Background: The 2019 European Society of Cardiology (ESC) guidelines recommend evaluation for right ventricular dysfunction in all normotensive patients with acute pulmonary embolism (PE). We compared the predictive performance of the 2019 and 2014 ESC risk stratification algorithms and the Pulmonary Embolism Severity Index (PESI).
Methods: We performed a post-hoc analysis of normotensive patients aged ≥65 years with acute PE from a prospective cohort. The primary outcome was overall mortality; secondary outcomes were PE-related mortality and adverse outcomes (PE-related death, cardiopulmonary resuscitation, intubation, catecholamine use, recurrent venous thromboembolism) at 30 days. We assessed outcomes in intermediate-high, intermediate-low, and low risk groups according to the 2019 and 2014 ESC algorithms and the PESI. Discriminative power was compared using the area under the receiver operating curve (AUC).
Results: Among 419 patients, 14 (3.3%) died (7 from PE) and 16 (3.8%) had adverse outcomes within 30 days. The 2019 ESC algorithm classified more patients as intermediate-high risk (45%) than the 2014 ESC algorithm (24%) or PESI (37%), and only 19% as low risk (32% with 2014 ESC or PESI). Discriminatory power for overall mortality was lower with the 2019 ESC algorithm (AUC 63.6%), compared to the 2014 ESC algorithm (AUC 71.5%) or PESI (AUC 75.2%), although the difference did not reach statistical significance (p=0.063). Discrimination for PE-related mortality and adverse outcomes was similar.
Conclusions: While categorizing more patients in higher-risk groups, the 2019 ESC algorithm for PE did not improve prediction of short-term outcomes compared to the 2014 ESC algorithm or the PESI.