Comparison of various prognostic scores for identification of patients with intermediate-high risk pulmonary embolism

Background: Various risk assessment tools have been proposed to identify stable patients with acute pulmonary embolism (PE) who are at high risk of early adverse outcome (i.e., intermediate-high risk).

Methods: We evaluated the ability of the 2019 ESC, Bova, modified FAST and PEITHO-3 models for predicting a 30-day complicated course (death, haemodynamic collapse, and/or recurrent PE) in a cohort of 848 stable patients with acute PE. We also tested whether replacement of echocardiographic right ventricle (RV) dysfunction by computed tomographic (CT) RV enlargement provides comparable prognostic information for identifying these patients.

Results: A complicated course occurred in 63 (7.4 %) of the 848 patients with PE during the first month of follow-up. The proportion of patients designated as having intermediate-high risk PE was 6.7 % with the ESC model, 4.4 % with the Bova score, 15.7 % with the FAST score, and 5.2 % with the PEITHO-3 model. However, among patients identified as intermediate-high risk, the 30-day complicated course rate was higher with the Bova score (21.6 %) than with the ESC model (17.5 %), the PEITHO-3 model (15.9 %), or the modified FAST score (14.3 %). When echocardiographic RV dysfunction was replaced by CT RV enlargement in the models, the proportion of patients classified as having intermediate-high risk PE and the rate of an adverse outcome among them slightly increased.

Conclusions: The Bova score might identify patients with intermediate-high risk PE slightly better than the ESC, PEITHO-3, and modified FAST score. When echocardiography is not readily available, CT-assessed RV enlargement might be used for identifying intermediate-high risk PE.

Clinicaltrials: gov number: NCT02238639.