Diabetes mellitus is associated with increased bleeding in pulmonary embolism receiving conventional anticoagulant therapy: findings from a "real-world" study.

Bleeding refers to the most important complication during anticoagulation therapy in patients with pulmonary embolism (PE). However, the incidence and risk factors of bleeding in Chinese population with anticoagulant therapy remains unknown. Although diabetes mellitus (DM) has been demonstrated to increase the risk of PE, little information of its influence on anticoagulation-associated bleeding risk can be available. In our study, 563 acute PE patients, who fulfilled the including criteria were enrolled from a single center and received conventional anticoagulant therapy. And there were 539 patients completed the 3 months following-up. The cumulative incidences of major bleeding (MB) and clinically relevant non-major bleeding (CRNMB) were 3.0% (95% CI 1.01-3.05) and 14.0% (95% CI 1.47-5.21), respectively. Besides, anemia (OR 3.52, 95% CI 1.12-11.41) and recent history of MB (OR 8.14, 95% CI 1.41-31.95) were independently associated with MB. Age >65 year (OR 1.51, 95% CI 1.12-3.11), cancer (OR 2.01, 95% CI 1.12-4.01) and therapeutic range (TTR) during 3 months (OR 0.93, 95% CI 0.91-0.98) were independently associated with CRNMB. Additionally, DM was an independent risk factor for both MB (OR 2.11, 95% CI 1.10-4.12) and CRNMB (OR 2.11, 95% CI 1.10-4.12). Notably, the incidence of MB or CRNMB was significantly higher in DM patients than non-DM patients. At the end of 3-month follow-up, the HbA1C in CRNMB group was 8.3%, yet it was 7.0% in non-CRNMB group among diabetic patients (p = 0.04). In conclusions, the bleeding rates are high in patients with acute PE who receive anticoagulant therapy. In addition to the already known bleeding risk factors, DM can also increase the bleeding risk significantly. Thus, good glycemic control may be essential after prescription of anticoagulant therapy.