Abstract
The performances of RIETE, VTE-BLEED, SWITCO65 + , and Hokusai-VTE scores for predicting major bleeding events in hospitalized elderly cancer patients with venous thromboembolism (VTE) have not been evaluated. This study validated the performances of these scoring systems in a cohort of elderly cancer patients with VTE. Between June 2015 and March 2021, a total of 408 cancer patients (aged ≥ 65 years) with acute VTE were consecutively enrolled. The overall rates of in-hospital major bleeding and clinically relevant bleeding (CRB) were 8.3% (34/408) and 11.8% (48/408), respectively. RIETE score could categorize patients with increasing rate of major bleeding and CRB into low-/intermediate- and high-risk categories (7.1 vs. 14.1%, p = 0.05 and 10.1 vs. 19.7%, p = 0.02, respectively). The discriminative power of the four scores for predicting major bleeding was poor to moderate, indicated by areas under the receiver operating characteristic curves (0.45 [95% confidence interval, CI: 0.35-0.55] for Hokusai-VTE, 0.54 [95% CI: 0.43-0.64] for SWITCO65 + , 0.58 [95% CI: 0.49-0.68] for VTE-BLEED, and 0.61 [95% CI: 0.51-0.71] for RIETE). RIETE score might be used to predict major bleeding in hospitalized elderly cancer patients with acute VTE.