Refining risk stratification in pulmonary embolism: Integrating glomerular filtration rate and simplified pulmonary embolism severity index as a potent predictor of patient survival
Background/Aim. Patients classified as belonging to simplified pulmonary embolism severity index (sPESI) class 0 are considered to have low-risk pulmonary embolism (PE). Yet, certain laboratory and echocardiographic parameters not accounted for in the sPESI score might suggest a likelihood of worse outcomes in PE cases. This study seeks to determine if the prognostic value of the sPESI score in acute PE can be improved, refined, and optimised by incorporating brain natriuretic peptide (BNP) and troponin I (TnI) levels, echocardiographic parameters, or glomerular filtration rate. Methods. The study encompassed 1,201 consecutive patients diagnosed with PE, confirmed by multidetector computed tomography (MDCT). Upon admission, each patient underwent an echocardiography exam, and blood samples were taken to measure B-type natriuretic peptide (BNP), troponin I (TnI), creatinine, and other routine laboratory markers. Results. The in-hospital mortality rate was 11.5%. The patients were categorized into three groups using the three-level sPESI model: sPESI 0, sPESI 1, and sPESI ≥ 2. Statistically significant differences were found among these groups regarding mortality rates, TnI values, BNP levels, estimated glomerular filtration rate (eGFR), and the presence of right ventricular dysfunction (RVD). Cox regression analysis identified eGFR as the most reliable predictor of 30-day all-cause mortality [HR 2.24 (CI 1.264-3.969); p = 0.006] across all sPESI categories. However, incorporating TnI, BNP, or RVD did not improve risk prediction beyond the three-level sPESI model. Conclusion. Renal dysfunction at the time of admission is closely related to an elevated risk of in-hospital mortality in patients with acute PE. The three-level sPESI score offers a more accurate method for prognostic stratification in these patients.