Echocardiography-Derived Stroke Volume Index Is Associated With Adverse In-Hospital Outcomes in Intermediate-Risk Acute Pulmonary Embolism: A Retrospective Cohort Study.

Abstract

There remains uncertainty in the optimal prognostication and management of patients with intermediate-risk pulmonary embolism (PE). Transthoracic echocardiography can identify right ventricular dysfunction to recognize intermediate-high-risk patients. Is echocardiographic-derived stroke volume index (SVI) associated with death or cardiopulmonary decompensation in intermediate-risk patients with PE? and Methods: We retrospectively evaluated echocardiographic-derived variables that included SVI in normotensive patients with acute PE who were admitted between January 2012 and March 2017. SVI was determined with the use of the Doppler velocity-time integral in the left or right ventricular outflow tract. The primary outcome was in-hospital PE-related death or cardiopulmonary decompensation. We used logistic regression to determine the association between SVI and outcomes and receiver operating characteristic analysis to compare the performance of SVI and other echocardiographic measures. The primary outcome occurred in 26 of the 665 intermediate-risk patients (3.9%) with PE. Univariate logistic regression showed an OR of 1.37 (95% CI, 1.23-1.52; P < .001) per 1-mL/m² decrease in SVI for the primary outcome. Bivariate logistic regression showed that SVI was independent of age, sex, heart rate, tricuspid regurgitation velocity, tricuspid annular plane systolic excursion, troponin, and Bova score. SVI had the highest C-statistic of 0.88 (95% CI, 0.81-0.96) of all echocardiographic variables with a Youden's J-statistic that identifies an optimal cut-point of 20.0 mL/m², which corresponds to positive and negative likelihood ratios of 6.5 (95% CI, 5.0-8.6) and 0.2 (95% CI, 0.1-0.5) for the primary outcomes, respectively. Low SVI was associated with in-hospital death or cardiopulmonary decompensation in acute PE. SVI had excellent performance compared with other clinical and echocardiographic variables.