Bethesda, MD - The rate of mechanical complications following ICD implantation is almost 50% higher when inexperienced physicians are performing the procedures, a new study shows [1]. Infections are even more common, Dr Sana M Al-Khatib (Duke Clinical Research Institute, Durham, NC) and colleagues write in the October 18, 2005 issue of the Journal of the American College of Cardiology.

The findings, say the authors, are the first to show an inverse relationship between volume of ICD implantations and patient outcomes. Of note, in an interview with heartwire, study coauthor Dr F Lee Lucas (Maine Medical Center, Portland, ME) emphasized that there was no particular cut point beyond which there was no improvement seen in outcomes. "For mortality, we found no difference, but for the other outcomes we saw a trend that decreased across the volume categories. So that means, the more you do, the better you are."

Al-Khatib et al reviewed Medicare files, Medicare Provider Analysis and Review hospital records, and ICD-9 diagnosis and procedure codes to examine the link between operator volume and complication rates within the first 30 and 90 days following device implant for all new ICD implantations between 1999 and 2001. Physician volume was divided into quartiles according to whether the physician performed 1-10, 11-18, 19-28, or more than 29 ICD implantation procedures per year.

After adjusting for potential confounders, Al-Khatib et al found that while mortality rates were more or less the same between patients treated by physicians across all quartiles, 30- and 90-day mechanical complications and 30- and 90-day infections were significantly more common among patients treated by physicians performing 10 procedures or less per year.

The authors propose several possible explanations for the inverse relationship between ICD implantation volume and patient outcomes, the most plausible, they say, being the fact that greater experience translates into better operative skills and that high-volume physicians likely are trained in electrophysiology, whereas low-volume physicians are not.

To heartwire, Lucas acknowledged that they did not collect data on electrophysiology training, but that it would be something to address directly in future research.

Dr Anne Curtis (University of South Florida, Tampa), who wrote the editorial accompanying the study, picks up on this point [2]. She notes that the recent expansion of Medicare coverage for ICDs means that many cardiologists without electrophysiology training are now implanting ICDs or intend to be implanting them in the near future.

Since the authors looked only at Medicare beneficiaries and Medicare patients typically account for only half of a physician's practice, the rates of complications and infections overall might be double that reported, Curtis notes. As well, since mechanical complications can lead to repeat operations that are in turn associated with even higher infection rates, the overall infection rates and, indeed, mortality rates beyond 90 days may be higher than that reported by Al-Khatib et al.

"Although there are cardiologists and surgeons who have attained the knowledge and expertise to achieve good outcomes with ICDs, clinical cardiac electrophysiologists have devoted their careers to arrhythmia management and by definition have the expertise required in this area," she writes. "Whatever the physician's background in this area, it is clear that experience counts, and it counts for a lot."