Defibrillator implantations performed by physicians with low volume are more likely to have adverse outcomes

As the implantation of cardioverter-defibrillator devices to prevent sudden cardiac death becomes increasingly popular, patients are likely to get the best results when they are treated by physicians who perform the procedures frequently, according to a new study in the Oct. 18, 2005, issue of the Journal of the American College of Cardiology.

"Defibrillator implantations performed by physicians with low volume are more likely to have adverse outcomes, including infections and mechanical complications. This suggests that defibrillator implantations should be directed toward physicians with a high volume of defibrillator implantations," said Sana M. Al-Khatib, M.D., M.H.S., F.A.C.C. from the Duke Clinical Research Institute in Durham, North Carolina.

This is the largest study of the relationship between the volume of implantable cardioverter-defibrillator (ICD) procedures done by physicians and the outcomes for their patients. The small devices are placed under the skin, with electrodes leading to the patient's heart. If the device detects a potentially life-threatening heart arrhythmia, it can automatically deliver a shock to jolt the heart back into a normal rhythm.

The researchers analyzed Medicare records for 1999 through 2001 to identify patients who received cardioverter-defibrillators and also how many of those patients received hospital treatment for complications within 90 days of the implant procedure. There was no significant difference in death rates between high-volume and low-volume physicians; but mechanical complications were more common among patients of the lowest-volume physicians. Among higher-volume physicians, those who implanted at least 11 devices per year into Medicare patients had similar rates of mechanical complications. Patients of the lowest-volume physicians also suffered significantly more infections around the site of their implants. Because this analysis included only procedures paid for by Medicare, the actual total volume of procedures done by the physicians may have been higher than the figures reported in the study.

Dr. Al-Khatib noted that although they looked at only Medicare patients, she didn't see any reason to believe the relationship between physician volume and outcomes would be different among patients who are not enrolled in Medicare. She also said that since they analyzed administrative data from Medicare, rather than looking directly at medical records, they did not have data on how sick the patients were; so it could be that the patients who had complications were sicker.

Nevertheless, the authors say the results point to physician volume as an important indicator of patient complication rates.

"Our findings suggest that ICD implantation should be directed toward high-volume physicians," the authors wrote.

Anne B. Curtis, M.D., F.A.C.C., from the University of South Florida and the President of the Heart Rhythm Society wrote in an editorial in the journal that, "Whatever the physician's background in this area, it is clear that experience counts, and it counts for a lot."

"Previous studies and the Al-Khatib article suggest that, in order to do the procedure safely and have good patient outcomes, an M.D. has to have a sufficiently high volume practice. That fact was key in the Heart Rhythm Society's alternative training pathway guidelines for implantation of cardioverter defibrillators and cardiac resynchronization devices, as referenced in the article," Dr. Curtis said. "Cardiologists need to be adequately trained to start doing procedures for which they did not train in fellowship. Some cardiologists in practice may not be happy with these conclusions, but the Heart Rhythm Society and I believe that patient outcome is the key determinant as to how we should handle this issue, and the Al-Khatib manuscript supports our guidelines."

In her editorial, Dr. Curtis wrote that Medicare enrollees often account for around half of a cardiologist's practice, so the total volume of implantations, including both Medicare and non-Medicare patients, by the physicians included in this study may be approximately double the levels reported.

"If so, then implanting fewer than two ICDs per month would be associated with a higher complication rate," Dr. Curtis wrote.

Stephen C. Hammill, M.D., F.A.C.C. from the Mayo Clinic in Rochester, Minnesota, who was not connected with this study said the results highlight the importance of efforts to train and credential physicians implanting these devices and then to track the outcomes of their patients.

"There is great concern in the medical and payer community that ICDs will be implanted by inexperienced physicians resulting in increased patient risk as supported by the Al-Khatib paper. Several things have happened to allow less experienced MDs to implant ICDs, including expanded coverage by Medicare, which greatly enlarges the number of patients who are candidates for the ICD, smaller and easier devices to implant, and reduced cost of the devices to hospitals," Dr. Hammill said.