By Joshua Moss, MD

Associate Professor of Clinical Medicine, Cardiac Electrophysiology, Division of Cardiology, University of California, San Francisco

Dr. Moss reports he is a consultant for Abbott, Boston Scientific, and Medtronic.

SYNOPSIS: In patients with left ventricular ejection fraction ≥ 35% undergoing dialysis, the insertion of a prophylactic implantable cardioverter defibrillator carries a relatively high risk for adverse events and does not significantly decrease sudden cardiac death or mortality.

SOURCE: Jukema JW, Timal RJ, Rotmans JI, et al. Prophylactic use of implantable cardioverter-defibrillators in the prevention of sudden cardiac death in dialysis patients. Circulation 2019;139:2628-2638.

Patients with end-stage renal disease undergoing dialysis are at a markedly higher risk for death than the general population. Arrhythmias and cardiac arrest have been estimated to account for about one-third of those deaths. Jukema et al sought to determine whether a prophylactic implantable cardioverter defibrillator (ICD) could provide a survival benefit in this population.

Physicians from 17 centers in the Netherlands referred their patients to a single implantation and coordination center, Leiden University Medical Center, to maximize safety in what is known to be a population at high risk for procedural complications. The authors used block randomization with stratification for age and referring center to randomize patients to receive either a Biotronik ICD (dual-chamber, unless there was chronic atrial fibrillation) or usual care. A total of 188 patients (median age, 67 years) were randomized; most were on hemodialysis (71%) rather than peritoneal dialysis. Exclusion criteria were age < 55 years or > 80 years, hemodialysis or peritoneal dialysis for < 90 days, impending renal transplant, a class I indication for ICD implant based on traditional LVEF criteria, New York Heart Association functional class IV heart failure, recent myocardial infarction, a medical condition that made one-year survival unlikely, or an indwelling central venous catheter. Of the 97 patients randomized to receive an ICD implant, 17 did not undergo the procedure for various reasons and were included in the intention-to-treat (ITT) analysis (but not a per-protocol analysis).

The ICD was implanted contralateral to the site of arteriovenous (AV) fistula in hemodialysis patients and on the side of the dominant arm in peritoneal dialysis patients (to allow for construction of an AV fistula in the nondominant arm in the future). In 58 of 80 implants, it was on the right side. Defibrillation threshold testing was performed, and devices initially were programmed with a monitor zone at 150 beats per minute, multiple bursts of antitachycardia pacing (ATP) prior to shocks between 188 and 214 beats per minute, and a single ATP and shocks above 214 beats per minute. Over a median follow-up of 6.8 years, 53.6% of patients in the ICD group and 51.6% of patients in the control group died, while 29.9% in the ICD group and 27.5% in the control group underwent kidney transplant (P = 0.71). The authors believed sudden cardiac (arrhythmic) death (SCD) accounted for 21% of deaths in the ICD group and 17% of deaths in the control group, a nonsignificant difference. Eleven patients in the ICD group received appropriate ICD therapy, including shocks for ventricular fibrillation in four patients, while four other patients received inappropriate therapy. Short-term procedural complications occurred in 12.5% of implants, with reinterventions required in 12.5% and explantation in 7.5% (mostly for recurrent bacteremia). A per-protocol analysis yielded similar results to the ITT analysis, and no prespecified subgroup was found to derive significant mortality benefit. The authors concluded that prophylactic ICD implantation in dialysis patients did not reduce the high rate of SCD or all-cause mortality.


The ICD2 trial represents the first randomized, controlled trial of ICDs for primary prevention of SCD in patients undergoing dialysis, and it decidedly showed no mortality benefit. At first glance, the results seem surprising considering the presumed high rate of SCD in this population. Jukema et al hypothesized a 70% reduction in SCD when powering the study, noting a 50% reduction in SCD in secondary prevention trials and a reduction that was “even greater in studies that compared ICD treatment with medical treatment for primary prevention.” However, for unclear reasons, they cited the DINAMIT trial to support the latter statement. But the results of DINAMIT showed no significant mortality benefit from ICDs in a high-risk population.1 The landmark MADIT-II trial, which concerned primary prevention ICDs in ischemic cardiomyopathy, showed only a 31% reduction in risk of death, while SCD-HeFT showed a 23% reduction.2,3 Thus, the Jukema et al study may have been substantially underpowered to show any significant benefit, particularly if the prevalence of arrhythmic death also was overestimated.

In other ways, the results of the trial seem quite predictable. While there was a significant history of coronary artery disease in the enrolled population (16% had a history of coronary angioplasty, 13% a history of bypass surgery, and 27% a history of myocardial infarction), most patients had normal LV systolic function (nearly all patients had an LVEF > 45%). No previous randomized trial of ICDs for primary prevention of SCD included patients with such minimal LV dysfunction. Even patients with LVEF up to 40% in the MUSTT trial had inducible sustained ventricular tachyarrhythmias.4 The results of prior studies suggest that the presence of LV dysfunction and myocardial scar represent the most predictive risk factors for ventricular arrhythmias in the absence of an inherited arrhythmia syndrome. The population in the Jukema et al study may have been predictably at lower risk. Indeed, only eight patients in the control group of 91 patients died suddenly over a median follow-up of nearly seven years. The results of autopsy-based studies have demonstrated that SCD may be attributable to ventricular tachyarrhythmias in fewer patients than often assumed.

The Jukema et al study adds valuable data to guide the care of patients with end-stage renal disease on dialysis, demonstrating that a proven intervention for life-threatening ventricular arrhythmias in other populations does not reduce mortality in this one. Unfortunately, the study leaves unanswered another important question, and one that is encountered perhaps more commonly: Do ICDs improve mortality in dialysis patients with a traditional indication for primary prevention? Generally, such patients were excluded from trials of primary prevention ICDs, and Jukema et al confirmed that this population is likely more susceptible to procedural- and device-related complications. Perhaps a randomized trial of prophylactic ICDs in dialysis patients with a LVEF ≤ 35% and heart failure symptoms will be the next logical step.


  1. Hohnloser SH, Kuck KH, Dorian P, et al. Prophylactic use of an implantable cardioverter-defibrillator after acute myocardial infarction. N Engl J Med 2004;351:2481-2488.
  2. Moss AJ1, Zareba W, Hall WJ, et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877-883.
  3. 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.
  4. Buxton AE, Lee KL, Fisher JD, et al. A randomized study of the prevention of sudden death in patients with coronary artery disease. N Engl J Med 1999;341:1882-1890.