Author + information
- Received July 28, 2015
- Revision received September 14, 2015
- Accepted September 21, 2015
- Published online December 15, 2015.
- Madhurmeet Singh, DO,
- Norman C. Wang, MD, MS,
- Sandeep Jain, MD,
- Andrew H. Voigt, MD,
- Samir Saba, MD and
- Evan C. Adelstein, MD∗ ()
- ↵∗Reprint requests and correspondence:
Dr. Evan C. Adelstein, Heart and Vascular Institute, University of Pittsburgh, 200 Lothrop Street, PUH B535, Pittsburgh, Pennsylvania 15213.
Background The wearable cardioverter-defibrillator (WCD) has emerged as a means of protecting patients with newly diagnosed nonischemic cardiomyopathy (NICM) or ischemic cardiomyopathy (ICM) against sudden cardiac death while awaiting re-evaluation of cardiac function.
Objectives This study sought to characterize the risk of appropriate WCD therapy in newly diagnosed NICM and ICM patients according to cardiomyopathy etiology in an independent study.
Methods Medical records of all patients prescribed a WCD between June 2004 and May 2015 at our institution (n = 639) were analyzed, focusing on 254 patients with newly diagnosed NICM and 271 patients with newly diagnosed ICM. Patients with a prior implantable cardioverter-defibrillator or sustained ventricular arrhythmias were excluded (n = 114). The primary endpoint was appropriate WCD therapy.
Results Median WCD wear time was 61 days (interquartile range [IQR]: 25 to 102 days) per patient and 22 h/day (IQR: 17 to 23 h/day). During 56.7 patient-years, 0 NICM patients received an appropriate WCD shock, whereas 3 (1.2%) received an inappropriate shock. During 46.7 patient-years, 6 (2.2%) ICM patients received an appropriate shock; 5 survived the episode, and 4 survived to hospital discharge. All 6 patients with an appropriate shock were male with QRS duration >120 ms. Two (0.7%) ICM patients received an inappropriate shock.
Conclusions In this independent, retrospective study, the risk of appropriate WCD therapies in patients with newly diagnosed NICM was minimal. Routine use of the WCD in this population should be prospectively evaluated. The risk of appropriate therapies in newly diagnosed ICM was comparable to that observed in prior observational studies.
Sudden cardiac death (SCD) is a devastating outcome in patients with cardiomyopathy of ischemic (ICM) or nonischemic (NICM) etiology. Current guidelines, based on data from several randomized clinical trials, recommend consideration of a prophylactic implantable cardioverter-defibrillator (ICD) for patients with left ventricular ejection fraction (LVEF) ≤35% (1–3). The recommended waiting period before ICD implantation for primary prevention varies from 40 to 90 days after cardiomyopathy diagnosis, depending on its etiology, presentation, treatment with maximally tolerated guideline-directed medical therapy (GDMT), and antecedent revascularization (4). During this waiting period, patients remain at risk of arrhythmic SCD. This risk may be mitigated with a wearable cardioverter-defibrillator (WCD), approved by the U.S. Food and Drug Administration in 2002. There is currently a single WCD manufacturer (ZOLL, Pittsburgh, Pennsylvania).
Routine WCD prescription has become widely accepted in patients with newly diagnosed NICM during the 90-day waiting period on GDMT before re-evaluation of left ventricular function, despite a dearth of data suggesting benefit of WCDs in this population. Use of the WCD in newly diagnosed ICM has been more thoroughly investigated, albeit not in a randomized fashion (5–7). We report long-term experience and outcomes with the WCD at a large academic tertiary care center in patients with a new diagnosis of NICM or ICM, using independently derived data.
This study is a retrospective review of all consecutive patients prescribed a WCD between June 1, 2004 and May 30, 2015 at the hospitals comprising the University of Pittsburgh Medical Center to which we had access to clinical data (n = 639). Patients with an explanted ICD awaiting reimplantation (n = 72), prior cardiac arrest unrelated to acute myocardial infarction (n = 10), or elevated risk of SCD for reasons other than ICM or NICM (e.g., possible channelopathy, n = 32) were excluded from this analysis. The remaining patients had either a newly diagnosed NICM (n = 254) or ICM (n = 271) with LVEF ≤35% by echocardiography. All patients were identified with ICM or NICM while hospitalized and were prescribed a WCD upon hospital discharge at the discretion of treating physician(s). Patient demographics and follow-up data were obtained from the electronic medical record. Patients were categorized as having NICM in the absence of prior coronary revascularization, major coronary artery stenosis ≥80%, definitive biomarker evidence of prior myocardial infarction, and imaging data suggesting prior infarction (8). The Institutional Review Board at the University of Pittsburgh approved this study.
All patients were prescribed maximally tolerated GDMT, including beta-adrenergic antagonists and inhibitors of the renin-angiotensin-aldosterone axis. Re-evaluation of LVEF by echocardiography or cardiac magnetic resonance imaging was performed after 40 days in patients with nonrevascularized acute myocardial infarction, and after 90 days in patients with coronary revascularization, new NICM diagnosis, or new ICM diagnosis without revascularization or acute myocardial infarction.
Patients were fitted with a WCD before hospital discharge and were counseled on its operation and the importance of wearing it. The ordering physician prescribed tachyarrhythmia rate cutoffs, usually with a ventricular tachycardia (VT) zone of 180 to 199 beats/min and a ventricular fibrillation (VF) zone ≥200 beats/min. Patient WCD shock data were obtained from the manufacturer and electronic medical record, and an electrophysiologist analyzed rhythm strips to determine their appropriateness. Patient compliance was assessed automatically by real-time monitoring.
Discrete variables are listed as absolute numbers and percentages and were compared using chi-square or Fisher exact tests, as deemed appropriate. Normally distributed continuous variables are listed as mean ± SD and were compared using Student t tests. Nonparametric continuous variables are listed as medians with interquartile ranges (IQRs), and were compared using the Mann-Whitney U test. All p values are 2-sided, and a p value ≤0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 22.0 (IBM, Armonk, New York).
Between June 2004 and May 2015, 525 WCDs were prescribed for newly diagnosed NICM (n = 254; 48%) and ICM (n = 271; 52%). Baseline characteristics for each group are noted in Table 1. Notably, nonsustained ventricular tachycardia (NSVT) was detected before WCD prescription during telemetry monitoring in 89 (35%) NICM and 108 (40%) ICM patients (p = 0.27). Use of WCDs increased during our decade-long experience (Figure 1). Specific clinical indications for WCDs within ICM and NICM groups are depicted in Figures 2 and 3, respectively.
The WCD was worn for a total of 37,765 patient-days (103 patient-years), with each patient using the WCD for a median of 61 days (IQR: 25 to 102 days), and a daily median compliance of 22 h (IQR: 17 to 23 h). Daily compliance was greater in ICM patients (median 23 vs. 21 h; p = 0.001), whereas NICM patients wore the WCD for a longer duration (median 71 vs. 53 days; p = 0.004).
During follow-up, 100 new NICM patients (39%) had an LVEF improvement to >35%, obviating consideration of a prophylactic ICD. Another 6 patients did not want an ICD. Other outcomes included lack of follow-up (n = 23), heart-lung transplantation (n = 1), ventricular assist device implantation (n = 2), persistent New York Heart Association functional class IV heart failure (n = 1), and death (n = 2). An additional 12 patients are currently awaiting LVEF recovery. Ultimately, 107 (42%) patients have undergone ICD implantation for primary prevention because of a persistently depressed LVEF.
Among ICM patients, 88 (32%) recovered their LVEF to >35% and did not receive a prophylactic ICD. Another 10 elected not to pursue device implantation. Other outcomes included lack of follow-up (n = 10), ventricular assist device implantation (n = 2), deferral of an ICD because of poor patient prognosis (n = 9), and death (n = 9). Eighteen patients are awaiting possible LVEF improvement. A similar percentage of ICM patients (n = 125, 46%) have been implanted with an ICD compared to NICM patients (p = 0.41).
No NICM patients (0 of 254; 0%) received an appropriate shock from the WCD during 20,711 patient-days (56.7 patient-years). Among ICM patients, 6 of 271 (2.2%) received an appropriate shock for VF during 17,056 patient-days (46.7 patient-years), which equates to 1 shock/2,843 patient-days (1 shock/7.8 patient-years). Four of these patients received a permanent ICD; 1 died from bradycardia after successful WCD defibrillation, and another died from complications 2 days after the WCD shock (Table 2). Another ICM patient had multiple recorded episodes of VT associated with pre-syncope, but deactivated the device several times before therapy was delivered and seeking medical attention. All ICM patients with an appropriate shock were male and had a QRS duration >120 ms and either right bundle-branch block or nonspecific intraventricular conduction delay. The mean time from initiating WCD use to the first appropriate shock was 34 ± 11 days (median 35 days, range 17 to 45 days).
In contrast, 3 of 254 (1.2%) NICM patients received an inappropriate shock, all of which were for artifacts misinterpreted as ventricular tachyarrhythmia. Two ICM patients (0.7%) received an inappropriate shock, 1 for rapidly conducted atrial fibrillation and the other for artifact.
In this study of patients with newly diagnosed cardiomyopathy prescribed a WCD, the incidence of appropriate shock(s) was 0.6% during a median wear time of 61 days, for a total of 103 patient-years, with no appropriate shocks delivered in new NICM patients during 56.7 patient-years. Six appropriate shocks were delivered by the WCD among newly diagnosed ICM patients, all of whom were male and had abnormal intraventricular conduction, although only 4 were discharged alive with an ICD (Central Illustration). Despite these low rates of appropriate WCD discharges, WCD prescriptions increased significantly during our 10-year experience.
Among newly diagnosed ICM patients, our data add to those from post-myocardial infarction studies in which the early risk of SCD is not insignificant (9). The 6 appropriate WCD shocks were delivered 34 ± 11 days after its prescription, somewhat longer than in a prior study (5). Yet the success of WCD defibrillation must be tempered by the immediate death of 1 of the 6 patients appropriately treated by the WCD, despite return of an organized rhythm, and another patient dying within 48 h of a WCD shock from complications. These deaths reduce the number of WCD “saves” to 4 of 271 (1.5%) and refocus attention on the DINAMIT (Defibrillator in Acute Myocardial Infarction Trial) and IRIS (Immediate Risk-Stratification Improves Survival) studies, demonstrating that ICDs decrease sudden, arrhythmic death, but do not diminish overall mortality (10,11). It is unclear whether the capacity to provide post-shock pacing, as occurs in ICDs, would alter such outcomes. Even in patients with an ICD, at least 25% of all deaths are sudden (12).
Data from previous studies investigating outcomes in patients with NICM suggest that the impact of defibrillators on reduction of SCD in patients during the first 90 days on GDMT is low (2,3). Although in DEFINITE (Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation), patients implanted with an ICD early (≤3 months) after NICM diagnosis demonstrated improved survival compared to patients without an ICD, this benefit was not observed early after implantation, and patient numbers were small (13). A recent retrospective study reported a low incidence of appropriate ICD shocks in patients who received an ICD before mandated waiting periods were complete (14).
Paradoxically, on the basis of our center’s experience, it appears that newly diagnosed NICM patients are prescribed the WCD as frequently as those with a newly diagnosed ICM, in whom published literature and our results indicate that the benefit is not negligible (6,7,15). The reason for this may be linked to factors such as the younger age of patients diagnosed with NICM and overestimation of the true risk of SCD in patients with a newly diagnosed NICM. Clinically, one of the more common reasons for prescribing a WCD in patients with newly diagnosed NICM is recognition of NSVT, which is often asymptomatic and incidentally noted during telemetry monitoring. Interestingly, despite a significant proportion (35%) of our NICM patient population with documented NSVT (likely an underestimate because of the variable intensity of monitoring), the incidence of SCD was 0%. Based on our data, there seems to be a disconnection between the clinician-perceived risk and the actual risk of SCD. Visceral reactions, incomplete knowledge of existing data, and concerns over litigation may color clinicians’ decisions to prescribe a WCD. The last concern may explain different WCD prescribing practices in Germany versus the United States (16). Prescribing a WCD may foster additional emotional trauma in newly diagnosed NICM patients; the WCD serves as a constant reminder of their illness, and patients often express concern over the cumbersome nature of the WCD (17).
Industry-sponsored publications have investigated WCD experience, but independent data assessing the WCD’s utility in patients with newly diagnosed cardiomyopathy, particularly NICM, have been lacking. No study has validated the role of WCDs in preventing early SCD compared to GDMT alone. The WEARIT-I/BIROAD registry enrolled 289 patients at high risk of SCD but who were not able to receive an ICD (17). The exact proportions of patients within pre-specified categories of SCD risk were not published. The risk of appropriate WCD shocks was 2.8%, although 2 of 8 patients who received a shock were not successfully defibrillated, and 6 (2.1%) patients received an inappropriate shock. Among WEARIT patients, 30% discontinued WCD use because of comfort and lifestyle issues, and in BIROAD, 11% discontinued the WCD. In the WEARIT II registry (18), 22 (1.1%) of 2,000 patients with ICM (40%), NICM (46%), or congenital heart disease (13%) were shocked for ventricular tachyarrhythmias, and the inappropriate shock rate was 0.5%. The prospective, randomized VEST (Vest Prevention of Early Sudden Death Trial) continued enrolling post-myocardial infarction patients with LVEF ≤35% until September 2015. In a published report, using data provided by ZOLL, Chung et al. (15) reported an overall incidence of 1.7% for appropriate shocks in a nationwide WCD registry that included primary prevention patients, secondary prevention patients, and patients with an explanted ICD. There were 4 VT/VF events in 546 new NICM patients who wore the WCD for 57 ± 59 days, corresponding to 1 appropriate shock over 7,712 patient-days. One NICM patient wearing a WCD died from VT/VF, but it is not clear whether this patient was shocked. The authors reported 12 patients shocked appropriately among 584 recent myocardial infarction or coronary artery bypass graft patients using the WCD for 47 ± 41 days, corresponding to 1 appropriate shock over 2,302 patient-days. Three ICM patients died from VT/VF while wearing the WCD, but it is unclear whether shocks were delivered. Another study reported 133 appropriate patient shocks among 8,678 post-myocardial infarction patients prescribed a WCD for 69 ± 61 days, yielding 1 shock per 4,502 patient-days (5).
Interestingly, in our study, all 3 shocks in patients with NICM and 1 of 2 shocks in ICM patients were delivered for artifacts misinterpreted as ventricular tachyarrhythmias. Although the WCD allows the patient to suppress a shock if he or she is conscious and, therefore, presumably not hemodynamically compromised, it is not clear why this did not occur in the patients with inappropriate shocks. One possibility is poor understanding of device function by the patient or patient reluctance to abort the shock out of fear of interfering with device function.
The cost associated with WCD use is substantial; the estimated rental cost is approximately $3,300/month (manufacturer data). During the past 10 years, approximately $2.2 million were spent on providing coverage for the WCD in patients with NICM at our center. Better predictors of early SCD in patients with newly diagnosed NICM need to be identified, instead of prescribing a WCD for all such patients, as is quickly becoming the standard of care. Larger, randomized trials should be pursued to address this issue. The increasing rates of WCD use over time (Figure 1) suggest that the WCD may be preventing inappropriate ICD implantations, as one-third of patients in our study experienced LVEF improvement and were therefore spared an ICD. In this manner, the benefit of WCDs cannot be based wholly on the incidence of appropriate shocks. Whether the cost savings of reduced inappropriate ICD implantations exceeds the cost of prescribing WCDs also requires additional study.
To our knowledge, this is the first study specifically investigating outcomes using the WCD in patients with newly diagnosed ICM and NICM that does not rely wholly on the manufacturer’s data. Prior WCD studies have focused on patients with ICM and have relied on the manufacturer’s database for data extraction. Although the present study is relatively small, independent investigators abstracted the data, and we have investigator-proven WCD indications and heart failure etiologies. The present study encompasses a heterogeneous group of patients and spans the use of WCDs over a decade. It is limited, however, because of its design as a retrospective chart review, and we cannot ascertain whether WCD use improves mortality compared to nonuse. We also cannot determine whether there were differences in new NICM or ICM patients prescribed a WCD compared to those not referred for a WCD at our institution. Additionally, some patients were lost to follow-up. Finally, although there were no appropriate shocks in newly diagnosed NICM patients in the present study, this does not imply that the risk of ventricular arrhythmias in this population is zero.
The practice of prescribing a WCD for all patients with newly diagnosed NICM is of limited utility, even in the presence of nonsustained ventricular arrhythmias. Further trials should be undertaken before this practice becomes the standard of care. In contrast, patients with newly diagnosed ICM have a small but significant risk of ventricular tachyarrhythmias, and the WCD may mitigate this risk.
COMPETENCY IN MEDICAL KNOWLEDGE: The risk of sustained ventricular tachyarrhythmias requiring therapy from a WCD is minimal in patients with newly diagnosed NICM and greater, but still small, in patients with newly diagnosed ICM.
TRANSLATIONAL OUTLOOK: Further studies are needed to identify optimal candidates for WCDs and improve the performance characteristics of these devices.
Dr. Wang receives limited research support from Boston Scientific. Dr. Jain has received limited research support from Medtronic. Dr. Saba has received research support from Boston Scientific, Medtronic, and St. Jude Medical. Dr. Adelstein has received limited research support from Medtronic and St. Jude Medical. Drs. Singh and Voigt have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- guideline-directed medical therapy
- implantable cardioverter-defibrillator
- ischemic cardiomyopathy
- left ventricular ejection fraction
- nonischemic cardiomyopathy
- nonsustained ventricular tachycardia
- sudden cardiac death
- ventricular fibrillation
- ventricular tachycardia
- wearable cardioverter-defibrillator
- Received July 28, 2015.
- Revision received September 14, 2015.
- Accepted September 21, 2015.
- American College of Cardiology Foundation
- Epstein A.E.,
- DiMarco J.P.,
- Ellenbogen K.A.,
- et al.
- Epstein A.E.,
- Abraham W.T.,
- Bianco N.R.,
- et al.
- Zishiri E.T.,
- Williams S.,
- Cronin E.M.,
- et al.
- Goldenberg I.,
- Moss A.J.,
- Hall W.J.,
- et al.,
- MADIT-CRT Executive Committee
- Mitchell L.B.,
- Pineda E.A.,
- Titus J.L.,
- et al.
- Kadish A.,
- Schaechter A.,
- Subacius H.,
- et al.
- Chung M.K.,
- Szymkiewicz S.J.,
- Shao M.,
- et al.
- Adler A.,
- Halkin A.,
- Viskin S.
- ↵Kutyifa V, Goldenberg I, Zareba W, et al. Results from the prospective registry of patients using the wearable defibrillator (WEARIT-II registry). Paper presented at: European Society of Cardiology Congress; September 3, 2014; Barcelona, Spain.