implantable cardioverter defibrillator shocks: a...

VOL. 2 NO. 2 2001 REVIEWS IN CARDIOVASCULAR MEDICINE 61

mplantable cardioverter defibrillators (ICDs) have achieved apreeminent role in the treatment of life-threatening ventriculararrhythmias. Because most ICD recipients have structural heart

disease, they receive care from cardiologists who are not electrophysiologists but are increasingly involved in the follow-up ofICD patients. The most common arrhythmia-related complaint inICD patients is “shocks.” This review describes the diagnosis andacute management of single, multiple, or repetitive shocks deliveredby ventricular ICDs. It supplements comprehensive reviews of ICDtroubleshooting.1-3 Management of shocks from atrial ICDs, whichcardiovert atrial fibrillation with a slow ventricular rate,4 are beyondthe scope of this review.

Implantable cardioverter defibrillators deliver shocks in response to electricalsignals that satisfy programmed criteria for detection of VT or VF. The firststep in diagnosis of inappropriate shocks in patients with ICDs is to determineif the shock was delivered in response to a true tachyarrhythmia by inspectingdata stored in the ICD. Shocks occur in the absence of tachyarrhythmias becausenonarrhythmic physiologic or nonphysiologic signals are oversensed by the ICDand detected as arrhythmias. Diagnosis and causes of oversensing are reviewed.The second step in diagnosis is to determine if the tachyarrhythmia stored inthe VT/VF episode log is VT/VF or SVT by analyzing stored electrograms.Frequent or repetitive shocks constitute an electrophysiologic emergency. Theapproach to this problem is reviewed. [Rev Cardiovasc Med. 2001;2(2):61–72]© 2001 MedReviews, LLC

Key words: Implantable cardioverter defibrillator • Ventricular tachycardia •Inappropriate shocks

Implantable CardioverterDefibrillator Shocks: ATroubleshooting GuideCharles D. Swerdlow, MD, Jianxin Zhang, MS

Division of Cardiology, Cedars-Sinai Medical Center, Los Angeles, CA

I

TREATMENT UPDATE

62 VOL. 2 NO. 2 2001 REVIEWS IN CARDIOVASCULAR MEDICINE

Implantable Cardioverter Defibrillator Shocks continued

To deliver appropriate and effectivetherapy, an ICD must sense cardiacsignals, detect tachyarrhythmias, anddiscriminate ventricular tachycardia

(VT) from supraventricular tachy-cardia (SVT). Sensing is the processby which an ICD determines the timing of electrogram signalscaused by cardiac depolarization.Detection is the algorithm by whichan ICD processes a series of electro-gram signals to classify the cardiacrhythm and determine if treatmentshould be delivered. The most basic detection algorithms measure and count RR intervals to detect any tachyarrhythmia that fulfills ventricular rate and duration criteria.Present detection algorithms include

programmable SVT-VT discrimina-tors (enhancements) that processeither single-chamber ventricularinformation or both atrial and ven-

tricular information to discriminateVT from SVT. Antitachycardia pac-ing or shocks must be withheld con-tinuously for SVT and confoundingnonarrhythmic signals but deliveredwithout delay for life-threatening VTor ventricular fibrillation (VF).

Basics of ICD TroubleshootingIdentifying the ICD. Diagnosingthe cause of shocks involves deter-mining the ICD’s programming,evaluating stored data, and performing specific troubleshooting.These activities require identification

of the ICD generator and electrodesfrom the patient’s identificationcard, medical alert bracelet, recordsfrom the implanting hospital orphysician, or chest radiograph. ICDgenerators from different manufac-turers have unique radiographicmarkers. Because all ICDs are regis-tered at implantation, they can beidentified by a phone call to the man-ufacturer’s 24-hour toll-free number(Table 1). Each manufacturer providestechnical assistance 24 hours per day.

ICD interrogation. Basic inter-rogation using the ICD’s program-mer can be performed without special training (Table 1). After theICD is interrogated, the initial settings and all stored data must bedocumented by printing them orsaving them electronically. Theseinclude bradycardia pacing parame-ters; rate, duration, and SVT-VT discrimination criteria used todetect VT and VF; programmedtherapies for VT and VF; measure-ments of lead integrity; and datafrom each device-detected tach-yarrhythmia episode.

Interpretation of stored ICD-detected tachyarrhythmiaepisodes. Stored electrograms of VTand VF episodes provide the mostimportant data for interpreting thecauses and results of ICD shocks.Physicians can interpret stored elec-trograms using skills learned fromreading surface ECGs. Single-cham-ber ICDs store one or more ventricu-lar electrograms; dual-chamber ICDsstore atrial and one or more ventricu-lar electrograms. ICDs use a closelyspaced, near-field (“rate”) dipole forsensing. On transvenous leads, thisdipole’s signal is recorded betweenthe right ventricular tip electrode andeither a sensing ring electrode or theright ventricular coil. A widelyspaced, far-field (“shock”) dipolemay be preferred for discrimination of

Guidant

800 227 3422

1. Turn on power switch on left side of 2930 programmer

1. Turn on power switch on left side of 9790 programmer

2. "Autoidentify" ICD

3. Interrogate and print or "Save to disk"

4. Select "Parameters" -> "Detection" from menu at right

1. Turn on power switch on left side of 3500/3510 programmer

2. Press Interrogate button in center above keyboard

3. Select "Tachy Parameters" from Menu

4. Program to "DEBFI OFF"

5. Program "VT Detection" and "VF Detection" OFF

Medtronic

800 325 2518

St. Jude

800 722 3423

CPI 106 GDT 104

2. Select " Quickstart" button at bottom of screen to identify device and begin interrogation

3. Select "Tachy Mode" button at top of screen and choose "OFF" from the pull-down menu. Device is programmed automatically.

ManufacturerRadiographic Logo Emergency Activation Procedure

Most cardiologists are not electrophysiologists, but they are increasinglyinvolved in the follow-up of ICD patients.

Table: 1. Radiographic logos and emergency deactivation procedures for major U.S. ICD manufactures.


Implantable Cardioverter Defibrillator Shocks

VT from SVT by analysis of electro-gram morphology. Marker channelsindicate ICD classification of atrialand ventricular events and device-measured intervals (P-P and R-R).

Tachyarrhythmia detection andclassification by ICDs have a signifi-cant error rate. Episodes stored asICD-detected SVT or VT/VF may notrepresent a true arrhythmia. If anepisode represents a true arrhythmia,the ICD may not accurately discrim-inate SVT from VT or VF. Thus, trueSVT episodes may be stored in theICD’s VT or VF log; instances of trueVTs may be stored in the ICD’s SVTlog. The physician must review allstored tachyarrhythmia episodes foraccuracy of classification.

Diagnosing the Cause of ShocksFigure 1 summarizes the approach to patients who present with ICD shocks. Rarely, a patient may

complain of a shock but no shock isrecorded in the ICD episode log. ICDsrecord delivered shocks so reliablythat the absence of a recorded shockindicates that the patient’s symptomis not caused by ICD shock(s). Thephysician must evaluate the cause of “phantom shocks.” Patients maydistinguish between “little” and “big”shocks. Patients almost never reportICD shocks as “little.” Little shocksmay be caused by pacing-inducedpectoral, diaphragmatic, or inter-costal muscle stimulation.

Shocks in the absence ofarrhythmias (oversensing). Asindicated in Figure 1, the first step indiagnosing the cause of shocks is todetermine if a tachyarrhythmiaoccurred. Because ICDs utilize feed-back mechanisms (automatic gaincontrol or auto-adjusting sensitivity)to ensure reliable sensing of low- andvariable-amplitude ventricular elec-trograms during VF1,5 they may senseelectrical signals they were not

Analyze Stored& Clinical Data

TachyarrhythmiaNo Tachyarrhythmia

(Oversensing)

IntracardiacSignals

ExtracardiacSignals

VT/VF(Appropriate

Detection)

SVT(Inappropriate

Detection)

RepetitiveVT/VF("VT Storm")

First Shock Success

Single VT/VFICD-ClassifiedShock Failure

Shock Failure to Terminate VT/VF

Successful ShockMisclassified by ICD

Frequent orRepetitiveShocks

Figure 1. Approach to the patient with shocks: Top panel: flow diagram for one or infrequent shocks. Bottompanel: diagram for multiple or repetitive shocks.

Intracardiac signals

Physiologic

T wave (low-amplitude R wave, prolonged QT)

P wave (consider ventricular lead dislodgment)

Double-counted R waves

Nonphysiologic

Separate electronic pacemaker

Mechanical contact of sensing lead and abandoned lead fragment

Extracardiac signals

Physiologic (myopotentials)

Lead insulation failure

Diaphragmatic myopotentials (bradycardia, positional)

Nonphysiologic

Signals originating in ICD system

“Chatter” in active fixation lead

Lead fracture

Loose set screw or adapter

Electromagnetic interference

Surgical electrocautery

Electronic article surveillance equipment

Magnetic resonance imaging

Lithotripsy

Table 2. Major Causes of inappropriate Shocks Due to Oversensing



intended to. Shocks occur in theabsence of tachyarrhythmias becausenonarrhythmic physiologic or non-physiologic signals are oversensed bythe ICD and detected as arrhythmias.Common causes of oversensing thatresults in shocks are summarized inTable 2. Clinical history is of limitedvalue in diagnosing oversensingexcept for specific identifiable causesof external electromagnetic interfer-ence.6 The presence of antecedentarrhythmic symptoms—palpitations,syncope, or near syncope—suggests atrue arrhythmia, but most shocks,appropriate or inappropriate, are notpreceded by symptoms.7 A pattern of consistent postural changes ormuscle activity preceding shockssuggests electrode-related oversens-ing. Audible tones from the ICD maybe an alarm that indicates a loss ofelectrical integrity in a lead.

Physiologic intracardiac signals.Patients with low-amplitude R wavesmay require higher sensitivity settings

to ensure reliable sensing of VF. Thismay result in T-wave oversensing,which can cause inappropriate inhibition of bradycardia pacing, antitachycardia pacing at the wrongrate, or inappropriate detection ofVT1 (Figure 2). P-wave oversensingcan occur if the ventricular sensingdipole is near the tricuspid valve.Except in the case of lead dislodg-ment, this is rare. R-wave doublesensing occurs if the duration of the sensing electrogram exceeds the ventricular refractory period of 120-140 ms. It may be exacerbatedby sodium-channel-blocking antiar-rhythmic drugs, particularly at highheart rates as use-dependent sodium-channel blockade increases.

In some cases, ICDs may oversenseintracardiac electronic signals fromindependent, implanted pacemakers.Pacemaker-ICD interactions havebeen described comprehensively.8

Oversensing of intracardiac elec-trograms can be recognized by char-

acteristic alternation of intervals andelectrogram morphology (P-R vs R-P,T-Q vs Q-T, and short R-R’ vs long R’-R separated by isoelectric base-lines. In contrast, the hallmark of oversensing of nonphysiologicsignals or extracardiac physiologicsignals (myopotentials) is thereplacement of the isoelectric base-line with high-frequency noise(Figure 3). Oversensing of physiologiccardiac signals may be detectedinappropriately as VT or VF, but non-physiologic signals are often over-sensed as R-R intervals close to theICD’s refractory period (120-140 ms)and are almost always detected as VF.

Myopotential oversensing. Sensinglead insulation failures in the pocketare a common cause of inappropriateoversensing of pectoral or abdominalmyopotential signals. A low pacinglead impedance (<200 Ω) indicatesthe presence of a parallel conductorpathway and supports the diagnosisof insulation failure. The risk of leadfailure depends on the lead type andthe pulse generator location. Failureis common in epicardial leads9 andtransvenous leads with abdominallyimplanted generators. With pectorallyimplanted generators, failure is morecommon in coaxial leads than inmultilumen leads.10 Oversensing ofdiaphragmatic myopotentials mayoccur in ICD systems with integratedbipolar electrodes in the right ven-tricular apex.11 It may be bradycardiadependent or positional and maycause both inappropriate inhibitionof bradycardia pacing and inappro-priate detection of VT/VF.

Extracardiac nonphysiologic signals.A lead fracture, loose set screw, oradapter malfunction may result inoversensing of nonphysiologic sig-nals. A high pacing lead impedance(>2000 Ω) indicates complete or partial interruption of the electricalcircuit and supports this diagnosis.

Figure 2. T-wave oversensing. T-wave (T) oversensing in the presence of a low-amplitude (3 mV) R wave (R).Atip-Aring, atrial sensing electrogram (EGM); Vtip-Vring, ventricular sensing EGM; AR, atrial refractory sense; TS,VT sense; TD, VT detection; Atip-Aring, atrial electrogram; Vtip-Vring, ventricular electrogram.



Except for medical electrocautery,external sources of electromagneticinterference are rare causes of over-sensing. They have been summarizedin recent reviews.6,12

Diagnosis of suspected oversensingmay be confirmed by real-timetelemetry of electrograms and markerchannels. Postural changes and deep breathing may be necessary toconfirm oversensing of diaphrag-

matic myopotentials. Posturalchanges, isometric exercises, andpocket manipulation may facilitatediagnosis of lead or connector-related oversensing.

ICD algorithms for discrimi-nation of SVT from VT. If stored electrograms indicate that a shockwas delivered in response to a truetachyarrhythmia, the second step indiagnosis is to determine if the

rhythm stored in the VT or VFepisode log is VT or SVT.

Clinical history is of limited value.A history of rapidly conducted atrial

Figure 3. Noise sensing. Top panel: High-frequency,high-amplitude signals caused by diaphragmatic over-sensing during sinus rhythm distort the baseline of theventricular sensing dipole (Vtip-Vcoil) on an integratedbipolar lead. Marker channels at the bottom showatrial (upper) and ventricular (lower) intervals. AS, atrialsense; TP, antitachycardia pacing; FS, VF sense; VS,ventricular sense in sinus zone; FD, VF detection; CE,charge ends. Values above and below marker channelindicate A-A intervals and V-V intervals, respectively;asterisk denotes charge end. Oversensed ventricularintervals in the VF zone are labeled FS. Note the shortintervals, close to the ICD’s refractory period of 120 ms.Oversensing of these signals resulted in repetitive inap-propriate shocks for device-detected VF. Atip-Aring, atrialEGM. Middle panel: Intermittent, high-frequency, low-amplitude signals caused by lead insulation failure of acoaxial integrated bipolar lead distort the baseline andare sensed in the VF zone during sinus rhythm. Lowerpanel: Ventricular (V-V) interval plot (ordinate) prior toand after inappropriate detection of VF (0 sec on abscis-sa). The horizontal lines at 320 and 370 ms denote VFand VT detection intervals, respectively. Sinus intervalsof 800-900 ms (67-75 bpm) are interrupted by inter-mittent clusters of short intervals. The minimum valueof these short intervals is the ICD’s refractory period of120 ms. Arrows indicate delivery of 5 repetitive inap-propriate shocks, with delivered energy values in joules(J). RV Coil-Can, far-field ventricular EGM. The coil ispart of the sensing circuit for this lead.

V-V Interval (ms)

• V-V VF = 320 ms VT = 370 ms

-50 -40 -30 -20 -10 0 10 -20 30 40 50 60 * * * *

180015001200

900600

400

200

29.3 J29.6 J29.3 J29.2 J

27.2 J

Time (sec)

•••••••••••••••••••••••••••••••••••••••••• •••••••••• •••••••••••••••

••••••••••••••••••••• • • ••••

•

•

•

•••••• •

••

•••

••

• ••••

••••• •

•••

•• •• •

••

• •••

••

•

•••• •• •• •••• ••••••••

••••

••••

•••••••••• ••••••••••••

••

••••••••

•

••••

••• • •

•

• •

•

•

••••

•• •

• ••

••••• • •

• •• • ••

••

•••••

••

•• •••••

••

•••••

• ••• •••

••

••

•

• ••••••• •• ••

•••

••••

••••••• ••• •••••• •

••••

••••

••••• •••••••• •••

••••••••••

••

••• ••

•••••••••••••• ••••



fibrillation suggests inappropriatetherapy for atrial fibrillation. A shockduring vigorous exercise suggestsinappropriate therapy for sinustachycardia. Inappropriate shocks

for SVT do not occur in pacemaker-dependent patients.

Three single-chamber SVT-VT dis-criminators have been implementedin ICDs13-19: stability, which discrimi-

nates monomorphic VT from atrialfibrillation based on regularity of theR-R interval; onset, which discrimi-nates VT from sinus tachycardiabased on suddenness of onset,enabling therapy to be withheldfrom tachycardias in which the rateincreases gradually; and morphology,which discriminates VT from anySVT based on morphologic differ-ences between electrograms in sinusrhythm and tachycardia. Dual-cham-ber algorithms may use atrial andventricular rates, chamber of origin,P:R pattern, and atrioventricularassociation (P-R interval stability) inaddition to the single-chamber criteria.

Interpreting the specific actionsof multiple, complex detectionenhancements may be daunting.Fortunately, this is largely unneces-sary for acute evaluation of patientswho present with shocks. Thus, thenonelectrophysiologist should focuson applying established principles ofsurface ECG analysis to intracardiacelectrograms of ICD-detected VT orVF. The diagrams in Figure 4 showmethods for analyzing single-cham-ber and dual-chamber electrograms,as discussed below.

Physician analysis of single-chamber data for SVT-VT discrimination. Morphology. Indiscriminating SVT from VT usingthe surface ECG, the first step is to

Analyze Atrialand Ventricular

Rates

A >

ConductedAFib/AFlu

VT+AFib/AFlu

SVF (1:1 Aconduction)

VT (1:1 VAconduction)

A = V > A

VT• Ventricular Morpholoyg• Ventricular Interval Stability• AV Associaion

• Ventricular Morpholoyg• AV Associaion• Chamber of Onset

Uniform andIdentical to Sinus

Morphology

SVT

Variable or MinimalDifference from

Sinus Morphology

Uniform andDistincly different

from SinusMorphology

VT• Abrupt Onsest -> VT• Irregulary Irregular -> iFib• Termination by one ATP -> VT

AnalyzeElectrogramMorphology

Figure 4. SVT-VT discrimination by analysis of stored EGMs. Top panel: method for single-chamber storedEGMs. Bottom panel: method for dual-chamber stored EGMs.

Figure 5. Rate-related bundle branch block in atrial tachycardia. Dual-chamber EGMs show abrupt onset of regular tachycardia in the atrium with 1:1 atrioventricular conduction. Theventricular EGM morphology in SVT is distinct from the morphology during sinus rhythm. If only a single-chamber ventricular EGM were available, the rhythm would be misdiagnosed asVT based on different ventricular EGM morphology, abrupt onset, and regular rhythm. Atip–Aring = atrial EGM; RV Tip-Coil, right ventricular integrated bipolar sensing EGM.



determine if the QRS morphology isventricular or supraventricular.Analogously, analysis of ventricularelectrogram morphology is the cor-nerstone of SVT-VT discrimination

based on visual inspection of storedepisodes in single-chamber ICDs.Whenever possible, electrogrammorphology should be analyzedfrom the signals of far-field dipoles.Visual inspection does not permitunequivocal discrimination of VTfrom sinus morphologies from near-field electrograms in 5% to 10% ofVTs.20 A real-time reference sinuselectrogram should be recorded withthe patient in the same posture inwhich the episode occurred.

Tachyarrhythmias are classified asSVT if the electrogram morphologyis uniform and identical to the sinusmorphology. Tachyarrhythmias are

classified as VT if the electrogrammorphology is uniform and distinctlydifferent from the sinus morphology.This approach necessarily classifiesrate-related bundle branch block

during SVT as VT (Figure 5). Onset. If electrogram morphology

is variable or differs only minimallyfrom sinus morphology, ancillarycriteria must be applied: abruptnessof onset, R-R interval stability, andresponse to therapy. Sinus tachycardiahas a gradual onset and is usuallydetected at the sinus-VT rate bound-ary (Figure 6). In contrast, the onsetof VT is usually abrupt. Exceptionsinclude VT that originates duringsinus tachycardia or other SVTs andVT that starts abruptly but at an initialrate below the programmed VT detec-tion rate. In the latter case, the ICDwill store the “onset” of the arrhyth-

mia as the gradual acceleration acrossthe sinus-VT rate boundary.

Stability. An irregularly irregularrhythm is characteristic of atrial fib-rillation. Subtle beat-to-beat variationin electrogram morphology, theintracardiac correlate of rate-relatedaberrancy, is common in rapidly con-ducted atrial fibrillation. Atrial fibrilla-tion is usually detected inappropriatelyduring an ongoing atrial fibrillationepisode when the ventricular rateexceeds the programmed rate criterion.Thus, stored intervals are irregularlyirregular prior to and during detec-tion. However, interval stability has 2important limitations as a criterion:(1) Because the ventricular rate in atrialfibrillation is more regular at fasterventricular rates, stability cannot reli-ably discriminate atrial fibrillationfrom VT at rates above 170/min. (2)Amiodarone or type IC antiarrhyth-mic drugs may cause monomorphicVT to become markedly irregular orpolymorphic VT to slow, causingunderdetection of VT.21,22

All ICDs are registered at implantation; each manufacturer providestechnical assistance 24 hours per day.

Figure 6. Inappropriate detection of sinus tachycardia. The top left tracings show surface ECG lead 2 and a reference far-field EGM (RV Coil-Can) during sinus rhythm. The topright tracing shows the stored EGM from the treated tachyarrhythmia. The EGMs in these 2 panels are identical. The lower panels are “flashback interval” plots of the R-R intervalsprior to detection of VF, which occurs at the right side of each panel. The interval number prior to detection is plotted on the abscissa, and the corresponding interval is plotted onthe ordinate. The lower left panel shows 2000 R-R intervals prior to detection. Tachycardia is present throughout. Shortly after the 400th interval, the rhythm accelerates graduallyuntil the interval decreases to below the programmed VF detection interval of 340 ms. The lower right panel shows this gradual acceleration on an expanded scale.

1000

600

<200

(ms)

-2000 -1600 -1200 -800 -400 0 -100 -80 -60 -40 -20 0

4 00

300

<200

(ms)

Interaval # Interaval #

Interaval Plot

Sinus Rhythm

Surface ECG

RV Coil - Can

Marker

Tachycardia

RV Coil - Can

Marker

VF Rx 1

Response to therapy. Terminationof a tachyarrhythmia by a singletrial of ventricular antitachycardiapacing favors a diagnosis of VT. This is not helpful in the analysis ofshocks, but termination of multiple other episodes with similar mor-phology and rate by a single trial ofanti-tachycardia pacing favors theconclusion that shocks are due toVT. The vast majority of VTs are

terminated by 1 or 2 shocks. Thus,failure of multiple shocks to termi-nate a regular tachycardia suggestsSVT, particularly sinus tachycardia.

Physician analysis of dual-chamber data for SVT-VT dis-crimination. Discrimination ofSVT from VT by analysis of the surface ECG is often hampered byuncertainty regarding the atrialrhythm diagnosis. Dual-chamber

electrograms permit accurate diag-nosis of the atrial rhythm if the atrial lead is functioning. Analysesof atrial and ventricular rates andatrioventricular relationships are thecornerstones of dual-chamber SVT-VTdiscrimination. If the ventricular rateis faster than the atrial rate in an ICDpatient, VT should be diagnosed.

1:1 A/V relationship. If the atrialand ventricular rates are equal,


Interval (ms)

• V-V AA VF = 300 ms VT = 400 ms

0 1 2 3 4 5 6 7 8

1800

1500

1200

900

600

400

200

Time (sec)

•• • • • • •• • • •• • • • •• • • • • • • •

Figure 8. Atrial flutter with 2:1 atrioventricular conduction. Left panel: Atrial (Atip-Aring) and ventricular (Vtip-Vring) EGMs. The ventricular EGMs have the same morphology as sinusEGMs (not shown). Right panel: Stable 2:1 ratio of atrial and ventricular EGMs with an atrial cycle length of 190 ms (rate 316 bpm) and a ventricular cycle length of 380 ms (rate 158 bpm).


Figure 7. Dual-chamber EGMs show onset of VT with minimal change in ventricular rate during sinus tachycardia. VT is diagnosed by onset in ventricle, change in morphologyof the ventricular EGM, and atrioventricular dissociation. The interval plot (bottom left) shows that atrial and ventricular rates are nearly equal in VT. The interval plot cannotidentify isorhythmic atrioventricular dissociation. The expanded insert of the marker channel (bottom right) emphasizes atrioventricular dissociation. In both bottom panels, onsetof VT (arrows) is identified by a change in the morphology of the ventricular EGM (Vtip-Vring). Atip-Aring, atrial EGM.


the physician must discriminatebetween VT with 1:1 ventriculoatrialconduction and SVT with 1:1 atrioventricular conduction. Sinus morphology strongly favors a diag-nosis of SVT. Sinus tachycardia has a

characteristic gradual atrial accelera-tion, usually with a stable PR interval.Analysis of the chamber of accelera-tion (where acceleration originates)may allow discrimination of atrialtachycardia from VT: Atrial tachy-

cardia begins with a short P-P inter-val followed by a short R-R interval,whereas VT usually begins with ashort R-R interval and a few beats ofatrioventricular dissociation until 1:1ventriculoatrial conduction stabilizes(Figure 7). Atrial electrogram mor-phology in sinus rhythm differsfrom the morphology of retrogradeatrial electrograms in VT and fromthat of ectopic atrial electrogramsduring atrial tachycardia; however,these differences may be subtle, and their absence should not beconsidered as confirmatory of sinusP waves.

VT during SVT. If the atrial rateexceeds the ventricular rate, thephysician must distinguish betweenconducted atrial fibrillation or atrialflutter (Figure 8) and VT during atrial

Figure 9. Far-field R waves in VT with 1:1 ventriculoatrial conduction. Dual-chamber EGMs (atrial bipole and ventricular sensing) show the onset of VT with 1:1 ventricu-loatrial conduction. Note the far-field R wave in sinus rhythm on the atrial channel (left asterisk). Tachycardia begins abruptly in the ventricle. The far-field R waves in VT(right asterisk) might be confused with atrial EGMs, resulting in misdiagnosis of atrial flutter with 2:1 conduction. Note that the ventricular EGM morphology varies slightly.The difference in ventricular EGM morphology between sinus tachycardia and VT is subtle and limited to the initial negative deflection. Given the variability of the ventricu-lar EGM, sinus tachycardia and VT morphologies cannot be distinguished with certainty. The morphologies of antegrade and retrograde atrial EGMs are also similar.


Figure 10. VT during atrial fibrillation (AF). The first conducted ventricular (Vtip-Vring) EGM is followed by aregular tachycardia with different morphology (VT). A single burst of antitachycardia pacing terminates VT. Thelast ventricular EGM has conducted supraventricular morphology. The interval plot displays the abrupt onset ofregular tachycardia during atrial fibrillation, with abrupt termination by antitachycardia pacing.

Interval (ms)

-16 -14 -12 -10 -8 -6 -4 4 5 6-2-2 0 -22

1800

1500

1200

900

600

400

200

Time (sec)

•••

•• • •

•• • •

••

••

• • • • • • • • • • • • • • • • ••• •• •• •

• ••

•• •

••

•••

• •• •

• V-V AA VF = 320 ms VT = 430 ms

Burst


arrhythmia. Rapid atrial electro-grams during atrial fibrillation andatrial flutter are easily identified bythe physician but may be underde-tected by the ICD due to atrialblanking periods.5 However, thephysician must not misinterpret far-field R waves on the atrial channelas atrial electrograms (Figure 9).Such misinterpretation is a possiblecause of incorrect classification ofVT as SVT by dual-chamber ICDs.5

In stored electrograms, abnormalventricular morphology and regularventricular rate are the best diagnosticcriteria for VT during atrial fibrilla-tion (Figure 10). Conducted atrialflutter may be diagnosed in the

presence of abnormal ventricularmorphology if consistent 2:1 atri-oventricular association or Mobitz 1atrioventricular block is present.

Even when all criteria for SVT-VTdiscrimination have been applied,experts are unable to diagnose somearrhythmias with certainty. It is betterto consider the diagnosis uncertainthan to make important therapeuticdecisions based on unreliable data.

Clinical Approach to thePatient with ShocksSingle or infrequent shocks.Single shocks or a few shocks sepa-rated by days do not constitute anemergency. The ICD should beinterrogated within 24 to 48 hr, and the logic in Figure 1 should be applied.

If oversensing is caused by externalelectromagnetic interference, thetreatment is to remove the offendingsource. Oversensing of physiologicsignals may require reprogramming,lead revision, or insertion of a newsensing lead.

Inappropriate therapy of SVT maybe corrected by reprogramming ratezones or SVT-VT discriminators.

Table 3: Drug Effects in Patient with ICD Shocks

Table 4: Causes of Repetitive Sustained VT (“Storm”)

Frequency of VT/VFProarrhythmic26,27

Antiarrhythmic drugsDrugs with proarrhythmic side effectsDrugs that interact with proarrhythmic drugs

Antiarrhythmic

Detection of VT/VFProgrammed Rate Boundaries

VT rate (decrease–IC, amiodarone)SVT ventricular rate

Decrease (beta blockers, amiodarone, sotalol)Increase (1:1 conduction of atrial flutter on IC drugs)

ICD SVT-VT discrimination algorithmsVT interval stability (more irregular on IC, amiodarone)Electrogram morphology

Therapy for VT/VFDefibrillation energy requirement

Increase (IA, IB, IC, amiodaroneDecrease (III except amiodarone)

Antitachycardia pacingEfficacy: variableUse-dependent increase in pacing threshold (IC)

Heart failure exacerbation(disopyramide, beta blockers, sotalol, IC, calcium antagonists)

Acute ischemia

Exacerbation of heart failure

Metabolic abnormalitiesHypokalemiaHypomagnesemiaHyperthyroldism (amiodarone-induced)

DrugProarrhythmiaChange in antiarrhythmic drug

Change in prescribed drugChange in preparation of prescribed drugNoncompliance


Alternatively, SVT may be treated with drugs or ablation.Reprogramming is usually preferablefor sinus tachycardia, whereas patho-logic atrial arrhythmias may requireboth approaches. The physicianshould consider the effect of repro-gramming and drugs on detection ofVT and of drugs on efficacy of VTtherapy. Comprehensive reviews ofdrug-ICD interactions have beenpublished.12,23,24 Relevant points aresummarized in Table 3.

Single or infrequent appropriateshocks for VT or VF do not require achange in therapy. Fast monomor-phic VT with rates up to 240/minmay be terminated by painless anti-tachycardia pacing without increasingthe risk of syncope.25 Thus, oneattempt at antitachycardia pacingshould be programmed to improvepatient acceptance of ICD therapy.If antitachycardia pacing is unreliable,an antiarrhythmic drug may berequired to decrease the frequencyof VT or permit successful antitachy-cardia pacing. Drug-ICD interactionsshould be considered (Table 3).12,23,24

Sotalol has been shown to reduce thefrequency of VT requiring shocks inICD patients.26

Frequent or repetitive shocks.More than 3 shocks in 1 day or repet-itive shocks constitute an electrophys-iologic emergency (Figure 1). Thesemay cause release of troponin I27 orsevere psychological complications.28

However, the logical approach is similar to that for infrequent shocks.

The diagnosis of repetitive shockscaused by oversensing is clear whenthe patient receives shocks during electrocardiographically documentedsinus rhythm. Emergency treatmentconsists of suspending or disablingVT/VF detection by placing a magnetover the ICD. Tachyarrhythmiadetection and/or therapy should thenbe turned off with a programmer. Thespecific steps are summarized inTable 1. The specific cause of over-sensing can be determined by analysisof real-time electrograms with ther-apies programmed off. Once the ICDis deactivated, the patient will notreceive appropriate ICD therapy ifVT or VF occurs; external shock willbe required.

Repetitive shocks caused by inap-propriate detection of SVT can alsobe addressed on an emergent basisby suspending detection with amagnet. The VT detection rate

can then be programmed to beabove the SVT rate. The definitiveapproach is the same as that for single inappropriate shocks for SVT.

Repetitive shocks for VT may becaused by recurring episodes of VT after successful shock termina-tion of VT (“VT storm”/”cluster”shocks) or by multiple unsuccessfulshocks for a single episode.Therapeuticapproaches differ.

VT storm. Known causes are sum-marized in Table 4.29,30 Frequently,the cause is unknown. Diagnosis ofacute coronary syndromes during VTstorm is problematic, since multiple(appropriate or inappropriate) shockscan cause repolarization changes aswell as elevations in troponin I up to15 ng/mL.27 Therapy may includereversing the precipitating cause,antiarrhythmic drugs (usually amio-darone, sotalol, and/or a beta blocker),and catheter ablation.

Rarely, ICDs deliver unnecessaryrepetitive shocks for recurring self-terminating episodes of VT. Shocksmay be prevented in such cases byprogramming the ICD to a noncom-mitted mode or increasing the num-ber of beats required for detection ofVT or VF.

Main Points• Shocks are a common complaint in patients with implantable cardioverter defibrillators (ICDs).

• ICDs deliver shocks when they detect ventricular tachycardia (VT) or ventricular fibrillation (VF).

• ICDs may deliver inappropriate shocks if supraventricular tachycardia (SVT) is misclassified as VT or nonarrhythmicphysiologic or nonphysiologic signals are oversensed and detected as VT/VF.

• The clinical history is of limited value in distinguishing appropriate from inappropriate shocks.

• The ICD data log of stored arrhythmia episodes must be reviewed to determine if a shock was delivered appropriatelyfor VT/VF or inappropriately.

• ICDs distinguish VT from SVT by use of specific SVT-VT discrimination algorithms (discriminators).

• Single-chamber algorithms are based on measures of stability, onset, and morphology. Dual-chamber algorithms utilize measures of atrial and ventricular rates or atrioventricular relationship. These discriminators are imperfect.

• Specific treatment may not be required for single shocks. Indicated treatment may include reprogramming the ICDor making changes in antiarrhythmic drugs.

• Frequent or repetitive shocks constitute an electrophysiologic emergency.




Failed shocks. Because defibrilla-tion is a statistical process, occasionalshock failures occur randomly, butfailure of ≥2 maximum-outputshocks should not occur. Shocksfrom chronically implanted ICDsystems may fail to terminate VT or VF because of patient-related or ICD system-related reasons.Pneumothorax and pleural effusiondecrease transmyocardial current,the former by adding intratho-racic resistance and the latter by providing a low-resistance parallel current path.

Misclassification of shock effica-cy. ICD shocks may terminate VT,but the ICD may not detect the ter-mination, resulting in incorrectclassification of the shocks as unsuc-cessful. This happens if VT recursbefore the ICD can detect post-shock sinus rhythm because ofimmediate post-shock recurrence ofnonsustained or sustained VT. Thelatter should be treated as repetitivesustained VT. Misclassification ofboth may be corrected by decreas-ing the number of beats for redetec-tion of sinus rhythm; misclassifica-tion of the former may be correctedby increasing the number of beatsfor redetection of VT.

ICDs may also misclassify shocksas ineffective if the post-shockrhythm is SVT in the VT rate zone(catecholamine-induced sinus tachy-cardia or shock-induced atrial fibril-lation). This should be treated as inappropriate detection of SVT. It may be corrected by applying SVT-VT discrimination algorithms toredetection of VT,increasing the VTdetection rate to above the rate ofpost-shock SVT, or increasing shockstrength to prevent shock-induced

atrial fibrillation. Pharmacologicapproaches include beta blockers to slow post-shock SVTs.

References1. Reiter MJ, Mann DE. Sensing and tach-

yarrhythmia detection problems inimplantable cardioverter defibrillators. J Cardiovasc Electrophysiol. 1996;7:542-558.

2. Lloyd M, Hayes D, Friedman P.Troubleshooting. In: Hayes D, Lloyd M,Friedman P, eds. Cardiac Pacing andDefibrillation: A Clinical Approach. Armonk,NY: Futura; 2000:347-452.

3. Rosenthal M, Alderfer J, Marchlinski F.Troubleshooting suspected ICD malfunction.In: Kroll MW, Lehmann MH, eds. ImplantableCardioverter Defibrillator Therapy. Norwell,MA: Kluwer Academic Publishers; 1996:435-476.

4. Wellens HJ, Lau CP, Luderitz B, et al.Atrioverter: an implantable device for thetreatment of atrial fibrillation. Circulation.1998;98:1651-1656.

5. Wood M, Swerdlow CD, Olson W. Sensingand arrhythmia detection by implantabledevices. In: Ellenbogen K, Kay G, Willkoff B,eds. Clinical Cardiac Pacing and Defibrillation.2nd ed. Philadelphia, PA: WB SaundersCompany; 2000:68-126.

6. Hayes D. Electromagnetic interference andimplantable devices. In: Hayes D, Lloyd M,Friedman P, eds. Cardiac Pacing andDefibrillation: A Clinical Approach. Armonk,NY: Futura; 1999:519-539.

7. Grimm W, Flores BF, Marchlinski FE.Symptoms and electrocardiographically docu-mented rhythm preceding spontaneousshocks in patients with implantable cardioverter-defibrillator. Am J Cardiol.1993;71:1415-1418.

8. Glikson M, Trusty JM, Grice SK, et al. A step-wise testing protocol for modern implantablecardioverter-defibrillator systems to preventpacemaker-implantable cardioverter-defibril-lator interactions. Am J Cardiol. 1999;83:360-366.

9. Brady PA, Friedman PA, Trusty JM, et al. Highfailure rate for an epicardial implantable car-dioverter-defibrillator lead: implications forlong-term follow-up of patients with animplantable cardioverter-defibrillator. J AmColl Cardiol. 1998;31:616-622.

10. Friedman PA. Transvenous defibrillator leadmalfunction: Predictors and mode of detec-tion. Am J Cardiol. 2001, in press.

11. Deshmukh P, Anderson K. Myopotential sensing by a dual chamber implantable cardioverter defibrillator: two case reports. J Cardiovasc Electrophysiol. 1998;9:767-772.

12. Goldschlager N, Epstein A, Friedman PA, et al.Environmental and drug effects on patientswith pacemakers and implantable cardiovert-er defibrillators: A practj72cal guide to patientmanagement. Arch Intern Med. 2001, in press.

13. Swerdlow CD, Chen PS, Kass RM, et al.Discrimination of ventricular tachycardiafrom sinus tachycardia and atrial fibrillationin a tiered-therapy cardioverter-defibrillator. JAm Coll Cardiol. 1994;23:1342-1355.

14. Bgada J, Mont L, Figueiredo M, et al.Enhanced detection criteria in implantabledefibrillators. J Cardiovasc Electrophysiol.1998;9:261-268.

15. Schaumann A, von zur Mühlen F, Gonska BD,Kreuzer H. Enhanced detection criteria inimplantable cardioverter-defibrillators toavoid inappropriate therapy. Am J Cardiol.1996;78:42-50.

16. Neuzner J, Pitschner HF, Schlepper M.Programmable VT detection enhancements inimplantable cardioverter defibrillator therapy.Pacing Clin Electrophysiol. 1995;18:539-547.

17. Swerdlow CD. Optimal programming of sens-ing and detection in single-chamber ventricu-lar ICDs. Cardiovasc Electrophysiol Rev.1998;2:300-304.

18. Duru F, Bauersfeld U, Rahn-Schonbeck M,Candinas R. Morphology discriminator fea-ture for enhanced ventricular tachycardia discrimination in implantable cardioverterdefibrillators. Pacing Clin Electrophysiol.2000;23:1365-1374.

19. Weber M, Bocker D, Bansch D, et al. Efficacyand safety of the initial use of stability andonset criteria in implantable cardioverterdefibrillators. J Cardiovasc Electrophysiol.1999;10:145-153.

20. Marchlinski FE, Callans DJ, Gottlieb CD, et al.Benefits and lessons learned from stored elec-trogram information in implantable defibril-lators. J Cardiovasc Electrophysiol. 1995;6:832-851.

21. Swerdlow C, Hwang C, Ahern T, et al.Incidence and significance of underdetectionof ventricular tachycardia by algorithms toenhance specificity in an advanced device[abstract]. Circulation. 1993;88:1-156.

22. Garcia-Alberola A, Yli-Mayry S, Block M, et al.RR interval variability in irregular monomor-phic ventricular tachycardia and atrial fibrilla-tion. Circulation. 1996;93:295-300.

23. Brode SE, Schwartzman D, Callans DJ, et al.ICD-antiarrhythmic drug and ICD-pacemakerinteractions. J Cardiovasc Electrophysiol.1997;8:830-842.

24. Carnes CA, Mehdirad AA, Nelson SD. Drugand defibrillator interactions. Pharmacotherapy.1998;18:516-525.

25. Wathen M, Sweeney M, Degroot P, et al.Prospective evaluation of efficacy of antitachy-cardia pacing for spontaneous rapid monomor-phic ventricular tachycardia [abstract].Circulation. 2000;102(suppl II):II-397.

26. Pacifico A, Hohnloser SH, Williams JH, et al.Prevention of implantable-defibrillator shocksby treatment with sotalol. d,l-SotalolImplantable Cardioverter-Defibrillator StudyGroup. N Engl J Med. 1999;340:1855-1862.

27. Hasdemir C, Shah N, Rao AP, et al. Effect ofout-of-hospital implantable-cardioverter-defibrillator shocks on cardiac troponin I levels [abstract]. Circulation. 2000;102(supplII):II-622.

28. Bourke JP, Turkington D, Thomas G, et al.Florid psychopathology in patients receivingshocks from implanted cardioverter-defibrilla-tors. Heart. 1997;78:581-583.

29. Credner SC, Klingenheben T, Mauss O, et al.Electrical storm in patients with transvenousimplantable cardioverter-defibrillators: inci-dence, management and prognostic implica-tions. J Am Coll Cardiol. 1998;32:1909-1915.

30. Miller JM, Hsia HH. Management of thepatient with frequent discharges fromimplantable cardioverter defibrillator devices.J Cardiovasc Electrophysiol. 1996;7:278-285.

31. Friedman PL, Stevenson WG. Proarrhythmia.Am J Cardiol. 1998;82:50N-58N.

32. Roden DM. Mechanisms and management ofproarrhythmia. Am J Cardiol. 1998;82:49I-57I.


implantable cardioverter defibrillator shocks: a...

Documents