November 2nd, 2010

Meta-Analysis Provides Little Support for Prophylactic ICD Use in Older Adults

Prophylactic implantable cardioverter-defibrillator (ICD) therapy does not appear to significantly improve survival among older adults with severe left ventricular dysfunction, according to a meta-analysis in the Annals of Internal Medicine.

Researchers analyzed data from five randomized trials comparing ICD therapy with medical treatment among some 5800 adults with cardiomyopathy. Older adults (defined as 60 and older in some trials and 65 and older in others) accounted for nearly half the patients.

The researchers found that while ICD therapy lowered mortality in younger patients, it did not have a significant impact on mortality in elders. Overall, 17% of patients experienced complications from their ICDs.

The authors point to previous research showing that cardiac resynchronization therapy (CRT) seems to benefit the young and old alike, adding: “Taken together, these findings support that CRT alone may be the best device therapy in elderly persons with severe left ventricular dysfunction.”

3 Responses to “Meta-Analysis Provides Little Support for Prophylactic ICD Use in Older Adults”

  1. Renato De Vecchis, public hospital physician says:

    I disagree.Market concerns are responsible for emphasis about CRT in older people with severe cardiomyopathy.30% only of CHF patients have left bundle branch block and among them 50% at least is recognized non responder to CRT.Thus,this therapy should be considered a valid therapeutic approach in about 15% of CHF patients…

  2. Recent studies indicate the proportion ICD recipients >80 years of age is as high as 18% (Swindle et al., Arch Int Med 2010). The original MADIT II and SCD-HeFT trials had very few patients at such an advanced age; in addition, the rate of comorbidities common to elderly individuals including cerebrovascular disease and advanced renal insufficiency was exceedingly low. There is the additional issue of quality-of-life, as elderly ICD recipients may be admitted to the hospital with inappropriate shocks, or receive shocks at the very end of life (Goldstein et al., Ann Int Med 2004).

    One wonders whether another randomized trial in a more “real-world” population would be appropriate.

  3. Robin Motz, M.D., Ph.D. says:

    I think most meta-analyses are fundamentally flawed. I was never impressed by Peto’s reasoning in his Lancet articles, and a blog by a Dr. George Thomas at, thoroughly convinced me that meta-analyses are biased in ways we can’t possible imagine or deduce from the published data. For instance, if some studies were discarded, would be have discarded those same studies? If they aren’t footnoted, we have no way of knowing. It’s similar to not publishing all the data in your notebook.