218: Can the “Bleeding Academic Research Consortium” (BARC) classification be applied to pulmonary embolism?

Abstract

Bleeding is associated with higher risk of adverse outcomes in pulmonary embolism (PE). The “Bleeding Academic Research Consortium” (BARC) developed a classification of events combining laboratory & clinical parameters, but based on acute coronary syndrome patients. We investigated whether this classification is applicable in the context of PE, where no standard bleeding definition exists. Prospective, single-center registry of patients with confirmed PE. We excluded BARC types 1 or 4 bleeding, considered not to be applicable to the context of PE. BARC type 2 bleeds were defined as any overt bleeding requiring non-surgical or medical care, or leading to hospitalisation. Type 3 bleeds were defined as drop of >3 g/dL in haemoglobin, any transfusion, tamponnade, intra-cranial hemorrhage, or bleeding requiring surgical intervention. Type 5 bleeds were defined as any fatal bleed. From 2007 to 2011, 666 patients with confirmed PE were included; average age 66±18 years; 52% women; 25% low-risk, 61% intermediate-risk and 14% high-risk PE. Treatment was: unfractionated heparin in 93 (14%), enoxaparin in 200 (30%), fondaparinux in 373 (56%). Thrombolysis was given in 167 (25%). Sixty patients (9%) experienced bleeding (n=13, 43, 4 for BARC types 2, 3, 5 respectively). Main in-hospital events are shown in table 1. By multivariate analysis, independent predictors of in-hospital death were: cardiogenic shock (OR 12.6 [4.8-20.8]); chronic obstructive pulmonary disease (OR 5.27 [2.25-8.43]); acute right ventricular dysfunction (OR 2.98 [1.25-6.96]) and any bleeding (BARC 2,3,5) (OR 3.15 [1.34-7.37]). Our data suggest that the BARC classification can be applied to acute PE and that bleeding is associated with unfavourable in-hospital outcome. We suggest use of the BARC as the standard for classification of bleeding events in PE. Table Results BARC type 2 (A) BARC type 3/5 (B) No bleed (n=606) C) p (A) vs (C) p (B) vs (C) Death 2 (15.4%) 7 (15%) 19 (3.1%) 0.034 0.0014 Recurrent PE 0 4 (8.5%) 5 (0.9%) 0.67 <0.0001 Treatment escalation 1 (7.7%) 2 (4.2%) 4 (0.7%) 0.017 0.028