- Email: [email protected]
Implantable cardioverter-defibrillator and wait-list outcomes in pediatric patients awaiting heart transplantation
Implantable cardioverter-defibrillator and wait-list outcomes in pediatric patients awaiting heart transplantation Iqbal El-Assaad, MD,* Sadeer G. Al-Kindi, MD,† Guilherme H. Oliveira, MD,† Gerard J. Boyle, MD,‡ Peter F. Aziz, MD, FHRS‡ From the *Department of Pediatrics, Cleveland Clinic Children’s, Cleveland Clinic Lerner College of Medicine, Cleveland, Ohio, †Advanced Heart Failure Center, Harrington Heart & Vascular Institute, University Hospitals Case Medical Center, Case Western Reserve University, Cleveland, Ohio ‡ Division of Pediatric Cardiology, Cleveland Clinic Children’s, Cleveland Clinic Lerner College of Medicine, Cleveland, Ohio. BACKGROUND Implantable cardioverter-defibrillators (ICDs) reduce the incidence of sudden cardiac death (SCD) in adults with end-stage heart failure; however, their efficacy in pediatric patients awaiting heart transplantation is not well established. OBJECTIVES This study sought to investigate the role of ICDs in preventing SCD and waiting list mortality as well as to determine risk factors for SCD in pediatric patients listed for heart transplantation. METHODS We queried the United Network for Organ Sharing database for all pediatric patients (age r18 years) listed for heart transplantation (2005–2014). The Cox proportional hazards model was used to identify risk factors for SCD and all-cause mortality. RESULTS A total of 5072 mostly White (55%) male (55%) patients (mean age 6.2 ⫾ 6.5 years) were identified, of whom 426 (8.3%) had ICD at listing. At 6 months, 65% underwent heart transplantation, 15% died (4% died of SCD), and 20% were alive. In a multivariable model, United Network for Organ Sharing status 1B (hazard ratio [HR] 0.52; 95% confidence interval [CI] 0.29–0.95; P ¼ .03), myocarditis (HR 0.19; 95% CI 0.05–0.77; P ¼ .02), restrictive cardiomyopathy (HR 0.19; 95% CI 0.05–0.76; P ¼ .02),
Introduction Although implantable cardioverter-defibrillator (ICD) use has been extensively validated in preventing sudden cardiac death (SCD) in adult patients with end-stage heart failure Dr El-Assaad and Dr Al-Kindi contributed equally to this work. This work was supported in part by the Health Resources and Services Administration (contract no. 234-2005-37011C). The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. Address reprint requests and correspondence: Dr Peter F. Aziz, Division of Pediatric Cardiology, Cleveland Clinic Children’s, Cleveland Clinic Lerner College of Medicine, 9500 Euclid Avenue, Cleveland, OH 44195. E-mail address: [email protected].
1547-5271/$-see front matter B 2015 Heart Rhythm Society. All rights reserved.
and dilated cardiomyopathy (HR 0.32; 95% CI 0.20–0.52; P o .001) were associated with lower SCD risk, while younger age at listing (HR 0.94 per year; 95% CI 0.90–0.98; P ¼ .003) was associated with higher SCD risk. ICD at listing was not associated with reduced SCD (P ¼ .12), all-cause mortality, or delisting (P ¼ .57). CONCLUSION In pediatric patients listed for heart transplantation, the risk of SCD remains low and does not differ between patients with and without an ICD at listing. KEYWORDS Pediatric; Heart transplant; Sudden cardiac death; Implantable cardioverter-defibrillator ABBREVIATIONS BMI ¼ body mass index; CHD ¼ congenital heart disease; CI ¼ confidence interval; DCM ¼ dilated cardiomyopathy; ECMO ¼ extracorporeal membrane oxygenation; HCM ¼ hypertrophic cardiomyopathy; HR ¼ hazard ratio; ICD ¼ implantable cardioverter-defibrillator; RCM ¼ restrictive cardiomyopathy; SCD ¼ sudden cardiac death; UNOS ¼ United Network for Organ Sharing (Heart Rhythm 2015;12:2443–2448) I 2015 Heart Rhythm Society. All rights reserved.
awaiting heart transplantation,1–3 little is known about its benefit in the pediatric population, especially those listed for heart transplantation. With an incidence of SCD of approximately 1.3%4 in pediatric patients listed for heart transplantation, the use of prophylactic ICDs in these patients is controversial because of the reportedly high complication rates5 and high costs.4,6 A single study that investigated the role of ICD in pediatric patients with dilated cardiomyopathy (DCM) suggested a sustained survival benefit with low complication rate. Dubin et al7 performed a multicenter retrospective review of ICD databases in 28 pediatric patients awaiting heart transplant and demonstrated a high incidence of appropriate ICD discharges (85%) with a mean time of 6 http://dx.doi.org/10.1016/j.hrthm.2015.07.036
2444 months from implantation to first shock. In the aforementioned study, the majority of patients had ICDs because of ventricular tachycardia or fibrillation and a small percentage had an ICD implanted secondary to poor myocardial function (17%). Because of limited and conflicting data7 on the role of ICDs in pediatric patients listed for heart transplant, we sought to investigate their role in preventing SCD and allcause mortality in this population. In addition, we sought to identify predictors of SCD in these patients in order to distinguish high-risk subgroups.
Methods Study population We conducted a retrospective cohort study using the United Network for Organ Sharing/Organ Procurement and Transplantation Network database (Health Resources and Services Administration). UNOS is a comprehensive online database that records details on all patients listed for solid organ transplantation in all centers across the United States. Deidentified data are obtained from the Organ Procurement and Transplantation Network contracted with the Health Resources and Services Administration. Data are collected at different time points: at listing (Transplant Candidate Registration Form), before transplantation (Transplant Recipient Registration Form), and regularly after transplantation (Transplant Recipient Follow-Up Form). The OPTN/UNOS Transplant Registry includes information on patient demographic characteristics, causes of cardiomyopathy, devices present, causes of removal from the waiting list, hemodynamics, and other peritransplantation and long-term data. The registry is continuously audited, and quality control is assured.8 The version used in this study recorded patients until June 30, 2014. We identified patients listed between January 2005 and June 2014 for heart transplantation and were 18 years or younger. All patients were followed from the time of listing until heart transplantation, death, or removal from the list. Heart transplant candidates were divided into 2 groups: those with ICD at listing and those without. The non-ICD group was used as the control group. Information was collected at the time of listing and before transplant by using a standardized transplant candidate registration form. We performed baseline comparisons between the 2 groups by using the following study variables: age, ethnicity, sex, UNOS status at listing, underlying cause of heart failure (DCM vs restrictive cardiomyopathy [RCM] vs hypertrophic cardiomyopathy [HCM] vs congenital heart disease [CHD] vs retransplant vs myocarditis), body mass index (BMI), baseline medications, and treatments including extracorporeal membrane oxygenation (ECMO), ventricular assist devices, ventilator use, hemodialysis, need for inotropes, and intra-aortic balloon pumps. Information on all-cause mortality, SCD, and transplantation rate was retrieved for analysis. The primary outcome was the cumulative rate of
Heart Rhythm, Vol 12, No 12, December 2015 SCD as defined by the UNOS database. Variables that were significantly different in univariable analysis were tested using a multivariable model to identify risk factors for SCD. Antiarrhythmic medications were not included in our model because data on this variable were not recorded in the database beyond 2007. The institutional review board at the Cleveland Clinic Foundation approved this study.
Statistical analysis Categorical variables were expressed as percentages and compared using Pearson’s χ2 test. Continuous variables were reported as mean ⫾ standard deviation and/or medians with interquartile ranges and compared using the independent t test. The Cox proportional hazards model was used to identify multivariable predictors of SCD. Kaplan-Meier analysis and log-rank test were used for unadjusted analyses. We selected and tested 15 potential predictors of SCD using the univariable analysis. We also performed a subgroup analysis based on the underlying etiology of heart failure, and adjustments were based on baseline differences between the ICD and the non-ICD group. All hazard ratios (HRs) were adjusted for baseline differences between the 2 groups. No assumptions were made for missing data because variables used in the models were available for 100% of the study population. All statistical analyses were performed using SPSS software, version 19.0 (IBM SPSS Statistics, IBM Corporation, Chicago, IL). All tests were 2 sided. Differences were considered statistically significant at P o .05.
Results A total of 5214 pediatric patients were identified. Patients with unknown ICD status (n ¼ 92, 1.7%) and those with UNOS status 7 (n ¼ 83, 1.6%) were excluded from the study, leaving a study population of 5072. Table 1 lists the baseline demographic and clinical characteristics in the study population, ICD group, and non-ICD group. Overall, the majority of patients were White (n ¼ 2783, 55%) male (n ¼ 2802, 55%) patients with mean age of 6.2 ⫾ 6.5 years. The major causes of end-stage heart failure were CHD (n = 2297, 45%) and DCM (n = 1656, 33%). Of all patients listed for transplant with a known ICD status, 426 patients (8.3%) had ICDs at listing.
ICD during listing Of the patients without an ICD at listing, 44 patients (0.9%) eventually received ICDs during their listing time. This was an intention-to-treat analysis; thus, these patients were analyzed in the non-ICD group. Of the patients who did not have an ICD at listing, the number (percentage) of patients who received an ICD after listing was as follows: 24 (0.75%) of Ia, 6 (1.1%) of Ib, 11 (1.3%) of II, 12 (0.55%) of CHD, 3 (1.4%) of myocarditis, 3 (3.6%) of HCM, 15 (1%) of DCM, and 6 (1.8%) of retransplant.
El-Assaad et al Table 1
ICD Use Before Pediatric Heart Transplant
2445
Baseline demographic and clinical characteristics in all patients, ICD group, and non-ICD group
Characteristic Age (y) Sex: male Ethnicity White African American Hispanic Asian Other* Cause of heart failure Dilated cardiomyopathy Myocarditis Restrictive cardiomyopathy Retransplant Hypertrophic cardiomyopathy Congenital heart disease Other UNOS status at listing 1A 1B 2 Ionotropes Treatment VAD ECMO IABP Hemodialysis Ventilator BMI (kg/m2)
Total population (N ¼ 5072)
ICD group (n ¼ 426)
Non-ICD group (n ¼ 4646)
6.2 ⫾ 6.5 2802 (55)
13.1 ⫾ 5.1 253 (59)
5.5 ⫾ 6.2 2549 (55)
2783 (55) 1059 (21) 917 (18) 176 (3.5) 137 (3)
243 92 66 14 11
(57) (22) (16) (3) (3)
2540 (55) 967 (21) 851 (18) 162 (3.5) 126 (3)
1656 (33) 228 (4.5) 267 (5) 348 (7) 135 (3) 2297 (45) 141 (3)
182 16 30 21 51 104 22
(43) (4) (7) (5) (12) (24) (5)
1474 (32) 212 (5) 237 (5) 327 (7) 84 (2) 2193 (47) 119 (3)
3390 (67) 645 (13) 1037 (20) 2279 (44.9)
173 81 172 148
(41) (19) (40) (34.7)
3217 (69) 564 (12) 856 (19) 2131 (46)
o.001
381 (7.5) 499 (9.8) 19 (0.4) 114 (2) 1124 (22) 17.4 ⫾ 5.1
33 (8) 12 (2.8) 3 (0.7) 7 (2) 26 (6) 21.3 ⫾ 5.6
348 (7.5) 487 (10.5) 16 (0.3) 107 (2) 1098 (24) 17.1 ⫾ 4.9
.85 o.001 .21 .49 o.001 o.001
P o.001 .075 .68
o.001
o.001
Categorical and continuous data are presented as n (%) and mean ⫾ SD, respectively. BMI ¼ body mass index; ECMO ¼ extracorporeal membrane oxygenation; IABP ¼ intra-aortic balloon pump; ICD ¼ implantable cardioverter-defibrillator; UNOS ¼ United Nation Organ Sharing; VAD ¼ ventricular assist device. * Other: American Indian, Alaskan Native, Native Hawaiian, Pacific Islander, multiracial. Bold P values represent statistical significance.
Baseline comparison between the control and the study group
Of the 426 patients who had an ICD at listing, 76.5% received heart transplantation whereas 8.5% died while waiting (2.3% died of SCD). In the control group, 67.4% of the transplant candidates received heart transplants, whereas 11.1% died while waiting (3.1% died of SCD).
Age, heart failure etiology, UNOS status at listing, BMI, need for intravenous inotropes, ECMO, and ventilator use were different between the 2 groups. In comparison to the control group, those with ICD at listing were older (13.1 ⫾ 5.1 years vs 5.5 ⫾ 6.2 years), less likely to have CHD (24% vs 47%), more likely to have HCM (12% vs 2.0%), and less likely to require inotropes (35% vs 46%), ECMO (2.8% vs 10.5%), and ventilator support (6.0% vs 24%). Furthermore, patients with ICD were less likely to be listed with status 1A (41% vs 69%) and more likely to be listed with status 2 (40% vs 19%). ICD recipients were also found to have a higher BMI (21.3 ⫾ 5.6 kg/m2 vs 17.1 ⫾ 4.9 kg/m2). There were no statistical differences in other studied variables: sex, ethnicity, use of ventricular assist devices, intra-aortic balloon pumps, or hemodialysis (Table 1).
Most SCD events (90%) occurred within 6 months of listing. In Cox proportional multivariable analysis, status 1B (HR 0.52; 95% confidence interval [CI] 0.29–0.95; P ¼ .03), myocarditis (HR 0.19; 95% CI 0.05–0.77; P ¼ .02), RCM (HR 0.19; 95% CI 0.05–0.76; P ¼ .02), and DCM (HR 0.32; 95% CI 0.20–0.52; P o .001) were associated with lower SCD risk while ventilator use (HR 1.73; 95% CI 1.17–2.56; P ¼ .006) and younger age at listing (HR 0.94 per year; 95% CI 0.90–0.98; P ¼ .003) were associated with higher SCD risk (Table 2).
Follow-up and outcome
Cumulative mortality and ICD use
Over a median follow-up of 1.8 months (interquartile range 0.6–4.8 months), 68% underwent heart transplantation, 10.9% died while waiting, and the remaining are still alive. Only 3.1% died of SCD, while the remainder of the population died of either cardiac or noncardiac etiology.
ICD was not associated with decreased all-cause mortality; delisting for deterioration, SCD, or transplant; or delisting for improvement (P ¼ .12, P ¼ .73, and P ¼ .57, respectively) (Figure 1). In subgroup analyses, ICD was not associated with decreased SCD rates in patients with
Risk factors for SCD before heart transplantation
2446
Heart Rhythm, Vol 12, No 12, December 2015
Table 2 Multivariate predictors of sudden cardiac death using the Cox proportional hazards model Variable ICD at listing ECMO at listing IV inotropes at listing Initial UNOS status 1A 1B 2 Age at listing (per year) Underlying cause Congenital heart disease Myocarditis Restrictive cardiomyopathy Hypertrophic cardiomyopathy Dilated cardiomyopathy Retransplant Other Ventilator use List year BMI
Adjusted HR (95% CI) 1.74 (0.87–3.74) 0.78 (0.41–1.47) 0.95 (0.66–1.38)
Multivariable P value .12 .45 .78
Reference Reference 0.52 (0.29–0.95) .033 0.64 (0.38–1.07) .09 0.94 (0.90–0.98) .003 Reference Reference 0.19 (0.05–0.77) .02 0.19 (0.05–0.76) .02 0.36 (0.09–1.51) .16 0.32 (0.20–0.52) o.001 1.49 (0.82–2.69) .19 0.59 (0.22–1.60) .30 1.73 (1.17–2.56) .006 0.94 (0.89–1.00) .06 0.99 (0.95–1.04) .78
BMI ¼ body mass index; CI ¼ confidence interval; ECMO ¼ extracorporeal membrane oxygenation; HR ¼ hazard ratio; ICD ¼ implantable cardioverter-defibrillator; IV ¼ intravenous; UNOS ¼ United Network for Organ Sharing.
DCM (P ¼ .62), RCM (P ¼ .95), myocarditis (P ¼ .92), retransplant (P ¼ .12), HCM (P ¼ .96), and CHD (P ¼ .15) (Table 3).
Discussion This is the largest study that investigates the risk of SCD in pediatric patients awaiting heart transplantation. We report,
for the first time, that the risk of SCD is not statistically different between pediatric cardiac transplant candidates who receive ICDs at or before listing and those who do not. Although these results are compelling, we are cautioned in concluding that ICD implantation for pediatric patients awaiting heart transplantation does not improve survival based on caveats discussed below. We also report that the overall risk of SCD in this patient population remains low at 3.1%. Our findings differ from adult studies that have consistently shown survival benefit from ICD as primary prevention therapy in patients with end-stage heart failure awaiting heart transplantation.1–3,9–11 This difference may largely be explained by the lower incidence of SCD in pediatric population in comparison to adults. Unlike pediatric patients, the leading cause of death in adult cardiac transplant candidates is SCD. In a retrospective adult study, SCD was found to be the major cause of death accounting for 66.7% of the total deaths.2 On the contrary, the leading cause of death in our population was non-SCD and SCD occurred only in 28% of the total deaths. Rhee et al4 reported an even lower rate of SCD in the pediatric heart transplant study (7.6% of the total deaths). Furthermore, the different survival benefit between adult and pediatric patients could be attributed to the difference in the underlying disease substrate. Based on our cohort, ischemic cardiomyopathy accounted for only a minority of patients (n ¼ 18 [o1.0%]), of whom none had a sudden cardiac event, as compared with 30%–40% of adult patients with heart transplant.1,3 The heterogeneous environment created by the presence of normal and abnormal scar tissue alters the electrophysiological properties of the myocardium, leading to the development of a reentry circuit,12 which predisposes to fatal ventricular arrhythmias and SCD.
Figure 1 Adjusted cumulative incidence of hazard ratio of sudden cardiac death (A) and all-cause mortality (B) before heart transplantation in implantable cardioverter-defibrillator (ICD) recipients and control subjects.
El-Assaad et al Table 3
ICD Use Before Pediatric Heart Transplant
Incidence of SCD at 6 months in subgroups Cumulative incidence of SCD (%)
Variable Initial UNOS status 1A 1B 2 Underlying cause Dilated cardiomyopathy Myocarditis Restrictive cardiomyopathy Retransplant Hypertrophic cardiomyopathy Congenital heart disease
Overall
ICD
Non-ICD
P
5.64 2.13 2.83
5.08 0 2.29
5.67 2.43 2.93
.32 .82 .06
1.94 1.52 1.14 4.76 2.34 6.94
0 0 0 6.45 0 8.28
2.19 1.64 1.29 4.64 3.85 6.87
.62 .92 .95 .12 .96 .15
ICD ¼ implantable cardioverter-defibrillator; SCD ¼ sudden cardiac death.
In addition, the discrepancy in survival benefit can be partially explained by the urgent listing status and shorter wait-list time in children compared with adults. In fact, approximately 70% of pediatric patients spent less than 3 months on the waiting list as compared with only 49.2% of adult patients who spent less than 1 year.10 The median waiting time to transplantation in our cohort was 1.6 months (interquartile range 0.6–3.7 months) as compared with 8.1 months in adults.2 The benefit of ICD therapy is a function of time, and shorter wait-list times may disguise that benefit. Lastly, pediatric patients are more likely to be listed with a UNOS status 1A than are adults.10 These patients are likely to be hospitalized until they receive a heart transplant and thus may receive appropriate treatments of ventricular arrhythmias without the need for ICDs. On the basis of a prospective randomized clinical trial examining the role of ICD in adult patients with heart failure, the American College of Cardiology/American Heart Association/Heart Rhythm Society committee has published clear guidelines for this specific cohort of patients.13 The Suden Cardiac Death in Heart Failure Trial (SCD-HeFT) trial in adults has convincingly demonstrated the favorable effects of ICD in primary prevention of SCD in patient with ischemic and non–ischemic DCM with New York Heart Association class II or III heart failure and a left ventricular ejection fraction of r35%.14 The International Society of Heart and Lung Transplantation guidelines state that in pediatric patients with DCM who have New York Heart Association class II or III and a left ventricular ejection fraction of o35%, physicians may consider ICD placement (class IIb, level of evidence C).15 However, there is lack of specific indications for pediatric patients with heart failure because of the paucity of prospective data in pediatric patients, which is due to the reportedly low incidence of SCD. The other important implication that warrants consideration in pediatric patients is the high complication rate due to ICD implantation. Multiple pediatric studies16,17 examining ICD role in patients who are at an increased risk of SCD have been conducted. A retrospective study18 that examined the incidence of appropriate shock delivery and adverse events
2447 found that in 140 pediatric patients who had ICDs for primary prevention, the percentage of adverse events was almost double the percentage of appropriate shocks over a mean follow-up of 4 years (36% vs 19%). Another multicenter retrospective study5 in pediatric ICD recipients reported a high frequency of appropriate as well as inappropriate shocks as high as 26% and 21%, respectively. We show that the incidence of SCD in pediatric patients is similar to that reported in previous studies.4,11 A cohort analysis that included 1803 pediatric patients with DCM estimated the 5-year incidence rate of SCD to be 2.4%.11 Rhee et al4 found that 1.3% of the total listed patients died suddenly. We also demonstrate that CHD is a strong predictor of SCD in accordance with previous studies.5,19 Furthermore, our study shows that patients who are younger at listing have a higher risk of SCD. However, these findings should be tempered by the fact that younger age at implantation has been found to be an independent predictor of lead failure, which is the main reason for inappropriate shock delivery.5,19 Interestingly, even in the high-risk population (those with CHD), ICD presence at listing was not associated with lower SCD rates.
Study limitations The findings of this study should be interpreted with caution because of significant database inherent limitations and loss of granularity. This is a retrospective study where data entry cannot be verified, and therefore the accuracy of the data is unknown. Although comprehensive, there are data points such as indications for ICD, cardiac function, and antiarrhythmic medications that impair our ability to adjust for confounding factors. A major limitation is the lack of information on the number of appropriate shocks delivered. Perhaps, patients with ICD at listing received appropriate shocks that prevented SCD, but given lack of data we were not able to adjust for this variable. It is plausible that patients with ICD had it implanted because of their suspected higher arrhythmogenic risk. Similarly, the lack of data on cardiac arrest events that received appropriate inpatient interventions in the non-ICD group may contribute to the lack of survival benefit in the ICD group. In addition, perhaps the most important clinical question for pediatric electrophysiologists is whether ICD implantation after listing affects survival. We could not analyze outcomes of patients who received ICDs after listing because of the lack of sufficient information on those patients. Data on ICD during listing is collected only on patients who receive transplantation using the Transplant Recipient Registration Form but not on patients who die or get delisted. Another significant limitation of the study is the lack of the specific underlying etiologies that led to SCD. The assumption in this study is that most of the patients had a presumed cardiac etiology; however, as in other analyses of cardiac arrests, it is hard to ascertain the exact etiology without autopsy results.20 If the primary etiology was nonarrhythmic, this may explain why the majority of the patients who had an ICD did not have improved survival. Despite these limitations, because of the
2448
Heart Rhythm, Vol 12, No 12, December 2015
rarity of SCD in pediatric patients awaiting heart transplantation, conducting a prospective study with high power may not be feasible.
Conclusion In pediatric patients listed for heart transplantation, the risk of SCD remains low and does not differ between patients with and without an ICD at listing. The utility of ICD implantation in these patients is masked by several confounders, which distinguishes pediatric patients from adults.
Acknowledgments We thank Mrs Penny Houghtaling, biostatistician in the Quantitative Health Department, Cleveland Clinic, Cleveland, Ohio for statistical contribution essential to this study. We would also like to thank Mrs Patricia Agatisa, research coordinator in the Department of Human Resources, Cleveland Clinic, Cleveland, Ohio for final manuscript editing essential to this study.
References 1. Ermis C, Zadeii G, Zhu AX, Fabian W, Collins J, Lurie KG, Sakaguchi S, Benditt DG. Improved survival of cardiac transplantation candidates with implantable cardioverter defibrillator therapy: role of beta-blocker or amiodarone treatment. J Cardiovasc Electrophysiol 2003;14:578–583. 2. Sandner SE, Wieselthaler G, Zuckermann A, Taghavi S, Schmidinger H, Pacher R, Ploner M, Laufer G, Wolner E, Grimm M. Survival benefit of the implantable cardioverter-defibrillator in patients on the waiting list for cardiac transplantation. Circulation 2001;104:I171–I176. 3. Frohlich GM, Holzmeister J, Hubler M, et al. Prophylactic implantable cardioverter defibrillator treatment in patients with end-stage heart failure awaiting heart transplantation. Heart 2013;99:1158–1165. 4. Rhee EK, Canter CE, Basile S, Webber SA, Naftel DC. Sudden death prior to pediatric heart transplantation: would implantable defibrillators improve outcome? J Heart Lung Transplant 2007;26:447–452. 5. Berul CI, Van Hare GF, Kertesz NJ, Dubin AM, Cecchin F, Collins KK, Cannon BC, Alexander ME, Triedman JK, Walsh EP, Friedman RA. Results of a multicenter retrospective implantable cardioverter-defibrillator registry of pediatric and congenital heart disease patients. J Am Coll Cardiol 2008;51:1685–1691. 6. Feingold B, Arora G, Webber SA, Smith KJ. Cost-effectiveness of implantable cardioverter-defibrillators in children with dilated cardiomyopathy. J Card Fail 2010;16:734–741.
7. Dubin AM, Berul CI, Bevilacqua LM, et al. The use of implantable cardioverterdefibrillators in pediatric patients awaiting heart transplantation. J Card Fail 2003;9:375–379. 8. Daily OP, Kauffman HM. Quality control of the OPTN/UNOS Transplant Registry. Transplantation 2004;77:1309–1310. 1309; author reply. 9. Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352: 225–237. 10. Colvin-Adams M, Smithy JM, Heubner BM, Skeans MA, Edwards LB, Waller C, Schnitzler MA, Snyder JJ, Israni AK, Kasiske BL. OPTN/SRTR 2012 annual data report: Heart. Am J Transplant 2014;14:113–138. 11. Pahl E, Sleeper LA, Canter CE, et al. Incidence of and risk factors for sudden cardiac death in children with dilated cardiomyopathy: a report from the pediatric cardiomyopathy registry. J Am Coll Cardiol 2012;59:607–615. 12. Ebinger MW, Krishnan S, Schuger CD. Mechanisms of ventricular arrhythmias in heart failure. Curr Heart Fail Rep 2005;2:111–117. 13. Epstein AE, Dimarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: executive summary. Heart Rhythm 2008;5:934–955. 14. Russo AM, Stainback RF, Bailey SR, Epstein AE, Heidenreich PA, Jessup M, Kapa S, Kremers MS, Lindsay BD, Stevenson LW. ACCF/HRS/AHA/ASE/ HFSA/SCAI/SCCT/SCMR 2013 appropriate use criteria for implantable cardioverter-defibrillators and cardiac resynchronization therapy: a report of the American College of Cardiology Foundation appropriate use criteria task force, Heart Rhythm Society, American Heart Association, American Society of Echocardiography, Heart Failure Society of America, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, and Society for Cardiovascular Magnetic Resonance. J Am Coll Cardiol 2013;61:1318–1368. 15. Kirk R, Dipchand AI, Rosenthal DN, et al. The International Society of Heart and Lung Transplantation guidelines for the management of pediatric heart failure: executive summary. J Heart Lung Transplant 2014;33:888–909. 16. Alexander ME, Cecchin F, Walsh EP, Triedman JK, Bevilacqua LM, Berul CI. Implications of implantable cardioverter defibrillator therapy in congenital heart disease and pediatrics. J Cardiovasc Electrophysiol 2004;15:72–76. 17. Silka MJ, Kron J, Dunnigan A, Dick M II; The Pediatric Electrophysiology Society. Sudden cardiac death and the use of implantable cardioverterdefibrillators in pediatric patients. Circulation 1993;87:800–807. 18. DeWitt ES, Triedman JK, Cecchin F, Mah DY, Abrams DJ, Walsh EP, Gauvreau K, Alexander ME. Time dependence of risks and benefits in pediatric primary prevention icd therapy. Circ Arrhythm Electrophysiol 2014;7: 1057–1063. 19. Atallah J, Erickson CC, Cecchin F, Dubin AM, Law IH, Cohen MI, Lapage MJ, Cannon BC, Chun TU, Freedenberg V, Gierdalski M, Berul CI. Multiinstitutional study of implantable defibrillator lead performance in children and young adults: results of the Pediatric Lead Extractability and Survival Evaluation (PLEASE) study. Circulation 2013;127:2393–2402. 20. Kurkciyan I, Meron G, Behringer W, Sterz F, Berzlanovich A, Domanovits H, Mullner M, Bankl HC, Laggner AN. Accuracy and impact of presumed cause in patients with cardiac arrest. Circulation 1998;98:766–771.
CLINICAL PERSPECTIVES Implantable cardioverter-defibrillators (ICDs) have shown improved survival in adult patients with heart failure. The risk of sudden cardiac death (SCD) in pediatric patients awaiting heart transplantation is low, and the role of ICD in this group is largely unknown. In this large retrospective study, we report that the rate of SCD is low but is highest in young patients with congenital heart disease. One in 12 children listed for heart transplant had an ICD present at listing. The presence of ICD at listing was not associated with reduced SCD, all-cause mortality, or transplantation. The results of this study make prophylactic use of ICDs in this high-risk population less appealing. Data extrapolation from adult studies should not be performed owing to the differences in underlying disease pathology, incidence of SCD, wait-list times, and listing status in pediatric patients.
Introduction Although implantable cardioverter-defibrillator (ICD) use has been extensively validated in preventing sudden cardiac death (SCD) in adult patients with end-stage heart failure Dr El-Assaad and Dr Al-Kindi contributed equally to this work. This work was supported in part by the Health Resources and Services Administration (contract no. 234-2005-37011C). The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. Address reprint requests and correspondence: Dr Peter F. Aziz, Division of Pediatric Cardiology, Cleveland Clinic Children’s, Cleveland Clinic Lerner College of Medicine, 9500 Euclid Avenue, Cleveland, OH 44195. E-mail address: [email protected].
1547-5271/$-see front matter B 2015 Heart Rhythm Society. All rights reserved.
and dilated cardiomyopathy (HR 0.32; 95% CI 0.20–0.52; P o .001) were associated with lower SCD risk, while younger age at listing (HR 0.94 per year; 95% CI 0.90–0.98; P ¼ .003) was associated with higher SCD risk. ICD at listing was not associated with reduced SCD (P ¼ .12), all-cause mortality, or delisting (P ¼ .57). CONCLUSION In pediatric patients listed for heart transplantation, the risk of SCD remains low and does not differ between patients with and without an ICD at listing. KEYWORDS Pediatric; Heart transplant; Sudden cardiac death; Implantable cardioverter-defibrillator ABBREVIATIONS BMI ¼ body mass index; CHD ¼ congenital heart disease; CI ¼ confidence interval; DCM ¼ dilated cardiomyopathy; ECMO ¼ extracorporeal membrane oxygenation; HCM ¼ hypertrophic cardiomyopathy; HR ¼ hazard ratio; ICD ¼ implantable cardioverter-defibrillator; RCM ¼ restrictive cardiomyopathy; SCD ¼ sudden cardiac death; UNOS ¼ United Network for Organ Sharing (Heart Rhythm 2015;12:2443–2448) I 2015 Heart Rhythm Society. All rights reserved.
awaiting heart transplantation,1–3 little is known about its benefit in the pediatric population, especially those listed for heart transplantation. With an incidence of SCD of approximately 1.3%4 in pediatric patients listed for heart transplantation, the use of prophylactic ICDs in these patients is controversial because of the reportedly high complication rates5 and high costs.4,6 A single study that investigated the role of ICD in pediatric patients with dilated cardiomyopathy (DCM) suggested a sustained survival benefit with low complication rate. Dubin et al7 performed a multicenter retrospective review of ICD databases in 28 pediatric patients awaiting heart transplant and demonstrated a high incidence of appropriate ICD discharges (85%) with a mean time of 6 http://dx.doi.org/10.1016/j.hrthm.2015.07.036
2444 months from implantation to first shock. In the aforementioned study, the majority of patients had ICDs because of ventricular tachycardia or fibrillation and a small percentage had an ICD implanted secondary to poor myocardial function (17%). Because of limited and conflicting data7 on the role of ICDs in pediatric patients listed for heart transplant, we sought to investigate their role in preventing SCD and allcause mortality in this population. In addition, we sought to identify predictors of SCD in these patients in order to distinguish high-risk subgroups.
Methods Study population We conducted a retrospective cohort study using the United Network for Organ Sharing/Organ Procurement and Transplantation Network database (Health Resources and Services Administration). UNOS is a comprehensive online database that records details on all patients listed for solid organ transplantation in all centers across the United States. Deidentified data are obtained from the Organ Procurement and Transplantation Network contracted with the Health Resources and Services Administration. Data are collected at different time points: at listing (Transplant Candidate Registration Form), before transplantation (Transplant Recipient Registration Form), and regularly after transplantation (Transplant Recipient Follow-Up Form). The OPTN/UNOS Transplant Registry includes information on patient demographic characteristics, causes of cardiomyopathy, devices present, causes of removal from the waiting list, hemodynamics, and other peritransplantation and long-term data. The registry is continuously audited, and quality control is assured.8 The version used in this study recorded patients until June 30, 2014. We identified patients listed between January 2005 and June 2014 for heart transplantation and were 18 years or younger. All patients were followed from the time of listing until heart transplantation, death, or removal from the list. Heart transplant candidates were divided into 2 groups: those with ICD at listing and those without. The non-ICD group was used as the control group. Information was collected at the time of listing and before transplant by using a standardized transplant candidate registration form. We performed baseline comparisons between the 2 groups by using the following study variables: age, ethnicity, sex, UNOS status at listing, underlying cause of heart failure (DCM vs restrictive cardiomyopathy [RCM] vs hypertrophic cardiomyopathy [HCM] vs congenital heart disease [CHD] vs retransplant vs myocarditis), body mass index (BMI), baseline medications, and treatments including extracorporeal membrane oxygenation (ECMO), ventricular assist devices, ventilator use, hemodialysis, need for inotropes, and intra-aortic balloon pumps. Information on all-cause mortality, SCD, and transplantation rate was retrieved for analysis. The primary outcome was the cumulative rate of
Heart Rhythm, Vol 12, No 12, December 2015 SCD as defined by the UNOS database. Variables that were significantly different in univariable analysis were tested using a multivariable model to identify risk factors for SCD. Antiarrhythmic medications were not included in our model because data on this variable were not recorded in the database beyond 2007. The institutional review board at the Cleveland Clinic Foundation approved this study.
Statistical analysis Categorical variables were expressed as percentages and compared using Pearson’s χ2 test. Continuous variables were reported as mean ⫾ standard deviation and/or medians with interquartile ranges and compared using the independent t test. The Cox proportional hazards model was used to identify multivariable predictors of SCD. Kaplan-Meier analysis and log-rank test were used for unadjusted analyses. We selected and tested 15 potential predictors of SCD using the univariable analysis. We also performed a subgroup analysis based on the underlying etiology of heart failure, and adjustments were based on baseline differences between the ICD and the non-ICD group. All hazard ratios (HRs) were adjusted for baseline differences between the 2 groups. No assumptions were made for missing data because variables used in the models were available for 100% of the study population. All statistical analyses were performed using SPSS software, version 19.0 (IBM SPSS Statistics, IBM Corporation, Chicago, IL). All tests were 2 sided. Differences were considered statistically significant at P o .05.
Results A total of 5214 pediatric patients were identified. Patients with unknown ICD status (n ¼ 92, 1.7%) and those with UNOS status 7 (n ¼ 83, 1.6%) were excluded from the study, leaving a study population of 5072. Table 1 lists the baseline demographic and clinical characteristics in the study population, ICD group, and non-ICD group. Overall, the majority of patients were White (n ¼ 2783, 55%) male (n ¼ 2802, 55%) patients with mean age of 6.2 ⫾ 6.5 years. The major causes of end-stage heart failure were CHD (n = 2297, 45%) and DCM (n = 1656, 33%). Of all patients listed for transplant with a known ICD status, 426 patients (8.3%) had ICDs at listing.
ICD during listing Of the patients without an ICD at listing, 44 patients (0.9%) eventually received ICDs during their listing time. This was an intention-to-treat analysis; thus, these patients were analyzed in the non-ICD group. Of the patients who did not have an ICD at listing, the number (percentage) of patients who received an ICD after listing was as follows: 24 (0.75%) of Ia, 6 (1.1%) of Ib, 11 (1.3%) of II, 12 (0.55%) of CHD, 3 (1.4%) of myocarditis, 3 (3.6%) of HCM, 15 (1%) of DCM, and 6 (1.8%) of retransplant.
El-Assaad et al Table 1
ICD Use Before Pediatric Heart Transplant
2445
Baseline demographic and clinical characteristics in all patients, ICD group, and non-ICD group
Characteristic Age (y) Sex: male Ethnicity White African American Hispanic Asian Other* Cause of heart failure Dilated cardiomyopathy Myocarditis Restrictive cardiomyopathy Retransplant Hypertrophic cardiomyopathy Congenital heart disease Other UNOS status at listing 1A 1B 2 Ionotropes Treatment VAD ECMO IABP Hemodialysis Ventilator BMI (kg/m2)
Total population (N ¼ 5072)
ICD group (n ¼ 426)
Non-ICD group (n ¼ 4646)
6.2 ⫾ 6.5 2802 (55)
13.1 ⫾ 5.1 253 (59)
5.5 ⫾ 6.2 2549 (55)
2783 (55) 1059 (21) 917 (18) 176 (3.5) 137 (3)
243 92 66 14 11
(57) (22) (16) (3) (3)
2540 (55) 967 (21) 851 (18) 162 (3.5) 126 (3)
1656 (33) 228 (4.5) 267 (5) 348 (7) 135 (3) 2297 (45) 141 (3)
182 16 30 21 51 104 22
(43) (4) (7) (5) (12) (24) (5)
1474 (32) 212 (5) 237 (5) 327 (7) 84 (2) 2193 (47) 119 (3)
3390 (67) 645 (13) 1037 (20) 2279 (44.9)
173 81 172 148
(41) (19) (40) (34.7)
3217 (69) 564 (12) 856 (19) 2131 (46)
o.001
381 (7.5) 499 (9.8) 19 (0.4) 114 (2) 1124 (22) 17.4 ⫾ 5.1
33 (8) 12 (2.8) 3 (0.7) 7 (2) 26 (6) 21.3 ⫾ 5.6
348 (7.5) 487 (10.5) 16 (0.3) 107 (2) 1098 (24) 17.1 ⫾ 4.9
.85 o.001 .21 .49 o.001 o.001
P o.001 .075 .68
o.001
o.001
Categorical and continuous data are presented as n (%) and mean ⫾ SD, respectively. BMI ¼ body mass index; ECMO ¼ extracorporeal membrane oxygenation; IABP ¼ intra-aortic balloon pump; ICD ¼ implantable cardioverter-defibrillator; UNOS ¼ United Nation Organ Sharing; VAD ¼ ventricular assist device. * Other: American Indian, Alaskan Native, Native Hawaiian, Pacific Islander, multiracial. Bold P values represent statistical significance.
Baseline comparison between the control and the study group
Of the 426 patients who had an ICD at listing, 76.5% received heart transplantation whereas 8.5% died while waiting (2.3% died of SCD). In the control group, 67.4% of the transplant candidates received heart transplants, whereas 11.1% died while waiting (3.1% died of SCD).
Age, heart failure etiology, UNOS status at listing, BMI, need for intravenous inotropes, ECMO, and ventilator use were different between the 2 groups. In comparison to the control group, those with ICD at listing were older (13.1 ⫾ 5.1 years vs 5.5 ⫾ 6.2 years), less likely to have CHD (24% vs 47%), more likely to have HCM (12% vs 2.0%), and less likely to require inotropes (35% vs 46%), ECMO (2.8% vs 10.5%), and ventilator support (6.0% vs 24%). Furthermore, patients with ICD were less likely to be listed with status 1A (41% vs 69%) and more likely to be listed with status 2 (40% vs 19%). ICD recipients were also found to have a higher BMI (21.3 ⫾ 5.6 kg/m2 vs 17.1 ⫾ 4.9 kg/m2). There were no statistical differences in other studied variables: sex, ethnicity, use of ventricular assist devices, intra-aortic balloon pumps, or hemodialysis (Table 1).
Most SCD events (90%) occurred within 6 months of listing. In Cox proportional multivariable analysis, status 1B (HR 0.52; 95% confidence interval [CI] 0.29–0.95; P ¼ .03), myocarditis (HR 0.19; 95% CI 0.05–0.77; P ¼ .02), RCM (HR 0.19; 95% CI 0.05–0.76; P ¼ .02), and DCM (HR 0.32; 95% CI 0.20–0.52; P o .001) were associated with lower SCD risk while ventilator use (HR 1.73; 95% CI 1.17–2.56; P ¼ .006) and younger age at listing (HR 0.94 per year; 95% CI 0.90–0.98; P ¼ .003) were associated with higher SCD risk (Table 2).
Follow-up and outcome
Cumulative mortality and ICD use
Over a median follow-up of 1.8 months (interquartile range 0.6–4.8 months), 68% underwent heart transplantation, 10.9% died while waiting, and the remaining are still alive. Only 3.1% died of SCD, while the remainder of the population died of either cardiac or noncardiac etiology.
ICD was not associated with decreased all-cause mortality; delisting for deterioration, SCD, or transplant; or delisting for improvement (P ¼ .12, P ¼ .73, and P ¼ .57, respectively) (Figure 1). In subgroup analyses, ICD was not associated with decreased SCD rates in patients with
Risk factors for SCD before heart transplantation
2446
Heart Rhythm, Vol 12, No 12, December 2015
Table 2 Multivariate predictors of sudden cardiac death using the Cox proportional hazards model Variable ICD at listing ECMO at listing IV inotropes at listing Initial UNOS status 1A 1B 2 Age at listing (per year) Underlying cause Congenital heart disease Myocarditis Restrictive cardiomyopathy Hypertrophic cardiomyopathy Dilated cardiomyopathy Retransplant Other Ventilator use List year BMI
Adjusted HR (95% CI) 1.74 (0.87–3.74) 0.78 (0.41–1.47) 0.95 (0.66–1.38)
Multivariable P value .12 .45 .78
Reference Reference 0.52 (0.29–0.95) .033 0.64 (0.38–1.07) .09 0.94 (0.90–0.98) .003 Reference Reference 0.19 (0.05–0.77) .02 0.19 (0.05–0.76) .02 0.36 (0.09–1.51) .16 0.32 (0.20–0.52) o.001 1.49 (0.82–2.69) .19 0.59 (0.22–1.60) .30 1.73 (1.17–2.56) .006 0.94 (0.89–1.00) .06 0.99 (0.95–1.04) .78
BMI ¼ body mass index; CI ¼ confidence interval; ECMO ¼ extracorporeal membrane oxygenation; HR ¼ hazard ratio; ICD ¼ implantable cardioverter-defibrillator; IV ¼ intravenous; UNOS ¼ United Network for Organ Sharing.
DCM (P ¼ .62), RCM (P ¼ .95), myocarditis (P ¼ .92), retransplant (P ¼ .12), HCM (P ¼ .96), and CHD (P ¼ .15) (Table 3).
Discussion This is the largest study that investigates the risk of SCD in pediatric patients awaiting heart transplantation. We report,
for the first time, that the risk of SCD is not statistically different between pediatric cardiac transplant candidates who receive ICDs at or before listing and those who do not. Although these results are compelling, we are cautioned in concluding that ICD implantation for pediatric patients awaiting heart transplantation does not improve survival based on caveats discussed below. We also report that the overall risk of SCD in this patient population remains low at 3.1%. Our findings differ from adult studies that have consistently shown survival benefit from ICD as primary prevention therapy in patients with end-stage heart failure awaiting heart transplantation.1–3,9–11 This difference may largely be explained by the lower incidence of SCD in pediatric population in comparison to adults. Unlike pediatric patients, the leading cause of death in adult cardiac transplant candidates is SCD. In a retrospective adult study, SCD was found to be the major cause of death accounting for 66.7% of the total deaths.2 On the contrary, the leading cause of death in our population was non-SCD and SCD occurred only in 28% of the total deaths. Rhee et al4 reported an even lower rate of SCD in the pediatric heart transplant study (7.6% of the total deaths). Furthermore, the different survival benefit between adult and pediatric patients could be attributed to the difference in the underlying disease substrate. Based on our cohort, ischemic cardiomyopathy accounted for only a minority of patients (n ¼ 18 [o1.0%]), of whom none had a sudden cardiac event, as compared with 30%–40% of adult patients with heart transplant.1,3 The heterogeneous environment created by the presence of normal and abnormal scar tissue alters the electrophysiological properties of the myocardium, leading to the development of a reentry circuit,12 which predisposes to fatal ventricular arrhythmias and SCD.
Figure 1 Adjusted cumulative incidence of hazard ratio of sudden cardiac death (A) and all-cause mortality (B) before heart transplantation in implantable cardioverter-defibrillator (ICD) recipients and control subjects.
El-Assaad et al Table 3
ICD Use Before Pediatric Heart Transplant
Incidence of SCD at 6 months in subgroups Cumulative incidence of SCD (%)
Variable Initial UNOS status 1A 1B 2 Underlying cause Dilated cardiomyopathy Myocarditis Restrictive cardiomyopathy Retransplant Hypertrophic cardiomyopathy Congenital heart disease
Overall
ICD
Non-ICD
P
5.64 2.13 2.83
5.08 0 2.29
5.67 2.43 2.93
.32 .82 .06
1.94 1.52 1.14 4.76 2.34 6.94
0 0 0 6.45 0 8.28
2.19 1.64 1.29 4.64 3.85 6.87
.62 .92 .95 .12 .96 .15
ICD ¼ implantable cardioverter-defibrillator; SCD ¼ sudden cardiac death.
In addition, the discrepancy in survival benefit can be partially explained by the urgent listing status and shorter wait-list time in children compared with adults. In fact, approximately 70% of pediatric patients spent less than 3 months on the waiting list as compared with only 49.2% of adult patients who spent less than 1 year.10 The median waiting time to transplantation in our cohort was 1.6 months (interquartile range 0.6–3.7 months) as compared with 8.1 months in adults.2 The benefit of ICD therapy is a function of time, and shorter wait-list times may disguise that benefit. Lastly, pediatric patients are more likely to be listed with a UNOS status 1A than are adults.10 These patients are likely to be hospitalized until they receive a heart transplant and thus may receive appropriate treatments of ventricular arrhythmias without the need for ICDs. On the basis of a prospective randomized clinical trial examining the role of ICD in adult patients with heart failure, the American College of Cardiology/American Heart Association/Heart Rhythm Society committee has published clear guidelines for this specific cohort of patients.13 The Suden Cardiac Death in Heart Failure Trial (SCD-HeFT) trial in adults has convincingly demonstrated the favorable effects of ICD in primary prevention of SCD in patient with ischemic and non–ischemic DCM with New York Heart Association class II or III heart failure and a left ventricular ejection fraction of r35%.14 The International Society of Heart and Lung Transplantation guidelines state that in pediatric patients with DCM who have New York Heart Association class II or III and a left ventricular ejection fraction of o35%, physicians may consider ICD placement (class IIb, level of evidence C).15 However, there is lack of specific indications for pediatric patients with heart failure because of the paucity of prospective data in pediatric patients, which is due to the reportedly low incidence of SCD. The other important implication that warrants consideration in pediatric patients is the high complication rate due to ICD implantation. Multiple pediatric studies16,17 examining ICD role in patients who are at an increased risk of SCD have been conducted. A retrospective study18 that examined the incidence of appropriate shock delivery and adverse events
2447 found that in 140 pediatric patients who had ICDs for primary prevention, the percentage of adverse events was almost double the percentage of appropriate shocks over a mean follow-up of 4 years (36% vs 19%). Another multicenter retrospective study5 in pediatric ICD recipients reported a high frequency of appropriate as well as inappropriate shocks as high as 26% and 21%, respectively. We show that the incidence of SCD in pediatric patients is similar to that reported in previous studies.4,11 A cohort analysis that included 1803 pediatric patients with DCM estimated the 5-year incidence rate of SCD to be 2.4%.11 Rhee et al4 found that 1.3% of the total listed patients died suddenly. We also demonstrate that CHD is a strong predictor of SCD in accordance with previous studies.5,19 Furthermore, our study shows that patients who are younger at listing have a higher risk of SCD. However, these findings should be tempered by the fact that younger age at implantation has been found to be an independent predictor of lead failure, which is the main reason for inappropriate shock delivery.5,19 Interestingly, even in the high-risk population (those with CHD), ICD presence at listing was not associated with lower SCD rates.
Study limitations The findings of this study should be interpreted with caution because of significant database inherent limitations and loss of granularity. This is a retrospective study where data entry cannot be verified, and therefore the accuracy of the data is unknown. Although comprehensive, there are data points such as indications for ICD, cardiac function, and antiarrhythmic medications that impair our ability to adjust for confounding factors. A major limitation is the lack of information on the number of appropriate shocks delivered. Perhaps, patients with ICD at listing received appropriate shocks that prevented SCD, but given lack of data we were not able to adjust for this variable. It is plausible that patients with ICD had it implanted because of their suspected higher arrhythmogenic risk. Similarly, the lack of data on cardiac arrest events that received appropriate inpatient interventions in the non-ICD group may contribute to the lack of survival benefit in the ICD group. In addition, perhaps the most important clinical question for pediatric electrophysiologists is whether ICD implantation after listing affects survival. We could not analyze outcomes of patients who received ICDs after listing because of the lack of sufficient information on those patients. Data on ICD during listing is collected only on patients who receive transplantation using the Transplant Recipient Registration Form but not on patients who die or get delisted. Another significant limitation of the study is the lack of the specific underlying etiologies that led to SCD. The assumption in this study is that most of the patients had a presumed cardiac etiology; however, as in other analyses of cardiac arrests, it is hard to ascertain the exact etiology without autopsy results.20 If the primary etiology was nonarrhythmic, this may explain why the majority of the patients who had an ICD did not have improved survival. Despite these limitations, because of the
2448
Heart Rhythm, Vol 12, No 12, December 2015
rarity of SCD in pediatric patients awaiting heart transplantation, conducting a prospective study with high power may not be feasible.
Conclusion In pediatric patients listed for heart transplantation, the risk of SCD remains low and does not differ between patients with and without an ICD at listing. The utility of ICD implantation in these patients is masked by several confounders, which distinguishes pediatric patients from adults.
Acknowledgments We thank Mrs Penny Houghtaling, biostatistician in the Quantitative Health Department, Cleveland Clinic, Cleveland, Ohio for statistical contribution essential to this study. We would also like to thank Mrs Patricia Agatisa, research coordinator in the Department of Human Resources, Cleveland Clinic, Cleveland, Ohio for final manuscript editing essential to this study.
References 1. Ermis C, Zadeii G, Zhu AX, Fabian W, Collins J, Lurie KG, Sakaguchi S, Benditt DG. Improved survival of cardiac transplantation candidates with implantable cardioverter defibrillator therapy: role of beta-blocker or amiodarone treatment. J Cardiovasc Electrophysiol 2003;14:578–583. 2. Sandner SE, Wieselthaler G, Zuckermann A, Taghavi S, Schmidinger H, Pacher R, Ploner M, Laufer G, Wolner E, Grimm M. Survival benefit of the implantable cardioverter-defibrillator in patients on the waiting list for cardiac transplantation. Circulation 2001;104:I171–I176. 3. Frohlich GM, Holzmeister J, Hubler M, et al. Prophylactic implantable cardioverter defibrillator treatment in patients with end-stage heart failure awaiting heart transplantation. Heart 2013;99:1158–1165. 4. Rhee EK, Canter CE, Basile S, Webber SA, Naftel DC. Sudden death prior to pediatric heart transplantation: would implantable defibrillators improve outcome? J Heart Lung Transplant 2007;26:447–452. 5. Berul CI, Van Hare GF, Kertesz NJ, Dubin AM, Cecchin F, Collins KK, Cannon BC, Alexander ME, Triedman JK, Walsh EP, Friedman RA. Results of a multicenter retrospective implantable cardioverter-defibrillator registry of pediatric and congenital heart disease patients. J Am Coll Cardiol 2008;51:1685–1691. 6. Feingold B, Arora G, Webber SA, Smith KJ. Cost-effectiveness of implantable cardioverter-defibrillators in children with dilated cardiomyopathy. J Card Fail 2010;16:734–741.
7. Dubin AM, Berul CI, Bevilacqua LM, et al. The use of implantable cardioverterdefibrillators in pediatric patients awaiting heart transplantation. J Card Fail 2003;9:375–379. 8. Daily OP, Kauffman HM. Quality control of the OPTN/UNOS Transplant Registry. Transplantation 2004;77:1309–1310. 1309; author reply. 9. Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352: 225–237. 10. Colvin-Adams M, Smithy JM, Heubner BM, Skeans MA, Edwards LB, Waller C, Schnitzler MA, Snyder JJ, Israni AK, Kasiske BL. OPTN/SRTR 2012 annual data report: Heart. Am J Transplant 2014;14:113–138. 11. Pahl E, Sleeper LA, Canter CE, et al. Incidence of and risk factors for sudden cardiac death in children with dilated cardiomyopathy: a report from the pediatric cardiomyopathy registry. J Am Coll Cardiol 2012;59:607–615. 12. Ebinger MW, Krishnan S, Schuger CD. Mechanisms of ventricular arrhythmias in heart failure. Curr Heart Fail Rep 2005;2:111–117. 13. Epstein AE, Dimarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: executive summary. Heart Rhythm 2008;5:934–955. 14. Russo AM, Stainback RF, Bailey SR, Epstein AE, Heidenreich PA, Jessup M, Kapa S, Kremers MS, Lindsay BD, Stevenson LW. ACCF/HRS/AHA/ASE/ HFSA/SCAI/SCCT/SCMR 2013 appropriate use criteria for implantable cardioverter-defibrillators and cardiac resynchronization therapy: a report of the American College of Cardiology Foundation appropriate use criteria task force, Heart Rhythm Society, American Heart Association, American Society of Echocardiography, Heart Failure Society of America, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, and Society for Cardiovascular Magnetic Resonance. J Am Coll Cardiol 2013;61:1318–1368. 15. Kirk R, Dipchand AI, Rosenthal DN, et al. The International Society of Heart and Lung Transplantation guidelines for the management of pediatric heart failure: executive summary. J Heart Lung Transplant 2014;33:888–909. 16. Alexander ME, Cecchin F, Walsh EP, Triedman JK, Bevilacqua LM, Berul CI. Implications of implantable cardioverter defibrillator therapy in congenital heart disease and pediatrics. J Cardiovasc Electrophysiol 2004;15:72–76. 17. Silka MJ, Kron J, Dunnigan A, Dick M II; The Pediatric Electrophysiology Society. Sudden cardiac death and the use of implantable cardioverterdefibrillators in pediatric patients. Circulation 1993;87:800–807. 18. DeWitt ES, Triedman JK, Cecchin F, Mah DY, Abrams DJ, Walsh EP, Gauvreau K, Alexander ME. Time dependence of risks and benefits in pediatric primary prevention icd therapy. Circ Arrhythm Electrophysiol 2014;7: 1057–1063. 19. Atallah J, Erickson CC, Cecchin F, Dubin AM, Law IH, Cohen MI, Lapage MJ, Cannon BC, Chun TU, Freedenberg V, Gierdalski M, Berul CI. Multiinstitutional study of implantable defibrillator lead performance in children and young adults: results of the Pediatric Lead Extractability and Survival Evaluation (PLEASE) study. Circulation 2013;127:2393–2402. 20. Kurkciyan I, Meron G, Behringer W, Sterz F, Berzlanovich A, Domanovits H, Mullner M, Bankl HC, Laggner AN. Accuracy and impact of presumed cause in patients with cardiac arrest. Circulation 1998;98:766–771.
CLINICAL PERSPECTIVES Implantable cardioverter-defibrillators (ICDs) have shown improved survival in adult patients with heart failure. The risk of sudden cardiac death (SCD) in pediatric patients awaiting heart transplantation is low, and the role of ICD in this group is largely unknown. In this large retrospective study, we report that the rate of SCD is low but is highest in young patients with congenital heart disease. One in 12 children listed for heart transplant had an ICD present at listing. The presence of ICD at listing was not associated with reduced SCD, all-cause mortality, or transplantation. The results of this study make prophylactic use of ICDs in this high-risk population less appealing. Data extrapolation from adult studies should not be performed owing to the differences in underlying disease pathology, incidence of SCD, wait-list times, and listing status in pediatric patients.