ORLANDO — Warfarin was not nearly as effective in the real world as it has been in clinical trials for reducing the risk of stroke in patients with atrial fibrillation, in a review of about 50,000 patients.
Analysis of the same database also showed that fewer than half of the atrial fibrillation patients who were apparently ideal candidates for warfarin received it, Stephen D. Sander, Pharm.D., said at the annual meeting of the American College of Cardiology.
The unexpectedly low benefit from warfarin therapy “indicates that even when prescribed, the level of anticoagulation achieved may not be optimal to obtain the dramatic effect [from warfarin] observed in clinical trials,” said Dr. Sander, associate director of health economics and outcomes research at Boehringer Ingelheim Pharmaceuticals Inc.
In the patients included in the review, warfarin cut the stroke rate by 15% relative to patients not on warfarin (a 1.2% absolute reduction) in an analysis that controlled for clinical and demographic differences among the patients in the two treatment groups. In contrast, clinical trial results showed that warfarin treatment drops the stroke rate by 60%–70% compared with no anticoagulant treatment, said Dr. Frederick A. Masoudi, a cardiologist at Denver Health Medical Center.
The study used data collected by HealthCore in its Integrated Research Database of more than 20 million commercially insured beneficiaries during January 2004–February 2008. The database included more than 100,000 patients aged 18 or older with atrial fibrillation, with at least two medical claims for the condition and continuous medical insurance coverage during at least 6 months before and at least 6 months after the index atrial fibrillation claim.
From this group, the analysis identified slightly more than 50,000 patients who had no transient cause of atrial fibrillation and no valvular disease. In this subgroup, 41% received at least two prescriptions for warfarin.
Dr. Sander and his associates further reduced the study group by focusing on the nearly 19,000 patients with no apparent precautions in their medical charts against warfarin use and with at least one risk factor for stroke based on guidelines from the American College of Chest Physicians. Among these patients who constituted an “ideal” population for warfarin treatment, 42% received two or more prescriptions and 57% received no prescriptions. (The remaining 1% received a single warfarin prescription.)
Patients were less likely to receive warfarin if they were women, and if they were older than 75. In addition, warfarin use fell with increasing CHADS2 score (congestive heart failure, hypertension, age greater than 75, diabetes, and prior stroke or transient ischemic attack), a measure of the likelihood of stroke occurring in atrial fibrillation patients. This finding is especially surprising because ideally warfarin use should increase as patients' CHADS2 scores increase, an indication of a higher stroke risk, Dr. Masoudi said.
Warfarin use was above average for patients located in the northeastern and western United States, and below average in the midwest and southern regions, Dr. Sander said.
The findings also inexplicably showed that the incidence of hemorrhagic strokes and of bleeding episodes requiring hospitalization were significantly lower in patients who received warfarin, compared with those who did not get the anticoagulant.
Data on testing for the international normalized ratio, available for 748 patients on warfarin, showed that on average these patients spent 55% of the time they were followed in the INR target range of 2.0–3.0, 30% of the time with an INR less than 2.0, and 15% of the time with an INR greater than 3.0. About a quarter of the patients were in the INR target range more than 70% of the time they were tested, and 46% of the patients were in the INR target range less than 40% of the time, Dr. Sander said.
The level of anticoagulation achieved may not be optimal to obtain the dramatic effect' seen in trials. DR. SANDER