Postgrad Med J 88:719-728 doi:10.1136/postgradmedj-2012-301861rep
  • Republished education in heart

Republished: The burden of inappropriate shocks in young people and how to avoid them

  1. Jagmeet P Singh
  1. Cardiac Arrhythmia Service, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
  1. Correspondence to Dr Jagmeet P Singh, Cardiac Arrhythmia Service, Massachusetts General Hospital Heart Center, Gray Building 109, 55 Fruit Street, Boston, MA 02114, USA; jsingh{at}

Implicated in up to half of all cardiovascular mortality among developed nations worldwide,w1 w2 sudden cardiac death (SCD) is associated with approximately 450 000 deaths annually in the USA and >700 000 deaths in Europe.w3 w4 In 23% of out-of-hospital and 17% of inhospital cardiac arrests, the initial observed rhythm was ventricular tachycardia (VT) or ventricular fibrillation (VF), both potentially treatable with defibrillation.w5 w6 Large scale trials have also demonstrated that implantable cardioverter defibrillators (ICDs) can improve survival through terminating malignant ventricular arrhythmias, particularly among SCD survivorsw7 and heart failure patients with systolic dysfunction.w8–w12 Based on these data, there has been a dramatic surge in the utilisation of ICDs over the past two decades, with 116 000 ICD device implantations in the USA in 2009 alone, of which 53 000 (46%) were implanted in patients younger than the age of 65 years.w6

With increasing use of ICDs, there has also been heightened awareness of device-related complications and associated risk–benefit trade-offs. Specifically, although it is clear that appropriate ICD therapies extend survival, inappropriate ICD shocks are common and occur in a similar magnitude of patients.w12 w13 Frequent ICD shocks are associated with diminished quality of life,w14 w15 and significant anxiety, depression and post-traumatic stress in a minority of patients.w16 w17 In addition, inappropriate ICD shocks have been linked to the possibility of proarryhthmiaw18–w20 and are associated with increased mortality in some patients, although any causal relationship remains controversial.1–3 w21–w23 Conversely, interventions to prevent inappropriate ICD shocks reduce morbidity,4 and also have been shown to reduce heart failure hospitalisation.5

As the use of ICDs continues to expand further into primary prevention and younger patient populations, identifying at-risk patients and devising a treatment plan for patients suffering from inappropriate shocks are critical. Strategies to avoid inappropriate shocks include alteration of device programming, optimisation of medical therapy, and also selective application of advanced therapies such as catheter ablation, in order to mitigate harms and maximise benefit of ICDs over the long-run. This article examines the data regarding the overall burden of inappropriate shocks and outlines a management algorithm to address and minimise unnecessary ICD shocks, with an emphasis on device programming for the primary prevention patient.

Burden of inappropriate shocks across patients

Appropriate and inappropriate ICD therapies are typically dichotomised based on their inciting rhythm. Conventionally, ICD therapies which are delivered to treat VT or VF are considered appropriate, while inappropriate therapies are those which are delivered in response to sinus tachycardia (ST), supraventricular tachyarrhythmias (SVTs; most commonly atrial fibrillation [AF]), arrhythmia misidentification (eg, due to T-wave or noise oversensing) or device malfunction (eg, lead fracture). While this definition preserves simplicity, some appropriate shocks delivered for VT may in fact be avoidable through the use of antitachycardia pacing (ATP) or reprogramming detection zones to permit non-sustained VT (NSVT) that might spontaneously terminate. As such, the overall, holistic burden of ‘unnecessary’ (inclusive of inappropriate and avoidable) shocks remains uncertain, although some reports suggest that up to half of all shocks may fall into these categories.6 w24 w25

However, even using the more conservative conventional definition, the burden of inappropriate shocks is substantial. The prevalence of inappropriate shocks reported in recent large scale secondaryw13 and primary prevention1 ,2 w12 trials varied between 12% and 24% of all patients receiving ICDs (figure 1). Furthermore, inappropriate therapies constituted between 19% and 31% of all ICD therapies delivered, often with multiple inappropriate shocks delivered to individual patients.w13 1 A recent population based registry of ICD patients implanted with modern ICD devices demonstrated a similar cumulative incidence of inappropriate ICD shocks in 18% of patients at 5-year follow-up.3 In another large, unselected population of 15 991 patients assembled using remote monitoring data, the burden of inappropriate shocks appeared even higher: 40% of all first-shocks were delivered inappropriately (ie, for AF, atrial flutter, SVTs, noise or artefact).6 Where reported, the age profile of patients across these trials has been similar, with a median age usually between 60 and 65 years.

Figure 1

(A) Prevalence of inappropriate shocks in large trials. (B) Patient characteristics and device programming parameters (nested table). ATP, antitachycardia pacing; bpm, beats/min; CHF, congestive heart failure; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NSVT, non-sustained ventricular tachycardia; NYHA, New York Heart Association; SVT, supraventricular tachyarrhythmia; VF, ventricular fibrillation; VT, ventricular tachycardia.

Importantly, none of the large scale primary prevention trials was explicitly designed to study older patients (more than 50% of patients were younger than 60 years), and a recent analysis suggests that the survival benefit is greater in patients who are younger.w26 Unfortunately, younger patients (defined variably as <70 years and <60 years) are also at a higher risk for inappropriate shocks.3 w27 7 This is consistent with data from even younger paediatric populations (most <30 years of age), as rates of inappropriate shocks range from 21% to as high as 47% reported in recent registries and case series of children and young adults with ICDs.w28–w36 (One investigatorw37 did report a much lower rate of 6% inappropriate shocks in a cohort of 33 patients; however, all other studies reported rates >20%.)

While some of the increased risk of inappropriate therapy in the very young may be due to increased activity levels, the underlying aetiology of cardiovascular illness is also quite different in paediatric patients with a higher proportion of congenital heart disease, inherited cardiomyopathies and channelopathies. Regardless of age, the adverse effect of inappropriate shocks extends beyond pain alone, and emphasis is shifting from assessment of psychiatric morbidity to also include awareness of potential impact on longer term mortality.

Inappropriate shocks and impact on mortality

While an independent, increased risk of mortality associated with inpatients receiving appropriate shocks for ventricular arrhythmias has been recognised previously,w38 the results of two randomised clinical trials1 ,2 and one registry3 have also reported on an increased risk of mortality with inappropriate shocks (table 1). In the MADIT II and SCD-HeFT trials, patients receiving a combination of appropriate and inappropriate shocks were found to be at the highest risk for death. Furthermore, patients receiving appropriate shocks only were at a higher risk of death than those receiving inappropriate shocks only. Finally, no increased mortality was found in those receiving no ICD shocks relative to controls.1 ,2 This stratification was not felt to be due to device therapy, but rather driven by underlying substrate for arrhythmia, that is, patients who manifest VT or VF (and are therefore eligible for appropriate shock) have a more diseased substrate and are at a higher risk for death than those who never require appropriate shock therapy to avert SCD.w39

Table 1

Inappropriate shocks and mortality

Challenging this understanding, however, are recent combined data from four ATP strategy trials enrolling 2135 patients in which patients demonstrating fast ventricular tachycardia (FVT; ≥182–200 beats/min (bpm)) terminated by ATP alone showed survival rates similar to those patients who never demonstrated malignant arrhythmia.8 Similar findings were also found in a longitudinal prospective cohort study of heart failure patients.9 Furthermore, in subset analysis of the MADIT II trial, patients demonstrating appropriate ATP therapy without shocks were also not at an increased risk of all-cause mortality.1 Taken together, these observations raise troubling questions with respect to why inappropriate shocks may be associated with an increased risk of mortality in some trials, presumably given that patients with inappropriate shocks alone do not demonstrate significant ventricular arrhythmic burden.

Device malfunction leading to fatal proarrhythmia has been observed in isolated cases, and may explain some of the increased risk.w40–w42 In patients with bradycardia, bradycardia pacing-induced short-long-short sequences may trigger ventricular arrhythmia, although this would lead to an ‘appropriate’ shock by the standard definition.w43 Based on a review of a primary prevention ICD population, however, this mechanism appears to occur only rarely.w44

Alternatively, AF—a common trigger for inappropriate shocks—itself is a marker of increased mortality in CHF patients and those with ICDs, and likely signals sicker patients who will, not unsurprisingly, be at a higher risk for inappropriate shocks.10 w45 Indeed, in data presented from the LATITUDE remote monitoring system, inappropriate shocks due to AF or flutter—but not those due to ST, noise, or artefact oversensing—were associated with increased mortality risk.11 Interestingly, however, in MADIT II patients who received inappropriate ATP therapy without shocks demonstrated a survival rate similar to those patients who never received therapy even when patients receiving inappropriate ATP therapy with ICD shocks were at an increased risk of death. This remained true despite adjustment for incidence of AF. Sample sizes for these groups were not reported, however, and the possibility remains that intrinsic patient characteristics and other unmeasured confounders beyond AF alone may drive differences in survival between the two groups.

Regardless of aetiology, the underlying pathophysiology driving the increased mortality of inappropriate shocks is also uncertain. Some observers have suggested that an increased rate of depression in patients receiving frequent ICD shocks might indirectly lead to increased mortality.w46 Others argue that ICD discharges may be the proximate cause of harm to patients through direct cardiac injury. Indeed, biomarker elevation, including creatinine kinase and cardiac troponin, has been well described during ICD device implantation and defibrillation threshold testing.w47–w50 In addition, long term endocardial fibrosis of the lead tip has been observed on gross and microscopic examination on pathology.w51 The link between these laboratory and pathological findings and clinical significance, however, remains uncertain.

It is known that high-energy defibrillation shocks cause post-resuscitation myocardial dysfunction,w52 and post-shock electromechanical dissociation has been implicated in rare cases of sudden death after appropriate ICD therapy.w53 Even the relatively low energies used during device testing are followed by significant short term reductions in cardiac output and blood pressure in patients with moderate or severely reduced left ventricular ejection fraction.w54 There has been a suggestion that end-organ tissue perfusion in vulnerable patients may be threatened by testing, and some investigators have found that cardiac troponin elevation after testing might predict mortality.w55 With that said, recent larger non-randomised cohort data did not show that defibrillation testing was associated with an adverse impact on survival.12 Clearly, a causal relationship between ICD shocks and mortality remains controversial, and mechanistic hypotheses remain speculative.

Irrespective of mechanism, unnecessary ICD shocks are clearly acutely painful, and there is growing evidence of more insidious long term harm. As such, there is unanimity of opinion that employment of strategies to minimise shocks is beneficial and may also lead to salutary benefits on survival. We will outline a management approach beginning first with a discussion of the aetiology of inappropriate shocks and then direct focus to modifications of device programming which help reduce inappropriate shock burden.

Causes of unnecessary shocks

Inappropriate shocks and appropriate but avoidable shocks together constitute a total burden of unnecessary shocks. The distinction is not simply semantic, as it pertains to the underlying aetiology. Fundamentally, inappropriate shocks are due to errors of identification (ie, ST, SVT, noise or artefact is inappropriately identified as VT/VF), whereas appropriate but avoidable shocks are due to errors of therapy (ie, correctly identified VT/VF is prematurely treated with defibrillation). Inappropriate shocks can be further broken down based on failure of SVT-VT discrimination, abnormal sensing or device malfunction. Avoidable shocks can be further disaggregated based on those which are intrinsically non-sustained and therefore require no treatment versus those which are sustained but treatable with an ATP strategy which avoids ICD shocks. We present a general framework for considering the overall aetiology of unnecessary shocks and common causes in figure 2.

Figure 2

Common causes of unnecessary implantable cardioverter defibrillator (ICD) shocks. ATP, antitachycardia pacing; SVT, supraventricular tachyarrhythmia; VF, ventricular fibrillation; VT, ventricular tachycardia.

Based on data from large studies and registries, the single most common cause of inappropriate shocks is AF with rapid ventricular response—accounting for between a third and a half of all inappropriate shocks in randomised trials—followed by failure of discrimination of non-AF SVT and ST as other leading causes.w13 2 ,3 Abnormal sensing due to overcounting of T waves, R wave double-counting, electromagnetic interference, noise or diaphragmatic myopotentials are comparatively less common, although may be more frequent in biventricular devices.3 w56

Importantly, inappropriate shocks due to lead malfunction or senescence are also a significant problem. During long term follow-up, the annual rate of lead failure can reach as high as 20% in 10-year-old leads.13 Patient age also plays a role. Lead dislodgement and failure is still a predominant aetiology of inappropriate shocks in paediatric populations in which patient activity levels are high.w33

Specific leads have also been identified as being associated with a higher risk of failure and inappropriate shock. Previously, the Medtronic 6936 and 6966 were associated with post-shock prolonged oversensing and lead failure, particularly 5 years after implantation.w57 In contrast, the pace-sense conductor fracture risk of the Medtronic Sprint Fidelis defibrillation leads (models 6930, 6931, 6948, 6949) was found to occur early and deserves special mention. Medtronic's voluntary suspension of sales of the Fidelis lead led to significant public scrutiny and also led to increased awareness of the risk of inappropriate shocks.w58–w60 Subsequently, downloadable software and an audible lead-integrity alert system have helped mitigate this risk somewhat, although inappropriate discharges in the absence of ventricular arrhythmia detection continue to occur.w61–w63

More recently, the St Jude Medical Riata family of silicone leads has been associated with ‘inside-out’ ethylene tetrafluoroethylene insulation failure and a higher risk of inappropriate shocks.w64 Up to 10% of all Riata leads may be affected,w65 and the US Food and Drug Administration has initiated a class 1 recall of leads.w66 Lead extraction has also been recommended for younger patients by some authors, although a watchful stance may be reasonable with respect to asymptomatic defects in patients without a history of inappropriate shocks.w67

In normally functioning devices and leads, the overall burden of appropriate but avoidable shocks for VT and VF also remains uncertain. Evidence from large trials suggests that rates of appropriate shocks for VT/VF are often substantially greater (up to twofold) than the frequency of SCD in control arms.w19 w20 w68 The implication is that up to half of all VT/VF events might spontaneously terminate and therefore do not require appropriate ICD discharge to avert SCD. This assertion, however, does not take into account the haemodynamic consequence of uninterrupted ventricular arrhythmia such as reduced cardiac output and ensuing dizziness or syncope, even if VT or VF is non-sustained. Clearly, rate and duration of VT or VF determine clinical impact and therefore also whether device therapy is mandated.

Based on data from the PainFREE Rx II trial, 34% of FVT terminated during capacitor charging (median time of 3.3 s) and did not require ICD shock.14 There was no difference with respect to syncope between patients. This is consistent with data from the PREPARE study group, in which VT/VF detection interval was extended to 30 of 40 number of intervals to detect (NID). Here, 290 of 753 episodes (39%) of tachycardia lasting 12–29 beats were due to NSVT.4 The overall rate of syncope attributed to device programming and arrhythmia was 1.6% of patients. Similar findings were also reported for biventricular ICD devices also programmed with fixed, long detection intervals (30/40 NID vs 12/16 NID in controls) in the RELEVANT trial. There, 66% of VF, 91% of FVT and 92% of VT episodes self-terminated within 31 beats.5 The incidence of syncope was 1.2% among patients which was not significantly different from the control group (1.9%). This complements prior data from MIRACLE-ICD in which 48% of all VF episodes terminated during capacitor charging (typically 8–12 s).w69 Together, these data suggest that a substantial burden of appropriate VT/VF episodes may be managed safely without requiring ICD therapy (ie, shocks or ATP), if devices are programmed with an extended time to detect duration.

The number of VT/VF episodes which will not spontaneously terminate but will require the use of ATP or shock for treatment is less certain, as this depends on device programming and cycle length (CL) of the VT present. ATP is delivered immediately after VT detection (with no time required for charging), and the reported success rates have been high; in PainFREE Rx II the ATP success rate for FVT was 72%. As has been observed by other commentators,w70 however, this was likely an overestimation of the efficacy of ATP since at least some of these arrhythmias would have terminated spontaneously (18/24 NID was used). In the PREPARE trial, 32 of 52 FVTs (median rate 207 bpm) were terminated with one sequence of ATP, leading to an ATP efficacy of 62% (unadjusted) or 49% when adjusted (using the generalised estimating equation method).w70 w71 The proportion of these arrhythmias which would spontaneously terminate with a longer wait interval is not known. Given that the PREPARE trial implemented an extended detection period (30/40 NID), however, it is reasonable to suspect that untreated FVTs of longer duration would be associated with adverse clinical effect (ie, syncope) if not treated by either ATP or shock.

In any event, when inappropriate and appropriate but avoidable ICD shocks are taken together, there appears to be a significant opportunity for unnecessary shock reduction. Computer modelling based data on using the more conservative EMPIRIC trial15 (which utilised an 18/24 NID) suggest that a combination of shock reduction strategies including NID extension, use of programmable SVT discriminators and ATP could reduce overall shocked episodes by 59%.w72 A simplified approach to device programming to reduce unnecessary shock burden is outlined below.

Device programming

The occurrence of any ICD shock mandates a diagnostic enquiry into cause in order to define treatment and prevent future potentially unnecessary therapies.w73 Given the expansion of ICD use into younger, more diverse patient prevention populations, we will focus on accessible programming tips to avoid unnecessary shocks for primary prevention patients. A generalised approach is presented below.

Generalised approach

There are four commonly manipulated programming domains which should be evaluated in patients undergoing ICD therapy (figure 3):

  • Rate detection cut-offs

  • Detection duration

  • Use of SVT discrimination enhancements

  • ATP.

Figure 3

Programming domains and recommendations. ATP, antitachycardia pacing bpm, beats/min; FVT, fast ventricular tachycardia; NID, number of intervals to detect; SVT, supraventricular tachyarrhythmia; VF, ventricular fibrillation; VT, ventricular tachycardia.

Each domain will be reviewed briefly with respect to practical, ‘hands-on’ recommendations for the prospective ICD patient.

Rate detection cut-offs

Data from the MIRACLE-ICD population suggested that patients in primary preventive populations demonstrated significantly faster CLs during VT/VF than patients receiving devices for secondary prevention (CL 303±54 ms vs 366±71 ms).w69 Similarly, VF only rarely occurred at rates <250 bpm (CL ∼240 ms) in primary prevention patients enrolled in the INTRINSIC RV trial.16 Given these data, an initial detection zone targeting very rapid VT/VF may be reasonable for primary prevention patients with no history of AF or other SVT. Indeed, in the SCD-HeFT study, a rate cut-off of 187 bpm was applied to a single VF zone.w8 The PREPARE trial, which evaluated multiple parameters of strategic programming in primary prevention patients, utilised a rate cut-off of 182 bpm. Both trials demonstrated low rates of inappropriate shocks at 1 year, and neither demonstrated significant rates of near syncope or syncope associated with higher rates.

Recent large, unselected patient data from the Medtronic Discovery Link database7 confirm the efficacy of these programming parameters, finding that patients with a detection threshold of ≥188 bpm have the lowest number of shocks. In addition, the recent results of the RIGHT trial even suggest an increased risk of inappropriate ICD shocks with a rate cut-off of <175 bpm.17 Taken together, an initial rate cut-off threshold of >188 bpm (CL ∼320 ms) appears to be safe and preferable for primary prevention patients.

Detection duration

The NID establishes the number of beats (or time) which must meet the rate cut-off threshold for the duration criterion to be satisfied. In the first PainFREE Rx trial and the recent PROVE trial, an NID of 12 was applied.18 w24 Both trials reported impressive ICD shock reduction through the use of ATP. As mentioned above, however, this likely did not reflect success of ATP by itself, as even a slightly longer detection duration in PainFREE Rx II (NID 18/24) was associated with spontaneous termination of FVT in over a third of cases. As such, the NID provides a means to avoid appropriate ICD therapies for NSVT as well as reduce inappropriate therapy for paroxysmal AF.

In the PREPARE and RELEVANT studies, the NID was safely extended to 30 of 40 beats with further inappropriate ICD shock reduction and with no increased incidence of syncope.4 ,5 In a large population of 106 513 patients enrolled in the Medtronic Discovery Link remote monitoring database, a longer programmed NID of 24/32 or 30/40 beats were both associated with a reduced risk of shocks.7 Based on these data it appears reasonable to implement longer NID (of up to 30/40) in patients without a history of known haemodynamically unstable VT.

Use of SVT discrimination enhancements

Three commonly used enhancements are used to discriminate SVT from VT in single-chamber ICD devices: sudden onset, interval stability and electrogram (EGM) morphology comparison. Historically, the sudden onset and stability criteria were developed initially in first-generation tiered-therapy ICDs.w74 w75 The sudden onset criterion relies on measuring the prematurity of an initial beat of tachycardia with respect to prior beats. This is used as a means to distinguish ST—which demonstrates ‘warm up’—from VT, which exhibits suddenness during initiation. The interval stability discriminator assesses variation in CL in order to identify AF. Although demonstrating impressive reductions in inappropriate detections of ST and AF, there was concern early on regarding the risk of underdetecting true VTw76 and the deterioration of reliability of these discriminators at higher heart rates (>190 bpm).w77

As a consequence, ‘safety’ timeouts were designed to trigger ICD shock in the event of any detection zone tachycardia persisting greater than a default time limit.w78 Examples include the Biotronik ‘Forced Termination Period’ (at 1 min), Guidant/Boston Scientific ‘Sustained Rate Duration’ (at 3 min), and the St Jude ‘SVT Discrimination Timeout’ (at 30 s); these were nominally set to ‘on’ in older devices. Particularly given the longevity of some SVTs, the risk of inappropriate shocks due to these timeouts may be substantial and many propose that they be set to ‘off’ should modern SVT discriminators be applied.w25 w79

During the past decade, proprietary EGM morphology analysis algorithms (eg, Guidant Rhythm ID, Medtronic Wavelet, St Jude morphology discrimination) have been developed to distinguish SVT from VT by comparing EGM during tachycardia with a template or reference EGM obtained at baseline (ideally in sinus rhythm). These algorithms demonstrate better sensitivity and specificity than the sudden onset or interval stability criteria alone, and have demonstrated excellent (>90%) sensitivity and specificity when used together to discriminate SVT from VT.w80–w83

An expanded suite of discrimination enhancements is available for dual-chamber devices that take advantage of comparison of atrial (A) and ventricular (V) EGM timing and relationships (eg, V=A, V>A, A<V). Examples include the ELA/Sorin PARAD algorithm, Medtronic Enhanced PR Logic and the St Jude Medical A-V Rate Branch. Initially, there had been enthusiasm suggesting that dual chamber devices were more effective than single chamber devices at reducing inappropriate shocks.19 Recent meta-analysis,w84 real-world registry data3 and a large dataset of remote monitoring data,7 however, suggest that while the total number of inappropriate shocks might be reduced with an atrial lead, a similar magnitude of patients are affected. Given the known increased risk of periprocedural complication and greater inhospital mortality with dual-chamber devices,w85 justification to use these devices to reduce inappropriate shocks alone (in the absence of another indication for dual chamber system) appears unwarranted.

The use of SVT discriminators in primary prevention patients has been limited, although they were safely applied in primary preventive patients in the PREPARE study, PROVE trial and in unselected patients enrolled in the Medtronic CareLink programme.4 ,7 w24 We feel that the use of SVT discriminators appears quite reasonable for VT/VF detection zones up to the threshold of 200 bpm based on the PREPARE study. Even if not activated, enabling discriminators in ‘passive’ mode would allow for data collection to better define patient-specific rate cut-off values based on acquired data.w86

Antitachycardia pacing

The track record of ATP as a means of reducing ICD shocks has been prospectively evaluated in multiple studies, including PainFREE Rx,18 PainFREE Rx II,14 EMPIRIC,15 PREPARE4 and the PROVE trial.w24 ATP only treated patients enjoy fewer appropriate and inappropriate ICD shocks with a low incidence of VT acceleration and syncope, and no increased risk of mortality, even with spontaneous, ‘fast’ VTs (CL 300–330 ms).4 ,8 ,14 ,15 ,18 In addition, among patients with Medtronic devices enrolled in remote monitoring, ATP was shown to perform well in a real-world population, with 38% reduction in shocks relative to those without ATP programmed.

With respect to the type of ATP used, there has been considerable variation in types of therapy implemented. Burst pacing at 88% of VT appears more effective than ramp ATP, with a lower risk of VT acceleration.20 Recent data also indicate that biventricular ATP may be implemented safely in patients undergoing cardiac resynchronisation therapy (CRT), with a trend towards greater benefit in CRT patients with a history of ischaemic cardiomyopathy.w87 Interestingly, any ventricular ATP may also be effective in terminating or slowing 1:1 SVTs through retrograde concealed conduction.w70 w88 There are a number of additional large ongoing studies regarding various ATP protocols currently underway which will hopefully elucidate ‘ideal’ parameters for ATP implementation.

Based on current data, we favour ATP for VT between 188 and 220 bpm in primary prevention patients. An important caveat, however, is that because shock confirmation/reconfirmation is less specific with VT redetection after ATP, up to 5% of successful ATPs are followed by unnecessary shock. Certainly, ATP parameters can and should be further revised after shocks based on a review of stored EGMs.w25 w70 w89

Role for medical and advanced therapies

Medical therapy to address inappropriate shock burden returns to the fundamental aetiology of unnecessary shocks. Vulnerability to unnecessary shocks increases during settings of increased SVTs (particularly AF with rapid ventricular response) and NSVT. Advanced heart failure status (New York Heart Association class III–IV) is associated with an increased risk of appropriatew90 and inappropriatew91 ICD therapies. Certainly, volume optimisation and optimal heart failure management reduce ICD shocks, and may also reduce inappropriate shock risk, although it has not been explored prospectively.

Antiarrhythmic therapy with amiodarone has been successful in reducing inappropriate shock burden and may be an effective additive therapeutic option for selected patients. Amiodarone was found to be efficacious when used in conjunction with β-blockers (as in the OPTIC trialw92) or on its own.w93 Toxicity, however, has limited widespread adoption. As an alternative, sotalol has also shown benefit in reducing inappropriate shocks, but was less effective than amiodarone.w92 Other medical options to reduce inappropriate shocks remain limited. Interestingly, a recent observational study indicated that the pleomorphic effects of statin therapy were also associated with reduced inappropriate ICD shocks, although a pathophysiological basis remains undefined, and confirmation in larger trials is yet to be performed.w94

With respect to advanced interventional therapies, the use of VT ablation has been evaluated as a means to reduce appropriate ICD shocks previously,w95 21 and may also reduce the risk of unnecessary shocks from NSVT, particularly in secondary prevention patients with a history of myocardial infarction. The use of radiofrequency ablation of non-AF SVT has also recently been shown to reduce inappropriate shock burden for some patients, and may be an effective ‘second-line’ approach to patients suffering inappropriate shocks from SVTs beyond medical titration and device programming.w96 The role of other procedures, including pulmonary vein isolation or atrioventricular node ablation, as a means to reduce overall ICD therapy delivery in patients with refractory, symptomatic AF has not been explored.

Future directions and integrated approach

There are multiple ongoing and planned clinical trials to evaluate the efficacy of various programming strategies to reduce inappropriate shocks. Some notable trials include: MADIT-RIT (NCT00947310), ADVANCE III (NCT00617175), ASAP (NCT01169246) and PainFree SST (NCT00982397). No single strategy has emerged yet as ‘optimal’ in reducing inappropriate shock burden. We favour an integrated, multidisciplinary strategy employing collaboration between electrophysiologists and heart failure specialists (figure 4).

Figure 4

Integrated management to reduce unnecessary implantable cardioverter defibrillator (ICD) shocks. AF, atrial fibrillation; ATP, antitachycardia pacing; AVNA, atrioventricular nodal ablation; CHF, congestive heart failure; EP, electrophysiology; HF, heart failure; NSVT, non-sustained ventricular tachycardia; PVI, pulmonary vein isolation; SVT, supraventricular tachyarrhythmia; VF, ventricular fibrillation; VT, ventricular tachycardia.

Management begins with stratifying patients according to risk, particularly primary prevention patients with a history of AF or other SVTs. Among at-risk patients, ‘out-of-box’ programming should be proactively adjusted to reduce risk of inappropriate shocks by electrophysiologists. Regular follow-up and medical optimisation of heart failure should be coordinated with general cardiologists. More frequent surveillance of device data with respect to SVT or NSVT burden by electrophysiologists can prompt joint discussion regarding medication titration or the use of antiarrhythmic therapy. Finally, any ICD shock should trigger examination of therapy (ie, appropriate, appropriate but avoidable, or inappropriate), root-cause analysis, expedited device reprogramming or revision, and a collaborative discussion regarding potential advanced treatment options, including catheter ablation.


SCD is a leading cause of cardiovascular mortality, and ICDs have driven significant improvements in survival among selected patients. Tempering their benefit, however, is the high incidence of unnecessary ICD shocks, particularly among patients below 60 years of age. Simple modifications to device programming can address causes and reduce the risk of unnecessary shocks substantially. These include using high rate cut-offs to avoid misidentification of SVT, extending detection duration periods to allow for spontaneous termination of NSVT, implementing SVT discriminators to improve arrhythmia classification and, finally, aggressively utilising ATP to pace-terminate susceptible VT or VF episodes. Individualising out-of-box ICD settings through employment of these features appears reasonable and safe in a majority of primary prevention patients, and ongoing trials will provide further data regarding optimal algorithms. When combined with multidisciplinary medical management and targeted use of advanced therapies such as catheter ablation, today's unnecessary shocks may be diminished to ‘unheard of’ in tomorrow's ICD patients.

The burden of inappropriate shocks in young people and how to avoid them: key points

Defining inappropriate shocks more broadly

  • Conventionally, implantable cardioverter defibrillator therapies which are delivered to treat ventricular tachycardia (VT) or ventricular fibrillation (VF) are considered appropriate, while therapies delivered in response to sinus tachycardia, supraventricular tachyarrhythmias (SVTs), or errors in sensing or lead malfunction are considered inappropriate.

  • Some appropriate shocks delivered for VT, however, may in fact be avoidable through the use of antitachycardia pacing (ATP) or reprogramming detection zones to permit non-sustained VT that might spontaneously terminate.

  • As such, the overall burden of unnecessary shocks includes both strictly inappropriate and avoidable appropriate therapies.

Impact of inappropriate shocks on mortality

  • There is a clear, independent, increased mortality risk associated with appropriate shocks for ventricular arrhythmias and mortality.

  • Recent data suggest that inappropriate shocks may also be associated with an increased risk of death, although a mechanistic understanding of this relationship remains speculative.

  • Importantly, ATP therapy is not associated with an increased risk of death, whether delivered appropriately or inappropriately.

Causes of inappropriate shocks

  • The single most common cause of inappropriate shocks is misidentification of atrial fibrillation with rapid ventricular response.

  • Failure of discrimination of non-atrial fibrillation SVTs and sinus tachycardia are other leading causes.

  • Abnormal sensing due to overcounting of T waves, R wave double-counting, electromagnetic interference, noise or diaphragmatic myopotentials is comparatively less common in modern devices.

  • Lead malfunction, however, remains significant. The Medtronic Sprint Fidelis leads and the St Jude Riata leads are of particular concern.

Device programming to reduce unnecessary shocks

  • Increase rate detection cut-offs to ≥188 bpm, and consider even higher rate cut-offs for younger (>200 bpm) or more active patients.

  • Implement longer detection duration before delivering implantable cardioverter defibrillator therapy (at least 24 of 32 number of intervals to detect or ∼12 s).

  • Programme SVT discriminators ‘on’ for all VT/VF zones up to 200 bpm and consider passive use up to 220 bpm. Consider setting forced time-out periods ‘off’ while using SVT discriminators.

  • Actively employ ATP strategies.

You can get CPD/CME credits for Education in Heart

Education in Heart articles are accredited by both the UK Royal College of Physicians (London) and the European Board for Accreditation in Cardiology—you need to answer the accompanying multiple choice questions (MCQs). To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to:

▶ RCP credits: Log your activity in your CPD diary online (—pass mark is 80%.

▶ EBAC credits: Print out and retain the BMJ Learning certificate once you have completed the MCQs—pass mark is 60%. EBAC/ EACCME Credits can now be converted to AMA PRA Category 1 CME Credits and are recognised by all National Accreditation Authorities in Europe (

Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into Heart with their journal's username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.


  • Competing interests In compliance with EBAC/EACCME guidelines, all authors participating in Education in Heart have disclosed potential conflicts of interest that might cause a bias in the article. Dr Upadhyay received salary support from the 2011 Max Schaldach Fellowship in Cardiac Pacing and Electrophysiology of the Heart Rhythm Society. Dr Mela reports speaker's honoraria from Boston Scientific Corp, Medtronic Inc, and St Jude Medical. Dr Singh reports receiving research grants from St Jude Medical, Medtronic Inc, Boston Scientific Corp and Biotronik. He also serves on advisory boards or as a consultant for Boston Scientific Corp, St Jude Medical, Sorin Group, CardioInsight Inc, Respicardia Inc and Medtronic Inc. Finally, he has received speaker's honoraria from Medtronic Inc, Biotronik, Boston Scientific Corp, St Jude Medical and the Sorin Group.

  • Provenance and peer review Commissioned; internally peer reviewed.


  1. Subgroup analysis of the MADIT II trial with focus on inappropriate shocks and increased mortality risk.
  2. Subgroup analysis of the SCD-HeFT trial with focus on inappropriate shocks and increased mortality risk.
  3. Large ‘real-world’ registry of ICD demonstrating risk of inappropriate shocks and increased mortality.
  4. Report of the results of the PREPARE study, evaluating modern rate cutoff and ATP parameters relative to historical controls.
  5. Results of the RELEVANT study, which compared long VT/VF detection and one fixed ATP burst for FVT relative to standard therapy in CRT-D devices.
  6. Large real-world study of patients enrolled in the Medtronic Discovery Link database evaluating impact of atrial fibrillation on shocks.
  7. Pooled analysis of four major ATP trials with attention to impact of shocks versus ATP on mortality.
  8. Results from PainFree Rx II, one of the first randomised trials comparing standardised empirical ATP versus shocks in treatment of spontaneous FVT.
  9. Results of the EMPIRIC study, comparing standardised, empiric programming to reduce shocks compared with physician tailored therapy as standard of care.
  10. First major SVT discriminator comparison trial, eavluating Guidant Rhythm ID performance relative to Medtronic Enhanced PR Logic or Wavelet algorithms.
  11. Results from PainFree Rx, one of the first major ATP trials, evaluating safety and effacy of ATP for FVT.

Free sample
This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of PMJ.
View free sample issue >>

Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.