TORONTO — Older patients with atrial fibrillation do not have worse warfarin control than do younger patients; international normalized ratio control and variability did not differ substantially in older and younger patients, Dr. Margaret C. Fang reported at the annual meeting of the Society of General Internal Medicine.
The Anticoagulation and Risk Factors in Atrial Fibrillation (ATRIA) study was a cohort study involving 13,559 patients with nonvalvular atrial fibrillation enrolled in Kaiser Permanente of Northern California. The study was led by investigators at the University of California, San Francisco; the Kaiser Permanente Division of Research; and Massachusetts General Hospital, Boston. Patients were monitored for a median follow-up of 6 years.
Warfarin use and anticoagulation intensity were determined using validated algorithms based on warfarin prescriptions and serial outpatient INR test results. The primary outcomes were the amount of time in therapeutic INR ranges, INR variability, and median number of days between INR measurements, excluding the first 4 weeks of therapy.
At baseline, patients aged 80 years or older were significantly less likely to receive warfarin than were those younger than 80 years (43% vs. 56%), and more likely to have stroke risk factors or prior strokes.
Patients aged 80 years and older were just as likely to be in a therapeutic INR range of 2.0–3.0 as younger patients were: 64.0% of patients 80 years or older, compared with 66.0% of those younger than 80 years (adjusted odds ratio 1.0; 95% confidence interval 0.9–1.0).
There were no significant differences by age in the proportion of INRs 4.0 and above (1.86% vs. 1.80%, respectively, adjusted OR 1.0 [CI 0.7–1.4]). The median number of days between INR measurements was 21 for both older and younger patients, indicating no need for more frequent INR testing in older patients.
In discussing the limitations of her study, Dr. Fang said most of the patients were managed by organized anticoagulation clinics, so the findings may not be generalizable to other clinical settings. Because this was not a randomized trial, more adherent and easier to manage patients may have been selected, possibly leading to “overly optimistic results.”
Patients aged 80 years and older were as likely to be in a therapeutic INR range of 2.0–3.0 as were younger patients. DR. FANG