[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Figure 1.
Cumulative Incidence of a Second Implantable Cardioverter-Defibrillator (ICD) Shock After Initial ICD Discharge
Cumulative Incidence of a Second Implantable Cardioverter-Defibrillator (ICD) Shock After Initial ICD Discharge

Among 5357 patients experiencing a second ICD shock, the cumulative incidence (95% CI) at relevant time points is reported.

Figure 2.
Cumulative Incidence of a Second Implantable Cardioverter-Defibrillator (ICD) Shock Stratified by Presumed Primary- and Secondary-Prevention Indications
Cumulative Incidence of a Second Implantable Cardioverter-Defibrillator (ICD) Shock Stratified by Presumed Primary- and Secondary-Prevention Indications

Patients with presumed secondary-prevention indications for implantation (ventricular tachycardia [VT] and ventricular fibrillation [VF]) had a higher incidence of receiving a second shock compared with ICD recipients with a presumed primary-prevention indication. Those with unknown implantation indications had an intermediate incidence.

Table 1.  
Baseline Characteristics
Baseline Characteristics
Table 2.  
Multivariate Regression Model for Time to Second ICD Shock After an Initial ICD Dischargea
Multivariate Regression Model for Time to Second ICD Shock After an Initial ICD Dischargea
Table 3.  
Estimated RH While Driving Among ICD Recipientsa
Estimated RH While Driving Among ICD Recipientsa
1.
Kramer  DB, Kennedy  KF, Noseworthy  PA,  et al.  Characteristics and outcomes of patients receiving new and replacement implantable cardioverter-defibrillators: results from the NCDR.  Circ Cardiovasc Qual Outcomes. 2013;6(4):488-497.PubMedGoogle ScholarCrossref
2.
Epstein  AE, Baessler  CA, Curtis  AB,  et al; American Heart Association; Heart Rhythm Society.  Addendum to “Personal and public safety issues related to arrhythmias that may affect consciousness: implications for regulation and physician recommendations: a medical/scientific statement from the American Heart Association and the North American Society of Pacing and Electrophysiology”: public safety issues in patients with implantable defibrillators: a scientific statement from the American Heart Association and the Heart Rhythm Society.  Circulation. 2007;115(9):1170-1176.PubMedGoogle ScholarCrossref
3.
Mylotte  D, Sheahan  RG, Nolan  PG,  et al.  The implantable defibrillator and return to operation of vehicles study.  Europace. 2013;15(2):212-218.PubMedGoogle ScholarCrossref
4.
Strickberger  SA, Cantillon  CO, Friedman  PL.  When should patients with lethal ventricular arrhythmia resume driving? an analysis of state regulations and physician practices.  Ann Intern Med. 1991;115(7):560-563.PubMedGoogle ScholarCrossref
5.
Vijgen  J, Botto  G, Camm  J,  et al; Task force members.  Consensus statement of the European Heart Rhythm Association: updated recommendations for driving by patients with implantable cardioverter defibrillators.  Europace. 2009;11(8):1097-1107.PubMedGoogle ScholarCrossref
6.
Thijssen  J, Borleffs  CJ, van Rees  JB,  et al.  Driving restrictions after implantable cardioverter defibrillator implantation: an evidence-based approach.  Eur Heart J. 2011;32(21):2678-2687.PubMedGoogle ScholarCrossref
7.
 Assessment of the cardiac patient for fitness to drive: 1996 update.  Can J Cardiol. 1996;12(11):1164-1170, 1175-1182.PubMedGoogle Scholar
8.
 Assessment of the cardiac patient for fitness to drive.  Can J Cardiol. 1992;8(4):406-419.PubMedGoogle Scholar
9.
Bänsch  D, Brunn  J, Castrucci  M,  et al.  Syncope in patients with an implantable cardioverter-defibrillator: incidence, prediction and implications for driving restrictions.  J Am Coll Cardiol. 1998;31(3):608-615.PubMedGoogle ScholarCrossref
10.
Freedberg  NA, Hill  JN, Fogel  RI, Prystowsky  EN; CARE Group.  Recurrence of symptomatic ventricular arrhythmias in patients with implantable cardioverter defibrillator after the first device therapy: implications for antiarrhythmic therapy and driving restrictions.  J Am Coll Cardiol. 2001;37(7):1910-1915.PubMedGoogle ScholarCrossref
11.
Grimm  W, Flores  BF, Marchlinski  FE.  Symptoms and electrocardiographically documented rhythm preceding spontaneous shocks in patients with implantable cardioverter-defibrillator.  Am J Cardiol. 1993;71(16):1415-1418.PubMedGoogle ScholarCrossref
12.
Kou  WH, Calkins  H, Lewis  RR,  et al.  Incidence of loss of consciousness during automatic implantable cardioverter-defibrillator shocks.  Ann Intern Med. 1991;115(12):942-945.PubMedGoogle ScholarCrossref
13.
Ruwald  MH, Okumura  K, Kimura  T,  et al.  Syncope in high-risk cardiomyopathy patients with implantable defibrillators: frequency, risk factors, mechanisms, and association with mortality: results from the Multicenter Automatic Defibrillator Implantation Trial–Reduce Inappropriate Therapy (MADIT-RIT) study.  Circulation. 2014;129(5):545-552.PubMedGoogle ScholarCrossref
14.
Tan  VH, Wilton  SB, Kuriachan  V, Sumner  GL, Exner  DV.  Impact of programming strategies aimed at reducing nonessential implantable cardioverter defibrillator therapies on mortality: a systematic review and meta-analysis.  Circ Arrhythm Electrophysiol. 2014;7(1):164-170.PubMedGoogle ScholarCrossref
15.
Larsen  GC, Stupey  MR, Walance  CG,  et al.  Recurrent cardiac events in survivors of ventricular fibrillation or tachycardia: implications for driving restrictions.  JAMA. 1994;271(17):1335-1339.PubMedGoogle ScholarCrossref
16.
Ezekowitz  JA, Armstrong  PW, McAlister  FA.  Implantable cardioverter defibrillators in primary and secondary prevention: a systematic review of randomized, controlled trials.  Ann Intern Med. 2003;138(6):445-452.PubMedGoogle ScholarCrossref
17.
Bunch  TJ, Anderson  JL.  Adjuvant antiarrhythmic therapy in patients with implantable cardioverter defibrillators.  Am J Cardiovasc Drugs. 2014;14(2):89-100.PubMedGoogle ScholarCrossref
18.
AlJaroudi  WA, Refaat  MM, Habib  RH,  et al; Genetic Risk Assessment of Defibrillator Events Investigators.  Effect of angiotensin-converting enzyme inhibitors and receptor blockers on appropriate implantable cardiac defibrillator shock in patients with severe systolic heart failure (from the GRADE Multicenter Study).  Am J Cardiol. 2015;115(7):924-931.PubMedGoogle ScholarCrossref
19.
Stockburger  M, Moss  AJ, Olshansky  B,  et al.  Time-dependent risk reduction of ventricular tachyarrhythmias in cardiac resynchronization therapy patients: a MADIT-RIT sub-study.  Europace. 2015;17(7):1085-1091.PubMedGoogle ScholarCrossref
20.
Kuck  KH, Schaumann  A, Eckardt  L,  et al; VTACH study group.  Catheter ablation of stable ventricular tachycardia before defibrillator implantation in patients with coronary heart disease (VTACH): a multicentre randomised controlled trial.  Lancet. 2010;375(9708):31-40.PubMedGoogle ScholarCrossref
21.
Reddy  VY, Reynolds  MR, Neuzil  P,  et al.  Prophylactic catheter ablation for the prevention of defibrillator therapy.  N Engl J Med. 2007;357(26):2657-2665.PubMedGoogle ScholarCrossref
22.
Gasparini  M, Proclemer  A, Klersy  C,  et al.  Effect of long-detection interval vs standard-detection interval for implantable cardioverter-defibrillators on antitachycardia pacing and shock delivery: the ADVANCE III randomized clinical trial.  JAMA. 2013;309(18):1903-1911.PubMedGoogle ScholarCrossref
23.
Sweeney  MO, Wathen  MS, Volosin  K,  et al.  Appropriate and inappropriate ventricular therapies, quality of life, and mortality among primary and secondary prevention implantable cardioverter defibrillator patients: results from the Pacing Fast VT Reduces Shock Therapies (PainFREE Rx II) trial.  Circulation. 2005;111(22):2898-2905.PubMedGoogle ScholarCrossref
24.
Wilkoff  BL, Williamson  BD, Stern  RS,  et al; PREPARE Study Investigators.  Strategic programming of detection and therapy parameters in implantable cardioverter-defibrillators reduces shocks in primary prevention patients: results from the PREPARE (Primary Prevention Parameters Evaluation) study.  J Am Coll Cardiol. 2008;52(7):541-550.PubMedGoogle ScholarCrossref
25.
Akiyama  T, Powell  JL, Mitchell  LB, Ehlert  FA, Baessler  C; Antiarrhythmics versus Implantable Defibrillators Investigators.  Resumption of driving after life-threatening ventricular tachyarrhythmia.  N Engl J Med. 2001;345(6):391-397.PubMedGoogle ScholarCrossref
26.
Curtis  AB, Conti  JB, Tucker  KJ, Kubilis  PS, Reilly  RE, Woodard  DA.  Motor vehicle accidents in patients with an implantable cardioverter-defibrillator.  J Am Coll Cardiol. 1995;26(1):180-184.PubMedGoogle ScholarCrossref
27.
Robinson  MR, Epstein  AE, Callans  DJ.  Secondary prevention in heart failure.  Heart Fail Clin. 2011;7(2):185-194, vii-viii.PubMedGoogle ScholarCrossref
28.
Varma  N, Piccini  JP, Snell  J, Fischer  A, Dalal  N, Mittal  S.  The monitoring and survival in pacemaker and defibrillator relationship between level of adherence to automatic wireless remote patients.  J Am Coll Cardiol. 2015;65(24):2601-2610.PubMedGoogle ScholarCrossref
Original Investigation
May 2016

Time Course of Subsequent Shocks After Initial Implantable Cardioverter-Defibrillator Discharge and Implications for Driving Restrictions

Author Affiliations
  • 1Cardiology Division, Department of Medicine, Emory University School of Medicine, Emory University Hospital Midtown, Atlanta, Georgia
  • 2St Jude Medical, Sylmar, California
  • 3Department of Biostatistics and Bioinformatics, Rollins School of Public Health, Emory University, Atlanta, Georgia
JAMA Cardiol. 2016;1(2):181-188. doi:10.1001/jamacardio.2015.0386
Abstract

Importance  Although guidelines recommend driving restrictions for 3 to 6 months after appropriate implantable cardioverter-defibrillator (ICD) shocks, contemporary data to support these recommendations are lacking.

Objective  To define the time course of subsequent shocks after an initial ICD discharge.

Design, Setting, and Participants  Retrospective analysis of a nationwide cohort of 14 230 ICD recipients enrolled in a remote monitoring program. Participants underwent ICD implantation from October 1, 2008, to December 31, 2013, and experienced at least 1 shock. The risk of driving after an ICD shock was estimated using the risk for harm (RH) formula, and an annual RH of less than 5 events per 100 000 ICD recipients was deemed safe. The likelihood of loss of consciousness associated with an ICD shock was estimated using a cautious value of 32% and an estimate of 14% based on contemporary data. Data were extracted and analyzed from December 17, 2014, to October 31, 2015.

Main Outcomes and Measures  Time course of subsequent shocks after an initial ICD discharge.

Results  Of 73 503 ICD recipients who underwent remote monitoring, 14 230 (19.4%) experienced at least 1 ICD shock and were included in this analysis (10 870 men [76.4%]; 3360 women [23.6%]; median age at device implantation, 68 years; interquartile range [IQR], 60-76 years). The cumulative incidence of receiving a second shock was 14.5% (IQR, 13.9%-15.1%) at 1 month and 28.7% (IQR, 27.9%-29.5%) at 6 months. The time from implantation to initial shock had an inverse association with the likelihood of receiving a second shock (lowest quartile of time at 6 months, 31.6% [95% CI, 30.2%-33.2]; highest quartile of time at 6 months, 25.3% [95% CI, 23.8%-26.9%]). The number of ICD therapy zones was also significantly associated with the incidence of a second shock (1 therapy zone, 20.8% [95% CI, 19.4%-22.3%] at 3 months to 51.5% [95% CI, 48.5%-53.7%] at 3 years; 3 therapy zones, 26.9% [95% CI, 24.8%-29.0%] at 3 months to 57.3% [95% CI, 54.1%-60.5%] at 3 years). When a likelihood of loss of consciousness of 32% associated with an ICD shock was used, the RH while driving fell below the accepted threshold at 4 to 6 months after an initial shock. However, when a contemporary estimate for loss of consciousness associated with an ICD shock of 14% was used, the RH fell below the threshold at 1 month after an initial shock.

Conclusions and Relevance  In this large cohort of ICD recipients, the incidence of a second shock after an initial ICD discharge was lower than previously reported and depended on several programmed ICD variables. These data, with future research to derive contemporary estimates of the likelihood of fatality resulting from an ICD shock while driving, should support the development of evidence-based guidelines for driving restrictions in ICD recipients.

Introduction

Implantable cardioverter-defibrillator (ICD) recipients are at risk for experiencing symptomatic ventricular arrhythmias resulting in syncope or impaired consciousness while driving, which poses a potential risk to themselves and others. Estimates from the National Cardiovascular Data Registry suggest that more than 120 000 ICDs are implanted annually in the United States,1 resulting in a significant number of people living with ICDs. As a result, an important public health need exists to develop evidence-based guidelines for activity restrictions, particularly driving restrictions, among ICD recipients. For noncommercial drivers (ie, drivers of private automobiles), the 2007 guidelines from the American Heart Association–Heart Rhythm Society recommend driving restrictions for 6 months among recipients of secondary-prevention ICDs or among recipients of primary-prevention ICDs who subsequently receive an appropriate ICD shock.2 Despite these guidelines, the duration of activity restriction recommended by physicians varies,3 and significant differences exist between states and countries with regard to laws governing driving restrictions in ICD recipients.4,5 In part, the variability in consensus statements and laws exists owing to a paucity of contemporary data on which to formulate evidence-based guidelines.

Box Section Ref ID

Key Points

  • Question What is the time course of subsequent shocks after an initial implantable cardioverter-defibrillator (ICD) discharge?

  • Findings The cumulative incidence of a second shock at 6 months after an initial discharge was lower than previously reported. The incidence of a second shock depended on several programmed ICD variables.

  • Meaning Contemporary estimates of second shock incidence reported in this cohort should support the development of evidence-based guidelines for driving restrictions in ICD recipients.

Methods

We analyzed remote follow-up data from ICD recipients who underwent implantation from October 1, 2008, to December 31, 2013, and were enrolled in a remote monitoring program (Merlin.net Patient Care Network; St Jude Medical). At the time of data extraction, 73 503 patients were undergoing remote monitoring, of whom 14 230 (19.4%) experienced at least 1 ICD shock and served as the cohort for this analysis. The decision to implant an ICD and determination of the device programming variables were performed at the discretion of the treating physician. Data sets from the remote monitoring system were deidentified before analysis and prepared for publication as defined by the Health Insurance Portability and Accountability Act (45 CFR Sec 164.514[b] implementation specifications: requirements for deidentification of protected health information). This study was reviewed by the institutional review board of Emory University, who waived the need for informed consent.

Data for this analysis were extracted from December 17, 2014, to October 31, 2015. The ICD shocks were clustered into discrete episodes such that all shocks occurring within a 24-hour window were counted as a single event. Data on the number of programmed ventricular tachycardia (VT) and ventricular fibrillation (VF) therapy zones and the heart rate detection cutoffs for each of those zones were collected from the ICD interrogation session closest to 6 months after device implantation. Shocks delivered for induced arrhythmias (ie, defibrillation threshold testing) were excluded.

When available, clinical information entered at the time of device implantation and available for query via Merlin.net was also analyzed. Data on left ventricular ejection fraction at the time of implantation were available for 4094 of the 14 230 participants (28.8%). An indication for device implantation was available for 5603 participants (39.4%). Among those with an indication available, 2019 had VT or VF listed, and, based on this information, these implants were presumed to be secondary-prevention ICDs. In the remaining 3584 participants with an indication available, the most common indications for implantation were ischemic cardiomyopathy, congestive heart failure, and dilated and/or nonischemic cardiomyopathy; these cases were presumed to represent predominantly primary-prevention devices. In the remaining 8627 patients, no data on implant indication were available.

The risk for harm (RH) while driving was estimated from the formula initially developed by the Canadian Cardiovascular Society Consensus Conference,68 which has been used in the subsequent development of US2 and European5 guidelines for driving restrictions in ICD recipients. The RH estimates the annual risk to other road users posed by a driver with an ICD and is calculated as RH = TD × V × Ac × SCI, where TD indicates the proportion of time spent driving; V, the type of vehicle; Ac, the probability that a sudden cardiac incapacitation event while driving will lead to an accident (crash) resulting in a fatality or injury; and SCI, the annual risk for sudden cardiac incapacitation.

Data predominantly from Canada have suggested that private (noncommercial) drivers spend approximately 4% of their time driving (TD = 0.04), whereas commercial drivers spend about 25% of their time driving (TD = 0.25).2,6 The risks for fatality or injuries associated with a motor vehicle crash are higher with commercial vehicles or passenger-carrying automobiles than private automobiles, and, based on these risk estimates, V = 1.0 for commercial drivers and V = 0.28 for noncommercial drivers for the RH formula. Among drivers who experience sudden cardiac death or loss of consciousness while driving, the risk for fatality or injury to other drivers or bystanders has been estimated at less than 2% (Ac = 0.02). Prior studies have estimated the risk for SCI in ICD recipients as the likelihood of receiving an ICD shock on an annual basis multiplied by the probability of experiencing impaired consciousness (syncope or near syncope) associated with the shock. The reported incidence of impaired consciousness associated with an ICD shock has varied across studies and ranges from 14% to 42%.913 Many of these estimates derive from early studies that predominantly included recipients of secondary-prevention ICDs, more liberal use of antiarrhythmics, and less capacity for delivering antitachycardia pacing to avert shocks.9,11,12 Contemporary data on the incidence of syncope associated with appropriate ICD shocks, including data from the Multicenter Automatic Defibrillator Implantation Trial–Reduce Inappropriate Therapy (MADIT-RIT), demonstrate a syncope incidence of approximately 14% associated with appropriate shocks.13,14 For the purpose of this analysis, we used the following 2 estimates for the likelihood of syncope or near syncope associated with an ICD shock: a cautious estimate of 32%6,10 and a contemporary estimate of 14% based principally on data in recipients of primary-prevention devicesfrom MADIT-RIT13 and also supported by studies of recipients of secondary-prevention devices.9,12 Last, although some studies suggest that the incidence of syncope or near syncope associated with inappropriate shocks is lower than that of appropriate shocks,11 given the relatively scant data on the incidence of impaired consciousness associated with an inappropriate shock, we used the same estimates for likelihood of syncope or near syncope associated with an inappropriate shock (32% and 14%) as with an appropriate shock. Although data defining an acceptable level of RH while driving among ICD recipients are scarce, in Canada an annual risk of 5 events per 100 000 ICD recipients (0.005%) was suggested as acceptable from a societal point of view, and this threshold has been used in the development of subsequent guidelines.2,5

Statistical Analysis

Data were analyzed from December 17, 2014, to October 31, 2015. Patient characteristics and characteristics of the device are reported as frequency and percentage for categorical variables and median and interquartile range for continuous variables. The time course of the second ICD shock after the initial ICD discharge was assessed using the Kaplan-Meier method, and cumulative incidence of a second ICD shock is reported at regular intervals after the initial ICD discharge. To assess the effect of covariates of interest on the time course of a second ICD shock, a stratified Kaplan-Meier analysis and a log-rank test were performed; continuous covariates were categorized into quartiles for the purpose of this analysis. Factors associated with time to a second shock after an initial ICD discharge were assessed by multivariate Cox proportional hazards regression models. P < .05 was considered significant. Statistical analyses were performed using SAS software (version 9.4; SAS Institute Inc).

Results

Of 73 503 ICD recipients who underwent remote monitoring, 14 230 (19.4%) experienced at least 1 ICD shock and were included in this analysis. Among these patients, median age at the time of device implantation was 68 (interquartile range, 60-76) years; 10 870 (76.4%) were male and 3360 (23.6%) were female. Among the 4094 patients with data available, the mean (SD) left ventricular ejection fraction at the time of device implantation was 27.8% (10.4%). Device characteristics are presented in Table 1. During a median follow-up of 785 days after the first shock, 5357 (37.6%) experienced a second shock. The cumulative incidence of a second shock after the initial ICD discharge is plotted in Figure 1. After an initial ICD discharge, the cumulative incidence of receiving a second shock was 14.5% (95% CI, 13.9%-15.1%) at 1 month, 22.5% (95% CI, 21.7%-23.2%) at 3 months, 28.7% (95% CI, 27.9%-29.5%) at 6 months, 37.0% (95% CI, 36.1%-37.9%) at 1 year, and 46.4% (95% CI, 45.4%-47.5%) at 2 years. By 3 years after the initial ICD shock, cumulative incidence of a second shock was 53.1% (95% CI, 51.9%-54.4%).

The time from implantation to first shock demonstrated a significant but modest inverse association with the incidence of receiving a second shock (lowest quartile of time at 6 months, 31.6% [95% CI, 30.2%-33.2]; highest quartile of time at 6 months, 25.3% [95% CI, 23.8%-26.9%]; eTable 1 in the Supplement). Device programming variables also had a significant effect on the time course of receiving a second shock. When stratified by the number of programmed VT and VF therapy zones, those with ICDs programmed with a single VF therapy zone were less likely to receive a second ICD shock at all points after an initial discharge (range, 20.8% [95% CI, 19.4%-22.3%] at 3 months to 51.5% [95% CI, 48.5%-53.7%] at 3 years) compared with recipients with ICDs programmed with 2 or 3 therapy zones (range, 26.9% [95% CI, 24.8%-29.0%] at 3 months to 57.3% [95% CI, 54.1%-60.5%] at 3 years) (eTable 2 in the Supplement). Similarly, the lowest programmed detected heart rate for which a shock could be delivered was also significantly associated with the likelihood of receiving a second shock (eTable 3 in the Supplement). In aggregate, the data in eTables 2 and 3 in the Supplement demonstrate that ICD recipients with a programmed single VF zone and relatively high cutoffs for therapy have a significantly lower incidence of receiving a second ICD shock, and the effect of device programming variables remains evident through at least 3 years of follow-up after the initial ICD discharge. In contrast to device programming variables, the year of device implantation and the type of device (single chamber vs dual chamber vs cardiac resynchronization) did not have a significant effect on the cumulative incidence of receiving a second ICD shock. Among those who received a second shock, we found a significant positive correlation between the detection heart rate for the first shock and the detection heart rate for the second shock (Spearman correlation coefficient, 0.59; P < .001; eFigure in the Supplement). Among patients who had data available on left ventricular ejection fraction and a presumed ICD indication (n = 2843), multivariate Cox proportional hazards regression models demonstrated that the most significant factors associated with time to the second shock after an initial ICD discharge were being younger at implantation, being female, having a lower programmed detection heart rate for which a shock could be delivered, a lower left ventricular ejection fraction, and a presumed secondary-prevention indication for implantation (Table 2).

Among the 2019 patients with a presumed secondary-prevention indication, the incidence of single-zone programming (380 [18.8%]) was significantly lower than among the 3584 patients with a presumed primary-prevention indication (889 [24.8%]) or the 8627 patients with no indication listed (2036 [23.6%]) (P < .001 for the comparison across groups). Similarly, the lowest detection heart rate for which a shock could be delivered was significantly lower among those with a presumed secondary-prevention indication (174.0 bpm; 95% CI, 173.2-174.8 bpm) compared with those with a presumed primary-prevention indication (177.8 bpm; 95% CI, 177.3-178.3 bpm) or no indication available (176.9 bpm; 95% CI, 176.5-177.2 bpm) (P < .001), although the absolute differences were small.

The RH while driving after an initial ICD shock for noncommercial drivers was estimated from the data in Figure 1 (Table 3). The RH was estimated from the formula given in the in the Methods section, with TD = 0.4, V = 0.28, and Ac = 0.02. For example, after an initial ICD shock, the cumulative incidence of recurrent shock at 3 months based on the data in Figure 1 is 22.5% (0.225). Because the RH formula uses the annual risk for SCI, the incidence at 3 months is multiplied by 4 to obtain an annual risk (0.225 × 4 = 0.9). The annual risk for SCI is then estimated by multiplying the annual risk for receiving a second ICD shock (0.9) by the probability of loss of consciousness (syncope or near syncope) associated with the shock. As evident from Table 3, using the cautious estimate of loss of consciousness associated with an ICD shock of 32%, the RH during noncommercial driving at 3 months after an initial ICD discharge is estimated as 6.5 events per year per 100 000 ICD recipients. By 4 months, the estimated RH decreases to 5.3 and at 6 months to 4.1 per 100 000 ICD recipients. Using the more contemporary estimate of loss of consciousness associated with an ICD shock of 14%, the annual RH during noncommercial driving is only 5.5 per 100 000 ICD recipients at 1 month after an initial ICD shock and decreases to 2.8 per 100 000 recipients by 3 months.

The cumulative incidence of a second shock after an initial ICD discharge, stratified by presumed primary or secondary implantation indication, is presented in Figure 2. The incidence of a second ICD shock was higher at all points assessed among those patients with a presumed secondary-prevention indication, lowest among those with a presumed primary-prevention indication, and intermediate in those with no indication (P = .002). However, even among those with a presumed secondary-prevention indication, the incidence of a second shock at 3 months was only 23.2% (95% CI, 21.3%-25.1%) and at 6 months was 30.3% (95% CI, 28.2%-32.4%). The corresponding values for the whole cohort (n = 14 230) were 22.5% (95%, 21.7%-23.2%) at 3 months and 28.7% (95% CI, 27.9%-29.5%) at 6 months (Figure 1). Among those with a presumed secondary-prevention indication, the RH while driving was 3.7 per 100 000 ICD recipients at 3 months and 4.3 per 100 000 ICD recipients at 6 months when we used a probability of loss of consciousness with an ICD shock of 32%. Using the contemporary probability of loss of consciousness of 14%, the corresponding values were 2.9 per 100 000 at 3 months and 1.9 per 100 000 at 6 months.

As is evident from Table 3, the Canadian formula for estimating the risk of RH while driving depended on estimates regarding the likelihood of fatality or injury owing to sudden cardiac incapacitation while driving. The traditionally used estimates for vehicle type and risk for fatality or injury (V = 0.28, Ac = 0.02, V × Ac = 0.005) were derived from data in Canada more than 3 decades ago. Changes in these estimates to reflect more contemporary vehicle types and driving practices could have an important effect on the RH while driving. The estimated risk of RH while driving at 6 months after an initial ICD shock based on varying the value for V × Ac over a hypothetical range from 0.0025 to 0.04 is reported in eTable 4 in the Supplement. These results highlight the wide changes in estimated risk of RH based on even small changes in the estimated likelihood of fatality or injury (ie, increasing V × Ac from 0.005 to 0.01).

Discussion

In this large, contemporary cohort of ICD recipients, the risk for receiving a second shock after an initial ICD discharge was significantly affected by several variables, including the time from implantation to initial shock and device-programmed variables, such as the number of therapy zones and lowest heart rate for which therapy is delivered. The incidence of a second shock after an initial ICD discharge was approximately 28.7%, which is lower than has been reported previously. These data, with contemporary estimates regarding the likelihood of syncope associated with an ICD shock and future research to derive updated estimates of the likelihood of fatality resulting from an ICD shock while driving, should support the development of evidence-based guidelines for driving restrictions in ICD recipients.

Current US guidelines recommend driving restrictions for 6 months after appropriate ICD therapy in recipients of primary-prevention devices,2 whereas the European guidelines have moved to a less restrictive recommendation of 3 months.5 These recommendations are based predominantly on event rates in patients presenting with ventricular arrhythmias who were treated primarily with antiarrhythmics guided by an electrophysiologic study15 or on event rates in recipients of secondary-prevention ICDs.9,10,12 However, neither of these estimates is likely to be an accurate reflection of contemporary practice. Beyond differences in patient characteristics and event rates between recipients of primary- and secondary-prevention ICDs,16 several other advances during the past 2 decades have likely had a significant effect on the risk for recurrent ICD shocks, including the use of evidence-based β-blockers17 and renin-angiotensin antagonists in heart failure,18 cardiac resynchronization therapy in appropriate candidates,19 and broader use of catheter ablation of ventricular tachycardia.20,21 In addition, evolution in ICD programming with a trend toward higher cutoffs for therapy rates, longer detection times, and more antitachycardia pacing has also reduced appropriate and inappropriate ICD shocks13,2224 without an increase in the frequency of syncope.14

Using a cautious estimate of loss of consciousness associated with an ICD shock of 32%, our data suggest that a period of driving restriction of 4 to 6 months after an initial ICD discharge would allow the RH for noncommercial drivers to fall below the currently accepted threshold of 5 per 100 000 ICD recipients per year. This result is consistent with a recent analysis from the Netherlands that included 2786 ICD recipients, of whom 1718 underwent implantation for primary prevention and 1068 for secondary prevention.6 After an initial appropriate ICD shock in recipients of primary-prevention devices, the RH for noncommercial driving using an estimate of loss of consciousness associated with an ICD shock of 0.31 (which is similar to the more cautious estimate used in our analysis) fell below the threshold of 5 per 100 000 ICD recipients at about 4 months after the initial ICD discharge.

Although estimates of syncope associated with shocks have varied widely,912 recent data from several studies assessing contemporary ICD programming variables have consistently suggested a much lower overall rate of syncope and reductions in the number of appropriate and inappropriate ICD shocks compared with earlier studies.13,14 Based on these findings, the RH while driving in our analysis using the lower estimate for likelihood of loss of consciousness (14%) may be a more accurate reflection of contemporary risk than estimates based on earlier data. eTable 4 in the Supplement also highlights the sensitivity of RH estimates to assumptions regarding the likelihood of fatality or injury while driving. Developing contemporary and possibly geographic region–specific data to update these values should be an area for future investigation that, coupled with the updated data in this report regarding the contemporary time course of ICD shocks, would provide a more robust and quantitative approach to estimating RH while driving in ICD recipients.

Ultimately, the impetus to estimate RH while driving is based on a need to balance individual rights with the societal need to protect others from harm caused by individuals who may be unable to operate a vehicle in a safe manner. The annual threshold of risk used in this analysis (5 per 100 000 ICD recipients) and others2,5,6 is based on data from Canada. Specific recommendations for driving restrictions may change significantly if the threshold for risk is believed to be too high or too low. Limited real-world data reporting on the frequency of motor vehicle collisions involving ICD recipients suggest a very low rate of crashes, even lower than the collision rate among the general population.25,26 Although these data may be subject to underreporting, they are consistent with the notion that current guidelines may be too restrictive in limiting driving among ICD recipients who experience shocks. Furthermore, many patients with ICDs continue to drive, and adherence with recommendations for driving restrictions is low.3,25 More limited and evidence-based recommendations for driving restrictions may improve adherence, although data in this regard are lacking.

Several important limitations of this analysis should be noted. First, we were unable to assess the impact of antiarrhythmic drugs or other interventions, such as VT catheter ablation, that may have influenced the likelihood of receiving subsequent shocks. In addition, the risk for ventricular arrhythmias is known to differ between primary- and secondary-prevention indications for implantation,27 and these indications should inform decisions about activity restriction. Although we made an attempt to assess primary- vs secondary-prevention indications for device implantation based on the information available through the Merlin.net database, implantation indications were not routinely available, and we cannot confirm implantation indications with certainty.

Beyond implantation indication, our data are also limited with regard to other aspects of the eletrophysiologic substrate that may be relevant with regard to assessing the RH while driving (eg, prior myocardial infarction, heart failure class, or presence of nonsustained ventricular arrhythmias). Given the large number of patients and ICD shocks, we were also unable to adjudicate appropriate vs inappropriate ICD shocks, and therefore we are unable to comment on the likelihood of receiving a subsequent shock after initial appropriate or inappropriate ICD therapy. However, by assuming the same risk for impaired consciousness associated with appropriate and inappropriate shocks, the need to distinguish between the causes of shock is reduced, although estimates of RH while driving in this study likely overestimate overall risk.

In addition, this cohort does not include data on the competing risk for mortality and the effect of death on the likelihood of receiving a second ICD shock. However, by treating patients who die before a second shock as censored events, our data tend to overestimate the cumulative incidence of a second ICD shock by assuming that these patients remain at risk for experiencing subsequent shocks, whereas in reality, they are no longer at risk for harm while driving. However, given the important public health implications of activity restriction in ICD recipients, we believe that conservative estimates are prudent until more data are available on the competing risk for death and its effect on the incidence of subsequent ICD shocks.

Our data also include only patients who were enrolled in a remote monitoring program. Patients who perform remote monitoring are known to have better outcomes than those who do not,28 and therefore the setting may limit the generalizability of our data to ICD recipients who do not use remote monitoring. Last, the focus of this analysis, and the RH formula while driving more generally, is on the risk for syncope or near syncope associated with ICD shocks. However, ICD shocks themselves may directly result in transient impairment, even if the antecedent arrhythmia did not result in loss of consciousness. Our analysis does not provide an estimate of the risk for impairment while driving directly attributable to experiencing an ICD shock.

Conclusions

In this large cohort of ICD recipients, the incidence of a second shock after an initial ICD discharge was lower than has been previously reported and is dependent on several programmed ICD variables, including the number of therapy zones and lowest heart rate for which therapy is delivered. These data, along with contemporary estimates regarding the likelihood of syncope associated with an ICD shock and future research to derive updated estimates of the likelihood of fatality resulting from an ICD shock while driving, should support the development of evidence-based guidelines for driving restrictions in ICD recipients.

Back to top
Article Information

Corresponding Author: Faisal M. Merchant, MD, Cardiology Division, Department of Medicine, Emory University School of Medicine, Emory University Hospital Midtown, 550 Peachtree St, Medical Office Tower, Sixth Floor, Atlanta, GA 30308 (fmercha@emory.edu).

Accepted for Publication: December 23, 2015.

Published Online: March 23, 2016. doi:10.1001/jamacardio.2015.0386.

Author Contributions: Dr Merchant had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Merchant, Hoskins, Benser, Roberts, Bastek, Langberg, El-Chami.

Acquisition, analysis, or interpretation of data: Merchant, Hoskins, Benser, Roberts, Knezevic, Huang, Leon, El-Chami.

Drafting of the manuscript: Merchant, Hoskins, Knezevic.

Critical revision of the manuscript for important intellectual content: Merchant, Hoskins, Benser, Roberts, Bastek, Huang, Langberg, Leon, El-Chami.

Statistical analysis: Merchant, Hoskins, Knezevic, Huang.

Administrative, technical, or material support: Benser, Roberts, Bastek, Huang, Langberg, Leon.

Study supervision: Merchant, Hoskins, Benser, Leon, El-Chami.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Drs Benser and Bastek and Mr Roberts report being employees of St Jude Medical. No other disclosures were reported.

Additional Information: The data for all noninduced ICD shocks were compiled from the Merlin.net database and clustered into discrete 24-hour windows by St Jude Medical. All subsequent cleaning of the dataset and analysis and presentation of the data were performed by investigators at Emory University. St Jude Medical assisted in the study design and approved the final manuscript.

References
1.
Kramer  DB, Kennedy  KF, Noseworthy  PA,  et al.  Characteristics and outcomes of patients receiving new and replacement implantable cardioverter-defibrillators: results from the NCDR.  Circ Cardiovasc Qual Outcomes. 2013;6(4):488-497.PubMedGoogle ScholarCrossref
2.
Epstein  AE, Baessler  CA, Curtis  AB,  et al; American Heart Association; Heart Rhythm Society.  Addendum to “Personal and public safety issues related to arrhythmias that may affect consciousness: implications for regulation and physician recommendations: a medical/scientific statement from the American Heart Association and the North American Society of Pacing and Electrophysiology”: public safety issues in patients with implantable defibrillators: a scientific statement from the American Heart Association and the Heart Rhythm Society.  Circulation. 2007;115(9):1170-1176.PubMedGoogle ScholarCrossref
3.
Mylotte  D, Sheahan  RG, Nolan  PG,  et al.  The implantable defibrillator and return to operation of vehicles study.  Europace. 2013;15(2):212-218.PubMedGoogle ScholarCrossref
4.
Strickberger  SA, Cantillon  CO, Friedman  PL.  When should patients with lethal ventricular arrhythmia resume driving? an analysis of state regulations and physician practices.  Ann Intern Med. 1991;115(7):560-563.PubMedGoogle ScholarCrossref
5.
Vijgen  J, Botto  G, Camm  J,  et al; Task force members.  Consensus statement of the European Heart Rhythm Association: updated recommendations for driving by patients with implantable cardioverter defibrillators.  Europace. 2009;11(8):1097-1107.PubMedGoogle ScholarCrossref
6.
Thijssen  J, Borleffs  CJ, van Rees  JB,  et al.  Driving restrictions after implantable cardioverter defibrillator implantation: an evidence-based approach.  Eur Heart J. 2011;32(21):2678-2687.PubMedGoogle ScholarCrossref
7.
 Assessment of the cardiac patient for fitness to drive: 1996 update.  Can J Cardiol. 1996;12(11):1164-1170, 1175-1182.PubMedGoogle Scholar
8.
 Assessment of the cardiac patient for fitness to drive.  Can J Cardiol. 1992;8(4):406-419.PubMedGoogle Scholar
9.
Bänsch  D, Brunn  J, Castrucci  M,  et al.  Syncope in patients with an implantable cardioverter-defibrillator: incidence, prediction and implications for driving restrictions.  J Am Coll Cardiol. 1998;31(3):608-615.PubMedGoogle ScholarCrossref
10.
Freedberg  NA, Hill  JN, Fogel  RI, Prystowsky  EN; CARE Group.  Recurrence of symptomatic ventricular arrhythmias in patients with implantable cardioverter defibrillator after the first device therapy: implications for antiarrhythmic therapy and driving restrictions.  J Am Coll Cardiol. 2001;37(7):1910-1915.PubMedGoogle ScholarCrossref
11.
Grimm  W, Flores  BF, Marchlinski  FE.  Symptoms and electrocardiographically documented rhythm preceding spontaneous shocks in patients with implantable cardioverter-defibrillator.  Am J Cardiol. 1993;71(16):1415-1418.PubMedGoogle ScholarCrossref
12.
Kou  WH, Calkins  H, Lewis  RR,  et al.  Incidence of loss of consciousness during automatic implantable cardioverter-defibrillator shocks.  Ann Intern Med. 1991;115(12):942-945.PubMedGoogle ScholarCrossref
13.
Ruwald  MH, Okumura  K, Kimura  T,  et al.  Syncope in high-risk cardiomyopathy patients with implantable defibrillators: frequency, risk factors, mechanisms, and association with mortality: results from the Multicenter Automatic Defibrillator Implantation Trial–Reduce Inappropriate Therapy (MADIT-RIT) study.  Circulation. 2014;129(5):545-552.PubMedGoogle ScholarCrossref
14.
Tan  VH, Wilton  SB, Kuriachan  V, Sumner  GL, Exner  DV.  Impact of programming strategies aimed at reducing nonessential implantable cardioverter defibrillator therapies on mortality: a systematic review and meta-analysis.  Circ Arrhythm Electrophysiol. 2014;7(1):164-170.PubMedGoogle ScholarCrossref
15.
Larsen  GC, Stupey  MR, Walance  CG,  et al.  Recurrent cardiac events in survivors of ventricular fibrillation or tachycardia: implications for driving restrictions.  JAMA. 1994;271(17):1335-1339.PubMedGoogle ScholarCrossref
16.
Ezekowitz  JA, Armstrong  PW, McAlister  FA.  Implantable cardioverter defibrillators in primary and secondary prevention: a systematic review of randomized, controlled trials.  Ann Intern Med. 2003;138(6):445-452.PubMedGoogle ScholarCrossref
17.
Bunch  TJ, Anderson  JL.  Adjuvant antiarrhythmic therapy in patients with implantable cardioverter defibrillators.  Am J Cardiovasc Drugs. 2014;14(2):89-100.PubMedGoogle ScholarCrossref
18.
AlJaroudi  WA, Refaat  MM, Habib  RH,  et al; Genetic Risk Assessment of Defibrillator Events Investigators.  Effect of angiotensin-converting enzyme inhibitors and receptor blockers on appropriate implantable cardiac defibrillator shock in patients with severe systolic heart failure (from the GRADE Multicenter Study).  Am J Cardiol. 2015;115(7):924-931.PubMedGoogle ScholarCrossref
19.
Stockburger  M, Moss  AJ, Olshansky  B,  et al.  Time-dependent risk reduction of ventricular tachyarrhythmias in cardiac resynchronization therapy patients: a MADIT-RIT sub-study.  Europace. 2015;17(7):1085-1091.PubMedGoogle ScholarCrossref
20.
Kuck  KH, Schaumann  A, Eckardt  L,  et al; VTACH study group.  Catheter ablation of stable ventricular tachycardia before defibrillator implantation in patients with coronary heart disease (VTACH): a multicentre randomised controlled trial.  Lancet. 2010;375(9708):31-40.PubMedGoogle ScholarCrossref
21.
Reddy  VY, Reynolds  MR, Neuzil  P,  et al.  Prophylactic catheter ablation for the prevention of defibrillator therapy.  N Engl J Med. 2007;357(26):2657-2665.PubMedGoogle ScholarCrossref
22.
Gasparini  M, Proclemer  A, Klersy  C,  et al.  Effect of long-detection interval vs standard-detection interval for implantable cardioverter-defibrillators on antitachycardia pacing and shock delivery: the ADVANCE III randomized clinical trial.  JAMA. 2013;309(18):1903-1911.PubMedGoogle ScholarCrossref
23.
Sweeney  MO, Wathen  MS, Volosin  K,  et al.  Appropriate and inappropriate ventricular therapies, quality of life, and mortality among primary and secondary prevention implantable cardioverter defibrillator patients: results from the Pacing Fast VT Reduces Shock Therapies (PainFREE Rx II) trial.  Circulation. 2005;111(22):2898-2905.PubMedGoogle ScholarCrossref
24.
Wilkoff  BL, Williamson  BD, Stern  RS,  et al; PREPARE Study Investigators.  Strategic programming of detection and therapy parameters in implantable cardioverter-defibrillators reduces shocks in primary prevention patients: results from the PREPARE (Primary Prevention Parameters Evaluation) study.  J Am Coll Cardiol. 2008;52(7):541-550.PubMedGoogle ScholarCrossref
25.
Akiyama  T, Powell  JL, Mitchell  LB, Ehlert  FA, Baessler  C; Antiarrhythmics versus Implantable Defibrillators Investigators.  Resumption of driving after life-threatening ventricular tachyarrhythmia.  N Engl J Med. 2001;345(6):391-397.PubMedGoogle ScholarCrossref
26.
Curtis  AB, Conti  JB, Tucker  KJ, Kubilis  PS, Reilly  RE, Woodard  DA.  Motor vehicle accidents in patients with an implantable cardioverter-defibrillator.  J Am Coll Cardiol. 1995;26(1):180-184.PubMedGoogle ScholarCrossref
27.
Robinson  MR, Epstein  AE, Callans  DJ.  Secondary prevention in heart failure.  Heart Fail Clin. 2011;7(2):185-194, vii-viii.PubMedGoogle ScholarCrossref
28.
Varma  N, Piccini  JP, Snell  J, Fischer  A, Dalal  N, Mittal  S.  The monitoring and survival in pacemaker and defibrillator relationship between level of adherence to automatic wireless remote patients.  J Am Coll Cardiol. 2015;65(24):2601-2610.PubMedGoogle ScholarCrossref
×