Abstract Body (Do not enter title and authors here): Background: Intermediate-risk pulmonary embolism (PE) is common and associated with significant mortality. We previously demonstrated that right ventricular global longitudinal strain (RVGLS) predicted short-term outcomes in intermediate-risk PE in a single-center retrospective cohort study using data up to 2018, identifying a best cutoff value of 17.7 %. However, limited evidence exists regarding the prognostic utility of RVGLS in more recent cohorts following the 2019 European Society of Cardiology guideline update. We conducted an external validation study using data from another institution between 2019 and 2022.
Methods: Retrospective cohort study of 119 patients with intermediate-risk PE from 2019 to 2022. The primary outcome was all-cause 30-day mortality. Echocardiographic parameters were compared between survivors and non-survivors. RVGLS and right atrial (RA) strain were applied using Philips software. Receiver operating characteristic (ROC) curves analysis and Kaplan-Meier analysis were performed to assess prognostic value.
Results: Among patients (mean age: 66.1 years, 48.7 % female), mortality at 30 days occurred in 6 patients (5.0 %). Image quality was sufficient to perform RV and RA strain analysis in 110 patients (92.4 %) and 105 patients (88.2 %). Non-survivors had significantly lower RVGLS (17.3 % vs. 19.6 %, p = 0.013) and RA strain (25.8 % vs. 31.4 %, p = 0.040) than survivors. Univariate analysis showed both RVGLS (odds ratio [OR] 0.567, p = 0.033) and RA strain (OR 0.865, p = 0.046) were associated with 30-day mortality. In ROC curves analysis, RVGLS had a higher area under the curve of 0.802 (Figure 1), compared to 0.751 for RA strain. Applying the previously reported cutoff of 17.7 %, the ROC curve provided 66.7 % sensitivity and 72.1 % specificity. When patients were divided into two groups utilizing the RVGLS value of 17.7 %, Kaplan-Meier curves demonstrated patients with low RVGLS had higher risk of 30-day mortality compared to those with high RVGLS (Log rank P = 0.042. Figure 2).
Conclusions: In this external cohort, reduced RVGLS at baseline was associated with increased 30-day mortality in patients with intermediate-risk PE. These data validate the prognostic significance of RVGLS previously reported in a single-center study. Prospective studies are warranted to evaluate whether incorporating RVGLS into treatment algorithms can impact care and improve patient outcomes.