ICD2: No Survival Benefit With ICDs in Dialysis Patients

Batya Swift Yasgur, MA, LSW

June 13, 2019

Prophylactic implantable cardioverter-defibrillators (ICDs) do not reduce the rate of sudden cardiac death (SCD) in dialysis patients, new research suggests.

The Implantable Cardioverter-Defibrillator in Dialysis Patients (ICD2) trial randomly assigned 188 patients who were receiving dialysis and whose left ventricular ejection fraction (LVEF) was ≥35% either to a group that received an ICD or to a control group that received treatment as usual during a median follow-up period of roughly 7 years.

There were more deaths due to SCD in the ICD group than in the control group; the cumulative SCD incidence at 5 years was 9.7% and 7.9%, respectively, and the survival probability was only 50.6% and 54.5%, respectively.

There were 25 adverse events related to ICD implantation.

The trial was stopped early on the recommendation of the data and safety monitoring board (DSMB), for reasons of futility.

"We wanted to know if patients on dialysis would benefit from an implantation of an ICD, so we set up a prospective, randomized controlled trial to figure this out, with half receiving the ICD and half not receiving it, and found that, on the whole, ICD implantation did not improve survival," lead author J. Wouter Jukema, MD, PhD, of the Department of Cardiology, Leiden University Medical Center, the Netherlands, told theheart.org | Medscape Cardiology.

"We demonstrated that if a patient suffers from a demonstrated cardiac arrhythmia, an ICD — if feasible — can be implanted; but routinely, preemptively, it doesn't work to improve survival," he said.

The study was published online June 4 in Circulation.

Meaningful Extension?

"Patients with end-stage renal disease (ESRD) requiring renal replacement therapy have a 14-fold increased risk of death compared with the general population," the authors write.

Arrhythmia and cardiac arrest are estimated to cause roughly one third of these fatalities. Thus far, no therapy has been shown to be effective in reducing mortality.

Patients who are receiving dialysis have been excluded from previous trials of ICDs, leaving unanswered the question of whether use of these devices could offer a survival advantage. Moreover, negative outcomes due to comorbidities or implantation-related complications might "mitigate the benefit" of the device, the authors note.

The researchers therefore sought "an evidence-based answer to the question of whether primary prevention of SCD by ICD implantation in dialysis patients with LVEF ≥35% is beneficial and safe."

"We had a long list of failures in improving survival in these patients, who have a bad prognosis even if they are properly dialyzed," Jukema commented.

"It was reported that they were probably dying from SCD, and if it's really the case that they were dying from SCD, then an antiarrhythmic device should work," he said.

"We selected on purpose a patient population that wasn't too ill and should really have had the possibility of significant enhancement of a worthwhile life, and hoped for a meaningful extension of life," Jukema said.

The trial was initiated in June 2007. The last patient was enrolled in January 2018. On the advice of the DSMB, the trial was stopped early (February 2018) because of futility.

Patients were randomly assigned in a 1:1 ratio to receive either an ICD or usual care (control group). Severe preexisting cardiac pathology was treated prior to randomization.

Eligible patients were required to be aged 55 years to <81 years and to have received treatment of ESRD with either hemodialysis or peritoneal dialysis for ≥90 days.

Patients who were receiving dialysis and who met the class I indication for ICD implantation were excluded, as were patients with heart failure (New York Heart Association functional class IV), a central venous catheter, or a medical condition that made 1-year survival unlikely.

Except for patients with chronic atrial fibrillation who received a single-chamber device, patients in the ICD group received a dual-chamber device, for the detection and treatment of ventricular fibrillation and ventricular tachycardia.

The primary endpoint was sudden cardiac (arrhythmic) death. The first secondary endpoint was all-cause death.

Not a Harmless Procedure

Of the original 220 dialysis patients (median age, 67 years; interquartile range [IQR], 62 – 74; 76.1% men; 71.3% undergoing hemodialysis) referred from participating medical centers, 188 were eligible for inclusion in the trial and the intention-to-treat analysis and were randomly assigned to receive either an ICD or usual care (n = 97 and n = 91, respectively).

Treatment groups were well balanced with respect to baseline and laboratory characteristics.

During a median follow-up of 6.8 years (IQR, 3.8 – 8.8 years), a total of 54 patients (28.7%) underwent kidney transplant (29 patients in the ICD group [29.9%] and 25 in the control group [27.5%], P = .71).

The median number of hospital admissions in both groups was 5 (ICD group: IQR, 3 – 7; range, 0 − 24; control group: IQR, 2 – 8; range, 0 − 21).

A total of 99 patients died (52.7%), 52 in the ICD group (53.6%) and 47 in the control group (51.6%).

SCD occurred in 19 of 188 patients (10.1%), 11 in the ICD group (11.3%) and eight in the control group (8.8%).

Cause of death was evenly distributed between treatment groups. Infectious disease was the most prevalent cause, followed by SCD.

The overall cumulative incidence of SCD at 5 years was 8.9% (95% confidence interval [CI], 4.4% – 13.3%): 9.7% (3.3% – 16.2%) in the ICD group and 7.9% (95% CI, 1.7% – 14.0%) in the control group (hazard ratio [HR], 1.32; 95% CI, 0.53 – 3.29; P = .55).

Overall survival probability at 5 years was 52.4% (95% CI, 44.5% – 60.3%).

In the ICD group, the survival probability was 50.6% (95% CI, 39.8% – 61.5%), vs 54.5% in the control group (95% CI, 43.0% – 66.0%) (HR, 1.02; 95% CI, 0.69 – 1.52; P = .92).

"Thus, ICD implantation was not associated with a reduced incidence of SCD or improved overall survival," the authors comment.

Moreover, there were no significant differences in the effect of ICD therapy on survival in subgroup analyses, stratified according to age, sex, history of diabetes mellitus, history of coronary artery disease, dialysis modality, and dialysis duration.

A total of 25 adverse events occurred in 22 patients (27.5%) who received an ICD. Short-term complications that were directly related to the implantation procedure occurred in 10 patients (12.5%); however, there were no cases of pneumothorax or death attributable to ICD implantation.

"After a median of 5.2 years of follow-up, the mortality was around 50% — half of the patients died, and we saw unfortunately that routine ICD implantation to avoid SCD doesn't help these types of patients," Jukema summarized.

"ICD implantation is not a harmless procedure and also has its complications, such as infection or dysfunction of the devices, so you should have some game [plan] to counteract the problems of implantation," he stated.

Proceed With Caution

Commenting on the study for theheart.org | Medscape Cardiology, Rod Passman, MD, MSCE, professor, Northwestern University Feinberg School of Medicine, Chicago, Illinois, said that the study "looked at patients who otherwise would not meet indications for ICD, based on today's standards, which is a previously unstudied population."

He distinguished between two types of patients: "Primary prevention patients who, we believe, are at high risk but have not yet manifested heart rhythm problems, and in these [we need] a lot of caution as to whether we have evidence that we'll be helping them," said Passman, who is the coauthor of an accompanying editorial and was not involved with the study.

"More difficult is a patient on dialysis who has already had a life-threatening event and was fortunate enough to survive it, and although we don't have a lot of data to suggest that these [ICDs] significantly prolong survival, we always have to advocate for the patient in front of us, and in those individuals, it would be hard to withhold this therapy," he said.

Nevertheless, he warned, "we should proceed with caution, understanding that we may not expect to see the benefit that we have already demonstrated in nondialysis patients," which "brings up the issue that the lessons we learned from the nondialysis population may not apply."

The authors note that the study findings "do not support the initiation of a larger study with all-cause death as the primary endpoint."

The ICD2 trial was supported by an unrestricted educational research grant from Biotronik GmbH & Co. Jukema and coauthors report no relevant financial relationships. Passman is on the advisory board for Abbott and Medtronic and receives royalties from UpToDate.

Circulation. Published online June 4, 2019. Abstract, Editorial

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.