Validating the Composite Pulmonary Embolism Shock Score for Predicting Clinical Deterioration

The Composite Pulmonary Embolism Shock (CPES) score was developed in an interventional cohort to identify normotensive shock in pulmonary embolism (PE) but has not been validated in emergency department (ED) populations. Existing risk stratification tools (e.g., simplified Pulmonary Embolism Severity Index [sPESI], European Society of Cardiology [ESC]) have limited ability to predict short-term deterioration. We evaluated CPES performance compared with sPESI and ESC for predicting clinically relevant outcomes.

We performed a retrospective external validation using prospectively collected multi-center PE registries (2016–2020). Adults with imaging-confirmed PE from six academic ED centers were included. CPES was calculated when ≥ 5 of 6 components were available (tachycardia, right ventricular dysfunction, elevated troponin, elevated brain natriuretic peptide, central thrombus, and concomitant deep vein thrombosis [DVT]). The primary outcome was in-hospital clinical deterioration (death, cardiac arrest, vasopressors or rescue respiratory intervention); secondary outcomes were advanced intervention use and 30-day mortality. Logistic regression and 10-fold cross-validated receiver operating curves (AUC) assessed predictive performance.

Among 1731 patients, clinical deterioration occurred in 193 (11.1%), advanced intervention in 123 (7.1%), and 30-day mortality in 124 (7.2%). For clinical deterioration, AUCs were CPES 0.68 (0.64–0.71), sPESI 0.60 (0.56–0.64), and ESC 0.55 (0.51–0.59). For advanced intervention, AUCs were CPES 0.78 (0.74–0.82), sPESI 0.52 (0.47–0.57), and ESC 0.52 (0.47–0.57). For 30-day mortality, AUCs were CPES 0.57 (0.52–0.61), sPESI 0.63 (0.59–0.68), ESC 0.53 (0.48–0.57). Central thrombus had odds ratios 0.93 (0.64–1.35), and 3.46 (1.49–8.93) for clinical deterioration and advanced intervention, respectively.

CPES outperformed sPESI and ESC for clinical deterioration and advanced intervention, but not mortality. Findings may support incorporating central thrombus location into ED PE risk stratification.