Predicting short term mortality after investigation for venous thromboembolism

Introduction Deaths following diagnosis of venous thromboembolism (VTE) often result from another concurrent illness. The specificity of mortality markers predicting death from pulmonary embolism is unknown. The aim of this analysis was to compare blood predictors of death in patients with confirmed VTE to patients with negative investigations for VTE. Materials and methods Consecutive patients investigated for VTE were prospectively consented from a single hospital over 9 months. VTE was diagnosed and excluded with a standard diagnostic algorithm. Blood was drawn for biomarker analysis and analyzed in batches for NT-proBNP, high sensitivity troponin T, C-reactive protein (CRP), fatty acid binding protein (FABP) and ischemia modified albumin (IMA). Participants were followed for 3 months. The cohort was analyzed in two groups: those diagnosed with VTE and those who had thrombosis excluded. Regression analysis for 3-month mortality was performed for each group. Results 16/153 patients diagnosed with VTE died within three months (10.5%) as did 23/606 patients who had negative investigations for VTE (3.8%). Predictors for death following VTE included cancer, NT-proBNP, troponin T, FABP, and Hb 500 pg/ml in acute cancer associated VTE predicted death with C-statistic of 0.89 (0.80-0.99). Cancer, NT-proBNP and troponin T also predicted death in patients with negative investigations for VTE. Conclusion Several blood markers are not specific for death from PE and may be surrogate markers of global declining health.