External validation of the PE-SARD bleeding score for early major bleeding in patients with acute pulmonary embolism: From the COMMAND VTE Registry-2.

ABSTRACT

BACKGROUND: There is no established risk score for anticoagulant-related bleeding during the acute phase in patients with pulmonary embolism (PE). The Syncope, Anemia, Renal Dysfunction (PE-SARD) bleeding score was developed to predict early major bleeding, but has not yet been fully externally validated. OBJECTIVES: To externally validate the PE-SARD bleeding score. PATIENTS/METHODS: Using the COMMAND VTE Registry-2 database, which enrolled 5197 consecutive acute symptomatic venous thromboembolism patients among 31 centers in Japan between January 2015 and August 2020, we identified acute PE patients. We divided those into 3 groups by the score high-risk (>2.5 points), intermediate-risk (1-2.5 points), and low-risk (0 points). The discriminating and calibration performances of the score for 30-day major bleeding were assessed. Subgroup analyses based on active cancer were also performed. RESULTS: Of 2781 eligible patients, the high-risk group accounted for 557 patients (20%), intermediate-risk group for 1412 (51%), and low-risk group for 812 (29%). Major bleeding occurred in 121 patients within 30 days. The cumulative 30-day incidence of major bleeding substantially increased in the higher risk categories by the score (high-risk group 8.2% [95%CI, 5.9%-10.5%], intermediate-risk group 4.6% [95%CI, 3.5%-5.7%], and low-risk group 1.8% [95%CI, 0.8%-2.7%]). The discriminating power of the score was modest with a C-statistic of 0.65 (95%CI, 0.61-0.70) with a good calibration performance with a score of <4 points except for in active cancer patients. CONCLUSIONS: The PE-SARD bleeding score had a modest discriminating performance with a limited calibration performance in acute PE patients without active cancer.