April 03, 2012

April 2, 2012 (Chicago, Illinois) — Patients due for elective replacement of an implantable cardioverter-defibrillator (ICD) they had received when their LVEF was <35% remain at risk for serious ventricular arrhythmias even if their LVEF has since improved to >35%, suggests a retrospective cohort study presented here at the American College of Cardiology 2012 Scientific Sessions [1].

Although in practice it is usually routine for ICDs nearing the end of their lifetime to be exchanged for new ones regardless of LVEF, if it's even measured at the time, conceivably the risk status of some patients could have improved such that ICD protection might no longer be needed, said the study's senior author, Dr Selcuk Adabag (Veterans Administration Medical Center and the University of Minnesota, Minneapolis), to heartwire .

Someone who received an ICD for secondary prevention or who had already experienced appropriate therapy from their device would likely be considered at future risk anyway, Adabag said. But what about someone with a primary-prevention ICD that has never delivered therapy--a shock or antitachycardia pacing (ATP)--and now their LVEF has improved to above the 35% threshold? The guidelines don't address this situation, according to Adabag, and his group's findings suggest ICDs could still be protective.

Consistent results were presented at a meeting by another group, as reported recently by heartwire . A cohort of 111 patients were followed with echocardiography and developed cardiac arrest; of 31 initially with an LVEF <35%, 11 patients had improved to >35%--including four whose LVEF had risen to >55%--over an average of about two years.

In the series reported here by Adabag's group, among 121 consecutive patients getting their ICDs exchanged from 2006 to 2010--excluding anyone receiving cardiac resynchronization therapy--LVEF had improved to >35% in about one-fourth. And in most of those cases, it didn't just edge above that threshold. In those 32 patients, average LVEF went from 32% at initial ICD implantation to 51% at the time of replacement. (For those with unimproved LVEF, the mean stayed at 36%.)

Yet 13 of those 32 patients experienced appropriate therapy from their replacement ICD; and in three of those 13 cases, it was the patient's first experience of an appropriate therapy, Adabag noted.

Incidence of Device Therapy Before and After ICD Replacement, by LVEF Group

Device therapy LVEF improved*, n=32 (%) LVEF unimproved, n=89 (%) p
Before ICD replacement      
Shock 31 26 0.56
Shock or ATP 41 36 0.64
After ICD replacement      
Shock 19 17 0.81
Shock or ATP 25 23 0.77

*Defined as LVEF >35% at ICD replacement and increase of >10% prior to ICD exchange

ATP=antitachycardia pacing

He said the analysis included "all comers," patients getting ICDs for primary or secondary prevention and for either ischemic or nonischemic heart disease. Those with improved LVEF were more likely to have nonischemic cardiomyopathy compared with those without improved LVEF (36% vs 19%, p=0.037).

Improvement in LVEF was defined as >35% at ICD change-out and an increase of >10% over the lifetime of the ICD now being explanted.

Over a mean follow-up of 6.2 years, the incidence of ICD shocks and combined incidence of any device therapy (shocks or ATP) were similar among those with improved and unimproved LVEF.

"We realize we're talking about very small numbers here, and that is our major limitation," Adabag said, noting also that the study was observational based on a single-center experience. But the data are novel and "intriguing, within their limitations, and raise the question of doing a larger study without the limitations."

The clinical question would be in patients who had received ICDs for primary prevention, especially those who had never received device therapy, he said. "Those are the ones who, one might think, would be least likely to get shocks after [ICD replacement]."

Even if such patients were found not to be at increased risk and conceivably not in need of ICD protection, any resulting change in practice would not come easily. Adabag thinks most electrophysiologists would opt to put in a new device anyway, and most patients are by then comfortable with them. "I haven't seen a patient yet who, once they had a device, didn't want to have it replaced."

Adabag discloses receiving investigator initiated research grants from Medtronic and Boston Scientific. Disclosures for the coauthors are listed in the abstract.