Washington, DC - Most patients with atrial fibrillation (AF) aren't getting prescriptions for warfarin, and of those who are on warfarin, most aren't being anticoagulated to the proper therapeutic extent; their international normalized ratios (INRs) are frequently outside the recommended range of 2.0 to 3.0, which puts them at significantly increased risk of intracranial hemorrhage (ICH) or embolic stroke, suggest data gleaned retrospectively from prescription reimbursement claims, hospital records, and similar sources from the years 1999 to 2005 [1]. There was no record of a prescription for an antiplatelet agent for almost half the patients in the analysis.

The findings, from a large community-based population, suggest that the firmly established evidence-based AF indication for warfarin in large part "hasn't translated into action at the primary-care level," according to Dr Alexander Walker (Harvard School of Public Health, Boston, MA), who coauthored the analysis with Dr Dimitri Bennett (GlaxoSmithKline, Philadelphia, PA). Walker told heartwire that their data on the scope and effectiveness of warfarin use in AF "are the only recent US numbers" but are consistent with what's been seen elsewhere.

That, Walker speculates, suggests that there is "a general behavioral phenomenon" shared by primary-care physicians most everywhere: warfarin underuse and inadequate titration to therapeutic INRs, stemming from the practical difficulties of such therapy and, perhaps, overly conservative management by some clinicians attempting to avoid the risks of overanticoagulation. "This has been documented again and again."

The analysis, which appears in the October 2008 issue of Heart Rhythm, is based on data from 116 969 patients with coverage from United Health Care who were at least 40 years old and had a diagnosis of AF or atrial flutter.

Overall in the analysis:

• Only 45% of the population was prescribed warfarin.
• A total of 52% had insurance claims for an anticoagulant or antiplatelet agent.
• Men were significantly more likely than women to be dispensed an anticoagulant (odds ratio 1.46; 95% CI 1.40-1.52).
• Patients aged 60 to 74 years were significantly more likely than those aged 40 to 59 years to be dispensed an anticoagulant (OR 1.68-1.73; 95% CI >1.57 to <1.85, depending on age subgroup).

In a substudy of the 13 115 patients for whom INRs were recorded, the median during follow-up was within the therapeutic range (2.2 INR). However, about one-third of the group was in the therapeutic range <20% of the time, and "only 19% of patients spent all or almost all of their time within the therapeutic range," write Walker and Bennett. The remainder were usually in the subtherapeutic range; INRs >3.0 were far less common.

Alarmingly, but not surprisingly, INR levels <2.0 were associated with significantly higher adjusted risks of embolic stroke and arterial thromboembolism (as defined by ICD-9 codes). INRs higher than the therapeutic range elevated the risk of intracranial bleeding.

"That women in our cohort were slightly less likely than men to receive warfarin is of concern, because untreated women not only have a higher risk for AF-related thromboembolism than men, they also respond as well as men to warfarin," according to Walker and Bennett. The finding hasn't been reported elsewhere, they write, and so "warrants further exploration."