Among patients with atrial fibrillation (AF) treated with warfarin, clinical risk scores for major bleeding and thrombotic events were more strongly associated with future clinical events than any International Normalized Ratio (INR) metrics for warfarin control. A post hoc analysis of ORBIT-AF study by Dr. Pokorney et al, published in JAMA Cardiology, revealed.
The Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF) Registry was a national community-based cohort study that included 51,000 visits of 10,137 AF patients (mean age 74.5 years; 57.4% males). Approximately half of the patients treated with warfarin were included in the bleeding and the other half were included in the thrombotic event analyses. Several quality metrics for warfarin provision were analyzed, including time in therapeutic range (TTR) of 2.0 to 3.0, the most recent INR, percentage of time having interpolated INR values of below 2.0 or above 3.0, INR variance, INR range, and percentage of INR values being in the therapeutic range. The investigators assessed the association of these INR metrics with the risk for bleeding and thromboembolic events in the subsequent 6 months, alone or in combination with clinical factors.
A total of 339 (6.1%) major bleeding events and 51 (0.9%) strokes occurred during a median follow-up of 1.5 years. Compared with any INR metric, the ORBIT-AF Bleeding Score demonstrated a stronger association with major bleeding events (OR=1.62 [95% CI: 1.47-1.79] per 1-SD increase; P<0.001; χ2=96). Among the warfarin control metrics, INR variance was most strongly associated with bleeding (OR=1.36 [95% CI: 1.23-1.52] per 1-SD increase; C-statistic=0.59 [95% CI: 0.56-0.62]). After adjusting for the ORBIT-AF Bleeding Score, three metrics remained associated with bleeding: INR variance (adjusted OR=1.32 [95% CI: 1.19-1.47]), maximum INR (adjusted OR=1.20 [95% CI: 1.10-1.31]), and TTR (adjusted OR=1.16 [95% CI: 1.02-1.32]). However, these metrics provided very limited incremental prognostic value to the risk score. Only the CHA2DS2-VASc risk score (adjusted OR=1.87 [95% CI: 1.42-2.46] per 1-SD increase; P<0.001) but none of the INR metrics demonstrated an association with stroke.
[perfectpullquote align=”full” bordertop=”false” cite=”” link=”” color=”” class=”” size=””]“Using the available information including multiple metrics of high-quality historical INR control, it was difficult to identify patients taking warfarin who were at low risk of future major bleeding and thrombotic events,” Dr. Fonarow told Cardiology Now.[/perfectpullquote]
Given the small number of strokes in the study, more research is needed to examine the relationship between INR metrics and thrombosis. The target INR range, INR measurement, and warfarin prescriptions were at the discretion of treating physicians, which may not be representative of all practitioners. Although visits with fewer than three INR values in the preceding 180 days were excluded, it is unclear whether medication nonadherence or temporary interruption of warfarin would affect the results.
“These findings further reinforce current ACC/AHA/HRS guidelines to preferentially utilize direct oral anticoagulants instead of warfarin for patients with non-rheumatic atrial fibrillation,” Dr. Fonarow noted.