Increased troponin I predicts in-hospital occurrence of hemodynamic instability in patients with sub-massive or non-massive pulmonary embolism independent to clinical, echocardiographic and laboratory information

Introduction: Whether in patients with acute central sub-massive or non-massive pulmonary embolism, mild troponin I increase (>0.03 mu g/L) predicts in-hospital occurrence of hemodynamic instability and death independent to prognostically relevant clinical, laboratory and echocardiographic information is not fully established. Methods and results: We evaluated consecutively patients admitted to the Emergency Room for pulmonary embolism; those in stable hemodynamics in whom central pulmonary embolism was confirmed by spiral-computed tomography were recruited. All participants underwent standardized study protocol, including clinical and diagnostic evaluation for assessment of severity of pulmonary embolism; therapy was established accordingly; troponin I was measured, but treatment protocol was not affected by knowledge of troponin I levels. Of 90 patients enrolled in the study, 33 (37%) developed hemodynamic instability during hospitalization (on average, 90 h :20 from admission). Troponin I was >0.03 mu g/L in 56% of the study population at admission, and predicted occurrence of hemodynamic instability during hospitalization (adjusted hazard ratio 9.8, 95% confidence interval 1.2-79.2), independent to age, gender, co-morbidity, systolic blood pressure, CK-MB, echocardiographic right ventricular dysfunction and other covariates. Twelve patients died during hospitalization (mean time to event 107 h +/- 24 from admission); troponin I >0.03 mu g/L predicted mortality in univariate analysis, but not after accounting for age, sex and clinical variables. Nevertheless, higher troponin as continuous variable correlated with higher likelihood of in-hospital death (adjusted likelihood ratio 2.2/mu g/L, 95% confidence interval 1.1-4.3) in multivariate analysis. In a further multivariate model, CK-MB predicted mortality independent of covariates and troponin I. Conclusions: In patients with acute central sub-massive or non-massive pulmonary embolism, even mild increase in troponin I >0.03 mu g/L may provide relevant short-term prognostic information independent to clinical, laboratory and echocardiographic data. (c) 2007 Elsevier Ireland Ltd. All rights reserved.