During a six-month period, only 26 percent of patients taking warfarin to treat atrial fibrillation had stable international normalized ratio (INR) values, according to a registry analysis. In addition, only 34 percent of those patients continued to have stable INR values in the following year.

Lead researcher Sean D. Pokorney, MD, MBA, of Duke University Medical Center in Durham, North Carolina, and colleagues published their results online in JAMA on Aug. 9.

Although previous studies have shown warfarin reduces the risk of stroke among patients with atrial fibrillation, the researchers mentioned the drug has a narrow therapeutic window with an INR value of between 2.0 and 3.0. The added that patients taking warfarin require monitoring and that the drug is associated with drug and food interactions.

For this analysis, the researchers 3,749 patients with atrial fibrillation who were taking warfarin and enrolled from June 2010 through August 2011 in the prospective Outcomes Registry for Better Informed Treatment of Atrial Fibrillation. They included patients who received warfarin at baseline and had three or more INR values in the first six months and six or more INR values in the subsequent year.

The researchers defined stability as 80 percent or more INR values in the therapeutic range of 2.0 to 3.0.

The mean age of patients was 75 years old, 57 percent were males and 26 percent had 80 percent or more of INR values in the 2.0 to 3.0 range during the first six months.

Of the patients with 100 percent of their INR values in range during the first six months, only 37 percent had stability over the subsequent year. Meanwhile, 36 percent of patients with 80 percent or more INRs in range at baseline had at least one well out of range INR.

Further, 33 percent of patients with 100 percent of baseline INR values in range had at least one well out of range INR in the subsequent year.

In addition, 32 percent of patients had a time in therapeutic range (TTR) of 80 percent or more during the first six months. Of those patients, 42 percent had a TTR of 80 percent of more during the subsequent year and 38 percent had at least one INR that was well out of range.

The researchers mentioned a few limitations of the study, including that they assumed the patients’ had a targeted INR range of 2.0 to 3.0. In fact, physicians could have targeted lower or higher INR ranges. They also said the results might not have mirrored community practice. In addition, they did not examine clinical end points with out of range INR values.

“A common belief has been that patients with stable INRs while taking warfarin would continue to be stable and derive less benefit from switching to non–vitamin K oral anticoagulants,” the researchers wrote. “This analysis suggests warfarin stability is difficult to predict and challenges the notion that patients who have done well taking warfarin should maintain taking warfarin.”