Reconsider When a Prophylactic ICD Is Needed in Nonischemic Dilated Cardiomyopathy

Better imaging parameters may exist for risk stratification in nonischemic dilated cardiomyopathy (NIDCM), based on a meta-analysis.

The presence and extent of late gadolinium enhancement (LGE) on cardiac MRI was a significant predictor of all-cause mortality, cardiovascular mortality, arrhythmic events, and heart failure events over more than 3 years of follow-up, reported Christoph Gräni, MD, PhD, of Inselspital, Bern University Hospital in Switzerland, and colleagues.

Meanwhile, the established prognosticator in NIDCM, left ventricular ejection fraction (LVEF), showed no association with mortality and arrhythmic events. Instead, LVEF had some modest prediction of heart failure events and composite major adverse cardiovascular events (MACE), the authors reported in JAMA.

“Overall, these findings cast further doubt on the status of LVEF as the pivotal imaging criterion in risk stratification and selection of patients for prophylactic ICD [implantable cardioverter-defibrillator] implantation in NIDCM,” they said. “Based on historic ICD trials, risk stratification in NIDCM continues to be centered around LVEF threshold values at or below 35% as the main indicator for primary prophylactic ICD implantation.”

Several years ago, pediatric electrophysiologists already ditched LVEF as the main selection criterion for a prophylactic ICD in NIDCM.

NIDCM is characterized by left-ventricular or biventricular dilatation and contractile dysfunction in the absence of significant coronary artery disease and abnormal loading conditions. To prevent sudden cardiac death, preventive ICDs are placed in 100,000 people with NIDCM each year based largely on LVEF criteria.

However, most device recipients don’t end up with a single defibrillator shock, according to data cited by Gräni’s group. Additionally, NIDCM patients with LVEF >35% are often still at risk of SCD.

“These circumstances illustrate that precise risk assessment for guiding surveillance, resource allocation, and therapeutic decision-making remain a major unmet clinical need for patients with this complex, heterogeneous disease,” the investigators stressed. Better selection of ICD candidates “may avert considerable costs related to unnecessary ICD implantations and subsequent lifetime management … while allowing the provision of life-saving treatment for SCD-prone patients with NIDCM not captured by current selection criteria.”

They suggested that additional studies and imaging protocols will be needed to define LGE’s role as a potential selector for prophylactic ICD implantation in patients with NIDCM.

Currently, two relevant existing studies are CMR-ICD and BRITISH, ongoing randomized trials examining the effect of prophylactic ICDs in NIDCM patients with LVEF ≤35% and evidence of LGE. “These trials will provide definitive answers to whether LGE assessment can tangibly improve therapeutic decision-making in patients with NIDCM and advanced contractile impairment,” according to Gräni and colleagues.

Other cardiac MRI measurements of interest for risk stratification include native T1 relaxation times, extracellular volume fraction, and global longitudinal strain. Notably, limited data suggest that higher native T1 relaxation times were associated with arrhythmic events and MACE in the present meta-analysis.

Gräni and colleagues had pooled 103 observational studies, both prospective and retrospective ones, that had nearly 30,000 people with NIDCM included. The median follow-up lasted 37.8 months.

The study population was a median 55 years of age, 71% men, with a LVEF of 29.5% indicating severely reduced systolic function.

Gräni’s team acknowledged the variable quality among the studies included in the report. They also cautioned against applying the findings to people with early-stage NIDCM with less advanced contractile impairment, who were not represented by these studies.