Validation of the Predictive Model of the European Society of Cardiology for Early Mortality in Acute Pulmonary Embolism

Abstract

Background Acute pulmonary embolism (PE) is burdened by high mortality, especially within 30 days from the diagnosis. The development and the validation of predictive models for the risk of early mortality allow to differentiate patients who can undergo home treatment from those who need admission into intensive care units. Methods To validate the prognostic model for early mortality after PE diagnosis proposed by the European Society of Cardiology (ESC) in 2014, we analyzed data of a cohort of 272 consecutive patients with acute PE, observed in our hospital during a 10-year period. Moreover, we evaluated the additional contribution of D-dimer, measured at PE diagnosis, in improving the prognostic ability of the model. All cases of PE were objectively diagnosed by angiography chest CT scan or perfusion lung scan. Results The overall mortality rate within 30 days from PE diagnosis was 10% (95% confidence interval [CI]: 6.4–13.5%). According to the ESC prognostic model, the risk of death increased 3.23 times in the intermediate-low-risk category, 5.55 times in the intermediate-high-risk category, and 23.78 times in the high-risk category, as compared with the low-risk category. The receiver operating characteristic analysis showed a good discriminatory power of the model (area under the curve [AUC] = 0.77 [95% CI: 0.67–0.87]), which further increased when D-dimer was added (AUC = 0.85 [95% CI: 0.73–0.96]). Conclusion This study represents a good validation of the ESC predictive model whose performance can be further improved by adding D-dimer plasma levels measured at PE diagnosis.