Evaluation of the SAMe-TT2R2 score to predict the quality of anticoagulation control in patients with venous thromboembolism treated with vitamin K antagonists: Findings from the RIETE registry

Background: The time in therapeutic range (TTR) of patients with venous thromboembolism (VTE) treated with vitamin K antagonists (VKA) is usually below recommended, leading to higher frequency of vascular events, bleeding and mortality. The SAMe-TT2R2 prediction score discriminates those patients with high or low probability of obtaining poor INR control and its use is recommended in patients with atrial fibrillation. Its usefulness has been evaluated in patients with VTE, with conflicting results. Method: We included consecutive patients enrolled in Registro Informatizado Enfermedad TromboEmbolica (RIETE), a prospective multicenter VTE registry, treated with VKA for >90 days and a minimum of 3 INR determinations. We analyzed the relationship between the SAMe-TT2R2 score and TTR, determined by the Rosendaal method and by the percentage of INR determinations (after excluding the first month). A ROC curve was calculated considering a cut-off point of TTR >= 65% for good anticoagulation control. Results: 3893 patients were included and classified in high (1411 patients) or low (2482 patients) probability of obtaining poor INR control according to the total score obtained (0-1 points versus 2 points, respectively). TTR, calculated by direct method and Rosendaal method, was 51.2 (+/- 23.4) and 55.4 (+/- 25.9) in the high probability group; and 54.4 (+/- 23.0) and 58.2 (+/- 25.6) in the low probability group, respectively (p < 0.001 for both comparisons). The outcomes were similar between groups. The predictive capacity of the SAMe-TT2R2 score showed an area under the ROC curve of 0.54 (CI 95% 0.52-0.56) and 0.53 (CI 95% 0.51-0.55). Conclusions: In patients with VTE treated with VKA, the SAMe-TT2R2 score discriminated those patients with high probability of obtaining poor INR control, but with a low predictive capacity. Further studies are required to assess the usefulness of the score in clinical decision-making.