Warfarin decreases stroke in elderly AF patients: results showed a 50% reduction in the primary end point of death or major disabling stroke.

VANCOUVER, B.C. -- Thromboprophylaxis using warfarin produced a "huge, highly significant, 50% reduction" in death and major disability from stroke in elderly patients with a history of atrial fibrillation, roundly outperforming aspirin prophylaxis in a randomized trial, Dr. Richard Hobbs reported at the annual meeting of the North American Primary Care Research Group.

Feared bleeding complications were no higher than with aspirin therapy in the Birmingham Atrial Fibrillation Treatment of the Aged (BAFTA) trial, conducted in 262 general practices in England and Wales.

"We think warfarin should be routinely prescribed to any patient presenting with atrial fibrillation, regardless of age," declared Dr. Hobbs, lead investigator for the trial and professor and head of primary care and general practice at the University of Birmingham (England) School of Medicine.

The prevalence of atrial fibrillation increases with age, and 50% of all cases are in patients older than 75 years. It raises the risk of stroke fivefold and accounts for a third of all strokes in the elderly, explained Dr. Hobbs during an oral paper presentation at the meeting.

However, existing trial data supporting warfarin as a prophylaxis focused heavily on younger patients with atrial fibrillation who are believed to have a very different risk-benefit profile than patients older than 75. Indeed, among the few elderly patients enrolled in large-scale trials, the risk of bleeding events was high enough that the number needed to treat (46) was close to the number needed to harm (55).

"As a consequence of that, many physicians around the world were in equipoise about whether you should treat patients over 75 with warfarin or give them aspirin. Hence, this trial," Dr. Hobbs said.

The BAFTA trial identified more than 58,000 adults over the age of 75 who had been diagnosed with atrial fibrillation or screened for the condition based on an irregular pulse. Of a total of 973 men and women, 488 were randomized to receive warfarin, and 485 were assigned to receive aspirin. They were followed for a mean of 2.7 years.

The trial differed from original large-scale warfarin/aspirin studies in several ways. Patients, obviously, were much older (mean age 81, compared with about 60). Reflecting their advanced age, they were also less healthy, with many comorbidities and a relatively high mean CHADS2 score (28% with a score from 3-6 on the stroke risk scale for AF patients of 0-6 points), which reflected an elevated baseline risk of stroke.

Few exclusions prohibited entry into the study "because we wanted it to be relevant to most providers' practices," he said.

Patients were not enrolled if they had active peptic ulcer disease, a history of rheumatic heart disease or intracranial hemorrhage, or a major nontraumatic hemorrhage in the past 5 years.

The primary outcome was specified as a fatal or disabling stroke, the latter defined as a stroke resulting in persistent symptoms after 48 hours, intracranial hemorrhage, or significant arterial embolism. Secondary outcomes included major extracranial hemorrhages or major nonstroke vascular events.

Aspirin was given at a dosage of 75 mg/day, and warfarin was prescribed at a target international normalized ratio (INR) of 2.5 within a range of 2-3. Dosages of both drugs were lower than those used in major trials and commonly prescribed by U.S. physicians, Dr. Hobbs said.

The results were clear, he said. "We showed a huge, highly significant, 50% reduction in the primary end point, in terms of death or major disabling stroke," he said.

The patient group receiving warfarin had 1.8 strokes per year, compared with 3.8 per year in the group receiving aspirin. One fatal or disabling stroke per year was prevented for every 50 patients receiving warfarin. "A very important question is, was that at the expense of bleeding?" he asked.

The answer was no, with every major bleeding variable equal between the two groups. A multivariate analysis showed that the results persisted when investigators controlled for age, gender, previous warfarin use, enrollment type (screening or previous atrial fibrillation diagnosis), CHADS2 score, and comorbidities.

A crucial element in the study was the low INR used to guide warfarin therapy, because bleeding risk rises sharply at an INR of 3.5 or higher, he added.

The absolute risk of a major adverse event was 1.9% for patients taking warfarin and 2% for those taking aspirin. Because there was no placebo group, it is unknown whether those events were related to the drugs or were reflective of the background event rate in that population, Dr. Hobbs said in an interview.

The BAFTA study was funded by the Medical Research Council.

BY BETSY BATES

Los Angeles Bureau Number of Strokes per Year In Atrial Fibrillation Patients Warfarin group (n = 488) 1.8 Aspirin group (n = 485) 3.8 Note: Based on data for adults over the age of 75. Source: Dr. Hobbs ELSEVIER GLOBAL MEDICAL NEWS Note: Table made from bar graph.