Comparative study of three clinical scoring systems in patients with acute pulmonary thromboembolism

Abstract

Objective To compare the prognostic value of simplified revised Geneva Prognostic Score (sGPS), Pulmonary Embolism Severity Index (PESI) and simplified Pulmonary Embolism Severity Index (sPESI) in patients with acute pulmonary thromboembolism(PTE). Methods A retrospective cohort study was carried out on 276 consecutive patients with identified acute PTE admitted to our hospital from January 1997 to December 2016. We dichotomized patients as low vs. high risk in all three scoring systems. The 30-day mortality of the patients were used as prognostic factors. The prognostic value of each scoring system was evaluated by the area under the receiver operating characteristic curve(ROC). Results (1) The overall 30-day mortality of 276 patients with acute PTE was 22.5%. The 30-day mortality of patients in low vs. high risk groups according to sGPS, PESI and sPESI were 7.6% vs. 47.1%, 1.0% vs. 34.3%, 2.4% vs. 30.9%, respectively. The 30-day mortality of patients in high risk groups according to sGPS, PESI and sPESI were significantly higher than those of patients in low risk groups(P<0.01). The 30-day mortality of patients in low risk groups according to sGPS and PESI were significantly different(P=0.020). The 30-day mortality of patients in high risk groups according to sGPS were significantly different from those of patients in high risk groups according to PESI and sPESI, respectively (P=0.033, P=0.006). (2) The areas under the receiver operating characteristic (ROC) curves for evaluating the prognosis of patients with acute PTE according to sGPS, PESI and sPESI were 0.824, 0.891 and 0.846, respectively. The specificity (84.6%), the accuracy (84.4%) and the positive predictive value (61.2%) of PESI were the highest among the three prediction rules, the sensitivity (83.9%) and the negative predictive value (94.8%) of PESI were also relatively high. The negative predictive value of sPESI (98.6%) was the highest among the three prediction rules. Conclusions PESI can be more accurate for the overall risk stratification of patients with acute PTE, while sPESI is more helpful for identifying those patients with acute PTE who can be discharged early. Key words: Acute pulmonary thromboembolism; Simplified revised Geneva Prognostic Score; Pulmonary Embolism Severity Index; Simplified Pulmonary Embolism Severity Index; Risk stratification