Skip to main content

14-12-2010 | Cardiology | Article

ICDs may raise post-MI death risk


Free abstract

MedWire News: In patients at high risk for sudden death after myocardial infarction (MI), an implantable cardioverter-defibrillator (ICD) does not reduce overall cardiac-related death risk in the subsequent year, study findings suggest.

The study authors explain that "factors associated with arrhythmia requiring ICD therapy are also associated with a high risk of nonsudden death, [therefore] negating the benefit of ICDs in this setting."

Paul Dorian (St Michael's Hospital, Toronto, Ontario, Canada) and team analyzed 653 recent (6 to 40 days) acute MI patients.

All patients had left ventricular dysfunction (ejection fraction <35%) and evidence of impaired cardiac autonomic function (standard deviation of N-N intervals ≤70 ms or average heart rate >80 bpm on 24-hour Holter monitor, 3 or more days after the MI).

Dorian and team randomly allocated patients to receive an ICD device (n=311) or standard medical therapy (n=342) over a mean follow-up period of 30 months.

Of those assigned to receive an ICD, 59 patients received appropriate therapy (high-energy shocks or antitachycardia pacing) from the device, and 252 patients did not.

Writing in the journal Circulation, Dorian and colleagues say that it is "possible that patients with an ICD who are saved from arrhythmic death have clinical characteristics different from patients having nonarrhythmic deaths in the absence of an ICD."

To account for this effect, they performed an analysis of competing-risks at baseline. They found that the factors that increased the risk for arrhythmic death, such as male gender and severe heart failure, also increased the risk for nonarrhythmic deaths.

After adjustment for these factors, the researchers observed that receiving an ICD decreased the risk for arrhythmic death by 67%, but increased the nonarrhythmic death risk by 70% compared with not receiving the device (p<0.001).

Of note, death from any cause occurred at a higher yearly rate among patients who received appropriate therapy from their ICD than among those with an ICD but no appropriate therapy, at 15.1% versus 5.2% (P<0.001).

In an accompanying editorial, Michael Sweeney from Brigham and Women's Hospital in Boston, Massachusetts, USA, hypothesized that Dorian et al's findings may arise from cardiac changes triggered by ICD shocks, which subsequently increase the risk for nonarrhythmic death in patients with these devices.

Dorian and colleagues conclude: "Although the mechanisms responsible for the excess non-arrhythmic deaths in the ICD appropriate therapy group cannot be established with certainty, this study highlights the fact that appropriate ICD therapies in some patient populations do not necessarily lead to a meaningful overall mortality reduction."

MedWire ( is an independent clinical news service provided by Current Medicine Group, a trading division of Springer Healthcare Limited. © Springer Healthcare Ltd; 2010

By Lauretta Ihonor

Related topics