Comparisons of clinical scoring systems among suspected pulmonary embolism patients presenting to emergency department.

Abstract

Pulmonary embolism (PE) is among the most severe cardiovascular disorders worldwide. Timely and appropriate diagnosis of PE remains an important step in reducing PE related mortality and morbidity. In this retrospective single-center cohort study, we comprehensively compared the screening performances of several clinical scoring systems (Wells score [WS], Revised Geneva score [RGS], WS + d-Dimer [D-D], RGS + D-D, WS + PE rule-out criteria [PERC] and RGS + PERC) among PE suspected patients. Failure rates across different PE severity grades as well as overall sensitivity/specificity were considered in evaluating each screening strategy. A total of 3437 patients were included in this study and 698 of them were diagnosed with PE. Patients with and without PE were similar in demographics, while significantly different in respiration-related characteristics. Compared with WS or RGS alone, Integrating PERC or D-D with WS or RGS significantly decreased the failure rates across all PE severity grades, and increased the overall sensitivity from 88.5% and 87.2% to 96.3% and 94.8% (D-D) to 99.4% and 99.6% (PERC), respectively. However, compared with other four scoring approaches, using WS or RGS alone increased the specificity from 8.3% and 7.2%, 38.3% and 21.3%, to 63.5% and 34.8%, respectively, and increased the AUC from 0.54 to 0.54, 0.70 and 0.69, to 0.8 and 0.76, respectively. In general, all screening approaches achieved better performances among PE patients with respiratory distress compared to those without respiratory distress. Combining PERC or D-D with WS or RGS, and the presence of respiratory distress provide significantly better PE rule-out performances.