Predictors of reduced cardiac index in patients with acute submassive pulmonary embolism.

Abstract

Determine the baseline clinical, laboratory, and echocardiographic values that predict reduced cardiac index (CI) among subjects with acute submassive pulmonary embolism (PE). Submassive PE represents a large portion of acute PE population and there is controversy regarding optimal treatment strategies for these patients. There is significant heterogeneity within the submassive PE population and further refinement of risk stratification may aid clinical decision-making. We identified subjects with normotensive acute PE who underwent echocardiogram and right heart catheterization (RHC) prior to catheter-directed thrombolysis (CDT). We sought to determine the predictors of reduced CI, defined as CI < 2.2 L min-1 m-2 . Thirty-two subjects met the inclusion criteria and 41% had reduced CI. Baseline variables did not distinguish subjects with reduced versus normal CI. Brain natriuretic peptide (BNP) was significantly different between the reduced versus normal CI groups (BNP 440 vs. 160 pg/ml, p = .004, respectively). Univariate logistic regression identified BNP, right ventricular (RV):left ventricular (LV) diameter ratio, tricuspid annular plane systolic excursion (TAPSE), and right ventricular systolic pressure as predictors of reduced CI. In a multivariate logistic regression model, only TAPSE was an independent predictor of reduced CI. ROC curve analysis identified the following optimal cut points for prediction of reduced CI: BNP > 216 pg/ml, RV:LV ratio > 1.41, or TAPSE <1.6 cm. Almost half of subjects with acute submassive PE have reduced CI, despite normal systemic blood pressure. Optimal cut points for BNP, RV:LV ratio, and TAPSE were identified to predict reduced CI among patients with acute PE. These findings may aid in clinical decision-making and risk stratification of patients with acute submassive PE.