Quinidine syncope in children · 2016-11-08 · children. (1 Am Coil Cardiol1987;9:1031-7) currence of quinidine syncope in children. To examine for factors among children that may
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
JACC Vol. 9, No.5May 1987:1031-7
PEDIATRIC CARDIOLOGY
Quinidine Syncope in Children
1031
CATHERINE L WEBB, MD, MACDONALD DICK II, MD, FACC, ALBERT P. ROCCHINI, MD,
A. REBECCA SNIDER, MD, FACC, DENNIS C. CROWLEY, MD,
ROBERT H. BEEKMAN, MD, FACC, ROBERT L SPICER, MD, AMNON ROSENTHAL, MD, FACC
Ann Arbor. Michigan
Quinidine syncopeand factors associated with it are wellknown among adult patients treated for cardiac arrhythmias. To define factors that may influence the occurrence of syncope in children taking quinidine, theclinical, anatomic, electrocardiographic, roentgenographic and pharmacologic data were compared in sixpatients with syncope (Group A) and 22 patients withoutsyncope (Group B). There was a significant (chi-square= 10.2, P = 0.001) relation between heart disease andquinidine syncope: all six Group A (syncopal) patientshad heart disease whereas 15 of the 22 Group B (nonsyncopal) patients had no structural heart disease. Incontrast, no significant difference was noted betweenGroup A and Group B patients in mean age (11.4 versus11.4 years), mean quinidine serum concentration (2.9versus 2.3 Itg/ml), mean corrected QT interval beforequinidine (0.43 versus 0.40 second) or mean correctedQT interval during quinidine therapy (0.46 versus 0.46second) or between those taking digitalis and those not.
Two of the six Group A (syncopal)patients died during therapy, one 6 days after initiating therapy and one
Syncope has been recognized as a complication of quinidineadministration since 1923 (I). Most previous reports in adults(2-8) have related the syncope to quinidine-induced ventricular tachycardia observed in cardiac patients who areusually being treated for atrial flutter or fibrillation. Structural heart disease, hypokalemia and prolonged correctedQT interval on the electrocardiogram (ECG) have been observed to increase the vulnerability to ventricular arrhythmias (9,10). However, there is no information on the oc-
From the Division of Pediatric Cardiology, C. S. Mott Children'sHospital and the Department of Pediatrics, University of Michigan, AnnArbor, Michigan. This study was presented in part at the 34th AnnualScientific Session of the American College of Cardiology, Anaheim, California, 1985.
Manuscript received April 22, 1986; revised manuscript received December 3, 1986, accepted December 16, 1986.
Address for reprints: Macdonald Dick II, MD, FI123 Box 0204 C. S.Molt Children's Hospital, University of Michigan Medical Center, 1500E. Medical Center Drive, Ann Arbor, Michigan 48109-0204.
suddenly at home 6 months after beginning quinidine.Another two of the six Group A patients exhibited hypokalemia (both 2.9 mEq/liter) at the time of syncope,2 weeks and 6 months, respectively, after initiation ofquinidinetherapy; bothsurvived.Syncope occurred within8 days of initiation of quinidine therapy in three of thesix patients. Sustained ventricular tachycardia was observed during quinidine associated arrhythmia in threeof six patients with syncope; nonsustained ventriculartachycardia or complexventricular ectopicactivity whileon this therapy was observed before syncopein the otherthree patients in Group A.
This experience indicates that quinidine-related syncope in children is more likely to occur in patients withstructural heart disease, may be associated with hypokalemia, is a result of ventricular tachycardia and maybe fatal. In the presence of heart disease, in-hospitalobservation during the first 14 days and maintenance ofnormokalemia is a prerequisite for quinidine therapy inchildren.
(1 Am Coil Cardiol1987;9:1031-7)
currence of quinidine syncope in children. To examine forfactors among children that may predispose them to quinidine syncope we examined clinical, anatomic, ECG, roentgenographic, hemodynamic and pharmacologic data in 28children treated with quinidine at our institution. This reportsummarizes that experience.
MethodsStudy patients. To determine the number of children
treated with quinidine at our institution, the computer-assisted patient data base of the Division of Pediatric Cardiology, C. S. Mott Children's Hospital, University ofMichigan was interrogated for all patients with tachyarrhythmias. Three hundred fifty-five individuals seen duringthe past IS years were detected; of these, 28 had been treatedwith quinidine for more than 2 to 3 days. The records ofthese 28 children were reviewed and then assigned to oneof two groups (Table I): Group A comprised 6 patients who
0735-1097/87/$3.50
1032 WEBB ET AL.QUINIDINE SYNCOPE IN CHILDREN
JACC Vol. 9, No.5May 1987:1031-7
Table 1. Clinical and Laboratory Data
IntervalQN
QTc (s)Arrhy. to Serum
Pt. Age (yr) Anal. Treated QN-Related Syncope Before After Dose Level
experienced syncope while taking quinidine; Group B comprised 22 patients who were free of syncopebut treated withquinidine. Patients who received quinidine for 2 to 3 daysbut did not tolerate it because of nonarrhythmic-related sideeffects, such as gastrointestinal symptoms, were excludedfrom this study.
Patient data. The following data for all children wererecorded: follow-up status while taking quinidine (dead oralive), age at time of quinidine administration, sex, anatomic diagnosis, arrhythmia treated with quinidine, intervalfrom institution of quinidine to the onset of syncope andthe type of quinidine-related arrhythmia at the time of syncope (when known). In all patients, ECGs were analyzedbefore and after institution of quinidine for corrected QTinterval. In those patients with syncope the ECG obtainedmost recently before the syncopal event was used; if available, the corrected QT interval at syncope was measured.The quinidinedose, serumquinidine level, thedigitalisdoseand serum digitalis level and the serum potassium level (ifavailable) were also tabulated. Holter monitor tracings obtained while the patient was receiving quinidine were examined for minimal heart rates. If syncope occurred, theHolter tracing immediately preceding syncope was used.Chest X-ray films taken at the time of initiationof quinidinetherapy were examined and the cardiothoracic ratio wascomputed. Hemodynamic data, available in all patientswithsyncope and in II of 22 patients who did not have syncope,were also reviewed.
Statistics. These clinical, electrocardiographic, radiologic, physiologic and pharmacologic characteristics werecompared between the two groups using a paired t test,Fisher's exact test (two-tailed) and multiple discriminativeanalysis and, where appropriate, expressed as mean ± SO;probability (p) values <0.05 were considered significant.
ResultsSyncope occurred in 21 % of the 28 children in our study
taking quinidine for treatment of arrhythmias. There wasno significant difference in either age at initiation of quin-
idine therapy or sex between the syncopal (Group A) andnonsyncopal (Group B) groups. In Group A, five patientshad atrial flutter, three in association with sick sinus syndrome (II) and one patient had high grade ventricular ectopic activity; in Group B: 19 children had paroxysmalsupraventricular tachycardia and two had atrial flutter associated with sick sinus syndrome. The patients were followed up for 4 months to 15 years (mean 3.2 years) afterthe initiation of quinidine therapy.
Structural heart disease. The presence of heart diseasewas strongly associated with syncope. In Group A threepatients had transposition of the great arteries after Mustardrepair and one patient each had double outlet right ventricleafter repair, tricuspid atresia after a Waterston shunt andatrioventricular (AV) septal defect after repair and mitralvalve replacement. In Group B, 15 of 22 patients had nostructural heart disease, 2 had transposition of the greatarteries after Mustard repair, 2 had tetralogy of Fallot afterrepair, I had cardiomyopathy, I had Ebstein's anomaly(mild) of the tricuspid valve and I had a ventricular septaldefect after repair with associated pulmonary artery hypertension and pulmonary vascularobstructivedisease. Patient23 died suddenly4 monthsafter quinidinewasdiscontinued;patient 27 died while taking quinidine with irreversible lowcardiac output not associated with arrhythmia (in an intensive care unit with continuous hemodynamic and rhythmmonitoring). These two patients were classified in Group B(nonsyncope). In short, all 6 patients with syncope hadstructural heart disease but only 7 of 22 without syncopehad heart disease (chi-square == 10.2, P == 0.001).
Role of atrial flutter and fibrillation. All 6 patientswith syncopehad structural heart disease and 5 of the 6 hadatrial flutter as well; in contrast, only 7 of the 22 patientswithout syncope had heart disease and only 2 of these hadatrial flutter. Becauseof the well knownassociationbetweenatrial flutter (or fibrillation), structural heart disease andsyncope/sudden death (12), the possibilitythat the observedsyncope in our patients was associated with atrial flutteritself rather than with quinidine must be considered. Toaddress this issue, the patient data base was examined forall patients with atrial flutter or fibrillation and structural
Figure l. Patient I. Sustained ventricular tachycardia demonstrated byHolter monitor. The patient was successfully resuscitated.
JACC Vol. 9, No.5May 1987:1031-7
WEBB ET AL.QUINIDINE SYNCOPE IN CHILDREN
1035
heart disease. Individuals with paroxysmal atrial flutter orfibrillation (with or without the Wolff-Parkinson-White syndrome) but with a normal structural heart were excluded.The 7 patients (5 syncopal patients and 2 nonsyncopal patients from the present study) were joined with 36 otherpatients who had both structural heart disease and atrialflutter (n = 31) or atrial fibrillation (n = 5) but who hadnot received quinidine. Two of the untreated patients diedwith low cardiac output and three died suddenly. For purposes of analysis, all five of these patients were classifiedin the group with syncope/suddendeath. Chi-square analysisof these 43 patients (7 treated with quinidine and 36 nottreated), all with atrial flutter or fibrillation as well as structural heart disease, demonstrated a strong association (Fisher's exact probability = 0.004) between quinidine use andsyncope/sudden death. Thus, even when atrial flutter orfibrillation is held constant, a strong relation exists betweenquinidine and syncope in young persons with structural heartdisease.
Quinidine-related arrhythmia. The quinidine-relatedarrhythmia was demonstrated by ECG monitoring to besustained ventricular tachycardia in three Group A patients(Fig. I). In two other Group A patients, ECG tracings identified nonsustained ventricular tachycardia before the syncopal event. These two patients (Patients 3 and 6), bothwith transposition of the great arteries and sick sinus syndrome with recurrent atrial flutter, died. The Holter tracingof Patient 3 demonstrated slowing of the heart rate (1.2second pause) just before initiation of nonsustained ventricular tachycardia 2 days before he died. Patient 6 hadbeen treated with quinidine, digoxin and a pacemaker forsick sinus syndrome during the 6 months before her death.Holter monitor tracings at the onset of quinidine therapyshowed frequent ventricular premature beats as comparedwith prequinidine tracings. Before discharge from the hospital couplets and triplets had resolved although Lown classification grade II ventricular ectopic activity remained, Theremaining four patients recovered and are well.
Serum potassium and duration of quinidine therapy.Serum potassium levels were available in only 9 of the 28patients. In two of the patients who experienced syncope,the serum potassium level was less than 3 mEq/liter. Bothpatients were also taking diuretics at the time of hypokalemiaand syncope.
The interval from initiation ofquinidine therapy to syncope ranged from 6 days to 6 months. Two of the threepatients who experienced syncope at an interval > 10 daysfrom initiation of therapy (one 14 days and one 6 months)had a serum potassium concentration of 2.9 mEq/liter. Onepatient died 6 months after the initiation of quinidine therapy. Her serum potassium level was 4.0 mEq/liter duringattempted resuscitation. The other three patients, two ofthree with documented normokalernia, had syncope 6 to 8days after initiation of quinidine therapy.
QT interval. Three of 6 Group A patients (Cases I, 4and 5) and 7 of the 22 Group B patients (Cases 7, 15, 17,21-24) had a corrected QT interval ~0.43 second beforequinidine therapy, The mean corrected QT interval in GroupA before quinidine therapy (0.43 second), although longer.because of the uniform presence of heart disease in GroupA, was not significantlydifferent from that of Group B (0.40second). In two patients in Group A (Patients I and 4) themean corrected QT interval shortened after initiation ofquinidine therapy. Patient I exhibited a minor decrease inthis interval; at syncope it was 0.55 second. Patient 4 wasa 15.5 year old girl whose corrected QT interval shortenedfrom 0.47 to 0.42 second after quinidine therapy. However,her quinidine level was only 1.0 JLg/ml, and she was hypokalemic at the time of syncope. All but one patient inGroup B showed an increase in corrected QT interval afterbeginning quinidine. The mean corrected QT interval afterquinidine therapy in Group A (0.46 second) was not significantly different from that in Group B (0.46 second).Interestingly, this interval did increase significantly afterquinidine therapy in those patients without syncope in contrast to those with syncope, an unexpected observation thatis explained, at least in part, by the decreased corrected QTinterval of Patient 4 after therapy.
Quinidine and digitalis dosage. The mean quinidinedosage was not significantly different between the two groups,ranging from 20 to 35 mg/kg per day in Group A and from20 to 65 mg/kg per day in Group B. Mean serum quinidinelevels also were not significantly different between GroupA (2.9 ± 1.7 JLg/ml) and Group B (2.3 ± 0.8 JLg/ml).Five patients in Group A and 10 in Group B were takingdigitalis preparations as well as quinidine. Three of the fivepatients in Group A took digoxin and the other two weretaking digitoxin, The digoxin dose ranged from 62.5 to 125JLg/day in Group A and 70 to 250 JLg/day in Group B. Thedigitoxin dose in the two Group A patients was 45 and 75/Lg/day, respectively. All patients had their digitalis dosereducedwhen quinidine was initiated. In Group A, digitoxinlevels in the two patients were not elevated. In the threeGroup A patients taking digoxin, the serum level was elevated in Patient 3 (3.8 ng/ml), who died. Mean serum digoxin level in Group B was 1.4 ng/ml (10 patients). Importantly, there was no significant relation between digitalisuse and syncope when the data were controlled for thepresence of heart disease.
Hemodynamics. The minimal heart rate during quinidine therapy for Group A (61 ± 4 beats/min) was notsignificantly different from that for Group B (68 ± 16beats/min). In addition, the difference in mean cardiothoracic ratio between Group A (0.59 ± 0.6) and Group B(0.49 ± 0.1) could be accounted for by the presence ofstructural heart disease. Finally, there was no differencebetween Group A and Group B with respect to systemicventricular end-diastolic pressure (9.5 ± 5 versus 11,6 ±
1036 WEBB ET AL.QUINIDINE SYNCOPE IN CHILDREN
lACC Vol. 9. No.5May 1987:1031-7
4 mm Hg, respectively) and cardiac index (4.5 ± I versus3.2 ± 0.8 liters/min per rrr', respectively).
DiscussionOur data are in accord with those obtained in adults
(9,10), and indicate that quinidine-related syncope in children results from ventricular tachycardia and is not infrequent in patients with serious structural heart disease orhypokalemia. In adults taking quinidine, the reported incidence of syncope ranges from 0.8 to 8% (2,5,9). Thecontrasting 21% incidence of quinidine syncope in our studymost likely resulted from patient selection (albeit as complete as possible) as well as the small number of patientsreceiving quinidine. None of our patients were treated solelyfor isolated ventricular premature beats. Nonetheless, thisincidence underscores the need for careful monitoring ofchildren given quinidine, especially those with serious structural heart disease.
Role of digoxin and prolonged QT interval. Severalinvestigators (2,5-7,9) have also suggested an increased riskfor syncope with concurrent digoxin use or in the presenceof a prolonged QT interval. Concurrent use of digitalis amongour patients was not clearly associated with quinidine syncope when the data were controlled for the presence of heartdisease. On the other hand, one Group A patient who died(Case 3) had an elevated digoxin level at the time of autopsy(3.8 ng/ml). We noted a prolonged corrected QT intervalbefore quinidine treatment in three (50%) of the six patientswith syncope, which approximates the 71% incidence inone review (9). However, in our pediatric group, the magnitude of the observed increase in the corrected QT intervalafter quinidine therapy was not associated with syncope,thus suggesting that the presence of heart disease is a strongerpredictor of syncope than is the change in corrected QTinterval.
Quinidine-related ventricular arrhythmias. Quinidinesyncope related to ventricular arrhythmias has been described in adults (1-10). In one report (9) all 31 adult patients with quinidine syncope had organic heart disease suchas ischemic heart disease, rheumatic heart disease, hypertensive heart disease, congestive cardiomyopathy, sick sinussyndrome, mitral valve prolapse or atrial septal defect. Inthese patients, quinidine was used for control of atrial fibrillation or atrial flutter, ventricular arrhythmias and paroxysmal supraventricular tachycardia. Ninety percent of thesepatients were also taking digoxin and all had therapeuticserum digoxin levels. Five patients had a serum potassiumlevel less than 3.5 mEq/liter at the time of syncope. Themajority (71%) of patients had a prolonged QT intervalbefore initiation of quinidine therapy and 91% had a prolonged corrected QT interval after beginning quinidine. Thestudy (9) concluded that quinidine syncope is most oftenidiosyncratic, not dose related, and usually occurs within I
week of initiation of therapy. Our data in children supportthese conclusions.
Recently Ruskin et al. (10) showed by programmed ventricular extrastimulation that adult patients with quinidinesyncope have an increased susceptibility to ventricular fibrillation in the presence of quinidine. Thus quinidine mayinherently increase the risk for life-threatening ventriculararrhythmia in those patients with heart disease, includingchildren, and in those with hypokalemia. Unfortunately, inpatients with quinidine-related arrhythmias, other type Iantiarrhythmic drugs as well as amiodaronehave also beenassociated with complex ventricular arrhythmias (13).
Several investigators (14,15) reported that adult patientswith underlying bradycardia in combination with a prolonged QT interval may be most vulnerable to ventriculartachyarrhythmias. In one of our patients (Case 3), nonsustained ventricular tachycardia noted in the Holter tracingwas heralded by a long pause. Further, the corrected QTinterval, as expected, increased after quinidine therapy in25 of our 28 patients. This combination of bradycardia anda prolongation of the corrected QT interval may also placepediatric patients at increased risk for complex ventriculartachycardia.
Role of underlying heart disease and surgical procedure. Children, after a corrective venous baffle operationfor transposition of the great arteries, are known to exhibitincreased age-related bradyarrhythmias as well as intermittent tachyarrhythmias compared with normal children (16),and thus may be especially vulnerable to quinidine-inducedsyncope. Indeed three of our five patients with transpositionreceiving quinidine for atrial flutter experienced syncope;two of the three (Cases 3 and 6) died suddenly while receiving therapy. On the other hand, when the 28 patientswere stratified for no structural heart disease, transpositionof the great arteries or other forms of heart disease, therewas no significant difference in either minimal heart rate orcorrected QT interval, before or after quinidine, either withineach group or among the three groups. Despite our inabilityto firmly link quinidine syncope to the bradyarrhythmiaassociated with venous operations for transposition, mostlikely owing to the small number of patients in our study,we recommend great care in the use of quinidine in thesepatients. Paradoxically, because syncope or sudden deathin patients on quinidine therapy may also be related to thenatural history of the underlying heart disease, pacemakerimplantation to prevent possible proarrhythmic effects ofbradycardia in the presence of quinidine may not be aneffective measure (Patients 4 and 6).
Quinidine safety in absence of heart disease. At thesame time, based on experience with our Group B patientsas well as the experience of other investigators, quinidine,in the absence of heart disease, is safe, and when effectivefor the prophylaxis of paroxysmal supraventricular tachycardia and other arrhythmias it can be given to outpatients,
JACC Vol. 9, No.5May 1987:1031-7
WEBB ET AL.QUINIDINE SYNCOPE IN CHILDREN
1037
provided that there is normokalemia, a normal corrected QTinterval, no atrial fibrillation, no second or third degree heartblock, no digitalis toxicity and no prior quinidine toxicity(17). The patients should be followed up with periodic Holter monitoring and serum potassium levels. If digitalis isused concurrently, the digitalis dose should be reduced byapproximately 50%.
Quinidine therapy in presence of heart disease. In achild with serious structural heart disease, the risks of quinidine therapy must be carefully weighed against the risksof no quinidine therapy. Our recommendations for the useof quinidine in a child with structural heart disease are asfollows: the patient should be observed in the hospital forapproximately 2 weeks with continuous ECG monitoringand serial measurements of both serum potassium level andcorrected QT interval. Appropriate adjustment of digitalisdosage should be made. Pacemaker implantation for prevention of bradycardia should be considered in selectedpatients. During this period, if no exacerbation of an arrhythmia is noted, if the underlying arrhythmia is controlledand if the serum potassium level remains within the normalrange, the patient may be discharged. Careful monitoringof serum potassium level and the underlying cardiac rhythmis required throughout therapy. Although other type I antiarrhythmic medications are not necessarily free of this complication, other medications such as long-acting procainamide may be considered.
We gratefully acknowledge the expert secretarial assistance of ColleenRauch and Sandra Suter.
ReferencesI. Viko LE, Marvin HM, White PD. Clinical report of the use of quin
idine sulphate. Arch Intern Med 1923;31:345-63.
2. Rokseth R, Storstein O. Quinidine therapy of chronic auricular fibrillation. Arch Intern Med 1963;III :184-9.
3. Selzer A, Wray HW. Quinidine syncope; paroxysmal ventricular fibrillation occurring during treatment of chronic atrial arrhythmias.Circulation 30;1964:17-26.
4. Reynolds EW, Vander Ark CR. Quinidine syncope and the delayedrepolarization syndromes. Mod Concepts Cardiovasc Dis 1976;45:117-22.
5. Cohen IS, Jick H, Cohen SI. Adverse reactions to quinidine in hospitalizedpatients: findingsbasedon data from the Bostoncollaborativedrug surveillance program. Prog Cardiovasc Dis 1977;20:151-63.
6. Denes P, Gabster A, Huang SK. Clinical, electrocardiographic andfollow-upobservationsin patientshavingventricularfibrillation duringHolter monitoring. Am J Cardiol 1981;48:9-16.
7. Velebit V, Podrid P, Lown B, Cohen BH, GrayboysTB. Aggravationand provocation of ventricular arrhythmias by antiarrhythmic drugs.Circulation 1982;65:886-94.
9. BaumanJL. BauernfeindRA, HoffJV, Strasberg B, Swiryn S, RosenKM. Torsade de pointes due to quinidine: observations in 31 patients.Am Heart J 1984;107:425-30.
\0. Ruskin IN, McGovern B, Garan H, DiMarco JP. Kelly E. Antiarrhythmic drugs: a possible cause of out-of-hospital cardiac arrest. NEngl J Med 1983;309:1302-6.
11. Greenwood RD, Rosenthal A, Sloss LS, LaCorte M. Nadas AS. Sicksinus syndrome after surgery for congenital heart disease. Circulation1975;52:208-13.
12. Garson A Jr. Bink-boelkens M, Hesslein PS, et al. Atrial flutter inthe young: a collaborative study of 308 cases. J Am Coil Cardiol1985;6:871-8.
13. Clark M. Friday K. Anderson N, Jackman W. Aliot E. Lazarra R.Drug induced torsade de pointes: high concordance rate among typeIA antiarrhythmic drugs and amiodarone (abstr). J Am Coil Cardiol1985:5:450.
\4. Smith WM, Gallagher JJ. "Les torsades de pointes": an unusualventricular arrhythmia. Ann Intern Med \980;93:578-84.
IS. Keren A. Tzivoni D. Gavish 0, et al. Etiology, warning signs andtherapy of torsade de pointes. Circulation 1981;64:1167-74.
16. Flinn CJ. Wolff GS, Dick M, et al. Cardiac rhythm after the Mustardoperation for complete transposition of the great arteries. N Engl JMed 1984:310:1635-8.
17. Morganroth J, Horowitz LN. Incidenceof proarrhythmiceffects fromquinidine in the outpatient treatment of benign or potentially lethalventricular arrhythmias. Am J Cardiol 1985;56:585-7.