A prognostic model for short term adverse events in normotensive patients with pulmonary embolism

Risk stratification of patients with PE has gained interest in terms of the identification of patients in whom treatment on an outpatient base can be considered. Previous studies are of limited value due to their focus on adverse clinical events within several months after diagnosis of PE. We developed a prognostic model, based on easily accessible, clinical, and laboratory parameters, to predict adverse events during the first 10 days after the diagnosis of acute PE. We have analyzed the data of 210 outpatients with confirmed PE. Collected data included medical history, pulse rate, blood pressure, NT-proBNP, and D-dimer concentrations. The primary outcome was the occurrence of adverse clinical events in a 10 day follow-up period. Our final prognostic model to predict short-term adverse events consists of NT-proBNP levels, D-dimer concentrations, pulse rate, and the occurrence of active malignancy; the total score ranges from 0 to 37 points. Patients with a low score (no active malignancy, pulse rate < 90 bpm, NT-proBNP < 500 pg/ml, and D-dimer < 3,000 mu g/l FEU) have a 10-day adverse event risk < 1.5%. This risk increases to over 30% in patients with a maximum score, based on high pulse rate, D-dimer concentrations, and NT-proBNP levels. Our prognostic model, once prospectively validated in an independent sample of patients, can be used in the early risk stratification of PE to estimate the risk of adverse events and to differentiate between candidates for in- or out-hospital treatment. Am. J. Hematol. 86: 646-649, 2011. (C) 2011 Wiley-Liss, Inc.