Use of SAMe-TT2R2 in a racially diverse anticoagulation clinic: prediction of optimal anticoagulation.

Abstract

The SAMe-TT2R2 score predicts optimal long-term oral Vitamin K Antagonist (VKA) anticoagulation for homogenous Caucasian and homogenous Asian populations for non-valvular atrial fibrillation but at different score thresholds. The score that predicts optimal VKA anticoagulation in significantly diverse populations for multiple indications for systemic anticoagulation has not been reported. We determined whether optimal VKA anticoagulation is predicted by SAMe-TT2R2 score in a diverse inner-city population for non-valvular atrial fibrillation (NVAF), unprovoked venous and pulmonary thromboembolic disease (VTE), mechanical prosthetic heart valves and all other indications. All patients on long term VKA's that attended an inner-city anticoagulation clinic between February 2016 and October 2017 were included in this study. Eligible patients were grouped according to oral anticoagulation indication: (1) NVAF, (2) VTE, (3) prosthetic valves and (4) other indications. Each patient's SAMe-TT2R2 score and percent time of INR in the therapeutic range (TTR) was calculated with optimal international normalized ratio (INR) control defined as TTR ≥ 65%. The correlation between SAMe-TT2R2 score and TTR was determined by logistic regression for each oral anticoagulant indication. Receiver operating characteristic curves were then used to identify the best cutoff for prediction of ≥ 65% TTR. Of 316 patients meeting study criteria, 54% were non-Caucasian and there was a significant negative correlation between the SAMe-TT2R2 score and TTR (coefficient - 0.35, P < 0.0001) for all patients. A SAMe-TT2R2 score < 4 was identified as the best threshold for predicting optimal TTR (Youden's J-statistics = 0.238) with accuracy and positive likelihood ratio of 63.4% and 1.73, respectively. The SAMe-TT2R2 score predicts optimal VKA anticoagulation for systemic anticoagulation for multiple indications in a diverse urban population at a higher score than the original report for non-valvular atrial fibrillation of a cohort where < 10% non-Caucasians.