Using implantable cardioverter-defibrillators (ICDs) to prevent sudden cardiac death (SCD) was more effective than not using ICDs at lowering the risk for mortality and SCD, a meta-analysis published online Jan. 21 in Annals of Internal Medicine found. However, evidence from multiple studies was too weak to suggest a mortality or SCD benefit among different subgroups.

In their analysis, Amy Earley, BS, of Tufts Medical Center in Boston, and colleagues included 14 studies from the U.S., Canada and Europe that included patients eligible for an ICD to prevent SCD. All participants were followed from the time of implantation. They compared patients with ICDs with or without cardiac resynchronization therapy to patients with no ICDs for all-cause mortality and death caused by SCD and also compared subgroups that included age, sex, race or ethnicity, New York Heart Association (NYHA) class, left ventricular ejection fraction, heart failure, left bundle branch block and QRS interval.

All 14 studies found a statistically significant benefit of ICD in reducing all-cause mortality rates. ICDs in patients with no recent heart attack and no concurrent coronary revascularization lowered mortality risk by 31 percent over a period of three to seven years after device implantation.

However, the 10 studies that analyzed subgroups offered weak evidence of a mortality benefit of ICDs based on age, sex, race and QRS interval yet found no difference between ICDs and no ICDs. Evidence related to the other subgroups was indeterminate. One study did find ICDs more effective in NYHA Class II patients vs. NYHA Class III.

Evidence from nine studies was strong enough to suggest ICDs offer a protective benefit against SCD. Using an ICD in patients with ischemic or nonischemic cardiomyopathy without a recent heart attack or concurrent coronary revascularization lowered the risk of SCD by about 63 percent over a time period of two to six years after implantation. Only two studies carried out subgroup analyses, and based on this small number of analyses, the authors found the evidence to be indeterminate.

Although there were few studies that evaluated ICDs among subgroups, “each patient’s prognosis has to be considered to individualize treatment decisions in clinical practice,” the authors argued.