BackgroundTime in therapeutic range (TTR), albeit the standard measure of quality of anticoagulation control for warfarin, is underused in everyday clinical practice because of its tedious calculation. In contrast, the percentage of international normalized ratio measurements in range (PINRR) is a convenient alternative. Our objective was to investigate the correlation between PINRR and TTR and whether PINRR has clinical utility for prediction of ischemic stroke and intracranial hemorrhage in a “real-world” atrial fibrillation (AF) cohort. MethodsThis is an observational study based on a hospital-based AF registry. ResultsAmong 1428 Chinese patients with AF who were taking warfarin (76.2 ± 8.7 years; mean CHA2DS2-VASc, 4.2 ± 1.6 and HAS-BLED, 2.3 ± 0.9), mean and median TTR values were 38.2% ± 24.4% and 38.8% (interquartile range, 17.9% and 56.2%), respectively. Patients with TTR ≥ 65% (14.8%) had a lower annual risk of ischemic stroke (3.04% per year) than did those with TTR < 65% (5.35% per year). Mean and median PINRR were 34.3% ± 17.1% and 34.2% (interquartile range, 22.7% and 46.0%), respectively. TTR significantly correlated with PINRR in a linear fashion (r = 0.81; P < 0.0001). A cutoff of PINRR ≤ 56.1% was a good discriminator of TTR < 65%, with a high sensitivity (98.3%) and positive predictive value (91.9%). The annual ischemic stroke risk in patients with PINRR > 56.1% was 2.56% per year, lower than those with TTR ≥ 65% (3.04% per year). Patients with PINRR > 56.1% had an annual incidence of intracranial hemorrhage comparable to those with TTR ≥ 65% (0.49% per year vs 0.68% per year). ConclusionsAmong patients with AF who are taking warfarin, the PINRR is a user-friendly alternative to TTR, having a high sensitivity and positive predictive value in predicting TTR. As with TTR, PINRR is associated with clinical adverse events, ie, ischemic stroke and intracranial hemorrhage.