ICD Fails to Reduce SCD/HSVA in CMR GUIDE Trial

Implantation of an implantable cardioverter-defibrillator (ICD) did not significantly reduce the composite outcome of sudden cardiac death (SCD) or hemodynamically significant ventricular arrhythmia (HSVA) compared with an implantable loop recorder (ILR) in patients with left ventricular ejection fraction (LVEF) of 36% to 50% and CMR-defined myocardial scar, according to results of the CMR GUIDE trial presented at the European Society of Cardiology (ESC) Congress 2026 and published simultaneously in JAMA.¹

Current guidelines recommend primary prevention ICDs only for patients with an LVEF of 35% or less, yet a substantial proportion of sudden cardiac deaths occur above this threshold, where no validated risk-stratification tool has existed to guide device therapy.¹ CMR GUIDE tested whether late gadolinium enhancement (LGE) on cardiovascular magnetic resonance (CMR), a marker of myocardial scar and a known arrhythmic substrate, could identify patients with LVEF 36%-50% likely to benefit from a primary prevention ICD.

“In the CMR GUIDE trial, we assessed whether an ICD could improve outcomes in patients with LVEF 36-50% who have evidence of myocardial scarring, a risk factor for sudden cardiac death,” said Joseph Selvanayagam, DPhil, principal investigator, Flinders University, Flinders Medical Centre, Adelaide, Australia.²

CMR GUIDE trial design and primary outcome results

CMR GUIDE was an investigator-initiated, multicenter, open-label randomized trial conducted at 18 sites in Australia, Germany, and the UK, enrolling adults with ischemic or nonischemic cardiomyopathy, LVEF 36%-50%, and CMR-confirmed myocardial scar between 2015 and 2022.¹ Of 353 patients randomized 1:1 to receive an ICD (n = 180) or an ILR (n = 173), median age was 65 years, 18% were female, and 72% had an ischemic etiology. Median follow-up was 6.3 years.

The trial’s primary composite outcome, SCD or HSVA, occurred in 14 patients (7.8%) in the ICD group compared with 16 patients (9.2%) in the ILR group, a difference falling short of statistical significance (hazard ratio [HR], 0.76; 95% CI, 0.37-1.58).¹

All-cause mortality, cardiovascular mortality, and heart failure-related hospitalization were also statistically similar between groups, and a prespecified hierarchical win ratio analysis incorporating SCD, HSVA, heart failure hospitalization, and quality of life likewise showed no significant between-group difference (win ratio, 0.97; 95% CI, 0.74-1.27; P = .82).

Divergent findings on SCD, HSVA, and the age subgroup signal

Despite the neutral primary result, the trial’s individual outcome components diverged: SCD occurred in 3 patients (1.7%) in the ICD group versus 10 patients (5.8%) in the ILR group (HR, 0.26; 95% CI, 0.07-0.95), while HSVA was numerically more frequent with an ICD (6.7% vs 3.5%; HR, 1.77; 95% CI, 0.65-4.81).¹

Investigators noted this pattern is consistent with ICD therapy terminating some otherwise-fatal arrhythmias, effectively converting potential SCD events into nonfatal HSVA events captured by the device; per the prespecified analysis plan, formal significance testing was not reported for these individual secondary components, and the trial lacked power to detect differences in them independently.

In an exploratory, prespecified subgroup analysis of 6 variables, patients younger than 70 years assigned to an ICD had fewer primary outcome events than those assigned to an ILR (3.3% vs 10.0%; HR, 0.28; 95% CI, 0.09-0.89; P = .01 for interaction), with no such difference seen in patients 70 years or older (16.9% vs 7.5%; HR, 2.33; 95% CI, 0.75-7.26).¹

Investigators cautioned subgroup analyses were not adjusted for multiplicity and should be interpreted as hypothesis-generating rather than definitive, though they noted the finding parallels a similar age interaction previously reported in the DANISH trial.

Patients randomized to ICD also reported modestly worse quality of life on the Minnesota Living with Heart Failure Questionnaire over follow-up (mean difference, 5.8 points; 95% CI, 2.4-9.3), and device-related adverse events, primarily lead-related complications, were more frequent in the ICD group.

“Overall, the primary endpoint was neutral,” said Selvanayagam. “However, younger patients appeared to derive clinical benefit from ICDs. We suggest data from the CMR GUIDE trial are discussed with younger patients in whom ICDs may be an option in a shared decision-making process.”²

The trial authors concluded ICD implantation did not significantly reduce the composite outcome of SCD or HSVA in this population, though a possible reduction in SCD specifically warrants further study, particularly in younger patients, where competing mortality risk from other causes is lower. Investigators noted CMR GUIDE’s event rates were lower than originally projected, a shortfall attributed in part to improvements in guideline-directed medical therapy since the trial began enrolling in 2015.

References

1. Selvanayagam JB, Cleland JGF, Hillis GS, et al. Cardiovascular magnetic resonance to guide defibrillator implantation for LVEF of 36% to 50%: the CMR GUIDE randomized clinical trial. JAMA. Published online August 28, 2026. doi:10.1001/jama.2026.17078

2. European Society of Cardiology. Implantable cardioverter-defibrillators may benefit certain younger patients with mild-moderate heart dysfunction. Published August 28, 2026. Accessed August 29, 2026.