Warfarin Less Effective in Real Life Than in Trials


NEW ORLEANS — Real world experience with warfarin suggests that it is not as good at preventing strokes in patients with atrial fibrillation as clinical trial results have suggested, especially among African Americans.

A review of more than 23,000 Medicare patients with atrial fibrillation showed that overall, warfarin prophylaxis cut the stroke rate by 34%, compared with a 65% cut in strokes that's been consistently seen in clinical trials, Brian F. Gage, M.D., reported at the 30th International Stroke Conference.

This lower efficacy in a real-world setting is “disappointing,” said Dr. Gage, an internist at Washington University in St. Louis.

But warfarin performed even worse in the African Americans in the study. In this group, warfarin use was associated with a trend toward more strokes, although this increase was not statistically significant, compared with African Americans not on warfarin.

In this analysis, the 95% confidence interval showed that, at best, warfarin prophylaxis in African Americans produced an 18% reduction in strokes, compared with untreated patients. This benefit is close to the 22% stroke reduction from aspirin prophylaxis in all patients, a suggestion that aspirin prophylaxis may be just as good as or better than warfarin prophylaxis in African Americans. This hypothesis should be tested by analyzing results already collected in prior randomized, controlled studies, Dr. Gage told this newspaper.

The study used a national sample of 23,657 Medicare patients with atrial fibrillation who were treated during April 1998 through March 1999. Warfarin prophylaxis was used by 43% of African Americans in the study and by 50% of white patients.

Information culled from the medical records of the patients on warfarin therapy showed that this prophylaxis was often used in a less-than-ideal manner. Patients who regularly receive warfarin should have their dosage adjusted based on their international normalized ratio (INR), a measure of clotting time. Ideally, INRs should be measured about every 28 days in patients who regularly take warfarin.

Among all white patients, the average time between INR measurements was 26 days, and among African Americans the average interval was 30 days. But 25% of the white patients on warfarin had an interval of 39 days or longer between INR measurements. Among African Americans on warfarin, 25% had an interval of 57 days or longer between measurements, Dr. Gage said at the conference, sponsored by the American Stroke Association. For these subgroups, the interval between INR measurements was “way too long,” he said.

But substandard INR monitoring was not the only reason patients got less benefit from warfarin prophylaxis, compared with the benchmark of clinical trials. The rate of protection from stroke remained unexpectedly low among African Americans even in an analysis that controlled for the frequency of INR monitoring as well as clinical variables that predispose patients to strokes.

Other factors that were not controlled for in this analysis, and that may help explain warfarin's underachievement, include poor compliance with the warfarin regimen, inadequacies in the health care setting, and inadequate access to anticoagulant services, said Dr. Gage, who is also medical director of the Blood Thinner Clinic at Barnes-Jewish Hospital in St. Louis.

The clinical trials that assessed warfarin's efficacy for preventing strokes in patients with atrial fibrillation were highly selective; more than 90% of patients who were initially assessed for these trials were eventually excluded. The clinical trial results therefore came from patients that were mostly white, less than age 75 years, and followed very closely, and these results may not be generalizable to other health care settings, Dr. Gage said.

Recommended Reading

TPA Safe, Effective for Stroke After Cardiac Cath
MDedge Cardiology
Advances in Imaging Renew Interest in Carotid Bypass Surgery
MDedge Cardiology