronary artery fistula with Amplatzer duct occluder. CatheterCardiovasc Interv 1999; 48: 188–190

12. Pedra CA, Pihkala J, Nykanen DG, Benson LN. Antegradetranscatheter closure of coronary artery fistula using vascularocclusion devices. Heart 2000; 83: 94–96

13. Okubo M, Nykanen D, Benson LN. Outcomes of transcatheterembolization in the treatment of coronary artery fistulas. CatheterCardiovasc Interv 2001; 52: 510–517

14. Masura J, Walsh KP, Thanopoulous B, Chan C, Bass J, Goussous Y, etal. Catheter closure of moderate-to large-sized patent ductusarteriosus using the new Amplatzer duct occluder: immediate andshort-term results. J Am Coll Cardiol 1998; 31: 878–882

IHJ-210-01.p65 10/17/01, 1:57 PM495

yncope, which accounts for 3% of emergency room 1Svisits and 6% of all hospital admissions, is an ominous

symptom in patients with pre-excitation since it may 2indicate a higher risk for sudden death. We describe an

interesting case of recurrent syncope in a young women

with pre-excitation.

Case Report

An 18-year-old female presented with a 4-year history of

multiple episodes of paroxysmal palpitation. These were

often accompanied by syncope lasting for a few seconds to

minutes. She had experienced nearly 40 episode of syncope

over the past few years. Her basal electrocardiogram (ECG)

demonstrated minimal pre-excitation, suggestive of a left-

sided accessory pathway (AP) (Fig. 1a). There was no

documented tachycardia. The clinical examination, chest

X-ray and two-dimensional (2-D) echocardiographic

examination were normal. Serum electrolyte levels and

routine biochemistry were also within normal limits. The

patient was not on any antiarrhythmic drugs prior to the

electrophysiological study (EPS).

The syncopal episodes were thought to be related to the

pre-excitation, either due to atrial fibrillation (AF) with

rapid conduction over the AP or by rapid circus movement

tachycardia leading to hemodynamic instability. She was

taken up for EPS with a plan for radiofrequency ablation

(RFA). Electrode catheters were placed in the coronary

sinus, bundle of His and right ventricle via the femoral

Pre-excitation with Syncope: A False Lead?

An 18-year-old girl with pre-excitation presented with a history of recurrent syncope preceded by palpitation. The accessory

pathway, which had a relatively long antegrade effective refractory period of 340 ms, was mapped and successfully ablated in

the left lateral region. However, after ablation, she had reproducible sustained polymorphic ventricular tachycardia, which

was found to be the cause of her syncope. Thus, alternate mechanisms of tachycardia need to be considered in patients with

pre-excitation when the presentation is atypical. (Indian Heart J 2001; 53: 496-498)

Key words: Syncope, Pre-excitation, Ventricular tachycardia

route. The basal intervals were: PR 118 ms; AH 74 ms; HV

19 ms; QRS 98 ms; and QT 346 ms. The ventricular

stimulation protocol revealed eccentric retrograde

activation earliest in the distal coronary sinus, confirming a

left lateral AP. Atrial-and ventricular-programmed extra

stimulation and continuous pacing protocols were

performed at different cycle lengths. The antegrade and

retrograde effective refractory periods (ERPs) of the AP

were 340 ms and 300 ms, respectively. Nonsustained

orthodromic tachycardia was inducible with atrial extra

stimuli. Mapping was performed via the transaortic route

with a 7F ablation catheter (Cordis-Webster). The earliest

activation (-19 ms) was found in the left lateral region on the

ventricular aspect of the mitral annulus. The pre-excitation

disappeared immediately after starting the RF energy which 0was delivered for 60s with a peak temperature of 60 C and

power of 48 Watts. The 12-lead ECG was normal after the

procedure (Fig. 1b).

Surprisingly, the patient developed sustained

spontaneous polymorphic ventricular tachycardia (VT) a

few seconds later (Fig. 2). This VT persisted despite

withdrawal of the catheters from the ventricles and

terminated spontaneously only after 90s. The systolic BP

during VT was 70 mmHg. Since the refractory periods in the

AP were relatively long, we considered the possibility of

idiopathic polymorphic VT as the cause of syncope. Hence,

a complete VT induction protocol was performed.

Ventricular fibrillation (VF) was induced using three-

programmed extra stimuli, each at a cycle length greater

than 200 ms (Fig. 3), and was promptly defibrillated.The patient was started on amiodarone (200 mg daily

after and initial oral loading dose of 600 mg daily for 10

Correspondence: Dr Yash Y Lokhandwala, Associate Professor

of Cardiology, KEM Hospital, Parel, Mumbai 400012

e-mail: [email protected]

Brief Report

Satish C Toal, Pratap J Nathani, Anoop K Gupta, Yash Y LokhandwalaDepartment of Cardiology, GS Medical College and KEM Hospital, Mumbai

days) and atenolol (25 mg) prior to discharge. She was

asymptomatic at 3-month follow-up.

Discussion

The prevalence of syncope in patients with Wolff-3Parkinson-White (WPW) syndrome is about 20%. The

cause of syncope is often assumed to be secondary to an

arrhythmia utilizing an AP. Atrial fibrillation with rapid

conduction over an AP is probably the most common cause

of syncope in these patients, although a rapid

atrioventricular reciprocating tachycardia may also cause a 4decrease in cardiac output and loss off consciousness.

Clinical VT is rare in patients with WPW syndrome and

usually occurs in association with an underlying cardiac 5disease. Ventricular tachycardia causing syncope in

patients with pre-excitation has rarely been documented by 6EPS. Wellens, while studying the electrophysiological

properties of 322 patients with pre-excitation, found that

only 2 had VT independent of their AP, confirming that this 4is a rare occurrence. Lloyd et al. reported 5 patients in

whom EPS revealed VT rather than supraventricular

tachycardia as the cause of syncope. However,

nonsustained VT during programmed ventricular

stimulation is a nonspecific finding and is not a marker of 5sudden cardiac death in patients with WPW syndrome.

In the present case, the antegrade refractory period

of the AP was not short enough to explain the occurrence of

AF which was rapidly conducted over the AP leading to VF,

as a cause of syncope. Moreover, numerous episodes of

syncope are extremely rare in WPW syndrome. On the

contrary, the reproducible induction of polymorphic VT

rapidly degenerating into VF was presumably the cause of

the syncopal episodes. Idiopathic VT/VF is a rare but well 7documented cause of sudden death in young adults. The

pre-excitation pattern was a false lead in our patient and the

true cause of syncope was discovered fortuitously.

References

1. Kapoor WN. Evaluation and management of syncope. JAMA

1992; 268: 2553-2560

2. Kapoor WN, Karpf M, Wieand S, Peterson JR, Levey GS. A

prospective evaluation and follow up of patients with

Fig. 1. (a) The 12-lead ECG Demonstrating Minimal pre-

excitation; and (b) 12-lead ECG after radiofrequency

ablation.

Fig. 2. Spontaneously induced sustained polymorphic VT

after Ablation of the accessory pathway.

Fig. 3. Induction of ventricular fibrillation with

programmed ventricular extra stimuli.

Indian Heart J 2001; 53: 496-498 Toal et al. Pre-excitation with Syncope: A False Lead? 497

syncope. N Engl J Med 1983; 309: 197-2043. Brooks R, Ruskin JN. Evaluation of the patient with

unexplained syncope. In: Zipes DP, Jalife J (eds). Cardiac

electrophysiology: from cell to bedside. 2nd ed. Philadelphia:

WB Saunders Company; 1995. pp. 1247-12614. Lloyd EA, Hauer RN, Zipes DP, Heger JJ, Prystowsky En.

Syncope and ventricular tachycardia in patients with

ventricular pre-excitation. Am J Cardiol 1983; 52: 79-82

5. Yee R, Klien GJ, Guiraudon Gm. The Wolff-Parkinson-White

Syndrome. In: Zipes DP, Jalife J (eds). Cardiac

electrophysiology: from cell to bedside. 3rd ed. Philadelphia:

WB Saunders Company; 2000. pp. 845-8616. Wellens HJJ. In: Kulbertus AE (ed.) Reentrant arrhythmias.

London; 1977.p. 1537. Reichenbach DD, Moss NS, Meyer E. Pathology of the heart

in sudden cardiac death. Am J Cardiol 1977; 39: 865-872

Indian Heart J 2001; 53: 496-498 498 Toal et al. Pre-excitation with Syncope: A False Lead?

Indian Heart J 2001; 53: 499–502 Kothari et al. Aneurysm of the Vein of Galen in Neonates 499

Aneurysm of the Vein of Galen in Neonates:Report of Four Cases

SS Kothari, N Naik, R Juneja, A SaxenaDepartment of Cardiology, Cardiothoracic Sciences Centre, All India Institute of Medical Sciences, New Delhi

Heart failure and cyanosis in infants almost alwaysresults from congenital heart disease. Systemic

arteriovenous malformations (AVMs) are a rare cause ofsuch a presentation and the diagnosis is often missed. Wereport 4 cases of AVMs related to malformation of thevein of Galen who presented to us in the past five years.The purpose of this report is to highlight the clinicalpresentation and increase awareness of this entity. Theclinical diagnosis was missed in the first case but correctlymade in the 3 subsequent cases.

Case Reports

Case 1: A full-term male child presented at 5 days of agewith severe congestive heart failure (CHF) and mildcyanosis. The peripheral pulses were bounding and therewas no radiofemoral delay. The first heart sound was normalwhile the second had a loud pulmonic component. Therewas a grade 3/6 systolic murmur heard best in thepulmonary area. The electrocardiogram (ECG) showedevidence of biventricular hypertrophy while the chestX-ray showed cardiomegaly. Echocardiography did notreveal any structural anomaly except that the pulmonaryvenous drainage was not clearly visualized. On cardiaccatheterization, the aortic root angiogram showed markedlyenlarged and tortuous neck vessels, suggesting anintracardiac AVM. The diagnosis was confirmed on imagingthe cranium (Fig. 1). On detailed clinical examination, agrade 2/6 systolic bruit that had been missed in the initialclinical evaluation was heard over the cranium.

Case 2: A full-term male child presented at 2 days of agewith severe CHF and mild cyanosis. His peripheral pulseswere bounding in all four extremities. The first heart soundwas normal while the second had an accentuated pulmoniccomponent. There was a grade 3/6 systolic murmur heardbest in the pulmonary area. Due to our experience with theprevious patient who had a similar presentation, a cranialbruit was sought and found. The ECG showed evidence ofbiventricular hypertrophy while the chest X-ray showedcardiomegaly with increased pulmonary vascularity.

Correspondence: Dr SS Kothari, Additional Professor of Cardiology,Cardiothoracic Centre, All India Institute of Medical Sciences, New Delhi110029. e-mail: [email protected]. net.in

Brief Report

In neonates, aneurysm of the vein of Galen often masquerades as cyanotic congenital heart disease. We report4 cases of neonates presenting with malformation of the vein of Galen at our insititution. An increased awarenessof this entity seems warranted. (Indian Heart J 2001; 53: 499–502)

Key Words: Aneurysm, Heart failure, Neonates

Fig. 1. Digital subtraction angiogram showing vein of Galen malformation(arrows), dilated carotid artery and jugular vein (arrowheads). AO denotesaortic arch.

IHJ-192-01.p65 10/17/01, 1:44 PM499

500 Kothari et al. Aneurysm of the Vein of Galen in Neonates Indian Heart J 2001; 53: 499–502

Echocardiography revealed dilated right heart chambers, apatent foramen ovale, mild tricuspid regurgitation and anenlarged superior vena cava (SVC) and arch of the aorta(Fig. 2). Pulse Doppler study of the SVC showed increasedforward flow with normal flow in the inferior vena cava(IVC). Doppler study of the ascending aorta and carotidarteries showed continuous forward flow (Fig. 3) while thatof the descending aorta suggested diastolic run-off. Real-time ultrasound examination of the cranium suggestedmalformation of the vein of Galen and a contrast-enhancedcomputed tomographic (CT) scan of the head revealed ananeurysm of the vein of Galen.

Case 3: A 6-day-old male child had a history of CHF andmild cyanosis from the first day of life. His clinical findingswere similar to those of the previous patients. The ECG was

normal for his age while the chest X-ray showedcardiomegaly with increased pulmonary blood flow.Echocardiography, transcranial ultrasound and contrast-enhanced CT scan of the brain identified a malformationof the vein of Galen.

All three neonates succumbed to their illness within aweek of being diagnosed.

Case 4: A 5-day-old child presented with findings similarto those of the 3 previous patients. Computed tomographicscan of the head showed a large malformation of the veinof Galen (Fig. 4). The parents were keen on aggressivetreatment despite the poor prognosis. The patient was takenup for embolization of the feeding vessels using N-butylcyanoacrylate. There was a decrease in the size of theaneurysm after embolization (Figs 5a and b) along with amarked improvement in CHF, and a significant reductionin cardiac size was seen in the chest X-ray (Figs 6a and b).However, the child succumbed to his illness the next day.

Discussion

Arteriovenous malformations are an uncommon cause ofCHF in the neonatal period.1 Intracerebral AVMs being themost common amongst them. Although malformations ofthe vein of Galen constitute only 1% of all cerebral vascular

Fig. 2. Two-dimensional echocardiogram showing enlarged arch of aorta (AO)and arch vessels. PA: pulmonary artery; CC: left common carotid artery.

Fig. 3. Pulse Doppler echocardiogram with cursor in the carotid artery showingcontinuous antegrade flow.

Fig. 4. CT scan of the cranium showing vein of Galen malformation(white arrows), dilated lateral ventricles (black arrows), periventricularcalcification (black arrowheads) and changes of encephalomalacia in bothcerebral hemispheres.

IHJ-192-01.p65 10/17/01, 1:44 PM500

Indian Heart J 2001; 53: 499–502 Kothari et al. Aneurysm of the Vein of Galen in Neonates 501

Fig. 6a and b. Chest X-ray of patient 4 before and after embolization of thevein of Galen malformation, showing immediate marked reduction incardiomegaly post-embolization.

malformations, they comprise up to 30% of all pediatricvascular malformations.1,2 The vein of Galen (formed fromthe median vein of the prosencephalon) is formed by theunion of the two internal cerebral veins and goes on to draininto the straight sinus. An AVM occurs when the vein ofGalen has an arterial input from one or more majorintracranial arteries, either directly or via an interposedangiomatous malformation.3

Symptoms and clinical signs vary with the age atpresentation.4 In the newborn, the commonest mode ofpresentation is intractable heart failure. The malformationis more extensive in these infants as they receive inputs fromnumerous arterial feeders, and the prognosis is extremelypoor with almost 100% mortality. In infants with a lesserdegree of shunt, the clinical picture is that of an enlarginghead (due to both obstructive and nonobstructivehydrocephalus) and mild heart failure. Older children andadults may present with focal neurological signs,developmental delay, seizures, headache or subarachnoidhemorrhage.

Diagnostic confusion in the neonate arises from thecombination of severe CHF and cyanosis, pointing towardsa cardiac disorder. Cyanosis results from the torrentialvenous return crossing the patent foramen ovale and adiagnosis of persistent fetal circulation may be consideredby the unwary. Detailed clinical examination, however,reveals evidence of high-output heart failure and a cranialbruit (up to 80% of the cardiac output can be directedtowards the cerebrovascular bed due to low resistancewithin the vascular malformation). Brisk upper limb pulsesin the presence of severe heart failure should suggest thepossibility of an AVM. Lower limb pulses may be relativelyfeeble due to steal by the cerebral fistula.5 In the extremelysick child, however, all pulses may be feeble. Although abenign systolic bruit is described in up to 15% of normalneonates and children,6 the presence of a systolic bruit overthe cranium in a child with CHF should strongly suggestthis disorder. All 4 of our patients had a systolic bruit overthe cranium.

Two-dimensional echocardiography usually revealsnormal cardiac anatomy with dilated right heart chambersand an enlarged SVC and arch vessels.7 Uncommonly,structural heart disease has been reported in these infantsincluding coarctation of the aorta, atrial septal defect,ventricular septal defect and partial anomalous pulmonaryvenous drainage.8 Pulse Doppler echocardiography of thearch vessels demonstrates a continuous forward flow ofhigh velocity which reflects the low peripheral resistancebed introduced by the AVM. High-velocity forward flow inthe SVC with normal flow in the IVC are also seen. Thesefeatures, in the absence of aortic regurgitation and a patentductus arteriosus, suggest the extracardiac nature of thelesion and its intracerebral location. Transcranialultrasonography is extremely useful in demonstrating theAVM while Doppler studies aid in demonstrating the flowin it. A semi-quantitative measurement of flow in thefeeding and draining vessels can also be obtained. ColorDoppler echocardiography helps to demonstrate theturbulent flow within the aneurysm.9 In addition, contrastvenous echocardiography has been shown to be of use asthe rapid return of microbubbles into the SVC is apathognomonic sign of an intracerebral AVM in the absenceof another peripheral left-to-right shunt in the upper partof the body. The contrast can be injected into a peripheralleg vein (which then crosses over from the right to the leftacross a patent foramen ovale) or directly into the aortathrough an umbilical artery catheter.

Although CT scan of the brain demonstrates the AVMand visualizes the feeding vessels, the details thus obtainedmay not be sufficient to guide surgical treatment.10

Fig. 5a and b. Selective pre- and post-glue embolization angiograms.Microcatheter in left vertebral artery with its tip in posterior choroidalartery (arrowheads) demonstrating a large choroidal type malformation.Marked reduction in aneurysm size is seen after embolization of feedervessel.

a b

a b

IHJ-192-01.p65 10/17/01, 1:45 PM501

502 Kothari et al. Aneurysm of the Vein of Galen in Neonates Indian Heart J 2001; 53: 499–502

Computed tomography also depicts associated brain lesionssuch as calcification and anoxic brain damage. Magneticresonance imaging (MRI) is superior to CT in its ability todemonstrate vascular anatomy, though it is unlikely tosupplant angiography in the immediate future11 sincesufficient details of arterial and venous anatomy whichpermit precise endovascular/surgical therapeutic decisionsare usually not obtained. Prenatal diagnosis of this disorderis also possible and one can precisely delineate the vascularmalformation.12,13

Embolization of the feeding arteries of the AVM (in asingle or staged sitting) is the preferred therapeutic modalityfor a patient in severe CHF. Transarterial embolization isthe preferred route although transvenous and transocularroutes have also been used. Solid materials, includingmicrocoils (mostly fibered or unfibered platinum coils),microballoons and silk sutures have been used to embolizethese vessels, with variable success.1,2,14 Liquid adhesivesthat have been used for embolization include cyanoacrylatemonomers such as I-butyl cyanoacrylate and N-butylcyanoacrylate, and polymers such as ethylene vinyl alcoholcopolymer.2,14 Even partial reduction in cerebral flow afterembolization is sufficient for controlling CHF in the neonateas it permits retrograde thrombosis of the AVM. Themortality rate in neonates remains extremely high (up to55%) even after embolization, partly due to the large extentof the malformation.15 Some recent reports have, however,indicated better immediate neonatal outcome.16,17

Embolization promptly alleviates the massive intracranialarteriovenous shunting and improves CHF. In infants inwhom endovascular therapy is delayed, clinical results maybe poor despite obliteration of the aneurysmalmalformation. This is believed to be the result of an acquiredocclusive venopathy affecting the dural venous sinuses dueto high flow. Gamma knife surgery is not a viable option inthe neonate as it does not produce any immediate reductionin the size of the AVM. Hence, it is not useful for treatingthe neonate in intractable CHF but is more appropriate forthe clinically stable infant or for managing a residual AVMafter endovascular therapy.18 Surgical clipping of theseaneurysms as the primary procedure is no longer done dueto the high procedural mortality (up to 90% in neonatesand 55% across all age groups) and postoperativemorbidity.1,19 However, surgical clipping may be performedif CHF persists after endovascular therapy.

In conclusion, aneurysm of the vein of Galen, a rarecause of cyanosis and heart failure in infants, can bediagnosed clincally in the appropriate setting. The extensivedistribution of the aneurysm usually precludes surgicalmanagement/endovascular therapy.

References

1. Johnston IH, Whittle IR, Besser M, Morgan MK. Vein of Galenmalformation: diagnosis and management. Neurosurgery 1987; 20:747–758

2. Lylyk P, Vinuela F, Dion JE, Duckwiler G, Guglielmi G, Peacock W, etal. Therapeutic alternatives for vein of Galen vascular malformations.J Neurosurg 1993; 78: 438–445

3. Litvak J, Yahr MD, Ransohoff J. Aneurysms of the great vein of Galenand midline cerebral arteriovenous malformations. J Neurosurg 1960;17: 945–954

4. Berenstein A, Lasjaunias P. Arteriovenous fistulas of the brain. In:Surgical neuroangiography. 1st ed. Berlin: Springer-Verlag; 1992.pp. 267–317

5. O’Donnabhain D, Duff DF. Aneurysms of the vein of Galen. Arch DisChild 1989; 64: 1612–1617

6. Hughes R, Todd RM. Intracranial bruits in infants and children. ArchDis Child 1953; 28: 198–199

7. Sivakoff M, Nouri S. Diagnosis of vein of Galen malformation by two-dimensional ultrasound and pulsed Doppler method. Pediatrics 1982;69: 84–86

8. McElhinney DB, Halbach VV, Silverman NH, Dowd CF, Hanley FL.Congenital cardiac anomalies with vein of Galen malformations ininfants. Arch Dis Child 1998; 78: 548–551

9. Fong LV, Lee SH, Salmon AP. Diagnosis of cerebral arteriovenousmalformation by color Doppler examination. Eur Heart J 1992; 13:415–417

10. Martelli A, Scotti G, Harwood-Nash DC, Fitz CR, Chuang SH.Aneurysms of the vein of Galen in children: CT and angiographiccorrelations. Neuroradiology 1980; 20: 123–133

11. Seidenwurm D, Berenstein A, Hyman A, Kowalski H. Vein of Galenmalformation: correlation of clinical presentation, arteriography andMR imaging. AJNR 1991; 12: 347–354

12. Lee TH, Shih JC, Peng SS, Lee CN, Shyu MK, Hsieh FJ. Prenataldepiction of angioarchitecture of an aneurysm of the vein of Galenwith three-dimensional color power angiography. Ultrasound ObstetGynecol 2000; 15: 337–340

13. Chisholm CA, Kuller JA, Katz VL, McCoy MC. Aneurysm of the veinof Galen: prenatal diagnosis and perinatal management. Am JPerinatol 1996; 13: 503–506

14. Schweitzer JS, Chang BS, Madsen P, Vinuela F, Martin NA, MarroquinCE, et al. The pathology of arteriovenous malformations of the braintreated by embolotherapy, II: results of embolization with multipleagents. Neuroradiology 1993; 35: 468–474

15. Halbach NV, Dowd CF, Higashida RT, Balousek PA, Ciricillo SF, EdwardsMS. Endovascular therapy of mural-type vein of Galenmalformations. J Neurosurg 1998; 89: 74–80

16. Horowitz MB, Jungreis CA, Quisling RG, Pollack I. Vein of Galenaneurysms: a review and current perspectives. AJNR 1994; 15:1486–1496

17. Borthne A, Carteret M, Baraton J, Courtel J, Brunelle F. Vein of Galenmalformation in infants: clinical, radiological and therapeutic aspects.Eur Radiol 1997; 7: 1252–1258

18. Payne BR, Prasad D, Steiner M, Bunge H, Steiner L. Gamma surgeryfor vein of Galen malformations. J Neurosurg 2000; 93: 229–236

19. Hoffman HJ, Chuang S, Hendrick EB, Humphreys RP. Aneurysms ofthe vein of Galen: experience at the Hospital for Sick Children,Toronto. J Neurosurg 1982; 57: 316–322

IHJ-192-01.p65 10/17/01, 1:45 PM502

Indian Heart J 2001; 53: 503–504 Mohan et al. Congenitally Unguarded Tricuspid Valve Orifice 503

Congenitally Unguarded Tricuspid Valve Orifice with a GiantRight Atrium and a Massive Clot in an Asymptomatic Adult

Jagdish C Mohan, Partho P Sengupta, Ramesh AroraDepartment of Cardiology, GB Pant Hospital, New Delhi

An unguarded tricuspid valve is a rare congenitalanomaly described in only a few antemortem case

reports.1–9 It is a variant of tricuspid valve dysplasia whereinthere is partial or complete agenesis of the tricuspid valvetissue.2 The leaflets are normally inserted on the ring andthere is variable dysplasia of the chordae tendineae andpapillary muscles. Most of the cases reported in theliterature are associated with pulmonary atresia. Anunguarded tricuspid valve orifice with a patent rightventricular outflow tract and dilated right ventricle isuncommon and needs to be differentiated from the morecommon entity—Ebstein’s anomaly of the tricuspid valve.The clinical presentation is usually in early childhood withcyanosis and/or congestive heart failure but there are a fewpatients in whom decompensation occurs during adult lifewith right ventricular failure and tricuspid regurgitation.7–9

The natural outcome, with or without medical treatment,is not too dismal and one such patient survived till the ageof 53 years.7 The advent of echocardiography has resultedin the diagnosis of a significant number of such patients,some of whom may present with associated atrialfibrillation.8,9 This report describes an asymptomatic youngman who on echocardiographic examination showed alarge right atrial clot, intense spontaneous echo contrastand a giant right atrium in association with an unguardedtricuspid valve orifice.

Case Report

A 25-year-old asymptomatic young man was evaluated by

echocardiography because of abnormal findings on a 12-lead electrocardiogram (ECG) and chest X-ray. He deniedthe presence of any symptoms on exertion. Physicalexamination revealed a healthy, afebrile young man with asupine blood pressure of 112/76 mmHg, a pulse rate of 76beats/min, irregular rhythm and elevated jugular venouspressure. Mild nontender hepatomegaly was present andprecordial examination revealed faint heart sounds with awide, variably split second sound and a grade 2/6 basalejection systolic murmur. Hematological parameters andserum biochemistry were normal. A 12-lead ECG revealedatrial fibrillation with an average ventricular rate of 68beats/min, complete right bundle branch block andnonspecific ST–T-wave changes. A plain chest X-ray showedcardiomegaly (cardiothoracic ratio—70%) and oligemiclung fields.

Two-dimensional echocardiography (Fig. 1) showedan elongated left atrium and a small left ventriclewith paradoxical ventricular septal motion. The rightatrium was dilated (12×10 cm) with intense spontaneousecho contrast and a large mobile thrombus (6×5 cm)was attached to its free wall. The septal leaflet of thetricuspid valve was intact and normally inserted, but theanterior and posterior tricuspid leaflets and the subvalvarapparatus were not visible. The inferior vena cava wasdilated and showed no respiratory variation, while theright ventricular outflow tract was dilated with normalpulmonary arteries. Doppler interrogation of theright ventricular inflow and outflow tracts showed a low-velocity, to-and-fro flow with a peak velocity of less than 1m/s. The patient declined further investigation andsurgery.

Brief Report

Correspondence: Dr JC Mohan, Professor of Cardiology, GB Pant Hospital,New Delhi 110002. e-mail: [email protected]

Congenitally unguarded tricuspid valve orifice, a variant of tricuspid valve dysplasia, is a rare malformationwith protean manifestations. This report describes an asymptomatic adult who, on echocardiographicexamination ordered in view of an abnormal 12-lead surface electrocardiogram and plain chest X-ray, wasfound to have an unguarded tricuspid valve orifice with a giant right atrium (12×10 cm), intense spontaneousecho contrast and a large right atrial clot. (Indian Heart J 2001; 53: 503–504)

Key Words: Congenital heart defects, Echocardiography, Thrombosis

IHJ-134-01.p65 10/17/01, 12:16 PM503

504 Mohan et al. Congenitally Unguarded Tricuspid Valve Orifice Indian Heart J 2001; 53: 503–504

Discussion

Congenitally unguarded tricuspid orifice is a rare anomalyand antemortem diagnosis has been reported in theliterature in about 17 cases only.1–9 An isolated unguardedtricuspid orifice with no other congenital abnormality hasbeen reported in a few patients so far.1,3,4,8,9 No case of anasymptomatic patient surviving till adulthood has beenreported. A giant right atrium with thrombus andspontaneous contrast, in the absence of any history of rightheart failure and the presence of atrial fibrillation, areunique features of the case we studied. Atrial fibrillationhas been reported in two of seven adult patients in aprevious series reported by us.9

Partial or complete absence of tricuspid valvar tissuediagnosed on fetal echocardiography was labeled asunguarded tricuspid valve orifice by Kanjuh et al.10 in 1964.This condition needs to be differentiated from Ebstein’smalformation, tricuspid dysplasia in association withpulmonary atresia and intact ventricular septum, and Uhl’sanomaly. Dysplasia of the leaflets along with displacementof the septal leaflet is an integral part of Ebstein’smalformation. However, the mural leaflet is always presentunlike in unguarded tricuspid orfice, in which case it maybe completely absent. Pulmonary atresia with an intactinterventricular septum may be associated with a variabledegree of tricuspid valve dysplasia but it needs to bedifferentiated from Ebstein’s malformation or an unguardedtricuspid orifice in which the right ventricle is dilated. Thetricuspid valve is structurally normal in Uhl’s anomaly.

Dysplasia of the tricuspid valve is probably the mostcommon cause of isolated tricuspid regurgitation and theunguarded tricuspid orifice is its most extreme form.Because of the poor right ventricular contractile function,

pulmonary circulation is maintained by the pumping actionof the right atrium or the outflow tract. As observed in ourpatient, the right atrium can assume enormousproportions.8–9 Right-to-left shunting can occur through thepatent foramen ovale. In some of these cases, functionalpulmonary atresia can result from a combination of aseverely abnormal tricuspid valve and markedly depressedright ventricular contractility. Unrecognized infectiveendocarditis is unlikely to be the cause of an unguardedtricuspid valve, as it is very difficult for such an extensivedestruction to remain subclinical. The absence ofvegetations goes against this diagnosis.

The natural history of this entity is variable. Severalpatients with a mild degree of right ventricular dysfunctionsurvive to adulthood and even reach old age. Such patientstolerate tricuspid regurgitation well and becomesymptomatic only when significant right ventriculardysfunction sets in, with or without atrial fibrillation. Thismakes surgical treatment a difficult option since surgicalresults are not encouraging.4

References

1. Mohan JC, Tatke M, Arora R. Rudimentary dysplastic valvar tissueguarding the tricuspid orifice with dilatation of the right ventricleand a patent outflow tract. Int J Cardiol 1989; 25: 136–139

2. Anderson RH, Silverman NH, Zuberbuhler JR. Congenitallyunguarded tricuspid orifice: its differentiation from Ebstein’smalformation in association with pulmonary atresia and intactventricular septum. Pediatr Cardiol 1990: 11: 86–90

3. Ozkutlu S, Gunal N, Caglar M, Alehan D, Gungor C. Unguardedtricuspid orifice: a rare malformation of tricuspid valve diagnosed byechocardiography. Report of two cases and review of the literature.Int J Cardiol 1996; 56: 125–129

4. Magotra RA, Agrawal NB, Mall SP, Parikh SJ. Severe dysplasia of thetricuspid valve (unguarded tricuspid annulus): clinical presentationand surgical treatment. J Thorac Cardiovasc Surg 1990; 99: 174–175

5. Munoz Castellanos L, Salinas CH, Kuri Nivon M, Garcia Arenal F.Absence of the tricuspid valve. A case report. Arch Inst Cardiol Mex1992; 62: 61–67

6. Gussenhoven EJ, Essed CE, Bos E. Unguarded tricuspid orifice withtwo-chambered right ventricle. Pediatr Cardiol 1986; 7: 175–177

7. Lagarde O, Garabedian V, Coignard A, Duret JC, Piwnica A, DroniouJ. Congenital tricuspid insufficiency due to valvular dysplasia. Reviewof the literature in light of a case in a 40-year-old adult. Arch MalCoeur Vaiss 1980; 73: 387–396

8. Mohan JC, Passey R, Arora R. Echocardiographic spectrum ofcongenitally unguarded tricuspid valve orifice and patent rightventricular outflow tract. Int J Cardiol 2000; 74: 153–157

9. Mohan JC, Passey R, Arora R. Unguarded tricuspid orifice and patentright ventricular outflow tract presenting with long standing severeright heart failure in an adult. Int J Cardiol 1998; 66: 85–87

10. Kanjuh VI, Stevenson JE, Amplatz K, Edwards JE. Congenitallyunguarded tricuspid valve orifice with coexistent pulmonary atresia.Circulation 1964; 30: 911–917

Fig. 1. Two-dimensional echocardiographic apical and parasternal 4-chamberviews (a and b) showing a dilated right atrium and right ventricle with a largeclot (T) and spontaneous echo contrast. The tricuspid orifice is unguarded withvirtually no anterior leaflet and a rudimentary septal leaflet which is attachedin the normal position (arrows).

IHJ-134-01.p65 10/17/01, 12:16 PM504

Indian Heart J 2001; 53: 505–507 Ray et al. Isolated Cardiac Aspergillosis 505Brief Report

Isolated Cardiac Aspergillosis

Ruma Ray, ZN Singh, HS Wasir, P ChopraDepartments of Pathology and Cardiology,

All India Institute of Medical Sciences, New Delhi

Invasive aspergillosis commonly occurs inimmunocompromised individuals, the lung being the

commonest site involved.1,2 Isolated involvement of extra-pulmonary organs such as the heart has rarely beendocumented in the literature.2,3 We report a rare case ofisolated cardiac aspergillosis involving the interatrialseptum, aortic valve and root of the aorta in an ‘apparentlyhealthy’ individual, where the diagnosis could be made onlyon autopsy.

Case Report

This 40-year-old male was admitted in an unconscious stateto the cardiology department with complaints of dyspneaand palpitation for the past three days. The patient wasdiagnosed at another hospital as a case of Wolff–Parkinson–White (WPW) syndrome two years earlier for which he wastreated intermittently. Previous records were not available.On clinical examination, the patient was found to becyanosed, dyspneic with a pulse rate of 240 beats/minute.His blood pressure was 90/60 mmHg and the jugularvenous pressure was raised. Cardiac auscultation revealedmarked tachycardia and chest auscultation showed signsof pulmonary edema. There was a nontender hepatomegaly3 cm below the costal margin. The electrocardiogramrevealed a heart rate of 250 per minute, a broad QRScomplex and left axis deviation. Echocardiography revealedan enlargement of both the ventricular chambers, moderateaortic regurgitation with pulmonary arterial hypertension.

The left ventricular systolic function was found to bereduced.

The patient was diagnosed clinically as a case of WPWsyndrome with atrial flutter. He was treated with directcurrent cardioversion, and injections of dopamine, heparinand furosemide. In spite of the medical treatment,terminally the patient developed bradycardia and died ofcardiac arrest within two hours of admission to the hospital.A thoracoabdominal autopsy was performed.

Autopsy findings: The heart weighed 350 g and wasunremarkable externally, with no evidence of a pericardialeffusion. On opening the right atrium, the smooth part ofthe septum, including the fossa ovalis, showed an elevatedgray-white plaque-like lesion falling short of the tricuspidannulus. The lesion encroached into the superior vena cavalorifice which was thus compromised. The interatrial septummeasured 2.5 cm and was hard and thickened (Fig. 1).Grossly, the lesion mimicked a tumour and involved theanatomical sites for the sinoatrial (SA) and atrioventricular(AV) nodes. The right ventricle and pulmonary artery wereunremarkable. The left atrium revealed a similar raisedplaque-like area over the interatrial septum correspondingto the lesion on the right side. The mitral valve wasunremarkable, but the left ventricle was mildly dilated. Theaortic valve cusps showed the presence of friable, red, angry-looking vegetations measuring 3–4 cm. The root of theaorta revealed the presence of friable hemorrhagic clotsloosely adherent to the surface. Multiple sections wereexamined from both the atria including the interatrialseptum, conduction system, ventricles, aortic valve and rootof the aorta. Sections from the interatrial septum and the

A 40-year-old man, a known case of Wolff–Parkinson–White syndrome, was admitted to the hospital in anunconscious state. In spite of medical treatment, the patient died within two hours of admission. At autopsy, thedeceased was found to have aspergillosis involving the interatrial septum, aortic valve and root of the aorta. Therest of the organs were unremarkable. The patient did not show any obvious signs of being immunocompromised.We report this case of isolated cardiac aspergillosis in an apparently healthy individual. (Indian Heart J 2001;53: 505–507)

Key Words: Cardiac aspergillosis, Wolff–Parkinson–White syndrome, Infection

Correspondence: Dr Ruma Ray, Assistant Professor of Pathology, All IndiaInstitute of Medical Sciences, New Delhi

IHJ-147-01.p65 10/17/01, 1:22 PM505

506 Ray et al. Isolated Cardiac Aspergillosis Indian Heart J 2001; 53: 505–507

plaque-like area in the right atrium showed widespreadinvolvement of the myocardium with areas of necrosis andforeign body giant cell reaction (Fig. 2). There werenumerous septate fungal profiles with parallel walls andacute angle branching which were better elicited byperiodic-acid Schiff (Fig. 3) and Grocott silver methenaminestains. Many giant cells showed the presence of fungalfragments within their cytoplasm. A section from the aorticvegetation revealed similar fungal profiles within a fibrinmeshwork. There was a contiguous spread to the root ofthe ascending aorta. Sections taken from the areas of theSA node, AV node and internodal region showed widespreadreplacement by the fungal granuloma. The diagnosis ofaspergillosis of the interatrial septum, aspergillusendocarditis of the aortic valve and aspergillus aortitis ofthe root of the aorta was made. Meticulous gross andmicroscopic examination of the other organs failed to revealany focus of fungal infection. The lungs had bilateralbronchopneumonia and the left kidney had a small area ofinfarction. Microscopic evaluation including special stainsfor fungi, done on lung and kidney sections, did not showthe presence of fungus.

Discussion

Isolated cardiac aspergillosis is rare. Fisher et al.2 in a seriesof 91 cases of invasive aspergillosis, have documented asingle case of a 78-year-old woman with sarcoma whodeveloped nodal tachycardia and myocardial infarctionseveral days prior to death. At autopsy, there was anaspergillus abscess in the interventricular septum. Thedisease has also been reported in a 15-year-old girl followingan allogenic bone marrow transplantation.3 Both thesecases had underlying risk factors which could be attributedto the occurrence of aspergillosis. In fact, invasiveaspergillosis is known to occur in patients with a debilitatingdisease, on steroid therapy, cytotoxic drugs and antibiotics,or undergoing radiation therapy.2,4 Our case is unique asthe patient did not have any such recognizable predisposingfactors, and the development of invasive aspergillosis in an

Fig. 1. Specimen of the heart viewed from the posterior aspect. The interatrialseptum (a) is thicker than the interventricular septum (b).

Fig. 2. Photomicrographi of the thickened interatrial septum showing agranulomatous reaction with giant cells. A few residual cardiac myocytes areseen at the periphery (H&E ×88)

Fig. 3. Photomicrograph showing the septate fungal profiles of aspergillus(PAS ×175).

IHJ-147-01.p65 10/17/01, 1:23 PM506

Indian Heart J 2001; 53: 505–507 Ray et al. Isolated Cardiac Aspergillosis 507

otherwise healthy individual is extremely rare.5 Anassessment of the neutrophil function in such a casemay bring out subtle abnormalities of phagocytosis andimpaired bactericidal activity which may contribute to arelatively immunodeficient state.5 In the present case,such investigations could not be carried out as thepatient succumbed to the disease following emergencyadmission.

Though our patient was diagnosed as a case of WPWsyndrome 2 years earlier, it is difficult to relate this clinicaldiagnosis with the autopsy finding of cardiac aspergillosis.Sections from the nodes and internodal areas showed thepresence of fungal granuloma without any recognizableremnant of conduction tissue. It is possible that the patientinitially had a cardiac arrhythmic disorder compatible withthe WPW syndrome and a superimposed fungal infectionwas acquired later, before his demise.

The portal of entry for the fungus in the present caseremains unknown. The possible sources include the lungsand the gastrointestinal tract. The lungs are the usual portalof entry for aspergillus, but in our patient there was nopulmonary involvement, though there was focalbronchopneumonia. However, multiple sections examinedfrom these foci did not show any fungal profile.

Our patient also had aspergillus endocarditis involvingthe aortic valve, and this, together with the root of the aorta,showed destruction with numerous fungal profilesembedded in a fibrin-rich background. There was a paucityof inflammatory infiltrate within these vegetations, and agranulomatous reaction was conspicuous by its absence. Itcan be speculated, therefore, that the infective vegetationsembolized in the intramyocardial artery and the organismswere lodged in the interatrial septum, evoking thesubsequent granulomatous response. Aspergillusendocarditis is thought to be an opportunistic infection andcan be seen in patients with a history of valvular cardiac

surgery.6 The endocardial vegetations of aspergillus muralendocarditis are usually contiguous with the underlyingmyocardial infection.7 The mural endocardium overlyingthe interatrial septal lesion in the present case wasunaffected, though the aortic valvular endocardium wasinvolved by a noncontiguous spread.

The ante mortem diagnosis of cardiac aspergillosis isdifficult to reach especially in a case like the present one,where the patient did not have any of the conventionalpredisposing factors. The diagnosis is often made atautopsy.2,6 It is possible that the infective fungal profiles aretoo large to traverse the systemic capillary bed and hencemay not even enter the venous system. In such a case,fungemia is likely to be missed even if the venous blood iscultured. However, one needs to consider the diagnosis in afebrile, immunocompromised patient with unexplainedcardiopulmonary decompensation, especially when thefungus is isolated or suspected to be the cause of infectionelsewhere.3 Very rarely, the disease can primarily affect theheart in apparently immunocompetent individuals, therebyposing a diagnostic difficulty, as seen in our case.

References

1. Rinaldi MG. Invasive aspergillosis. Rev Infect Dis 1983; 5: 1061–10772. Fisher BD, Armstrong D, Yu B, Gold JW. Invasive aspergillosis. Progress

in early diagnosis and treatment. Am J Med 1981; 71: 571–5773. Johnson RB, Wing EJ, Miller TR, Rosenfeld CS. Isolated cardiac

aspergillosis after bone marrow transplantation. Arch Intern Med1987; 147: 1942–1943

4. Williams AH. Aspergillus myocarditis. Am J Clin Pathol 1974; 61:247–256

5. Ascah KJ, Hyland RH, Hutcheon MA, Urbanski SJ, Pruzanski W, StLouis EL, et al. Invasive aspergillosis in a healthy patient. Can MedAssoc J 1984; 13: 332–335

6. Kammer RB, Utz JP. Aspergillus species endocarditis. The new face ofa not so rare disease. Am J Med 1974; 56: 506–521

7. Walsh TJ, Hutchins GM. Aspergillus mural endocarditis. Am J ClinPathol 1979; 71: 640–644

IHJ-147-01.p65 10/17/01, 1:23 PM507

508 Anjaneyulu et al. Myocardial Stunning Following Anesthesia Indian Heart J 2001; 53: 508–510

Acute Reversible Left Ventricular DysfunctionFollowing General Anesthesia

A Anjaneyulu, V Koti Reddy, P Krishnam Raju, A Rajagopalaraju,A Sreenivas Kumar, R Ravichandra

Departments of Cardiology and Cardiac Anesthesia, Care Hospital, Hyderabad

Cardiogenic shock as a complication of generalanesthesia is usually secondary to a myocardial

infarction (MI) in patients with underlying coronary arterydisease (CAD). When this complication occurs in patientswithout pre-existing CAD, it could be due to severe coronaryspasm. This may result in an infarct or stunning of themyocardium leading to left ventricular (LV) dysfunction. Atransmural infarct may heal with scarring andthereby residual LV dysfunction, whereas a stunnedmyocardium can regain its normal contractility. Bedsideechocardiography helps in recognizing this complicationearly, and aggressive management to support thedysfunctional myocardium by inotropic drugs, intra-aorticballoon pump (IABP) or LV assist devices could restore theLV function to normal. We report a case of stunnedmyocardium following endotracheal intubation.

Case Report

A 45-year-old woman, mildly hypertensive with no othercoronary risk factor and with a normal restingechocardiogram, was taken up for nasal polypectomy undergeneral anesthesia. She was premedicated with 75 mg ofpethidine intravenously (i.v.). General anesthesia wasinduced with i.v. pentothal sodium and vecuronium wasused as a muscle relaxant. Endotracheal intubation wasdifficult and required three attempts which took about 2–3

minutes. Oxygen saturation ranged between 85% and 90%during this period. Soon after intubation, she had an episodeof ventricular tachycardia which was converted to sinusrhythm with a bolus of 60 mg i.v. lignocaine. Within thenext 5 minutes, she developed sinus tachycardia of 160beats/min with an S

3 gallop and hypertension (blood

pressure 180/120 mmHg). She developed pulmonaryedema, with frothy fluid gushing out of the endotrachealtube. There was a fall in systemic arterial oxygen saturation(SaO

2 78%). Initially her oxygen saturation improved to

85% with 60 mg i.v. furosemide and a nitroglycerine drip ata rate of 5 µg/min. The systemic pressure fell to 70 mmHgsystolic within 10 minutes of starting the nitroglycerinedrip and she was put on inotropic support (dopamine 10µg/kg/min and dobutamine 7.5 µg/kg/min). Theelectrocardiogram (ECG) taken within 15 minutes of theepisode showed sinus tachycardia with no ST- or T-wavechanges. Arterial blood gas (ABG) levels deteriorated andwere: PaO

2 54 mmHg; PaCO

2 38 mmHg; and the pH was

7.210. At this stage, an echocardiogram revealed severe LVdysfunction (LV end-systolic dimension [LVESD]:3.4 cm, andLV end-diastolic diastolic dimension [LVEDD]: 4 cm) with anakinetic anterior wall and dyskinetic interventricularseptum. The left ventricular ejection fraction (LVEF) was25% and there was mild mitral regurgitation (Fig. 1). Shewas shifted to the ICCU and a central venous pressure (CVP)line as well as an arterial line for pressure monitoring wereestablished along with continued ventilatory and inotropicsupport. Her CVP was 10 mmHg. An ABG analysis after 3hours of support showed improvement (PaO

2 108 mmHg;

PaCO2 32 mmHg; pH 7.350). Metabolic acidosis was

Correspondence: Dr Anne Anjaneyulu, Department of Cardiology,Care Hospital, Road No. 1, Banjara Hills, Hyderabad 500 034.e-mail: [email protected]

Brief Report

Acute reversible left ventricular dysfunction due to myocardial stunning is a known phenomenon during acutemyocardial infarction, coronary angiography, coronary angioplasty or after coronary artery bypass surgery. Wereport a rare case of acute reversible dysfunction of the myocardium as a complication of general anesthesia ina patient with normal coronary arteries. This is a potentially fatal complication unless recognized early andtreated aggressively. (Indian Heart J 2001; 53: 508–510)

Key Words: Myocardial stunning, Complications, Anesthesia

IHJ-151-01.p65 10/17/01, 1:24 PM508

Indian Heart J 2001; 53: 508–510 Anjaneyulu et al. Myocardial Stunning Following Anesthesia 509

corrected by i.v. sodium bicarbonate. The systemicblood pressure was 96/50 mmHg and the urine output40 ml/hour. The patient was put on L-carnitine. After 12hours, she developed another bout of pulmonary edemawhile on maximal inotropic (dobutamine 15 µg/kg/minand dopamine 15 µg/kg/min) and ventilatory support. Thesystemic blood pressure was 60 mmHg (systolic), ABGmeasurements were: PaO

2 64 mmHg; PCO

2 40 mmHg and

the pH 7.345. A repeat echocardiogram showed severe LVdysfunction involving all the segments with an LVESD of3.5 cm; LVEDD of 3.9 cm and EF of 15%. At this stage,IABP support was initiated. She became hemodynamicallystable with a mean arterial pressure of 82 mmHg and anaugmented pressure of 106 mmHg. The ABG levelsimproved over the next 6 hours—PaO

2 124 mmHg, PCO

2

28 mmHg at an FIO2 of 80%. Cardiac enzyme levels were

mildly elevated (CK 640 IU, CK-MB 32 IU) on the first day.Serial CK and CK-MB readings over the next two days wereCK 210 IU and 146 IU, CK-MB 20 IU and 18 IU, respectively.Repeat echocardiograms over the next 4 days showedprogressive improvement in LV contractility with moresegments regaining normal contractility. Serial ECGs didnot show any evolving changes of transmural MI. Smalldaily doses of 3.125 mg carvedilol and 0.25 mg of digoxinwere added on the third day. She maintained stable arterialpressure and good urine output (50–60 ml/hour). By thefifth day, the LVEF improved to 45% (LVESD 3 cm, LVEDD4.1 cm). Chest X-ray showed no evidence of pulmonary

edema and she was weaned away from IABP support. Bythe seventh day, she was weaned away from the ventilatorand inotropic support. At this time, the echocardiagramshowed a normal-sized LV (ESD 2.8 cm; EDD 4 cm) with anEF of 50%. There was hypokinesia of the lateral wall anddistal part of the septum. The ECG at this stage showed STcoving with T-wave inversion in leads I and aVL. The patientwas fully ambulant with no cardiac symptoms by the tenthday. Her echocardiogram showed normal LV function withno regional wall motion abnormality (Fig. 2). A diagnosisof pheochromocytoma was ruled out by the estimation ofurinary vanillylmandelic acid (VMA) and metanephrine,and a negative abdominal ultrasound for adrenal and para-aortic masses. Coronary angiography done after an intervalof 3 weeks revealed normal epicardial coronary arteries.She was discharged on diltiazem and was doing well at 6months follow-up with a normal ECG and echocardiogram.

Discussion

This 45-year-old woman with no significant coronary riskfactor developed cardiogenic shock following endotrachealintubation for general anesthesia. This case wascharacterized by acute severe LV dysfunction in a previouslynormal heart with normal LV function. She developedtransient hypertension and extreme sinus tachycardiafollowed by hypotension and pulmonary edema. The ECGsdid not show any acute MI pattern (ST elevation) and there

Fig. 1. Two-dimensional echocardiogram (parasternal long axis view)showing end-systolic image of LV—dilated left ventricle with globalhypokinesia and severe LV dysfunction (EF 15%).

Fig. 2. End-systolic image of LV on the seventh day (parasternal long axisview) showing improved LV function with LV size returning to normal(EF 50%).

IHJ-151-01.p65 10/17/01, 1:24 PM509

510 Anjaneyulu et al. Myocardial Stunning Following Anesthesia Indian Heart J 2001; 53: 508–510

was no significant elevation of cardiac enzyme levels, thusruling out an acute transmural MI. The fact that themyocardial contractility improved slowly with inotropicsupport shows that the myocardium was metabolicallyactive.

Stunned myocardium is characterized by reversiblecontractile dysfunction, when the myocardial blood flow isfully or almost fully restored. In this condition, no metabolicdeterioration occurs during inotropic stimulation. Ifmyocardial stunning is severe, involving large parts of theLV and thus impairing global LV function, it can be reversedwith inotropic agents and procedures. Our patient requiredIABP support.

What is the trigger for acute LV dysfunction in thesecases? In our case, no surgical procedure had been started;only endotracheal intubation was done which was precededby the administration of pentothal sodium and vecuronium.However, intubation was not easy and required multipleattempts. A number of agents used during generalanesthesia may have negative effects on myocardialcontractility.1 Halothane and enflurane have mild negativeinotropic effects, especially in pre-existing LV dysfunction.However, our patient did not receive any of these agents.Intravenous thiopental and methohexital may also depressmyocardial contractility in patients with impaired LVfunction or elderly subjects. Though our patient receivedi.v. pentothal, it is unlikely that this would have contributedto the LV dysfunction because prior LV function was normal.In the absence of a specific myocardial depressant and asthe patient had persistent sinus tachycardia with an initialelevation of blood pressure, we postulate that an acutehyperadrenergic state developed following endotrachealintubation. This could have produced transient severecoronary spasm2 resulting in global LV dysfunction andcardiogenic shock.

An increase in plasma adrenaline and noradrenalineconcentrations has been observed following trachealintubation and surgery.3 Oral premedication withmetoprolol attenuates the hypertensive response to trachealintubation and reduces arrhythmias and operative bloodloss during hysterectomy.3 Post-tachycardia-relatedcardiomyopathy was considered a possibility, but the intervalbetween the tachycardia and LV dysfunction noted by echo-cardiography was very short (15 minutes). Studies usingambulatory ECG monitoring4 have also demonstratedmyocardial ischemia during tracheal intubation andextubation. There have been reports of intraoperativecoronary spasm during noncardiac surgery and one ofthese cases required several hours to becomehemodynamically stable.5 Since our patient had a normal

coronary angiogram, coronary spasm is a possibility whichmight have resulted in acute ischemic LV dysfunction.Structure-independent epicardial vasospasm can be animportant element in serious cardiac ischemic events,particularly the focal persistent vasospasms that occurwithout plaques or injury.6 There was a case report of a 23-year-old woman without previous CAD developing an acutenon-Q MI and stunned myocardium following topicallignocaine and phenylephrine used during nasalseptoplasty.7 However, our patient did not receive any topicalnasal medication. The possibility of severe LV systolicdysfunction secondary to an acute increase in afterload(severe hypertension) was also considered. In this event,the systolic dysfunction should resolve within a few hoursof correcting the hypertension.

Our patient required more than 96 hours of inotropicand IABP support which would indicate an ischemia-induced LV dysfunction. Since the majority of LV myocardialsegments were involved in the dysfunction (akinesia,dyskinesia) and all of them returned to normal contractility,it would fit in with the phenomenon of “stunnedmyocardium”. Although acute MI due to anesthesia-induced coronary spasm has been reported,8 only a fewcases of stunned myocardium have been reported so farfollowing noncardiac surgery.9

References

1. Mckinney MS, Fee JP, Clarke RS. Cardiovascular effects of isofluraneand halothane in young and elderly adult patients. Br J Anaesth1993; 71: 696–701

2. Yamasaki F, Sato T, Takata J, Chikamori T, Ozawa T, Sasaki M, et al.Sympathetic hyperactivity in patients with vasospastic angina. JpnCirc J 1996; 60: 10–16

3. Jacobsen CJ, Blom L. Ef fect of perioperative metoprolol oncardiovascular and catecholamine response and bleeding duringhysterectomy. Eur J Anaesthesiol 1992; 9: 209–215

4. Edwards ND, Alford AM, Dobson PM, Peacock JE, Reily CS.Myocardial ischemia during tracheal intubation and extubation.Br J Anaesth 1994; 73: 537–539

5. Yammanoue T, Horibe M, Izuri H, Tsuchiya T. Intraoperativecoronary spasm: retrospective review of 10 cases. Masui 1990; 39:376–382

6. Homsmeyer K, Miyagawa K, Kelly ST, Rosch J, Hall AS, AxthelmMK, et al. Reactivity-based coronary vasospasm independent ofatherosclerosis in Rhesus monkeys. J Am Coll Cardiol 1997; 29: 671–680

7. Aschi M, Wiedmann HP, James KB. Cardiac complications from useof cocaine and phenylephrine in nasal septoplasty. Arch OtolaryngolHead Neck Surg 1995; 121: 681–684

8. Zainea H, Duvernoy WF, Chauhan A, David S, Soto E, Small D. Acutemyocardial infarction in angiographically normal coronaryarteries following induction of general anesthesia. Arch Intern Med1994; 154: 2495–2498

9. Shirakami G, Shingu K, Tamai S, Ando A, Suga S, Nakao K, et al.Case of stunned myocardium after noncardiac surgery. AnesthesiaAnalg 1994; 79: 175–179

IHJ-151-01.p65 10/17/01, 1:24 PM510

Indian Heart J 2001; 53: 511–518 Singh et al. Dyslipidemia and the Renin–Angiotensin System 511

Interactions Between the Renin–Angiotensin System andDyslipidemia: Relevance in Atherogenesis and

Therapy of Coronary Heart Disease

BK Singh, JL MehtaDivision of Cardiovascular Medicine, Department of Internal Medicine, University of Arkansas for

Medical Sciences and the Central Arkansas Veterans Healthcare System, Little Rock, AR, USA

Hypercholesterolemia and hypertension are major riskfactors for coronary heart disease, and both are often

present in the same patient. It is thought that interactionsbetween dyslipidemia and activation of neurohumoralsystems such as the renin–angiotensin system (RAS) maynot only explain the frequent coexistence of hypertensionand dyslipidemia, but may also play an important role inthe pathogenesis of atherosclerosis. Experimental datasuggest that there is a correlation between the effects ofangiotensin II (Ang II) and lipoproteins on atherogenic risk.Data from recent experimental and clinical studies suggestthat the pathways by which Ang II and low-densitylipoprotein (LDL)-cholesterol lead to vascular disease mayfrequently overlap. Interventions directed at lowering totalcholesterol, LDL-cholesterol and triglyceride levels, andraising high-density lipoprotein (HDL)-cholesterol levelsresult in a reduction in cardiovascular events. Control ofblood pressure results in a similar decrease in cardiovascularevents. Angiotensin-converting enzyme (ACE) inhibitors orangiotensin type 1 (AT

1) receptor blockers modulate RAS

and are beneficial in reducing cardiovascular events inpatients with vascular disease. There is a suggestion thatthe combined use of cholesterol-lowering drugs along withagents that modulate RAS may have additive benefit.

In this review, we discuss the results of experimental andclinical studies on the interaction between RAS anddyslipidemia. These observations may have an impact onthe therapy of patients with coronary heart disease.

Renin–Angiotensin System and CholesterolBiosynthesis

Cholesterol accumulation in the macrophages and theirtransformation into foam cells are major events in the

Indian Heart J 2001; 53: 511–518 Current Perspective

development of atherosclerosis. Cellular cholesterolaccumulation can result from increased uptake of LDL oroxidatively modified forms of LDL,1 as well as by enhancedmacrophage cholesterol synthesis. Using macrophagesharvested from the peritoneum after injection of Ang II,Keidar et al.2 were able to demonstrate that Ang IIdramatically increased macrophage cellular cholesterolbiosynthesis with no significant effect on blood pressure oron plasma cholesterol levels. The ACE inhibitor fosinopriland the AT

1 receptor blocker losartan decreased cholesterol

biosynthesis in response to Ang II. Further, in cells that lackthe AT

1 receptor (RAW macrophages), Ang II did not

increase cellular cholesterol synthesis. These observationsconfirm the role of the AT

1 receptor in Ang II-mediated

cholesterol synthesis by macrophages. Other studies byNickenig et al.3 have shown accumulation of LDL-cholesterol in cultured vascular smooth muscle cells andthis effect is mediated via AT

1 receptor activation.

Angiotensin II-mediated increase in macrophagecholesterol influx has been demonstrated, and attributedto the oxidant stress contributing to and facilitating LDLoxidation by arterial wall components.4 Angiotensin II canalso bind to LDL and form modified lipoprotein, which istaken up at an enhanced rate by the macrophages scavengerreceptor, leading to cellular cholesterol accumulation.5 Liet al.6 studied the kinetics of oxidized LDL (ox-LDL) uptakein endothelial cells and observed that Ang II, in aconcentration-dependent fashion, enhanced the uptake ofI 125 labeled ox-LDL in these cells. The AT

1 receptor blocker

losartan, but not the AT2 receptor blocker PD 123319,

blocked the enhanced uptake of ox-LDL.Fluvastatin, a competitive inhibitor of 3-hydroxy-3-

methyl glutaryl coenzyme A (HMG-CoA) reductase, blocksthe stimulatory effect of Ang II on macrophage cholesterolbiosynthesis.2 Further, Ang II has been shown to upregulatemacrophage mRNA for HMG-CoA reductase.2 Thebiochemical site of action for Ang II along the cholesterolbiosynthesis pathway is probably HMG-CoA reductase, therate-limiting enzyme in cholesterol biosynthesis.7

Correspondence: Dr JL Mehta, Stebbins Chair in Cardiology, Professor ofInternal Medicine and Physiology, Director, Division of CardiovascularMedicine, University of Arkansas for Medical Sciences, 4301 W Markham St,Slot 532, Little Rock, AR 72205,USA. e-mail: [email protected]

IHJ-007-01.p65 10/17/01, 12:00 PM511

512 Singh et al. Dyslipidemia and the Renin–Angiotensin System Indian Heart J 2001; 53: 511–518

Thus it appears that stimulation of cholesterolbiosynthesis in macrophages, uptake of LDL in smoothmuscle cells and ox-LDL in macrophages and endothelialcells requires, or is at least facilitated by, AT

1 receptor

activation. In this process, alteration in the expression ofHMG-CoA reductase may play an important role.

Renin–Angiotensin System, Dyslipidemia andReactive Oxygen Species (ROS)

Griendling et al.8 first documented that Ang II increasesnicotinamide adenine dinucleotide (phosphate) hydroxide(NADH/NADPH) oxidase activity in macrophages via AT

1

receptor activation. Increased oxidative stress is nowregarded as an important feature of hypercholesterolemicatherosclerosis. In this context, antioxidants have beenshown to reduce the extent of progression of atherosclerosisin experimental animals and, in some studies, in humansas well.

Warnholtz et al.9 studied superoxide production in theaorta of rabbits fed on a diet containing 0.5% cholesterol.In their first study, they looked at the effects of endotheliumremoval on vascular superoxide production in control andWatanabe rabbits (hypercholesterolemia secondary to anLDL receptor defect). The rate of superoxide production wasincreased approximately two-fold in aortic segments fromWatanabe rabbits compared with rabbits fed a normal diet(controls). This increase in superoxide production wasabolished by removal of the endothelium from the arterialsegments. In these segments, NADH oxidase but notNADPH activity was significantly increased. These findingssuggested that hypercholesterolemia is associated withincreased superoxide production secondary to activation ofvascular NADH oxidase. These authors then measured theeffects of an AT

1 receptor blocker (Bay 10-6734) on

superoxide production and NADH oxidase activity in aortasfrom the controls and rabbits fed a high-cholesterol diet.The administration of an AT

1 receptor blocker reduced

superoxide production and inhibited NADH oxidase activityin cholesterol-fed animals. The investigators concluded thatin hypercholesterolemic animals, NADH oxidase representsa major vascular source of superoxide and that increasedvascular levels of Ang II may cause increased NADH oxidaseactivity. Hypercholesterolemia is associated with AT

1

receptor upregulation, endothelial dysfunction andincreased NADH-dependent vascular superoxideproduction. The improvement in endothelial dysfunction,inhibition of the oxidase and reduction of early plaqueformation by an AT

1 receptor antagonist suggests that Ang

II-mediated superoxide production plays a crucial role in

the early stage of atherosclerosis. Clinical and experimentalstudies have identified a marked attenuation inendothelium-dependent vasodilatation as one of the earlystages in atherosclerosis.10,11 In some cases, this is relatedto enhanced inactivation of endothelium-derived nitricoxide (NO) by superoxide,12 rather than a consequence ofdecreased NO production.13 It is known that AT

1 receptor

activation leads to membrane-associated NADH-dependentoxidase.8 Low-density lipoprotein enhances AT

1 receptor

expression in cultured smooth muscle cells14 andatherosclerotic lesions are associated with increased ACEexpression,15 which may serve as a source for localproduction of Ang II and, ultimately, increased stimulationof vascular superoxide production.

A number of studies have shown that AT1 receptor

blockade normalizes the activity of NADH oxidase, reducesplaque area and macrophage infiltration, andsimultaneously improves the endothelial surface in animalsfed a high-cholesterol diet.16 These findings suggest thatRAS plays a pathogenic role in both the initiation andacceleration of the atherosclerotic process and thatinhibition of RAS may benefit the treatment of this malady.

Long-term treatment with ACE inhibitors has beenshown to improve endothelial vasomotor function inpatients with coronary artery disease (Trial on ReversingEndothelial Dysfunction, TREND),17 possibly because ofdecreased superoxide-mediated inactivation of NO.Importantly, the benefits of ACE inhibitor therapy are morepronounced in patients with hypercholesterolemia.

Hypercholesterolemia and RAS activation

Experimental studies have shown that hyperlipidemiaenhances RAS activity. All components of increased RASactivation have been identified in hyperlipidemicatherosclerotic lesions. These include, in particular,increased expression of ACE and AT

1 receptors.18,19 A

number of recent studies of human atherosclerotic tissueshave confirmed the upregulation of ACE and AT

1 receptors,

particularly in the regions that are prone to plaquerupture.2 0 Importantly, these areas show extensiveinflammatory cell deposits, macrophage accumulation andapoptosis.

In vitro studies have shown that incubation of vascularsmooth muscle cells with LDL increases expression of AT

1

receptors.21 Li et al.22 examined the expression of Ang IIreceptors in human coronary artery endothelial cells, andobserved that ox-LDL increases the mRNA and protein forAT

1, but not AT

2 receptors, implying that ox-LDL increases

AT1

expression at the transcriptional level. In this process,

IHJ-007-01.p65 10/17/01, 12:01 PM512


activation of the redox-sensitive transcription factor NF-κBplays a critical role. To define the relationship of RAS andlipids in humans, Nickenig et al.3 administered Ang II innormocholesterolemic and hypercholesterolemic men, andfound that blood pressure was increased in thehypercholesterolemic group and this response could beblunted by LDL-cholesterol lowering agents. Further, theseinvestigators found that there was a linear relationshipbetween AT

1 receptor density on platelets and LDL-

cholesterol concentration in plasma. Treatment with statinsdecreased the AT

1 receptor expression in this study. Statin-

mediated downregulation of AT1 receptor expression has

also been shown in vascular smooth muscle cells.23 A recentstudy has shown that statins directly decrease AT

1 receptor

expression in endothelial cells.24

The expression of genes for chymases—enzymes bywhich Ang II can be formed independent of ACEactivation—has been shown to increase in atheroscleroticlesions of the aorta of monkeys fed a high-cholesterol diet.25

The functional significance of chymase in the developmentof atherosclerosis, however, remains uncertain.

Role of Ang II in HypercholesterolemicAtherosclerosis

Activation of RAS with formation of Ang II and activationof Ang II receptors, particularly AT

1 receptors, has been

implicated in the pathobiology of atherosclerosis, plaquerupture, myocardial ischemic dysfunction and congestiveheart failure.26 Several studies show that ACE inhibitorsdecrease progression of atherosclerosis in a variety ofanimal species.27,28 Since a number of different ACEinhibitors exert similar anti-atherosclerotic effects, one canassume that this represents a class effect. In concurrencewith slowing of the progression of atherosclerosis, ACEinhibitors decrease markers of inflammation and LDLoxidation in the atherosclerotic regions.

A variety of AT1 receptor blockers have also been shown

to reduce the progression of atherosclerosis in differentanimal models.28,29 The effects are particularly evident athigh doses of AT

1 receptor blockers, which suggests

that either high doses block AT1 receptors more completely

than lower doses, or that high doses reduce atherosclerosisby some nonspecific effect. We recently reported the anti-atherosclerotic effect of losartan (25 mg/kg) inrabbits fed a high-cholesterol diet and showed that losartantherapy suppressed the expression of adhesion moleculesas well as NF-κB by activating its regulatory proteinIκBα.29

To determine the specificity of the role of RAS inhibitors

(v. the blood pressure-lowering effect), Leif et al.28 conducteda study with low doses of fosinopril (5 mg/kg/day) orlosartan (5 mg/kg/day) that did not lower blood pressure.Control animals were given either a placebo or a dose ofhydralazine which lowered blood pressure. Low-densitylipoprotein oxidation, as measured by levels ofthiobarbituric acid-reactive substances (TBARS) or byformation of conjugated dienes, was suppressed by low-dosefosinopril, suppressed only modestly by losartan andunaffected by the placebo or hydralazine. Atherosclerosiswas inhibited by fosinopril and losartan, suggesting that theanti-atherosclerotic effects of RAS inhibitors may be due,at least in part, to direct inhibition of LDL oxidation andother effects of Ang II on the vessel wall.

Bavry et al.30 from our laboratory showed that the ACEinhibitor quinapril decreased intra-arterial thrombusformation, whereas the AT

1 receptor blocker losartan had

a minimal effect. The inhibitory effect of ACE inhibitors onthe generation of plasminogen activator inhibitor-1 maybe relevant in this differential effect of ACE inhibitors andAT

1 receptor blockers. This is especially relevant since

thrombosis is intimately involved in atherogenesis.31

The role of Ang II in promoting atherosclerotic lesionsand aneurysms in apolipoprotein (apo) E-deficient mice hasbeen recently examined by Daugherty et al.32 Theseinvestigators showed that a 1-month infusion of Ang IIenhanced the severity of aortic atherosclerotic lesionscompared to a placebo. Interestingly, there was extensiveformation of abdominal aortic aneurysms in apo E-deficientmice infused with Ang II. Further, the presence ofhyperlipidemia was necessary for the development ofaneurysms. These observations suggest that increasedplasma concentrations of Ang II have profound effects onvascular pathology when combined with hyperlipidemia,and inhibitors of RAS may have a therapeutic benefit,especially in the hyperlipidemic state.

Endothelial function, RAS and Dyslipidemia

Endothelial dysfunction in hypercholesterolemic animalshas been shown to be improved by ACE inhibitors.33

Bradykinin antagonists can diminish some of this benefit,suggesting that inhibition of bradykinin breakdown ratherthan inhibition of Ang II formation may be important inthis effect.34 Recently, Mancini et al.17 showed that treatmentof patients with coronary artery disease with quinaprilimproved coronary vasomotion. Quinapril had greaterefficacy in improving endothelial function in patients withLDL-cholesterol >130 mg/dl than in patients with LDL-cholesterol <130 mg/dl.13–18

IHJ-007-01.p65 10/17/01, 12:01 PM513


Role of Dyslipidemia and RAS in AtherogenesisRole of Dyslipidemia and RAS in Atherogenesis

ECEC O2-NO

Ang IIAng II

Ox-LDLOx-LDL

LDLLDL

SRsSRs

Growth factorsGrowth factors

FibroblastsFibroblasts

Smooth muscleSmooth musclecellscells

Dyslipidemia, smoking, hypertension, diabetesDyslipidemia, smoking, hypertension, diabetes

AT1R

↑ Ang IIOxidation of LDL↑ Cytokines↑ Platelet Aggregation↑ Monocyte adhesion

LOX-1

Adhesionmolecules

Fig. 1. Interaction of dyslipidemia and the renin–angiotensin system (RAS)in the development of atherosclerosis. The amplification of RAS by dyslipidemia,in particular oxidized-LDL, may enhance the growth-promoting effect of AngII (and possibly other growth factors) and upregulate the expression of AT1

receptor. The pro-oxidant effect of Ang II may stimulate oxidation of LDL,degrade nitric oxide and may be related to its pro-inflammatory properties.Ang II promotes expression of genes for LOX-1 and other scavenger receptors.Ang II: angiotensin II; LDL: low-density lipoprotein; Ox-LDL: oxidized LDL;SRs: scavenger receptors; AT1R: angiotensin type 1 receptors; LOX-1: receptorsfor oxidized LDL; NO: nitric oxide; O2

-: superoxide anion; EC: endothelial cells.

Acetylcholine stimulates release of the potentvasodilator species NO, which is broken down by ROS. Oneof the mechanisms responsible for improvement inacetylcholine-mediated vasodilatation may be inhibition ofAng II-sensitive, NADH-dependent, superoxide-producingenzymes, resulting in a reduction of NO inactivation.Warnholtz et al.9 showed that AT

1 receptor blockade

inhibited NADH oxidase activity and simultaneouslyimproved endothelial dysfunction in animals fed a high-cholesterol diet. These findings cannot be attributed tolowering of cholesterol levels because treatment with theAT

1 receptor blocker has no ef fect on total or LDL-

cholesterol level.

Interaction between ox-LDL and RAS:Role of Receptors for ox-LDL (LOX-1)

We have recently identified high-af finity lectin-likereceptors for ox-LDL (LOX-1) in cultured human coronaryartery endothelial cells by reverse transcriptase-polymerasechain reaction (RT-PCR), Western blot, and radioligandbinding.35,36 Native LDL does not bind to this receptor.Vascular endothelial cells in culture37 and in vivo38

internalize and degrade ox-LDL through this putativereceptor-mediated pathway which does not seem to involvethe classic macrophage scavenger receptor. Recent studiesshow that the cytokine TNF-α39 and fluid shear stress40

markedly upregulate LOX-1 gene expression. Activation ofLOX-1 is involved in apoptosis (programmed cell death) inresponse to ox-LDL,41,42 mitogen-activated protein kinase(MAPK)-1 activation, and expression of adhesion moleculesand attachment of monocytes to activated endothelialcells.43 A critical role is played by NF-κB in the effect ox-LDL has on endothelial cells. 23 The pro-apoptotic effect ofAng II in human coronary artery endothelial cells and therole of AT

1 receptor and protein kinase C (PKC) activation

have also been shown by our group.44

Li et al.36 from our laboratory have demonstrated thatAng II upregulates LOX-1 expression as well as the uptakeof ox-LDL in human coronary artery endothelial cells viaactivation of the AT

1 receptor. The effects of Ang II were

blocked by the AT1 receptor blockers losartan and

candesartan, but not by the AT2 receptor blocker PD

213319. Angiotensin II and ox-LDL exerted a cumulativeinjurious effect on cells, measured as lactic dehydrogenase(LDH) release and cell viability. Again, AT

1 receptor blockers

reduced the cumulative injurious effect of Ang II and ox-LDL. Importantly, the chain-breaking antioxidant α-tocopherol also attenuated the injurious effect of ox-LDLand Ang II, emphasizing the importance of redox-sensitivepathways in the cross-talk.45

The cross-talk between ox-LDL and Ang II is furtherevident from the work of Chen et al.29 from our laboratory,who showed intense immunostaining for and upregulationof the gene for LOX-1 in the atherosclerotic tissue of rabbitsfed a high-cholesterol diet. Losartan therapy not onlyreduced atherosclerosis, but also blocked the upregulationof LOX-1. Recent unpublished studies from our laboratoryshow marked upregulation of LOX-1 in concert withapoptosis in human atherosclerotic plaques, particularly inthe regions that are prone to rupture. Figure 1 shows theinteraction of dyslipidemia and RAS in atherogenesis.

Dyslipidemia and RAS in Hypertension

The association of hypertension with hyperlipidemia hasbeen noted in several population studies. The prevalence ofhypertension is greater in populations with high cholesterollevels.46 Dyslipidemia may be another metabolic factor thatinfluences blood pressure. However, these studies used older,less rigorous definitions than are currently recommended.Recently, Lloyd-Jones et al.47 evaluated 4962 subjects fromthe Framingham Heart Offspring Study and cross-clarifiedthem according to the sixth Report of the Joint NationalCommittee on the Prevention, Detection, Evaluation, andTreatment of High Blood Pressure (JNC VI). Data werecollected from subjects examined between 1990 and 1995.

IHJ-007-01.p65 10/17/01, 12:01 PM514


The prevalence of dyslipidemia (defined as totalcholesterol >240 mg/dl, HDL-cholesterol <35 mg/dl, orcurrently receiving lipid-lowering therapy) increaseswith increasing blood pressure in men and women. Onan average, over 40% of men and 33% of women withblood pressure >145/>90 mmHg were also dyslipidemic.These data demonstrate that hypertension andhypercholesterolemia are frequently associated, even whencurrent rigorous definitions are used. These observationsalso suggest that individuals with hypertension may bemore likely to become dyslipidemic over time.

Sung et al.48 examined the blood pressure response to astandard mental arithmetic test in 37 healthy normotensivesubjects with hypercholesterolemia (mean total cholesterol263 mg/dl) and 33 normotensive subjects with normalcholesterol levels. None of the hypercholesterolemic groupwas receiving lipid-lowering therapy prior to induction inthe study. In the first part of the study, blood pressureresponse during the arithmetic test was determined andfound to be significantly higher in the hypercholesterolemicgroup compared with the normocholesterolemic group (18v. 10 mmHg, respectively, p<0.005). In the second part ofthe test, the hypercholesterolemic group was divided into 2subgroups which received either 6 weeks of lovastatin or 6weeks of placebo in a double-blind, cross-over design. Therewere 26 evaluable patients in this part of the study. Statintreatment resulted in significant reduction from baselinein total and LDL-cholesterol levels and was associated withlower mean systolic blood pressure prior to (119±11 v.122±9 mmHg, p=0.07) and during the arithmetic test(133±12 v. 141±10 mmHg, p<0.05). Diastolic bloodpressure changes were not significantly correlated withlowering of lipid levels. These observations demonstrate thatindividuals with hypercholesterolemia have an exaggeratedsystolic blood pressure response to mental stress and thelowering of lipid levels improves the systolic response tostress. Although the effects of elevated cholesterol levels onatherosclerosis are well documented, the modest change inthe degree of stenosis demonstrated by angiographic studiesis not sufficient to explain the benefit of reduction incholesterol levels. It may well be that lowering cholesterollevels alters the activity of some neurohumoral mediatorssuch as Ang II and improves vascular tone.

Nazzaro et al.49 made an interesting observation of thecombined and distinct vascular effects of ACE inhibitorsand statins on lowering blood pressure. They examinedthe effects of lowering of lipid levels on blood pressure in astudy of 30 subjects with coexisting hypertension andhypercholesterolemia. Subjects received a placebo for 4weeks and were then divided into 2 groups. Each group of

15 patients received monotherapy with either simvastatin10 mg or enalapril 20 mg for 14 weeks. After themonotherapy phase, each group received both drugs for anadditional 14 weeks. Blood pressure was measured duringstressful stimuli such as the Strop color test and the coldpressor forehead test. As expected, enalapril lowered bloodpressure. Interestingly, however, simvastatin also loweredblood pressure (although to a lesser extent) and thecombination of both medications achieved greater bloodpressure reduction than either alone. These observationssuggest a close interplay of RAS and lipid metabolism.

Nazzaro et al.49 also measured post-ischemic forearmblood flow and minimal vascular resistance to evaluate theeffects of mental stress on vasodilatative capacity andvascular structure, respectively. These parametersdemonstrated the same trends as blood pressure. Bothmonotherapies improved these parameters, but thecombination therapy was associated with a greaterimprovement than either monotherapy. These findingssuggest a cross-talk between dyslipidemia and RAS relativeto vascular dynamics. Table 1 shows the common effectsof dyslipidemia and RAS in atherosclerosis.

Table 1. Common effects of dyslipidemia and RAS inatherosclerosis

1. Both stimulate formation and release of ROS2. Both cause apoptosis3. Both cause activation of redox-sensitive transcription factor

NF-κB4. Both are pro-inflammatory (cause expression of adhesion

molecules and cytokines, upregulate the gene for monocytechemoattractant protein-1 and induce monocyte adhesion)

5. Both degrade or decrease endothelial nitric oxide synthase(eNOS) expression and hence decrease endothelium-dependent vasodilation.

Clinical Benefit of Modulation of RAS andDyslipidemia in Coronary Artery Disease (CAD)

Although numerous epidemiological studies have shownthat elevated levels of LDL are associated with the onset ofhypertension and atherosclerosis,50 the underlyingmechanisms remain unclear. Angiotensin-convertingenzyme inhibition has been shown to promote regressionand even prevent atherosclerosis, suggesting a link betweenatherosclerosis and RAS.51

The clinical benefits from simultaneous modulation ofRAS and dyslipidemia are summarized in Table 2. Indirectevidence for an interaction between dyslipidemia and RAScomes from some clinical studies such as Evaluation ofLosartan In the Elderly (ELITE)52 and Lipoprotein andCoronary Atherosclerosis Study (LCAS).53 There are studies

IHJ-007-01.p65 10/17/01, 12:01 PM515


which suggest that RAS may affect responses to lipid-lowering agents. Observations from unpublished data fromstudies such as the ELITE trial support the hypothesis thatcombination treatment with ACE inhibitors and statins mayhave incremental benefit in reducing mortality.

The LCAS was conducted in 429 patients with CAD andat least 1 lesion with 30%–75% diameter stenosis. Subjectswere randomized to statin (fluvastatin) or placebo for 2.5years and the primary end-point was a change in theminimum lumen diameter as assessed by quantitativecoronary angiography. Marian et al.53 studied the responseto statin therapy according to ACE insertion/deletion (I/D)genotype in the LCAS population. The subjects with DD, ID,or II genotypes achieved reductions of 31%, 25%, and 21%,respectively. There was a significant genotype-by-treatmentinteraction (p=0.005). A similar result was obtained forreduction in total cholesterol. Subjects with the DDgenotype also had a higher rate of regression and a lowerrate of progression than subjects with the other 2 genotypes.

The effect of ACE inhibition on CAD progression was the

subject of the Quinapril Ischemic Events Trial (QUIET). Thisstudy showed that quinapril had only a slight effect on theprogression of CAD.54,55 However, in patients with LDL-cholesterol levels >130 mg/dl, there was significantly lessprogression in the quinapril group. Thus, the rapidprogression of disease seen in patients given a placebo withhigher LDL-cholesterol levels did not occur in patientstreated with quinapril. As in the TREND study,56 ACEinhibitors appeared to have greater efficacy in patients withhigher LDL-cholesterol levels.

Angiotensin-converting enzyme inhibitors are beneficialin a variety of clinical situations, such as hypertension,diabetes and congestive heart failure. Recent long-termstudies with ACE inhibitors in patients with decreased leftventricular function57–60 have shown a decrease in cardiacischemic events and/or a need for revascularization. Onepathogenic factor common to both heart failure andischemic heart disease is endothelial damage or activation,which may explain the reduction in ischemic events seenin these trials. More so, other clinical studies such as theHeart Outcomes Prevention Evaluation (HOPE) trial61 havefurther confirmed the benefit of reducing vascular eventsand death even in patients with normal ventricular functionand normal blood pressure with pre-existing vascular orcoronary disease. The study to evaluate carotid ultrasoundchanges in patients treated with ramipril and vitamin E(SECURE) trial, a substudy of the HOPE trial, demonstratedthe beneficial effect of ramipril in preventing progressionof carotid atherosclerosis.62 Similarly, irbesartan, an AT

1

receptor blocker, has been shown to regulate markers ofinflammation in patients with premature atherosclerosis;this may retard the inf lammatory process seen inatherosclerosis.63 These findings suggest the potential roleof RAS in the development and progression ofatherosclerosis.

No large randomized study has yet examined thehypothesis of whether treatment by modulation of RASwith drugs (ACE inhibitors or AT

1 blockers) combined with

lipid-lowering drugs exerts additive or incremental benefits.The ongoing randomized trial may shed light in thisdirection.64

Summary

Hypertension and hypercholesterolemia, two major riskfactors for atherosclerotic disease, frequently coexist inpatients with hypertension and CAD. Data from clinicalstudies suggest the existence of lipoprotein–neurohormonalinteractions that may adversely affect vascular structureand reactivity. Data from preclinical studies suggest that

Clinical trials

TREND56

QUIET54,55

LCAS53

ELITE II52

Study objective

Effects of ACE inhibitor(quinapril) inacetylcholine-inducedendothelial response ofcoronary artery accordingto LDL-cholesterol level

Effect of ACE inhibitor(quinapril) on ischemicevents and angiographicprogression of coronarydisease assessed in patientswho underwentpercutaneous intervention

Effect of statin (fluvastatin)on minimal lumendiameter assessed byquantitative coronaryangiography according toACE genotype. LDLreduction according toACE genotype wasanalyzedEffect of captopril andlosartan on cardiac eventsin elderly patients withCHF

Results

Quinapril 40 mg/day hadgreater efficacy inimproving endothelialfunction in the group withLDL-cholesterol >130mg/dl than in the groupwith LDL-cholesterol <130mg/dlOverall effect on theprimary ischemic andangiographic end-pointswas neutral. However,patients with LDL-cholesterol >130 mg/dlhad significantly lessprogression of diseasecompared to patients withLDL-cholesterol <130mg/dlThere was a significantdifference in the reductionof LDL-cholesterolaccording to the ACEgenotype. Subjects with DD,ID, or II genotype achievedreductions of 31%, 25%,and 21%, respectively

Both captopril and losartandecreased crude mortalitysimilarly. However, patientswho were on statins hadadditional ~50% reductionin mortality

Table 2. Summary of the results of clinical trials suggest-ing interaction between RAS and dyslipidemia

IHJ-007-01.p65 10/17/01, 12:01 PM516


RAS may be upregulated by abnormal lipids, most likely viaproduction of ox-LDL. On the other hand, activation of RASleads to release of ROS and transcriptional upregulation ofLDL and ox-LDL uptake in macrophages, smooth musclecells and endothelial cells. These findings extend ourunderstanding of the interplay among risk factors tosynergistically increase cardiovascular risk, and of the anti-atherosclerotic effects of local ACE inhibition to reducecardiovascular risk. Trials aimed at modifying RAS alongwith drugs lowering total- and LDL-cholesterol levels andinhibitors of oxidative modification of LDL-cholesterol willaddress the clinical relevance of this biological interaction.

References

1. Aviram M. Modified forms of low-density lipoprotein andatherosclerosis. Atherosclerosis 1993; 98: 1–9

2. Keidar A, Attias J, Heinrich R, Coleman R, Aviram M. Angiotensin IIatherogenicity in apolipoprotein E deficient mice is associated withincreased cellular biosynthesis. Atherosclerosis 1999; 146: 249–257

3. Nazzaro P, Manzari M, Merlo M, Triggioni R, Scararo A, Ciancio L, etal. Distinct and combined vascular effects of ACE blockade and HMG-CoA reductase inhibition in hypertensive subjects. Hypertension 1999;33: 719–725

4. Keidar S. Angiotensin, LDL peroxidation and atherosclerosis. Life Sci1998; 63: 1–11

5 Keidar S, Kaplan M, Aviram M. Angiotensin II-modified LDL is takenup by macrophages via the scavenger receptor, leading to cellularcholesterol accumulation. Arterioscler Thromb Vasc Biol 1996; 16:97–105

6. Li DY, Zhang YC, Philips MI, Sawamura T, Mehta JL.Upregulation ofendothelial receptor for oxidized low-density lipoprotein (LOX-1) incultured human coronary artery endothelial cells by angiotensin IItype 1 receptor activation. Circ Res 1999; 84: 1043–1049

7. Goldstein JL, Brown MS. Regulation of the mevalonate pathway.Nature 1990; 343: 425–430

8. Griendling KK, Minieri CA, Ollerenshaw JD, Alezander RW.Angiotensin II stimulates NADH and NADPH oxidase activity incultured vascular smooth muscle cells. Circ Res 1994; 74:1141–1148

9. Warnholtz A, Nickenig G, Schulz E, Macharzina R, Brasen JH,Skatchkov M, et al. Increased NADH-oxidase-mediated superoxideproduction in the early stages of atherosclerosis: evidence forinvolvement of the renin–angiotensin system. Circulation 1999; 99:2027–2033

10. Freinman PC, Mitchell GG, Heisted DD, Armstrong ML, Harrison DG.Atherosclerosis impairs endothelium-dependent vascular relaxation toacetylcholine and thrombin in primates. Circ Res 1986; 58: 783–789

11. Zeiher AM, Drexler H, Wollschlager H, Just H. Endothelial dysfunctionof the coronary microvasculature is associated with coronary bloodflow regulation in patients with early atherosclerosis. Circulation1991; 84: 1984–1992

12. Schwemmer M, Sommer O, Koeckerbauer R, Bassenge E.Cardiovascular dysfunction in hypercholesterolemia associated withenhanced formation of AT1-receptor and of eicosanoids. J CardiovascPharmacol Ther 2000; 5: 59–68

13. Minor RL, Myers PR, Guerra R, Bates JN, Harrison DG. Diet-inducedatherosclerosis increases the release of nitrogen oxides from rabbitaorta. J Clin Invest 1990; 86: 2109–2116

14. Nickenig G, Sachinidis A, Michaelsen F, Bohm M, Seewald S, Vetter

H. Upregulation of vascular angiotensin II receptor gene expressionby low-density lipoprotein in vascular smooth muscle cells. Circulation1997; 95: 473–478

15. Diet F, Prat RE, Berry GJ, Momose N, Gibbons GH, Dzau VJ. Increasedaccumulation of tissue ACE in human atherosclerotic coronaryartery disease. Circulation 1996; 94: 2756–2767

16. Strawn WB, Chappell MC, Dean RH, Kivlighn S, Ferrario CM.Inhibition of early atherogenesis by losartan in monkeys with diet-induced hypercholesterolemia. Circulation 2000; 101: 1586–1593

17. Mancini GB, Henry GC, Macaya C, O’Neill BJ, Pucillo AL, Carer RG, etal. Angiotensin-converting enzyme inhibition with quinaprilimproves endothelial vasomotor dysfunction in patients withcoronary artery disease: the TREND (Trial on Reversing ENdothelialDysfunction). Circulation 1996; 94: 258–265

18. Mitan H, Bandoh T, Kimura, M, Totsuka T, Hayashi S. Increasedactivity of vascular ACE related to atherosclerotic lesions inhyperlipidemic rabbits. Am J Physiol 1996; 271: H1065–H1071

19. Strehlow K, Wassmann S, Bohm M, Nickenig G. Angiotensin AT1

receptor over-expression in hypercholesterolaemia. Ann Med 2000;32: 386–389

20. Schieffer B, Schieffer E, Hilfiker-Kleiner D, Hilfiker A, Kovanen PT,Kaartinen M, et al. Expression of angiotensin II and interleukin 6 inhuman coronary atherosclerotic plaques: potential implicationsfor inflammation and plaque instability. Circulation 2000; 101:1372–1378

21. Nickenig G, Sachinidis A, Seewald S, Bohm M, Vetter H. Influence ofoxidized low-density lipoprotein on vascular angiotensin II receptorexpression. J Hypertens 1997; 15 (Suppl): S27–S30

22. Li D, Saldeen T, Romeo F, Mehta JL. Oxidized LDL upregulatesangiotensin II type 1 receptor expression in cultured human coronaryartery endothelial cells: the potential role of transcription factorNF-kappa B. Circulation 2000; 102: 1970–1976

23. Wassmann S, Nickenig G, Bohm M. HMG-CoA reductase inhibitoratorvastatin downregulates AT1 receptor gene expression and cellproliferation in vascular smooth muscle cells. Kidney Blood Press Res1999; 21: 392–393

24. Nickenig G, Baumer AT, Temur Y, Kebben D, Jockenhovel F, Bohm M.Statin-sensitive dysregulated AT1 receptor function and density inhypercholesterolemic men. Circulation 1999; 100: 2131–2134

25. Takai S, Shiota N, Kobayashi S, Matsumura E, Miyasaki M. Inductionof chymase that forms angiontensin II in the monkey atheroscleroticaorta. FEBS Lett 1997; 412: 86–90

26. Griendling KK, Murphy TJ, Alexander RW. Molecular biology of therenin–angiotensin system. Circulation 1993; 87: 1816–1828

27. Hayek T, Attias J, Coleman R, Brodsky S, Smith J, Breslow JL, et al.The angiotensin-converting enzyme inhibitor, fosinopril, and theangiotensin II receptor antagonist, losartan, inhibit LDL oxidationand attenuate atherosclerosis independent of lowering bloodpressure in apolipoprotein E deficient mice. Cardiovasc Res 1999;44: 579–587

28. Jacobsson LS, Persson K, Aberg G, Anderson RG, Karlberg BE, OlssonAG. Antiatherosclerotic effects of the angiotensin-converting enzymeinhibitors captopril and fosinopril in hypercholesterolemic minipigs.J Cardiovasc Pharmacol 1994; 24: 670–677

29. Chen H, Li D, Sawamura T, Inoue K, Mehta JL. Upregulation of LOX-1 expression in aorta of hypercholesterolemic rabbits: modulationby losartan. Biochem Biophys Res Commun 2000; 276: 1100–1104

30. Bavry AA, Li D, Zander DS, Phillips MI, Mehta JL. Inhibition of arterialthrombogenesis by quinapril, but not losartan. J Cardiovasc PharmacolTher 2000; 5: 121–127

31. Conti CR, Mehta JL. Acute myocardial ischemia: role ofatherosclerosis, thrombosis, platelet activation, coronary vasospasm,and altered arachidonic acid metabolism. Circulation 1987; 75:V84–V95

32. Daugherty A, Manning MW, Cassis LA. Angiotensin II promotes

IHJ-007-01.p65 10/17/01, 12:01 PM517


atherosclerotic lesions and aneurysms in apolipoprotein E-deficientmice. J Clin Invest 2000; 105: 1605–1612

33. Becker RH, Wiemer G, Linz W. Preservation of endothelial functionby ramipril in rabbits on a long-term atherogenic diet. J CardiovascPharmacol 1991; 18: S110–S115

34. Farhy RD, Carretero OA, Ho KL, Scicli AG. Role of kinins and nitricoxide in the effects of angiotensin converting enzyme inhibitors onneointima formation. Circ Res 1993; 72: 1202–1210

35. Mehta JL, Li DY. Identification and autoregulation of receptor for Ox-LDL in cultured human coronary artery endothelial cells. BiochemBiophys Res Commun 1998; 248: 511–514

36. Li DY, Zhang YC, Philips MI, Sawamura T, Mehta JL. Upregulation ofendothelial receptor for oxidized low-density lipoprotein (LOX-1) incultured human coronary artery endothelial cells by angiotensin IItype-1 receptor activation. Circ Res 1999; 84: 1043–1049

37. Bickel PE, Freeman MW. Rabbit aortic smooth muscle cells expressinducible macrophage scavenger receptor messenger RNA that isabsent from endothelial cells. J Clin Invest 1992; 90: 1450–1457

38. Kume N, Arai H, Kawai C, Kita T. Receptors for modified low densitylipoproteins on human endothelial cells: different recognition foracetylated low-density lipoprotein and oxidized low-densitylipoprotein. Biochim Biophys Acta 1991; 1091: 63–67

39. Kume N, Murase T, Moriwaki H, Aoyama T, Sawamura T, Masaki TA, et al. Inducible expression of receptor-1 lectin-like oxidized lowdensity lipoprotein, in vascular endothelial cells. Circ Res 1998; 83:322–327

40. Murase T, Kume N, Korenaga R, Ando J, Sawamura T, Masaki T, et al.Fluid shear stress transcriptionally induces lectin-like oxidized LDLreceptor-1 in vascular endothelial cells. Circ Res 1998; 83: 328–333

41. Li DY, Yang BC, Mehta JL. Ox-LDL induces apoptosis in culturedhuman coronary artery endothelial cells: role of PKC, PTK, bcl-2,and Fas. Am J Physiol 1998; 275: H568–H576

42. Li D, Mehta JL. Upregulation of endothelial receptor for oxidized LDL(LOX-1) by oxidized LDL and implications in apoptosis of humancoronary artery endothelial cells: evidence from use of antisenseLOX-1 mRNA and chemical inhibitors. Arterioscler Thromb Vasc Biol2000; 20: 1116–1122

43. Li D, Mehta JL. Antisense to LOX-1 inhibits oxidized-mediatedupregulation of monocyte chemoattractant protein-1 and monocyteadhesion to human coronary artery endothelial cells. Circulation2000; 101: 2889–2895

44. Li D, Yang B, Philips MI, Mehta JL. Proapoptotic effects of Ang II inhuman coronary artery endothelial cells: role of AT1 receptor andPKC activation. Am J Physiol 1999; 276: H786–H792

45. Mehta JL, Li D. Facilitative interaction between angiotensin II andoxidized LDL in cultured human coronary artery endothelial cells. JRen Ang Ald Syst 2001; 2: D70–S76

46. Ross R. The pathogenesis of atherosclerosis: a perspective for the1990s. Nature 1993; 362: 801–809

47. Lloyd-Jones DM, Evans JC, Larson MG, O’Donnell CJ, Wilson PW, LevyD. Cross-classification of JNC VI blood pressure stages and risk groupsin the Framingham Heart Study. Arch Intern Med 1999; 159:2206–2212

48. Sung BH, Izzo JI, Wilson MF. Effects of cholesterol reduction on BPresponse to mental stress in patients with high cholesterol. Am JHypertens 1997; 10: 592–599

49. Nazzaro P, Manzari M, Merlo M, Triggiani R, Scarano A, Ciancio L, etal. Distinct and combined vascular effects of ACE blockade and HMG-CoA reductase inhibition in hypertensive subjects. Hypertension 1999;33: 719–725

50. Steinberg D, Witztum JL. Lipoproteins and atherogenesis. JAMA 1990;264:3047–3052

51. Becker RH, Wiemer G, Linz W. Preservation of endothelial functionby ramipril in rabbits on a long-term atherogenic diet. J CardiovascPharmacol 1991; 18: S110–S115

52. Pitt B, Poole-Wilson PA, Segal R, Martinez FA, Dickstein K, Camm AJ.et al. Effect of losartan compared with captopril on mortality inpatients with symptomatic heart failure: randomized trial—theLosartan Heart Failure Survival Study ELITE II. Lancet 2000; 355:1582–1587

53. Marian AJ, Safavi F, Ferlic L, Dunn JK, Gotto AM, Ballantyne CM.Interactions between angiotensin-1 converting enzyme insertion/deletion polymorphism and response of plasma lipids and coronaryatherosclerosis to treatment with fluvastatin: the Lipoprotein andCoronary Atherosclerosis Study. J Am Coll Cardiol 2000; 35: 89–95

54. Cashin-Hemphill L, Dinsmore RE, Chan RC, Williams SN, HombangG, Haver HE, et al. for the QUIET investigators. LDL cholesterol andangiographic progression in the QUIET trial. J Am Coll Cardiol 1997;29: 85A

55. Pitt B, O’Neill B, Feldman R, Ferrari R, Schwartz L, Mudra H, et al.The Quinapril Ischemic Event Trial (QUIET): evaluation of chronicACE inhibitor therapy in patients with ischemic heart diseaseand preserved left ventricular function. Am J Cardiol 2001; 87:1058–1063

56. Pitt B, Pepine C, O’Neill B, Haber H, Pressler M, Mancini GBJ, for theTREND investigators. Modulation of ACE inhibitor efficacy oncoronary endothelial dysfunction by low-density lipoproteincholesterol. J Am Coll Cardiol 1997; 29: 70A

57. Yusuf S, Pepine CH, Garces C, Pouleur H, Salem D, Kostis J, BenedictC, et al. Effect of enalapril on myocardial infarction and unstableangina in patients with low ejection fractions. Lancet 1992; 340:1173–1178

58. Pfeffer MA, Braunwald E, Moye LA, Basta L, Brown EJ Jr, Cuddy TE,et al. Effect of captopril on mortality and morbidity in patients withleft ventricular dysfunction after myocardial infarction. N Engl J Med1992; 327: 669–677

59. Lonn EM, Yusuf S, Jha P, Montague TJ, Teo KK, Benedict CR, et al.Emerging role of angiotensin-converting enzyme inhibitors in cardiacand vascular protection. Circulation 1994; 90: 2056–2069

60. Rutherford JD, Pfeffer MA, Moye LA, Davis BR, Flaker GC, Kowey PR,et al. Ef fects of captopril on ischemic events after myocardialinfarction. Results of the Survival and Ventricular Enlargement trial.SAVE Investigators. Circulation 1994; 90: 1731–1738

61. Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects ofan angiotensin-converting enzyme inhibitor, ramipril, oncardiovascular events in high-risk patients. The Heart OutcomesPrevention Evaluation Study Investigators. N Engl J Med 2000; 342:145–153

62. Lonn E, Yusuf S, Dzavik V, Doris C, Yi Q, Smith S, et al. Effects oframipril and vitamin E on atherosclerosis: the study to evaluatecarotid ultrasound changes in patients treated with ramipril andvitamin E (SECURE). Circulation 2001; 103: 919–925

63. Navalkar S, Parthasarathy S, Santanam N, Khan BV. Irbesartan, anangiotensin type 1 receptor inhibitor, regulates markers ofinflammation in patients with premature atherosclerosis. J Am CollCardiol 2001; 37: 440–444

64. Pepine CJ. Ongoing clinical trials of angiotensin-converting enzymeinhibitors for treatment of coronary artery disease in patientswith preserved left ventricular function. J Am Coll Cardiol 1996; 27:1048–1052

IHJ-007-01.p65 10/17/01, 12:01 PM518

Indian Heart J 2001; 53: 519–520 Duggal et al. Right Atrioventricular Metastasis of Hypernephroma 519

Right Atrioventricular Metastasis of Hypernephroma

Bhanu Duggal, Sandeep Seth, Pradeep K Kumar, Anil Bhan,Sandeep Aggarwal, Subodh Kumar

Department of Cardiology, All India Institute of Medical Sciences, New Delhi

Indian Heart J 2001; 53: 519–520 Cardiovascular Images

Correspondence: Dr Sandeep Seth, Assistant Professor of Cardiology, AllIndia Institute of Medical Sciences, New Delhi

Fig. 1. Transthoracic subcostal view showing the hypernephroma infiltratingthe IVC and extending into the right atrium.

Fig. 2. Transeophageal longitudinal view of the atrial septum and vena cava,showing the mass extending from the IVC–RA junction, and filling the rightatrial cavity.

A 37-year-old man was admitted to our hospital forevaluation of weight loss and pain in the right lumbar

region. Ultrasound of the abdomen, computerizedtomography and magnetic resonance imaging showed a

Fig. 3. Transesophageal four-chamber view showing the mass prolapsing acrossthe tricuspid valve during diastole.

Fig. 4. Intraoperative photograph showing a large solid tumor at the upperpole of the right kidney.

large mass over the upper pole of the right kidney withextension of the tumor into the inferior vena cava (IVC)extending up to the right atrium. A large, mobile right atrialmass was seen on transthoracic echocardiography and thesubcostal view showed the tumor extending into the rightatrium from the IVC and prolapsing into the right ventricle(Fig. 1). There were no obstructive signals across the

IHJ-Images-01.p65 10/17/01, 3:58 PM519

520 Duggal et al. Right Atrioventricular Metastasis of Hypernephroma Indian Heart J 2001; 53: 519–520

tricuspid valve. In view of the young age of the patient andthe absence of lymphadenopathy and liver metastasis onimaging, it was decided to subject the patient to rightnephrectomy and IVC thrombectomy.

Preoperative transesophageal echocardiography (Figs. 2and 3) confirmed the findings of transthoracicechocardiography. Intraoperatively, there was a large tumorabout 20×20 cm in size with a bosselated surface at theupper pole of the right kidney, adherent to the inferiorsurface of the liver (Fig. 4). Mobilization of the tumorresulted in generalized oozing from the inferior surface ofthe liver and the patient developed severe hypotensionperoperatively. The retroperitoneal space was packedwith sterile towels but we lost the patient in the earlypostoperative period.

Metastatic tumors in the right atrium as a result of directinvasion and extension up the IVC are rare.1 We report acase of renal cell carcinoma (RCC) which involves the venacava, extending up to the right atrium and right ventricle.Renal cell carcinoma, a unique neoplasm because of itspropensity to propagate into the renal vein and IVC, mayinvolve the heart in two ways. First, in 5%–10% of cases,the primary tumor invades the IVC and in 10%–40%, itfurther extends up the vena cava into the right atrium.Second, the primary tumor metastasizes to the heart in10%–20% of patients, who die of systemic spread of theRCC. Right atrial masses can be benign or malignant,

primary or secondary.2 Echocardiographically, primarytumors commonly originate from the interatrial septum,the most common being the right atrial myxoma. Secondaryinfiltrative tumors of the heart are by far more commonand usually extend into the right atrium from the IVC (andless commonly from the right atrial free wall). Tumorextension along the IVC and into the right atrium is themechanism of intracardiac tumor spread, most frequentlydescribed in RCC, Wilm’s tumor, hepatoma and uterineleiomyoma.3 One case of a hypernephroma prolapsingthrough the tricuspid valve, causing tricuspid inflowobstruction and a right-to-left shunt through a patentforamen ovale has also been reported.

Metastatic involvement of the right atrium by an RCCvia the IVC is one possible cause of a right atrial mass.Imaging of the IVC (subcostal view) can show the tumororiginating from the IVC, thus avoiding confusion withtumors which have a primary origin in the heart.

References

1. Montie JE, el Ammar R, Pontes JE, Medendorp SV, Novick AC, StreemSB, et al. Renal cell carcinoma with inferior vena cava tumor thrombi.Surg Gynecol Obstet 1991; 173: 107–115

2. Yadav BS, Agrawal S. Hypernephroma presenting as right atrial mass.J Assoc Physicians India 1998; 46: 307–308

3. Errichetti A, Weyman AE. Cardiac tumors and masses In; WeymanAE. Principles and practice of echocardiography. 2nd ed. Pennsylvania:Lea and Febiger; 1994. pp. 1166–1168

IHJ-Images-01.p65 10/17/01, 3:58 PM520

Indian Heart J 2001; 53: 521–523 Letters to the Editor

Egg Shell Calcification of theAorta in Aortoarteritis

The report by Sivakumar et al.1 of unusual, extensive,dense and nodular calcification of the aorta in

aortoarteritis made interesting reading. In this context, itmay be relevant to present a different pattern of calcificationin aortoarteritis.

The patient was a 22-year-old woman with type IIInonspecific aortoarteritis. A rim of calcification of the aortaextended from the ascending to the abdominal aorta(Fig. 1). No treatment was offered. Egg shell calcification ofthe aorta is uncommon in young adults. Our case highlightsthe unusual presentation of the condition. The appearanceis so typical that the diagnosis can usually be suspected froma simple chest X-ray.

Fig. 1. Chest X-ray chest in postero-anterior view showing egg shell calcificationfrom the ascending aorta to the abdominal aorta (arrows).

Pacemaker “Like” Syndrome

An 82-year-old male had a history of long-standingbronchial asthma, hypertension, coronary artery

disease (single-vessel disease), and presented withcomplaints of brief episodes of giddiness with slurring ofspeech for the past few weeks. He had been admitted twoyears ago with paroxysmal atrial fibrillation with leftventricular systolic dysfunction and pulmonary edema.

During one of the episodes, a cardiologist noted a normalpulse. Electroencephalogram (EEG) and Holter monitoringduring another episode were normal. We noted a normalpulse with exaggerated neck pulsations. Repeat Holtermonitoring captured two 30 s long symptomatic episodesof junctional rhythm [isorhythmic atrioventricular (AV)dissociation] (Fig. 1). His medications were modified tosuppress automaticity.

Hypotension and neurological symptoms due to cerebralhypoperfusion occurred in the patient due to loss of AVsynchrony. A normal pulse during “symptom” is rare inpatients with symptomatic arrhythmias except in thepacemaker syndrome. The symptoms of pacemaker

Fig. 1. Sinus rhythm (average; RR 830 ms) and transition to slightly fasterjunctional rhythm (average; RR 780 ms) (arrow) which lasted for 45 s.

Reference

1. Sivakumar V, Rao S, Lakshmi AY, Reddy CK. Extensive arterialcalcification in aortoarteritis. Indian Heart J 2001; 53: 372–373

KM KrishnamoorthyDepartment of Cardiology

Sri Sathya Sai Institute of Higher Medical SciencesPuttaparthi, Andhra Pradesh

IHJ-170-01.p65 10/17/01, 1:34 PM521

522 Letters to the Editor Indian Heart J 2001; 53: 521–523

Computerized averaged beats for that period (Fig. 1a andb), however, revealed a narrow QRS complex. Broad QRStachycardia was missed when only averaged complexes wereanalyzed.

Case 2: A 42-year-old male was subjected to computerizedstress testing for breathlessness on ef fort. Duringhyperventilation he developed isolated unifocal prematureventricular contractions (Fig. 2A and B). The computerselected one such beat and reproduced it as an averagedcomplex (Fig. 2a and b). When only averaged complexeswere analyzed, it was incorrectly interpreted ashyperventilation-induced intraventricular conductiondefect.

Fig. 2A and B. Raw electrocardiogram showing unifocal prematureventricular beats during stress test. a and b. Showing a prematureventricular beat as a computer-averaged beat.

Case 3: A 57-year-old male was subjected to computerizedstress testing for atypical chest pain. His resting ECGrevealed a right bundle branch block with a normal QRSaxis. During step 4, he developed left posterior hemiblock(Fig. 3A and B). Computer-averaged beats for that period,however, did not show the change in QRS axis (Fig. 3a andb). An analysis of averaged beats alone missed the leftposterior hemiblock.

Computerized systems use the following methods:1

1. Selecting the best complex (QRS complex with least noiseand baseline wander) for analysis and rejecting the restof the data.

2. Averaging of all morphologically similar, dominant QRScomplexes and rejection of beats with a differentconfiguration.

Irrespective of the method used, the computer rejectspart of the recording while producing an averaged complex.This rejection process can involve clinically usefulinformation. Averaged beats selected by the computer,

syndrome include orthostatic hypotension, near syncope,light headedness and disturbed mentation.1,2 Themechanism of symptoms in our patient is similar to thatseen in the pacemaker syndrome.

References

1. Barold SS, Zipes DP. Cardiac pacemakers and antiarrhythmicdevices. In: Braunwald E (ed). Heart disease: a textbook ofcardiovascular medicine. Vol. 1. Philadelphia: WB Saunders; 1997.p. 722

2. Ellenbogen KA, Stambler BD. Pacemaker syndrome. In:Ellenbogen KA, Kay GN, Wilkoffs BL (eds). Clinical cardiacpacing. Philadelphia: WB Saunders; 1995. p. 419

Dr Rana Verma and Dr Rajiv BajajDepartment of Cardiology

Batra HospitalNew Delhi

Limitation of ComputerizedAveraging During Stress Test

Use of computerized averaging of an electrocardiogram(ECG) in stress testing has facilitated the removal of

motion artifacts and baseline shifts. It also facilitatescontinuous on-line analysis of ECGs during the test period.This process, however, introduces some errors which arenot widely appreciated by clinicians. We report three casesin which computerized averaging of the QRS complexresulted in incorrect interpretation of data. All tests wereperformed on ECT WS 2000 (Medtronix) machine.

Case 1: A 50-year-old male was subjected to computerizedstress testing for chest pain. During step 2, he developeda run of broad QRS tachycardia (Fig. 1A and B).

Fig. 1A and B. Raw electrocardiogram showing a run of broad QRStachycardia during the stress test. a and b. Computer-averaged beats withnormal QRS complex.

IHJ-170-01.p65 10/17/01, 1:34 PM522

Fig. 3A and B. Raw electrocardiogram showing left posterior hemiblockduring stress test. a and b. Computer-averaged beats not showing the leftposterior hemiblock.

therefore, do not always reflect the raw ECG data.2 It is thusimportant that a print-out of raw data should be takenduring each step of exercise and compared with averagedcomplexes before making a final decision.3

Indian Heart J 2001; 53: 521–523 Letters to the Editor 523

References

1. Willems JL. Computer analysis of the electrocardiogram. In:Mecfarlane PW, Veitch Lawrie TD (eds). Comprehensiveelectrocardiology. New York: Pergamon Press; 1989. pp. 1139–1176

2. Chaitman BR. Exercise stress testing. In: Braunwald E (ed). Heartdisease. Philadelphia: WB Saunders; 1997. pp. 153–176

3. Jain R, Mittal SR. Limitation of computerized averaging of ECGduring stress testing. Indian Heart J 1997; 49: 221

Dr SR Mittal and Dr Anil MathurDepartment of Cardiology

JLN Medical CollegeAjmer, Rajasthan

IHJ-170-01.p65 10/17/01, 1:34 PM523

Selected Summaries Indian Heart J 2001; 53: 524–527

Platelet Glycoprotein IIb/IIIa Inhibition with Coronary Stenting forAcute Myocardial Infarction (ADMIRAL Trial)

Montalescot Giles et al. N Engl J Med 2001; 344: 1895–1903

Summary

This multicenter, double-blinded trial involving 300patients with acute myocardial infarction (AMI) studied theefficacy of glycoprotein (Gp) IIb/IIIa inhibitor (abciximab)given before primary stenting. Inclusion criteria were broadand comprised patients who had the first symptoms of AMIwithin 12 hours before enrollment. All the patients receivedaspirin and were randomized to receive either abciximab ora placebo as soon as possible (in a mobile ICU, the casualtydepartment, in the ICCU or in the catheterizationlaboratory), in all cases before sheath insertion andcoronary angiography (CART). Patients received eitherabciximab as a bolus of 0.25 mg/kg, followed by a 12-hourinfusion at a rate of 0.125 µg/kg/min or a matched placebo.Heparin was given as an initial bolus of 70 U/kg (maximum7000 U) with additional doses given as necessary tomaintain an ACT of 200 s. After coronary angioplasty(PTCA), a continuous infusion of heparin 7 U/kg/hour wasinitiated and maintained till check angiography wasobtained 24 hours after the procedure. Sheaths wereremoved 4–6 hours after the check angiography andticlopidine administered till 30 days after the procedure. Astent was implanted if the diameter of the infarct-relatedartery (IRA) was greater than 2.5 mm with no unsuitableanatomic features. The primary end-point was a compositeof death, reinfarction or urgent target vesselrevascularization (TVR) at 30 days. It was significantlyreduced in the abciximab group (6%) compared to theplacebo group (14.6%, p=0.01). At 6 months also, it wassignificantly reduced in the abciximab group (7.4% v.15.9%, p=0.02). The superior response in the abciximabgroup could perhaps be attributed to greater TIMI grade 3flow before the procedure (16.8% v. 5.4%, p=0.01),immediately afterwards (95.1% v. 86.7%, p=0.04), and 6months later (94.3% v. 82.8%, p=0.04). The primary end-point at 30 days had occurred in only 7.4% of patients withTIMI grade 3 flow at the end of the procedure as comparedwith 35.3% of the patients with TIMI grade 0, 1 or 2 flow(p<0.001). This difference persisted at 6 months (p<0.001).Minor bleeding was higher in the abciximab group (12.1%v. 3.3%, p=0.04) and there was 1 major bleed in this groupversus none in the placebo group. The authors concludedthat as compared with placebo, early administration ofabciximab before stenting improves coronary patency and

clinical outcomes and the response persists for at least 6months.

Comments

The superiority of a strategy of primary angioplasty withstenting over conventional thrombolysis in patientspresenting with AMI has now been well established. In arecent study, Le May et al. have shown that while a strategyof primary stenting was clearly superior to that ofaccelerated t-PA in terms of the composite end-point ofdeath, reinfarction, stroke or repeated TVR at 6 months(24.2% v. 55.7%, p<0.001), mortality was actually higherin the primary stenting group, (4.8% v. 3.3%) (thoughstatistically not significant). One possible reason could be ahigh thrombus load. In their study, the routine use ofabciximab was discouraged. It is logical to assume that theeffects of abciximab and stenting will be complementary.At present, data on the use of abciximab in patientsundergoing primary angioplasty and stenting are limited.In the present study, abciximab in combination with stentplacement reduced both the incidence of acute ischemicevents and subacute events related to clinical restenosis. Thebeneficial effect of this combination accrues probably frombetter procedural success, better patency of the IRA,reduced rate of reocclusion, and better left ventricularfunction at follow-up. Left ventricular functioning may alsohave been better preserved due to other beneficalmechanisms attributable to abciximab, i.e. less distalembolism, less side-branch closure and improvement inmicrocirculation. The benefits of abciximab were alsodemonstrated in the high-risk subgroup of patients withdiabetes, small coronary arteries or cardiogenic shock.However, an inordinately high rate of bleedingcomplications is a cause of concern. In contrast to theprevailing practice of using low-dose heparin, in the presentstudy, a higher dose of heparin was given (with excessivelyhigh ACT, mean of 316, and APT, mean of 2.2, in theabciximab group). Furthermore,a prolonged infusion ofheparin was given and the sheaths were removed more than24 hours after the procedure. Although the characteristicsof the patients enrolled in the two groups appear to bematched overall, there were more diabetics, hypertensivesand patients with CHF in the placebo group and, therefore,some bias may have crept into the results.

Selected Summary.p65 10/17/01, 5:24 PM524

Summary

This was a prospectively designed study of 2658 patientswho underwent percutaneous coronary interventions (PCI)in the Clopidogrel in Unstable angina to prevent RecurrentEvents (CURE) trial and were randomly assigned to double-blind treatment with clopidogrel (n=1313) or placebo(n=1345). The aim was to test the efficacy of treatment withclopidogrel in addition to aspirin prior to performing PCI inpreventing major ischemic events afterwards and determinethe possible benefits of its long-term use following PCI inpatients with non-ST segment elevation acute coronarysyndromes (ACS). Inclusion criteria were: patientspresenting with symptoms indicative of ACS within the last24 hours in the absence of >1 mm ST segment elevationon the electrocardiogram (ECG), but with ECG evidence ofnew ischemia, or cardiac enzymes or troponin levels at leasttwice the upper limit of normal. Exclusion criteria were:standard contraindications for antithrombin drugs, NYHAclass IV, previous coronary artery bypass grafting (CABG),PCI within 3 months or use of glycoprotein (Gp) IIb/IIIainhibitors within the last 3 days. Clopidogrel or a matchingplacebo was started in addition to aspirin afterrandomization till the time of PCI, following which open-label thienopyridine (clopidogrel or ticlopidine) was givenfor 2–4 weeks. After this, clopidogrel or placebo wascontinued for the next 3–12 months. The primary end-pointwas a composite of cardiovascular (CV) death, myocardialinfarction (MI) or urgent target vessel revascularization(TVR) within 30 days of PCI. Significantly fewer patientsdeveloped MI on clopidogrel prior to PCI (3.6% v. 5.5%,p=0.04). On follow-up at 30 days after PCI, there was asignificant reduction in the primary end-point in theclopidogrel group (4.5% v. 6.4%, p=0.03). The compositeof cardiac death or MI was also lower (2.4% v. 4.4%,p=0.04), mainly because of a significant reduction in allMIs (2.4% v. 3.8%). After excluding nearly 25% of patientswho received open-label thienopyridine prior to PCI, the42% reduction in the primary end-point was still highlysignificant in the clopidogrel group (4.2% v. 7.2%,p=0.005). This benefit occurred regardless of the use ofstents (6.1% v. 3.5%, p=0.016). There was a nonsignificantbut consistent reduction in the rate of CV death or MI (3.1%v. 3.9%) from 30 days to 8 months post-procedure. However,the composite reduction in CV death, MI or re-hospitalization (25.3% v. 28.9%) reached significance

Effects of Pretreatment with Clopidogrel and Aspirin Followed by Long-Term Therapy inPatients Undergoing Percutaneous Coronary Intervention: The PCI-CURE Study

Mehta SR et al. Lancet 2001; 358: 527–533

Indian Heart J 2001; 53: 524–527 Selected Summaries 525

during this period. Overall (including events before or afterPCI), there was a 31% reduction in CV death or MI. Takenseparately, there was a significant reduction in the primaryend-point in the period before PCI, from PCI to 30 days post-procedure and from PCI to the end of follow-up. The lowerrate of primary end-points was seen as early as 2 days afterPCI, indicating that the early benefit was mainly due to theeffects of pre-treatment with clopidogrel. There was nosignificant difference in the complications, including majorbleeding, though there was an excess of minor bleedingepisodes (3.5% v. 2.1%, p=0.03) in the clopidogrel group

Comments

This study has shown for the first time the long-termincremental benefit of using another antiplatelet agentalong with aspirin in the reduction of death or MI, bothbefore and after PCI for ACS. The significant benefit seen inthe clopidogrel group even during the 30 days after PCI,when all the patients received open-label thienopyridine,shows that the average pre-procedure 6-day therapy withclopidogrel also had a significant benefit in terms ofpreventing platelet thrombus-mediated events. The impactwas seen as early as 48 hours after PCI and increased over7 and 14 days, even though there may be some masking ofbenefit due to the use of open-label thienopyridine in about25% of patients pre-PCI in both groups. This benefit ismaximum for reduction in MI, especially Q-wave MIs. The32% reduction in death or MI in the period prior to PCIstrongly supports the use of this drug electively in patientswith ACS in whom an invasive strategy is being considered.The continued benefit of the use of additional clopidogrelover 8 months of follow-up was due to the reduction in therisk of rupture of the atherosclerotic plaques in areas otherthan the culprit lesion site. Moreover, long-term additionalclopidogrel use may prevent late stent occlusion in somecases as neointimal and endothelial coverage of the stentrequires 3–6 months. All these benefits are accompaniedby an increase in minor but not major or life-threateningbleeding events. The only issue is the cost of this new drug,which continues to be high and beyond the reach of a vastmajority of our population, especially if long-term use isadvocated.


Summary

The GUSTO V study was a randomized open-label trialwhich compared the efficacy of reteplase alone withreteplase plus abciximab in patients with acute myocardialinfarction (AMI). Overall, 16 588 patients with evolvingST segment elevation AMI presenting within 6 hours ofonset of chest pain were randomized to receive either full-dose reteplase (two 10 U boluses, 30 min apart), or acombination of half-dose reteplase (two boluses of 5 U, 30min apart) plus a standard abciximab dose. All the patientswere given aspirin and heparin; however, the dose ofheparin was reduced in the group given the combination.The primary end-point was 30-day mortality andsecondary end-points included the various complicationsof AMI. All the baseline characteristics, including the useof adjunctive drugs, were matched in the two groups.However, the time of initiation of thrombolysis was earlierin the reteplase arm (1.6 v. 2.2 hours in the combinationarm). Although 30-day mortality was lower in thecombination group (5.6% v. 5.9%), it was not statisticallysignificant (p=0.43; 95% CI: 0.83–1.08). However, thecomposite of death and nonfatal MI was lower (7.4% v.8.8%, p<0.001) in the combination group and there was alesser need for urgent percutaneous coronary intervention(PCI) and fewer major nonfatal ischemic complications ofAMI with the combination than reteplase alone. The rateof spontaneous hemorrhage, predominantly fromgastrointestinal sources, was significantly higher in thecombination group (4.6% v. 2.3%, p<0.0001). Except inthe elderly (>75%), the risk of intracranial hemorrhage wasnot different (1% v. 0.9%, p=0.55). The authors concludethat although the combination of low-dose reteplase andabciximab was not superior to reteplase alone as per 30-day mortality, there was a consistent reduction incomplications of AMI, such as reinfarction or need forurgent PCI or CABG. The beneficial ef fects of thecombination therapy were counterbalanced to some extentby an increase in nonintracranial bleeding.

Comments

Since their initiation in 1990, the GUSTO series of studieshave broadened our understanding of the pathophysiologyof AMI and have made an impact on prevailing clinicalpractice. The present study, GUSTO V addresses thelimitations of fibrinolytic therapy alone and investigates theefficacy of a combination of fibrinolytic therapy with potent

antithrombin agents in reducing the hard end-point of 30-day mortality. The limitations of fibrinolytic therapy alone(delayed and incomplete restoration of TIMI grade 3 flow,reocclusion of opened artery and lack of tissue reperfusion)are well recognized. One of the strategies to overcome thisproblem is to perform primary angioplasty. However,because of limited accessibility and the high cost involved,primary PTCA may not be a viable option in the vastmajority of patients. One of the mechanisms limitingef fective tissue perfusion in thrombolysis has beenmicroembolization of platelet-rich clots and the paradoxicalprothrombotic effect of fibrin degradation products (arisingfrom fibrinolysis). Aspirin, as such, is only a mildantithrombin agent. In this context, potent antithromboticagents such as glycoprotein IIb/IIIa receptor antagonistsmay hold a key to this problem. Several phase II trialsevaluating a combination of these two agents have providedthe pathophysiological basis of this hypothesis. The TIMI14 trial enrolled 888 patients of AMI and randomized themto receive either accelerated t-PA, abciximab bolus plusinfusion, or a combination of abciximab with reduced dosesof t-PA or STK. A prompt and significant increase in therate of TIMI grade 3 flow at 90 min in the combinationgroup of low dose t-PA with abciximab was demonstratedand this improvement occurred without an increase in therisk of major bleeding. Similarly, other studies, such asSPEED and INTRO-AMI, also showed the beneficial effectsof this combination. However, none of these phase II trialswere powered to investigate mortality reduction. Incontrast, GUSTO V is a megatrial involving >16 000patients. In this study too, although there was a significantreduction in all of the nonfatal complications of AMI, theprimary end-point of mortality reduction was not affected.There could be several reasons for this. Perhaps, due tobetter management of AMI (in both the groups), mortalityitself is low; therefore, a still larger number of subjects maybe required to achieve a significant difference in mortality.Secondly, 30-day mortality may not be the right time-framefor this parameter. The benefits accrued by this combinationare likely to manifest over a longer period of time, perhaps1 year. Finally, thrombolysis was achieved earlier in thereteplase group (1.6 v. 2.2 hours in the combination group)perhaps due to technical reasons and this maycounterbalance the beneficial effect of the combination. Theopen-label design of the study is a technical, thoughunavoidable, flaw in the trial.

Reperfusion Therapy for Acute Myocardial Infarction with Fibrinolytic Therapy orCombination Reduced Fibrinolytic Therapy and Platelet Glycoprotein

IIb/IIIa Inhibition: The GUSTO Randomized Trial

Topol Eric J et al. Lancet 2001; 357: 1905–1914

526 Selected Summaries Indian Heart J 2001; 53: 524–527


Summary

This study was a randomized, placebo-controlled trialcomparing the short-term efficacy of immunosuppressivetherapy (prednisone and azathioprine) with a placebo inpatients with idiopathic dilated cardiomyopathy (IDC).Eighty-four patients of chronic heart failure (>6 months)with an ejection fraction <40% and increased expressionof human leucocyte antigen (HLA) on endomyocardialbiopsy (EMB) specimens were included in the present study.Patients suspected of having coronary artery disease (CAD)based on the clinical findings or having two or more riskfactors underwent coronary angiography to rule it out andthose with more than 50% coronary artery stenosis wereexcluded from the present study. On the basis of the Dallascriteria, 7 patients (8.3%) had active myocarditis, 16 (19%)had borderline myocarditis and 61 (72.6%) had nomyocarditis. All the patients received conventional therapy,including digitalis, diuretics, an angiotensin-convertingenzyme (ACE) inhibitor and β-blockers. Theseimmunohistologically verified chronic myocarditis patientswere then randomized to placebo or immunosuppressivetherapy with steroids, and azathioprine. Prednisone wasstarted at a dose of 1 mg/kg/day initially and then taperedto a maintenance dose of 0.2 mg/kg/day for a total of 90days. Azathioprine was given at a dose of 1 mg/kg/day for100 days. The patients were followed up over 2 years. Theprimary end-point was a composite of death, hearttransplantation and hospital re-admission. At the end of 2years, there was no difference in the composite end-pointin the immunosuppressive therapy group (22.8%)compared with the placebo group (20.5%). Left ventricularejection fraction (LVEF) increased significantly both at 3months and at the end of 2 years in the immunosuppressivegroup. Furthermore, protocol-specified definition ofimprovement was significantly higher in theimmunosuppressive group, both at the end of 3 months(71.8% v. 20.9%, p<0.001) and at the end of 2 years (71.4%v. 30.8, p<0.001).

Comments

Inflammatory dilated cardiomyopathy is a heterogenousdisease. The diagnosis may be based on clinical,histopathological or immunohistological criteria. Whileconventional therapies like diuretics, ACE inhibitors andβ-blockers provide symptomatic benefit and retard theprogression of the disease, they are not really helpful in

reversal of the disease process. As cardiomyopathy is animmunological disorder, immunosuppressive therapy seemslogical if ongoing inflammation is detectable byimmunohistochemistry. In practice, however, the usage hasbeen controversial. While some investigators, like Parrilloet al. and Mason et al., have failed to show any benefit withimmunosuppressives, others like Maisch and Shultheisshave shown the benefit of this type of therapy. The differencein the results can probably be explained by the difference inthe diagnostic criteria utilized for the inclusion of patientsin these studies. Parrillo et al. utilized clinical criteria fordiagnosis and included 102 patients. Patients receivingprednisone showed no overall improvement in the clinicalcriteria and LVEF. Mason et al. randomized 111 patientsby histological Dallas criteria (lymphocyte infiltrationand myocyte necrosis). At the end of 12 months, bothimmunosuppressive and control groups showed anequivalent improvement in EF. However, there were severallimitations of these studies: (i) histological features ofmyocarditis (unlike immunohistological features) tend tobe focal and therefore can be missed in EMB; (ii) clinicalfeatures were not very specific; active myocarditis wasdetected in only 8% of patients enrolled on the basis ofclinical criteria; (iii) there was widespread disagreement onthe same biopsy specimen among different pathologists(when the Dallas criteria were used) to the tune of 40%;(iv) as many as 30% of patients were lost to follow-up. Inthis context, immunohistological techniques are not onlymore sensitive but also more specific. In the present study,upregulation of human leucocyte antigen (HLA) has beenused as a diagnostic criterion. Identification of HLA proteinson the surface of cells has been correlated with autoimmunereactions and is invariable in inflammatory IDC. Thiscriterion may represent an advancement in diagnostictechniques based on the better understanding of thepathogenesis of IDC and its application in clinicalmanagement. In the present study, although the primaryend-point was not different at the end of 2 years, patientson immunosupressive therapy demonstrated an impressiveincrease in LVEF at 3 months and 2 years. However, thereare some limitations of the present study. The criterion ofHLA upregulation requires validation in a larger study.Patients with acute heart failure (<6 months) wereexcluded. The authors reasoned that CHF in a largeproportion of these patients would anyway be reversible.The study had a high drop-out rate (31%). Finally, the studycould not be truly blinded as steroid therapy is identifiableclinically by cushingoid features.

Randomized, Placebo-Controlled Study for Immunosuppressive Treatment ofInflammatory Dilated Cardiomyopathy

Wojnicz Romuald et al. Circulation 2001; 104: 39–45

Indian Heart J 2001; 53: 524–527 Selected Summaries 527


November 11–14, 2001, 74th Scientific Session,American Heart Association, Anaheim, California, USAContact: American Heart Association, 7320, Greenville

Avenue, Dallas, Texas 75231, USAFax: 1 214 373 3406

December 6–9, 2001, 53rd Annual Conference ofCardiological Society of India, Hyderabad, IndiaContact: Professor P Krishnam Raju, Organizing Secretary,

Care Hospital, D. No. 6-3-248/1, Former HotelBhaskara Palace, Road No. 1, Banjara Hills,Hyderabad 34Fax: 040-6668888e-mail: [email protected]

or

Dr Anjan Lal Dutta,Chairman, Scientific Committee, Indian HeartHouse, P-60, C.I.T. Road, Scheme VII M,Kankurgachi, Kolkata 700 054Fax: (033) 355 6308e-mail: [email protected]

January 4–6, 2002, 5th Annual NewEra Cardiac Care2002: Innovation & Technology, California, USAContact: Dr W Randolph Chitwood, 1835, South Center

City Parkway, PMB 513 Escondido, CA 92025,USAFax: 1 760 839 1250

February 1–3, 2002, Annual Conference ofInternational Society for Heart Research – IndianSection, Thiruvananthapuram, IndiaContact: Professor CC Kartha, Division of Cellular &

Molecular Cardiology, Sree Chitra TirunalInstitute for Medical Sciences & Technology,ThiruvananthapuramFax: 91-471-446433/550728e-mail: [email protected]

February 8–10, 2002, VIth World Congress ofEchocardiography and Vascular Ultrasound,New Delhi, IndiaContact: Dr (Col) SK Parashar, Secretary General, C-144,

Sarita Vihar, New Delhi 110044Fax: 6942222e-mail: [email protected]

March 17–20, 2002, 51st Annual Scientific Sessions,American College of Cardiology, Atlanta, Georgia, USAContact: American College of Cardiology, 9111 Old

Georgetown Road, Bethesda, MD 20814, USAFax: 1 301 897 9745

Calendar of Conferences Indian Heart J 2001; 53: 528

May 5–9, 2002, XIV World Congress of Cardiology,Sydney, AustraliaContact: The Congress Secretariat, QVB Post Office Locked

Bag Q4002, Sydney, NSW 1230, AustraliaFax: 61 2 9290 2444e-mail: [email protected]

June 8–11, 2002, Heart Failure Update 2002 – FromDamage to Defence, Oslo, NorwayContact: European Society of Cardiology, The European

Heart House, 2035 Route des Colles,Les Templiers – BP 179, Sophia AntipolisCedex 06903, FranceFax: 33 4 9294 7601e-mail: [email protected]

June 19–22, 2002, A World Congress in CardiacElectrophysiology, Nice, FranceContact: Dr Jacques Mugica, Cardiostim 12 rue Pasteur,

Saint-Cloud 92210, FranceFax: 33 1 4602 0509e-mail: [email protected]

July 17–21, 2002, 14th Asean Congress of CardiologyKuala Lumpur, MalaysiaContact: Dr David KL Quek, Chairman, Organizing

Committee, c/o Letter Box 1502, 15th Floor,Menara Merais, 1 Jalan 19/3 Petaling Jaya46300, Selangor, MalaysiaFax: 60 3757 8363

August 31–September 4, 2002, XXIV Congress of theEuropean Society of Cardiology, Berlin, GermanyContact: European Society of Cardiology (ECOR), B.P. 174,

Sophia Antipolis Cedex F-06903, FranceFax: 33 49244 7601

September 24–29, 2002, Transcatheter Cardio-vascular Therapeutics 2002, Washington, USAContact: The Course Directors, 55 East 59th Street, 6th

Floor, New York, NY 10022–1112, USAFax: 212 434 6386e-mai: [email protected]

November 17–20, 2002, 75th Annual Session,American Heart Association, Chicago, Illinois, USAContact: American Heart Association, 7320 Greenville

Avenue, Dallas, TX 75231, USAFax: 1 214 373 3406

Cal of Conference.p65 10/17/01, 11:57 AM528

Coping with Syncope: Tilt Towards Pacingindianheartjournal.com/pdf/July-Aug2001.pdf · Coping with Syncope: Tilt Towards Pacing VK Bahl, Gautam Sharma Department of Cardiology, All

Documents