"Decisions regarding whether or not to perform the operation must consider the overall small, but significant, benefit found in the study, the possibility that the benefit may be further reduced with modern medical therapy, individual patient characteristics, the surgeons' experience and complication rates, and patient preferences," he added.
This modest reduction in risk and the number of patients needed to treat to prevent one stroke may not be enough to justify performing the surgery on most patients, Goldstein said.
"This means that 24 operations would need to be performed to prevent one stroke over five years, or 22 operations to prevent one stroke over 10 years. This includes a 3 percent risk [of stroke from] the operation," he said, adding that there was no reduction in total mortality.
In a comment published with the study, Pierre Amarenco of Bichat Hospital in Paris, France, and colleagues wrote: "... it should be investigated further why the net benefit is not significant in patients older than 75 years, to distinguish whether some older patients could still benefit. Until then, decision making for these patients can only be on the basis of medical judgment for individual patients with in-depth discussions with the patient."
For more information on stroke, visit the U.S. National Library of Medicine.
SOURCES: Larry B. Goldstein, M.D., professor, neurology, and director, Duke Stroke Center, Duke University Medical Center, Durham, N.C.; Sept. 25, 2010, The Lancet
All rights reserved