Cardiac Event Recorder

8/8/2019 Cardiac Event Recorder

1/5

Reprintedrom NNALSF |NTERNALnM,:l i :U.:f t . 124;No.1, part ),1 Januarytee6

CardiacEventRecorders ield More Diagnoses nd Are MoreCost-effectivehan 48-HourHolter Monitoring n PatientswithPalpitationsA Controlled Clinical TrialScottKinlay,MBBS,PhD,FRACP;JamesW. Leitch,MBBS,FRACP;AmandaNeiI,BSc;BarryL. Chapman,MBBS,FRACP;DavidB. Hardy,DMU, RDCS;andPeterJ. Fletcher,MBBS,PhD,BMed(Sci), RACp

Objective: To compare he diagnostic ield an d cost-effectivenessf transtelephoni cventmonitors it h hoseof Holtermonitoringn patients it h ntermittent alpita-t ions.Design: Randomizedrossoverr ia l .Setting: Diagnostic ervice f a teachinghospitalan dsurrounding rimary arepractices.Patients: 43 patients ith previouslyninvestigatedal -pitationswh o were eferred or Holtermonitoring.Measurements: Patients ere andomly llocatedo re-ce ivean eventmonitoror 48-hourHoltermonitorandthen o receivehe otherdevice. ventmonitors ereusedfor 3 months r unt i l wo recordingsereobtained hi lesymptoms ccurred.hemainend pointwasan electro_gram recorded uringsymptoms. he ncrementalost-effectivenessf obtaininga diagnostichythm trip romeventmonitors ascompared i th that of Hol termoni-tor ing.Resul ts:Themean *SD)pat ient gewas45 + 19years;37 patients 88%)were women. Eventmonitorsweretwiceas ikely o provide diagnostichythm tripelectro-cardiogramur ing ymptomss48-hour ol termonitor-in g (2 9patients 67o/olnd15patients35%].espectively;P < 0.001). ventmonitors etected pat ients19%) i thclinicallymportant rrhythmias6patients it hsupraven-tr icularachycardiand2 withatr ia l br i l lat ionr lut ter) ,whereas he Holtermonitorsdetected o signif icant r-rhythmia P< 0.005).With the eventmonitors,mostpa-tients ransmitted n electrocardiogramecording y 6weeks. ventmonitors er edominant nd hereforemorecost-effectivehan48-hourHoltermonitoring,esult ingna cost savings f 9213 or eachaddit ional iagnost icrhythmstripobtained uringsymptoms.Conclusions:Holtermonitoring sa poordiagnostices tfo r intermittent alpitations. vent ecordersrovide et-te r dataan dar emorecost-effective.

An n Intem Med. 1996;124(lpt l) :16-20.From John t lunter Hospit i r l and an d Ccntrc fo r Cl inical Enit le-miokrgy an d Biostat ist ics. er v South Walcs,Austral ia.Fo r cur-rcnt author addrcsses. ce end of text.16 O [99(r American College of Physicians

'patpitation is a common symptom hat sometimesI results from a substantialcardiac arrhythmia.Establishing he causeof palpitations may be diffi-cult because istoricalcluesar e not alwaysaccurate(1). A 24-hour ambulatory Holter) monitor is usu-ally used, but the yield of this instrument s low inpatients whose symptomsoccur infrequently (2-5).Another instrument used to study palpitations sa transtelephonic ost-event ecorder. These hand-held devices re given o patientsand are applied othe chestwhen symptomsoccur.The patient pressesa button to record about 30 secondsof the cardiacrhythm, which is stored in the memory of the cle-vice. Th e recording is later transmitted over thetelephone or printing an d interpretation.Although this instrument has been available ormany years,concerns av ebeen expressed bout thequaliry of recording and th e extent to which anunselectedgroup of patients ca n provide diagnosticrecordings.Because o randomized, ontrolled rialshave compared hese deviceswith Holter monitor-ing, we compared the yields of Holter monitoringwith those of event recorders n diagnosingpalpita-tions.in an unselected opulation.

MethodsPatients

We consideredor our study all 634 men andwomen referred o the cardiovascularnit at theJohn Hunter Hospital or Holter monitoring.Weexcluded atients eingmonitored or silent sch-emia (7%), assessmentf therapy 187o), yncope(18%),or other esearchtudies r inp atientmon-itoring (87o);patients onsidered oo ol


2/5


3/5

Var iab le

Table 1. Numberof patientswith an ElectrocardiogramRecorded uring Symptoms r with Clin ici l lySignifcantArrhythmia*clinically signif icantarrhythmia wa s documented neach of B patients l9Vo) (Tables 1 and 2; Figure l).

Of th e 29 patients wh o recorded an electrocar_diogram with the event monitor, 20 (69%) madetwo recordingsan d returned he monitor before themaximum 3 months allowed.Fo r i3 patients 657o)wh o sent in tw o recordings, he tw o recordingsshowed the same rhythm. Fo r the remaining 7 pa _tients, he discordant hythms were supraventriculartachycardia nd sinus achycardia n 3 patients 2 ofwhom ha d supraventricular achycardiaon the firstrecording), atrial flutter and ventricular prematurebeats n 1 patient (atrial f lutrer on the fi ist record-ing), sinus tachycardiaand ventricular prematurebeats n 1 patient, and sinus hythm an d ventr icularpremature beats n 2 patients.Of th e 15 patients n whom symptomsoccurredduring 48-hour Holter monitoring, 4 (27Vo) ha d 2separatesymptomaticepisodes. n 2 of these4 pa-tients, th e rhythm recorded during symptomswa sth e same (sinus tachycardia).Both discordant pa _tients had ventricular premature beats with on esymptomatic episode and sinus rhythm with theother.Figure 2 shows the cumulative number of pa_

tients wh o sent an electrocardiogram btained bythe event recorder during symptoms, according othe time taken to record a symptomaticepisode.Most patientswh o eventually ent in recordingsdi dso by 6 weeks.Table 3 shows the costs an d incremental cost_effectiveness f event monitors comparedwith thoseof Holter monitors.Event record.r , * . r " dominantin obtainingan electrocardiogram uring symproms,result ing n a $213 cost savingsper additional out_come detected. Event recorders were dominantcomparedwith Holter monitors in al l scenarios x_

amined in the sensit ivityanalyses. abor costs ac _counted for 52Voof Holter monitor costs an d 41Zaof event monitor costs.

Figure 1. Electrocardiogram ecording from an event recordershowingatr ia l ibr i l lat ionduring palpi t l t ions.12 4 . Number I (Part l)

Hol lerMonitor Event oni torn(ok)

Recordinguring 15 35 \[21% o 49ok] 29 (67)153V.o 81% tsymptoms95 %Cl lCl in ical lyigni{ icant 0 (0 )[O % o 3%l 8 (19)t8% o 0%]+arrhythmia95 %CI ]'Forty-threepat ients art ic ipatedn th e study.t P < 0.001 rom paired est.+ P< 0.005 ro m pairedest.

attributed o Holter monitoruse ro m 17Voo 107a.The effectivenessas alsovariedby the 95ToCIs.

Results. Of the 108eligiblepatients, e asked 5 (40Ta)to participaten the study.We did no t ask theothers o participate ecausehe tw o chief nvcsti_gatorswereno t availableo interviewhem.Of the45 patientsasked,43 (96qo) ompleted he study.Two personswithdrewbeforereceivinghe eventmonitorbecausehe Holtermonitor ea-ds er e oouncomfortable.Thirry-eight articipants88To) er ewomen, ndthe mean ag e of all patients tSD) was 45 -+-19years. hirty-fourpatients 81%)reported hat pal_pitationsoccurredat leastevery2 weeks,ancl24(56Vo) elievedha t theirpalpitations er e egular;th e averageestimateof the longestattack wa s74 -r 159minutes. he mean esting ulse ate wa s76'r 15 beats/min;systolic blood pressurewasl3 l ! 26 mm Hg; and diastolic loodpressure as77 -+-12mm Hg. Four patientseported history fischemic eartdisease, 4 (33Vo)eportecl hisioryof hypertension,nd 7 (16%)weresmokers. wen_ty-fourpatients 567o)were randomlyallocatedoreceivehe 48-hourHolter monitorbefore eceivinsthe event ecorder.Table 1 lists he resultsor the primaryandsec-ondaryend points.Thirty patients702o) ent n atleaston e electrocardiogramecorried y the eventmonitor while symptomsoccurred.Two patientshad a technicallynadequate lectrocardiograme-cording transmission, ut I patient subsequentlysent wo rhythmstrips howing inus hythm. here_fore,29 patients 67Vo) entat leaston e recordingfrom an event recorder hat could be interpreteJcomparedwith th e approximatelyne third of pa_tients who obtained recordings uring symptomswith th e Holter monitor(p < 0.001).Furthermore.the Holtermonitordetected o cl inical lyignif icantarrhythmias. uring event recordermonitoring,l8 1 January 1996 . Annals of Intental Medicne . Volume


4/5

5upraventr icularachycardiaAtr ial ibr i l lat ionr lutterVentricularctopic eatsSinusachycardiaSinus hythmTotalelectrocardiogramecordings*

Table 2. Typesof CardiacRhythm rom Electrocardio.gramsRecorded uring SymptomsHol ter Mon i t o r EventMon i t o r

In ou r study, 6 rveeks va ssufl lcient or using th eevent monitor. In a report of a company databaseof pat ientswho used event monitors,96Voof pa_tients sent their first recording by 4 weeks (i2).Although 4 to 6 weeks is usually aclequate, om epatients vith very infrequent symptomsmay neecl ous e the event monitor fo r a longer period.Although the event recorders hatl a sreater di_agnost ic ie ld,somediscorclanceo, , . .n (35%) inthe rhytlrms sent by patientswh o hacl tw o record_ings.Discordancewa s alsoseenwith 48-hour t{oltermonitoring in hv o of the four patients n whom twosynrptom episodeswere recorded. Although mostcl in ical lysigni f icant r rhythmiaswcre seen on thefirst recording frclrn event monitors, obtaining atIeast tw o recordingswlrile syrnptomsare occurringn ray re p rudcn t .

Our study had adequate tat ist ical ower to ad_dress hc nrain study hypotheses, ut it was small. na larger study, sonre patients with clinicallysignif i_cant arrhythmias ecorded us ing 4g-houi Holtermonitoring would havebeen dentifiecl;we found nosuch paticnts n ou r study.Other studieshave ounclthat I lo l ter monitors dcnt i fy smal l numbersof pa-t ients with tachyarrhythrnias.n one stuclyof 40pat ients, 4-hourLlol ter monitor ing dent i f ie


5/5

Table 3. Costsand Cost_Effectivenessof EachDevrceVariable

HolterMonitor EventMonitorCost,{or 3 patjentsor 2 1.5weeks.rangeromsensrtivttynalysis)lncJemntal_costtfectjvenessor eventmonitors 4245 2650 o 4gg4t 1258 1049ro 1471 )5r l2s8 54245_)9__ lE_ _ _ nz peradditional lectrocardjogramecorded uring ymptoms*5 1 2 5 8 $ 4 2 4 5 - { 3 7 r n o . ,_ __rr/rperaddit ionalcl in ical lysigni f icantarrhythmiadetected** This epresentscost-sdvingit h eventmontto.s

than n otherstudies f patients it h cardiac ymp_toms (2Voto 20o/o) 2_5). However, hesesrudiesincludedHolter monitoring or 12 und 24 hou.r,an d the diference.in diagnostic ield may reflectthese horter ecording .i iodr.The longer ecording eriodwith the event e_corders s almostcertainly he main reason ha tthesedevices acia higherdiagnosticietA han he- Holter monitors. he averageumber f days totalnumberof daysmonitoredn all patients/numberofpatients)wa smuchhigherwith tile euen'fecorders(3 4 days) han with the Holtemonito., (2 days).Presumably,f Holter monitoring ouldbe contin_ued fo r a similarnumberof day"s,t wouldhaveadiagnosticield similar o thatof ifr" ""*, record_ers. lowever, prolongedHolter monitorings un _comfortableor patientsantl ha s highercosts ortechniciannd carci iologistime.Eventrecorders annot eplace ll the functionsof Holter monirors"l o.l. siudy,only ll,fo of pu-tientswereconsidered ligible '- ;" ; recorclers.Another 16%owouldhave- een uUt. io us e theseclevices ut were excludedbecausetiev had hatlpreviousmonitoring. he remaining atients ithercouldnot use he event ecorderbeciuse f investiga_ljol fo..syncope r because f physical r cognirivelimitations)or hatl indicationsf,ut n..ersitot.Amonitoring.-br asymptomatic lectrocardiographychangessilent schemia, ssessmentf pu.emote.functionor antiarrhythmicrugs, nA olitorlng ofinpa.tients).s manyason e thlrdof patientsn ourteaching ospitalpopulationwouldbe able to userncevenI ecorder:hisproport ion aybehighcr nother centers r practiccs here he assessmentftherapeuticevices r drugs s a less ommon ndi-cation or cardiacmonitoiing ha n is investigationof palpi tat ions.Ou r study shows that comparedwith Holtermonitors, vent ecordersesult n adclit ionalene_fits at a lowercostwhenused or diagnosingnter_mittent palpitations.Even if costs"u." ignor.,t,Holter monitoringis a poor diagnosticest n pa_

11L1tsitn palpirations.venr ecorders rovidebetterdata,are more cost_effective,nd should e__placeHolter monitoring or thispufior""*t.n.u".possible.Acknowledgmenls;The authors hank HelenBednar,FionaBro-!:i:l:,r.l"I Key,Jodiepettrt,Tammiepowell,and Mia Tolfree::"i,,^.".jr"'".,assisrancend TammieU"a!., i"."*.retarial as -I:1,,.::,t ft,.ltynnrs: Scott Kinlay,MBBS, phD, FRACp, Car-rovascularnit,John HrrnrerHospita l ,;;k;; 'Brg" t , IIrnt.,Mai l Cenrre, ew SouthWales i l 'U, Ar; ; ; i i ; . "' "

Kinlay. Leitch, Chapman, Hardy,^ Un r t , John Hun te r Hosp i ra l ,Centre, Ne w South Wales 2i10,

Current Author Atldresses: Drs.and Fletcher: CardiovascularLocked Ba g 1, Hunter Mai lAustra l ia.Ms . Nei l :UniversityNc w Southfi i:..lji*,",',^t",l,it$:m:"flfr ,,,*,fi:if:;

References1. Leitch., , Krein G, ye e R. can pa-trents iscnmrnate triar ibr, iarron ro megutar upravenrrcutarachycardia?, I carj," i . " iS9t,Ul,n[, u. DiMar(o Jp , philbrick JT. i- lo.n,t91nsnn nrern"o';-rr?,iiit-' j:toryelectrocardiosraphicHolter)5. 5aie AF , Maxwell RT . Trandetectronan d rrealment "f t- lu] : l1"l l : .ul: t t rocardrographicmonrtoring orl l 0 -2 . ca rd rac r rhy t hmra .os t g radMe d J . 199d , 66 :4. Thomas LE , Shapiro LM , p^e-r.r in:EJ , Fo x KM . Detect jonor arrnythmra:

i;-:td;:i;liil!jrout'"ntctrvatedecordrn;.'i..'"bi ,redCrinest lifil,i;,i$si1,#;3:l:: Yt: Evaruationfoutpatientrrhvrhm,as6. cofton-r. st.tlti'-.tl^ it.r].ironng Am Heart 1979:97:759-61t:il;;'w-i,t:iil::il'iliil:,i:il?'t,illli;,Xlll,l"l.n'J;L'nf;i;n0".fr:!|!i?l of transientardiachyrhm ,rtuo.n.ui. utuioirn proc rgaz;8. clair wK, wilkinsonwe {-cc:rthy EA,pageRL pritchettEL sponra_eous ccurrencef symptomaticaroxysmaltiral ,triir.ir"i^i paroxysmalupraventricularachycardjanuntreatedut,unts-iririar,Jn.-inil,gr,,, ,o -" trrlt t, " MA, Hesstein , DunniganA. Efficacyf rransretephoniclec-flojldiooruohi.moniroringn pediamcatients.m I os Cnril B9O;144tO 53:y*,a LeendersCM,-Roelandt,J. setuloessf rransrelephon,cocumentationof theelectroc.. ::1:l cardra-crrhyrhmras",','?llJ,'J Jli: jil?.?: svmptomsusees-". :J1:.T:n RS, aponeRJ,Most aS. annytnmiauru-e,ltuncey rransrele-nonlcmonitoring:ompdlsor

.r ourarory.parrenrsArnHear!, lrL!:,iig &nrtorrnsIn svmptomaticm-''l::fi t,'*Iff"';i?I,",i,i.1f"""1:'?l:,i;1,,,'.,lli'_i;#tlfiransrenrympromaricvenr erecrronef".,r.rrli"i"r sd , ll,ur-,

20 I January 1996 ' '4nnarsof IntemarMetricine. Vorumer2,l . Number (part 1)

Cardiac Event Recorder

Documents