BLEEDING RISK ASSESSMENT WITH THE PE-SARD, PEBSI, AND RIETE SCORES IN PATIENTS WITH ACUTE PULMONARY EMBOLISM DURING THE EARLY ANTICOAGULATION PERIOD

Abstract

A retrospective study was conducted, including patients treated for acute PE at the Department of Pulmonology, Hospital of the Lithuanian University of Health Sciences Kauno Klinikos, between December 1, 2023, and September 1, 2024. Demographic data, risk factors, and comorbidities were collected. Bleeding risk was assessed using the RIETE (sp. Registro Informatizado de la Enfermedad TromboEmbólica), PEBSI (Pulmonary Embolism Bleeding Score Index), and PE-SARD (Syncope, Anemia, Renal Dysfunction) scores. Associations between major bleeding (MB) events and high bleeding risk were analyzed. MB was defined as bleeding causing organ damage (intracerebral or internal organ bleeding) or a hemoglobin drop ≥20 g/L. Results. During the study period, 124 cases of acute PE were identified, of which 23 were excluded due to incomplete data. Among the included patients, 58 (57.4%) were men and 43 (42.6%) women; mean age was 66.8 ± 15.9 years, with no significant sex-related differences. High bleeding risk was identified in 22 (21.8%) patients by PEBSI, 18 (17.8%) by RIETE, and 6 (5.9%) by PE-SARD. MB occurred in 14 patients (13.9%). The sensitivity and specificity for predicting MB were: RIETE – 28% and 90.7%; PEBSI – 45.5% and 94.9%; PE-SARD – 33.3% and 93.2%. The odds of MB increased 3.77-fold for a RIETE score higher than 3.5 (p=0.02, AUC 61.6%), 3.47-fold for a PEBSI score higher than 1.5 (p=0.057, AUC 64.5%), and 6.9-fold for a PE-SARD score higher than 2.25 (p=0.002, AUC 72%). A positive correlation was observed only between RIETE and PE-SARD scores (r = 0.481, p<0.001). Conclusions. All three scoring systems demonstrated high specificity but low sensitivity. This study confirmed the predictive accuracy of the MB risk score and identified its highest effectiveness in assessing MB.