Risk score accurately predicts major bleeding risk in COVID-19 patients, study finds
A bleeding risk score had an area under the receiver-operating characteristic curve of 0.84 when predicting major bleeding risk in hospitalized COVID-19 patients 90 days after admission, a study of 3,886 patients in China found.
The International Medical Prevention Registry on Venous Thromboembolism (IMPROVE) bleeding risk score effectively predicted major bleeding risk in COVID-19 patients and provided guidance for low-molecular-weight heparin (LMWH) thromboprophylaxis in older patients, according to results of a multicenter prospective study.
Researchers assessed deidentified data from 3,886 hospitalized COVID-19 patients (median age, 74 years; 60.8% males) across six tertiary hospitals in China between December 2022 and January 2023. All patients were diagnosed by pathogen or antibody detection and followed for 90 days. Major bleeding and clinically relevant nonmajor bleeding (CRNMB) were the primary outcomes, measured at 14, 30, and 90 days after admission. Findings were published by the Journal of General Internal Medicine on Feb. 20.
A total of 42 major bleeding events (1.1%) and 47 CRNMB events (1.2%) occurred during follow-up. The IMPROVE risk score performed well in predicting major bleeding events, with an area under the receiver-operating characteristic curve (AUC) of 0.84 (95% CI, 0.77 to 0.91) at 90 days. The score had a sensitivity of 0.83 (95% CI, 0.71 to 0.94) and a specificity of 0.77 (95% CI, 0.76 to 0.78); the positive predictive value was 0.037 (95% CI, 0.025 to 0.049), and the negative predictive value was 0.998 (95% CI, 0.996 to 0.999). The score's predictive performance decreased for CRNMB and had the highest discrimination ability at 90 days (AUC, 0.73; 95% CI, 0.66 to 0.80).
The 515 patients classified as high risk based on their IMPROVE score had a significantly higher risk of bleeding than the 3,371 low-risk patients. This association remained after adjustment for LMWH thromboprophylaxis (adjusted hazard ratios [HRs], 6.63 [95% CI, 3.62 to 12.15] for major bleeding and 3.69 [95% CI, 2.04 to 6.71] for CRNMB). A total of 37.8% of patients used LMWH for thromboprophylaxis, which significantly increased major bleeding risk in high-risk patients who were at least 65 years of age (adjusted HRs, 5.45 [95% CI, 1.15 to 25.94] at 14 days and 4.16 [95% CI, 1.11 to 15.53] at 30 days; P<0.05 for both comparisons).
Thromboprophylaxis may have influenced the results, and the findings may not be generalizable to general medical patients, the researchers cautioned. They noted that more research is needed to validate the applicability of the risk score in different populations and refine threshold values to improve predictive accuracy.
“Our study supports using the IMPROVE [bleeding risk score] as a bleeding risk assessment tool in COVID-19 patients and recommends caution when using LMWH for thromboprophylaxis in elderly patients at high bleeding risk,” the researchers concluded.
