Author + information
- Received July 17, 2017
- Revision received September 18, 2017
- Accepted September 27, 2017
- Published online February 19, 2018.
- Jordan S. Leyton-Mange, MDa,
- William J. Hucker, MD, PhDa,
- Nino Mihatov, MDa,
- Matthew Reynolds, MD, MScb,
- Christine Albert, MD, MPHc,
- Steven A. Lubitz, MD, MPHa,d and
- David J. Milan, MDa,∗ ()
- aCardiovascular Research Center, Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts
- bDepartment of Cardiology, Lahey Hospital and Medical Center, Burlington, Massachusetts
- cCenter for Arrhythmia Prevention, Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
- dProgram in Medical Population Genetics, Broad Institute of Harvard and Massachusetts Institute of Technology, Cambridge, Massachusetts
- ↵∗Address for correspondence:
Dr. David J. Milan, Cardiovascular Research Center, Massachusetts General Hospital, 55 Fruit Street, GRB-109, Boston, Massachusetts 021241.
Objectives This study sought to characterize the experience in a cohort of patients prescribed a wearable cardioverter-defibrillator (WCD) over a 2-year interval at 2 academic medical centers.
Background The WCD is available for patients felt to be at high risk of sudden cardiac death. However, there is a lack of randomized data to guide its use and prescribing patterns vary.
Methods We retrospectively reviewed indications and therapies of all WCD prescriptions over a 2-year period from 2 large academic medical centers. Data on compliance and treatment events of patients wearing the WCD were reviewed.
Results Among the 147 patients prescribed a WCD, 80% were male with an age of 59 ± 14 years. The WCD was prescribed for the following reasons: primary prevention in the setting of a left ventricular ejection fraction ≤35% (53%), secondary prevention when an implantable cardioverter-defibrillator was not implanted (16%), implantable cardioverter-defibrillator explantation (23%), and other high-risk scenarios for arrhythmic sudden death (9%). The median wear duration was 50 days (interquartile range [IQR]: 25 to 85 days) with a median of 21.0 h of wear per day (IQR: 15.0 to 22.8 h). High-voltage treatment was delivered in 3 separate patients, 2 of whom died. The third patient received 3 WCD shocks without restoration of a perfusing rhythm and ultimately was resuscitated by emergency responders. No patients received inappropriate therapies.
Conclusions Events requiring therapy were rare and no lives were directly saved by the WCD. Future efforts are needed to improve identification of patients most likely to benefit from a WCD.
The wearable cardioverter-defibrillator (WCD), manufactured by ZOLL (Pittsburgh, Pennsylvania), is available for patients felt to be at high risk of sudden cardiac death (SCD). There is general acceptance for its use in circumstances when there is both an implantable cardioverter-defibrillator (ICD) indication and concomitant but temporary contraindication, such as during the interval before reimplantation after removal of an infected ICD (1,2). However, the device is being used more frequently in circumstances when the risk of SCD remains uncertain over the long term, such as during the early “window” period following myocardial infarction (MI) or newly diagnosed cardiomyopathy (3), for which ICD therapy is not indicated and prior studies have not demonstrated a benefit to early ICD implantation (4,5). In these scenarios, the WCD provides protection from SCD prior to final risk stratification. The American Heart Association recently published recommendations regarding WCD use; however, there are limited data available to guide patient selection (2).
Ideally, WCD use would be informed by randomized trial data such as that anticipated from VEST (Vest Prevention of Early Sudden Death Trial) (NCT01446965), but given the availability of the WCD outside the trial as an approved device, enrollment has proven challenging. In the absence of unbiased randomized data, analysis of WCD use in contemporary clinical practice may provide helpful insight into the effective use of this therapy (6–8). Several studies have examined prescribing patterns and event rates by analyzing data from the U.S. and German WCD registries (3,9–12). Studies that also incorporate review of the primary medical records provide additional perspective and a greater granularity of data in respect to prescription indications and compliance. In this study, we investigated all WCD prescriptions over a 2-year period from 2 large tertiary care centers by primary chart review to evaluate patient outcomes and WCD compliance using both the electronic medical records and manufacturer data.
All patients at the Massachusetts General Hospital and the Brigham and Women’s Hospital prescribed a WCD from January 1, 2012, to December 31, 2013, were identified for the purposes of this study (n = 147). The electronic medical records were independently reviewed by 3 study investigators, and the primary prescribing indications and respective patient outcomes were adjudicated by consensus. Inpatients for whom WCD prescriptions were filed but were not discharged with the device were excluded from further analysis (n = 8). Information on event detection, wear duration, hourly wear, device therapies, and rhythm tracings were provided by ZOLL). Details regarding ICD implantation and vital status were determined for patients for whom follow-up information was available from the longitudinal electronic medical records (n = 111). Wear duration was defined as the time between the first day of WCD wear and the final day of wear. The prescription duration was defined as the time from the onset of the “wear duration” until either death or documented clinician instructions to stop wearing the WCD (e.g., ICD implantation, demonstration of ejection fraction improvement). Average daily wear was computed as the cumulative wear hours divided by the wear duration. Daily compliance was computed as the proportion of distinct calendar wear days divided by the prescription duration. Total compliance was computed as the cumulative wear hours divided by the prescription duration (Online Figure 1). Statistical comparisons were made with linear regression, or the Wilcoxon rank-sum or Pearson chi-squared tests with α = 0.05, adjusted for multiplicity where appropriate with a Bonferroni correction. All research was approved by the Partners Review Board representing both institutions.
We identified 147 prescriptions for the WCD across our institutions. The majority were male (118, 80%), with an average age of 59 ± 14 years (Table 1). Sixteen (11%) had a history of ventricular tachycardia (VT)/ventricular fibrillation (VF) and 104 (71%) had congestive heart failure. Prescription indications included primary prevention for a cardiomyopathy with a left ventricular ejection fraction (LVEF) ≤35% (53%), secondary prevention (16%), ICD explantation (23%), and patients felt to be at otherwise high risk for SCD (9%) (Figure 1). Overall, there was 87% concordance between the indications known to the manufacturer and those adjudicated by chart review. There were 78 WCD prescriptions for primary prevention of SCD: 31 for newly diagnosed ischemic cardiomyopathy (ICM) and LVEF ≤35% in the waiting period prior to assessing ICD candidacy, 37 in the setting of new non-ICM with LVEF ≤35% in the waiting period prior to assessing ICD candidacy, and 10 in patients with established cardiomyopathy and an indication for a primary prevention ICD, but at least 1 simultaneous contraindication. Additional high-risk features listed in Online Table 1, such as the presence of nonsustained VT or a family history of sudden death, were documented in the medical records in 71% of the patients wearing a WCD for a new cardiomyopathy and may have further motivated the prescription of a WCD by the treating clinicians.
There were 23 WCD prescribed for secondary prevention indications including for 9 patients with an incomplete work-up to assess reversible causes of the arrest. Thirty-four WCD were prescribed for patients after ICD explantation during an interval before reimplantation. Finally, 12 were prescribed for patients who were considered to be at otherwise high risk for SCD. Characterizing details regarding the patients described are listed in Table 2.
Of the 147 prescribed WCD, 8 prescriptions were discontinued prior to wear because of transitioning goals of care (n = 3) or death (n = 1) after unforeseen clinical deterioration, leaving against medical advice (n = 1), re-evaluation of the indication following LVEF recovery (n = 1), inability to understand operation of the WCD after fitting (n = 1), and implantation of an ICD prior to discharge (n = 1). These patients were removed from further analysis. The wear duration varied by prescribing indication among the remaining 139 patients (Figure 2). The median wear time was 50 days (interquartile range [IQR]: 25 to 85 days; range 0 to 356 days), with the longest wear durations for primary prevention indications with wide variation within each group. The median daily wear was 21.0 h/day (IQR: 15.0 to 22.8 h/day) on days the patients wore the device. One-third of patients wore the device for fewer than 16.5 h daily (Figure 3A). Of note, 2 patients never wore the WCD and 3 additional patients wore it for < 1 day.
Information pertaining to the WCD prescription duration was available for 111 of 139 patients. Patients wore the WCD for a median of 97% (IQR: 60% to 100%) of days during the prescription duration, with notably one-third of patients wearing the WCD for fewer than 75% of the prescription duration (daily compliance) (Figure 3B). The patients in the lowest tertile for daily compliance also wore the WCD for fewer hours per day with a median wear time of 10.2 h/day (tertile 2: 21.0 h/day, tertile 3: 22.9 h/day, p ≤ 0.002 for all group comparisons). Similar to prior reports, average daily wear was greater in patients with the longer wear durations (p < 0.001 between all tertiles). However, this relationship was no longer evident when stratified by prescription duration (p = NS for all tertiles) and patients with longer prescription durations were more likely to wear the WCD for fewer hours daily (<16 h daily: tertile 1: 14%, tertile 2: 32%, tertile 3: 46%, p < 0.01 by chi-square test; <12 h daily: 8% vs. 24% vs. 35% respectfully, p = 0.02 by chi-square test). Patients with the longest prescriptions also had worse daily compliance compared with those with the shortest (p = 0.006 tertiles 1 vs. 3) and were more likely to wear for fewer than 50% (p = 0.005) or 70% (p = 0.01) of days during the prescription duration. We found no relationship between total compliance (cumulative wear hours over the prescription duration) and age by linear regression (R2 = 0.016). Stratifying between patients with <70% and ≥70% total compliance, there was no difference between patients cared for between our institutions (Massachusetts General Hospital vs. Brigham and Women’s Hospital, p = 0.93). Female patients were less likely to be <70% compliant than were men (22% vs. 46%, p = 0.03). Patients wearing the WCD for primary prevention indications were more likely to be <70% compliant when compared with patients with nonprimary prevention indications (49% vs. 29%, p = 0.03).
Of a total of 139 patients, 29 (21%) were lost to follow-up. The majority (n = 24) had documented plans to receive follow-up care elsewhere, including 12 who were prescribed a WCD after ICD explantation and had follow-up elsewhere due to referral patterns for the procedure. Four patients died during the WCD prescription period prior to the decision regarding an ICD (3%) (Table 3). Among nondeceased patients for whom follow-up was available, ICD were eventually implanted in 54 of 106 patients (51%) (Table 3). ICD were also recommended in 12 additional patients who either declined implantation or did not attend their ICD pre-procedure appointment. Among the 61 patients with primary prevention indications, 26 (43%) experienced LVEF (n = 25) or symptomatic (n = 1) improvement obviating a primary prevention ICD indication. The remaining 34 patients with primary prevention indications either received (n = 26) or declined (n = 8) an ICD. Among 17 WCD patients with secondary prevention indications, 2 had reversible causes treated, respectively, with coronary revascularization and VT ablation in a structurally normal heart. The remaining 15 patients either received (n = 14) or declined (n = 1) an ICD. In contrast, none of the patients wearing a WCD in the “otherwise high-risk” category received an ICD at the end of the WCD-wearing period.
In aggregate, the devices detected over 11,000 high ventricular rate events over a wear time of 7,888 person-days, a median of 0.42 arrhythmic events per person-day (IQR: 0.14 to 1.60), which was similar across indications (p > 0.10 for all). In this cohort, 3 patients received a total of 6 appropriate WCD treatments. The first patient was prescribed a WCD in the setting of acute orthotopic heart transplant rejection while on inotropic support. She was successfully defibrillated for salvos of polymorphic VT; however, the arrhythmia recurred and the patient died (Figure 4A). The second patient wore the device for secondary prevention during an active infection. He received 2 shocks for VF, each converting the rhythm to a wide complex bradyarrhythmia; this patient also died (Figure 4B). The third patient wore the WCD after a recently missed MI and new ICM without revascularization. His baseline rhythm was sinus with first-degree atrioventricular block, left anterior fascicular block, and intraventricular conduction delay with QRS duration 132 ms. Following several minutes of frequent ventricular ectopy (Figure 5A), he developed sustained VT, triggering the audible alarm on the WCD and prompting him to call emergency medical services. After 178 s of sustained monomorphic VT, he received a shock after which the monomorphic VT immediately recurred (Figure 5B). A second shock 20 s later converted monomorphic VT to polymorphic VT, which degenerated to VF, and a third shock delivered 56 s later converted VF to asystole (Figure 5C). The patient was fortunately resuscitated after the timely arrival of emergency medical personnel. He went on to receive percutaneous coronary revascularization and ICD implantation and has now survived 3 years after the event. No patients received inappropriate therapies.
Our study examined prescribing indications, compliance, and outcomes for 147 patients for whom WCD were prescribed over a 2-year period at two tertiary medical centers. A slight majority of prescriptions were for primary prevention. We found that patient compliance varied substantially with one-third of patients wearing the device <16.5 h/day when calculated by average daily wear, and for <75% of the prescribed duration. WCD delivered 6 appropriate high-voltage therapies to 3 patients, yet no lives were directly saved. Just over one-half of patients went on to receive ICD.
Although randomized trial data are lacking in the study of WCD use, aggregate national registry results reported that 1.7% of patients who wore a WCD received appropriate therapy and that the device was successful in terminating 100% of unconscious VT/VF episodes after the first shock with 90% survival (9). Inappropriate shocks have been reported to occur in 1.9% of WCD patients (9). Outside of the U.S. experience, Wassnig et al. (12) reported on 6,043 German WCD patients and found 1.6% and 0.4% rates of appropriate and inappropriate therapy, respectively, and a median daily use of 23.2 h (IQR: 21.0 to 23.7 h). Focusing on primary prevention, Epstein et al. (11) reported on 8,453 registry patients prescribed a WCD in the first 3 months after MI and found a 1.6% rate of appropriate therapies, with 91% resuscitated. One study reported a survival advantage for registry patients prescribed a WCD after coronary artery bypass grafting (10), yet the absolute mortality benefit exceeded the rate of appropriate WCD therapy, raising concerns about the adequacy of the control comparisons (13).
Among single-center studies (6–8), the largest by Singh et al. (8) followed 525 patients prescribed a WCD for primary prevention with newly diagnosed cardiomyopathy and found no appropriate therapies for non-ICM and an appropriate therapy rate of 2.2% for ICM, resulting in 1.5% survival post-shock to hospital discharge. Further study is warranted to identify patients most likely to benefit from WCD therapy, ideally in the form of randomized data. The VEST trial is currently ongoing, randomizing post-MI patients to either a WCD or no WCD. Any benefits found for the WCD must be considered in light of the cost of WCD therapy, which is approximately $3,300 per month (8).
Prescriptions and use
In agreement with national registry–based data, the WCD prescriptions in this study reflect a mix of indications, including 45% of WCD prescribed for primary prevention with LVEF ≤35% during the “window period” (9). This group of patients, encompassing those in the early phase after MI or revascularization, or with a newly recognized non-ICM, is recognized to be at elevated risk for SCD (14–16) but also ineligible for ICD by current guidelines. The prophylactic use of ICD after recent MI was evaluated in the DINAMIT (Defibrillator in Acute Myocardial Infarction trial) (5) and IRIS (Immediate Risk Stratification Improves Survival trial) studies (4), with each trial demonstrating a decrease in arrhythmic SCD in the device arms, but also an offsetting increase in nonarrhythmic death without a net difference in mortality by the end of the follow-up period. Similarly neutral results for ICD have been found in patients with recent revascularization (17). Consistent with ICD data, the available WCD data also suggests utility in the primary prevention of SCD for patients in the window period, but as randomized mortality data is lacking, a recent American Heart Association advisory statement was measured in the endorsement of WCD use in primary prevention settings, providing a Class IIb, Level of Evidence: C recommendation (2). It is therefore notable that in this study, the primary prevention population accounted for the majority of WCD wear time (187 of 263 patient-months of use).
Prevention of SCD during a period of risk stratification in patients with an uncertain benefit from ICD implantation is an important role for the WCD. Patients with a history of ventricular arrhythmias but a temporary ICD contraindication, unclear recurrence risk, or incomplete work-up to assess for a reversible cause accounted for 23 prescriptions (16%) in this study. All but 2 of these patients were referred for ICD. Secondary prevention patients composed a similar fraction (16.1%) of patients in a U.S. registry study (9). WCD may also be useful for patients otherwise eligible for primary prevention ICD after 3 months of optimal therapy if their LV function is believed to possibly recover further with longer monitoring (18).
Median WCD wear time has been reported previously in U.S. registry studies (21.7 h/day  and 22.5 h/day ) and is similar to the median wear time observed in this study (21 h/day). The study by Chung et al. (9) reported “satisfactory compliance” with >90% use on WCD wear days in 52% of patients and >80% use in 71% of patients. The respective figures calculated similarly from our study were very close: 47% and 67% of patents. However, these statistics belie the fact that one-third wore the WCD for fewer than 16.5 h/day over the duration that they wore the WCD and wore it for fewer than 75% of the prescribed days. Five patients did not comply with use beyond even the first day of wear. The German registry study by Wassnig et al. (12) reported greater compliance (23.1 h/day). Reasons for the discrepancy with our study are unclear but may relate in part to differences in patient education or selection. Interestingly, we found a similar association to these studies with improved daily wear compliance associating with increased total days of use (3,9,12). However, the total days of use underestimates the prescription duration in those patients who skip days or stop wearing the WCD prior to the end of their prescription. When accounting for the total prescription durations in our study, decreased compliance is associated with progressively longer prescription durations. The inclusion of the prescription duration extracted from the medical record highlights a strength of our study design, complementing the interpretation of large-scale registry data. We also found that female sex and nonprimary prevention WCD use indications were both associated with improved compliance. Several factors may contribute to nonadherence including device discomfort or the disturbances caused by inappropriate event alarms. The anxiety or fear of being shocked was rated as “very” or “moderately” important by 93% of patients according to a recent survey (19). Thus, although the WCD has been proposed as a “bridge” to a decision regarding ICD candidacy (20), the imperfect compliance we observed limits the level of protection compared with that of an ICD for a significant proportion of patients and underscores the importance of proper patient education and support. In light of the psychosocial and financial costs associated with WCD use and limitations with compliance, careful attention must be made for WCD prescription for any given clinical circumstance, a challenging and often dynamic process well illustrated by the 8 patients whose prescriptions were discontinued prior to discharge in this study.
Events requiring therapy were rare in our cohort (2.1%) and were similar to national registry numbers despite the fact that it was a highly selected population felt to be at elevated risk for SCD by their treating physicians. Nevertheless, no therapies restored perfusing rhythms, nor were any lives directly saved by the WCD in this study, which contrasts to the registry-reported survival of 90% after VT/VF events in WCD patients (9). Whereas part of this discrepancy may be due to the limited size of this study, analysis of the treatments delivered is illustrative. For instance, despite successfully terminating ventricular tachyarrhythmias, 2 of the 6 shocks observed in this cohort resulted in asystole or severe bradycardia, highlighting the absence of a pacing capability of the WCD. In comparison, transvenous and subcutaneous ICD are engineered to pace following defibrillation for any post-shock bradycardia and future iterations of the WCD may benefit from a similar functionality. Furthermore, the treatment efficacy of defibrillation depends on the underlying clinical substrate and the patients in Figure 4 may not have been inherently able to be resuscitated.
This study must be interpreted in the context of its observational and retrospective nature. Furthermore, there are likely to be unidentified selection biases in the prescription of the WCD. The sample size limits the generalizability of our observations. Lastly, we cannot exclude the possibility that our prescription patterns or patient population are different from those of other centers.
In summary, our experience illustrates contemporary practice with the WCD. Our data suggest that patient compliance with the device is low for a significant fraction of patients, limiting the protection offered by the device. We report a low rate of appropriate therapies and no lives were directly saved in this cohort of highly selected patients. We did not observe any inappropriate shocks. Further efforts are needed to further define the patient populations most likely to benefit from the WCD.
COMPETENCY IN MEDICAL KNOWLEDGE: Although approved and available for patients felt to be at high risk of SCD, prescribing patterns of WCD vary, particularly in the period immediately following MI or newly recognized cardiomyopathy. Without randomized data, current WCD guidelines are based on the results of large registry data. Our data adds to the presently published reports with an increased level of granularity about both the patient characteristics and compliance data. We found that primary prevention indications accounted for roughly one-half of WCD prescriptions and the majority of wear time. A significant fraction of patients prescribed a WCD for secondary prevention had incomplete work-ups for a potentially reversible etiology, yet 90% of them eventually received ICD. Treatment events were infrequent and not always life-saving. Compliance was limited in one-third of patients.
TRANSLATIONAL OUTLOOK: These data underscore the importance of proper patient selection and education for the prescription of WCD. Future studies, preferably with randomized trials, are needed to improve identification of patients most likely to benefit from a WCD.
The authors thank Steve Szymkiewicz, Brad Carlson, and Noelle Wehle from ZOLL for providing data regarding WCD wear time, compliance, arrhythmia detection, and treatment events.
Dr. Hucker was supported by Award Number T32HL007208 from the National Heart, Lung, and Blood Institute. Dr. Reynolds has served as a consultant for Medtronic and St. Jude Medical. Dr. Lubitz has received support from the National Institutes of Health/National Heart, Lung, and Blood Institute (grant no. K23HL114724) and the Doris Duke Charitable Foundation (Clinical Scientist Development Award no. 2014105); has served as a consultant to St. Jude Medical; and has received sponsored research support from Boerhinger Ingelheim. Dr. Milan has received support from the National Institutes of Health/National Heart, Lung, and Blood Institute (grant nos. R01 HL127692 and R01 HL132905). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Leyton-Mange and Hucker contributed equally to this work.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors' institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- implantable cardioverter-defibrillator
- ischemic cardiomyopathy
- interquartile range
- left ventricular ejection fraction
- myocardial infarction
- sudden cardiac death
- ventricular fibrillation
- ventricular tachycardia
- wearable cardioverter-defibrillator
- Received July 17, 2017.
- Revision received September 18, 2017.
- Accepted September 27, 2017.
- 2018 American College of Cardiology Foundation
- Piccini J.P. Sr..,
- Allen L.A.,
- Kudenchuk P.J.,
- Page R.L.,
- Patel M.R.,
- Turakhia M.P.
- Kutyifa V.,
- Moss A.J.,
- Klein H.,
- et al.
- Kondo Y.,
- Linhart M.,
- Andrie R.P.,
- Schwab J.O.
- Singh M.,
- Wang N.C.,
- Jain S.,
- Voigt A.H.,
- Saba S.,
- Adelstein E.C.
- Chung M.K.,
- Szymkiewicz S.J.,
- Shao M.,
- et al.
- Zishiri E.T.,
- Williams S.,
- Cronin E.M.,
- et al.
- Epstein A.E.,
- Abraham W.T.,
- Bianco N.R.,
- et al.
- Wassnig N.K.,
- Gunther M.,
- Quick S.,
- et al.
- Kron J.,
- Ellenbogen K.A.
- Kadish A.,
- Schaechter A.,
- Subacius H.,
- et al.
- Duncker D.,
- Konig T.,
- Hohmann S.,
- Bauersachs J.,
- Veltmann C.
- Francis J.,
- Reek S.