Right ventricular and left ventricular diameters are independent predictors of death or cardiopulmonary resuscitation in intermediate and high-risk pulmonary embolisms

Background

Increased right ventricular (RV) dilation measured by the ratio of RV diameter (RVD) to left ventricular (LV) diameter (LVD) (RV:LV) is associated with pulmonary embolism (PE) severity and mortality. Data regarding the individual contributions of RVD and LVD are limited.

Purpose

To examine RVD and LVD as independent contributors to death or need for cardiopulmonary resuscitation (CPR) in intermediate- or high-risk PEs treated with surgical or catheter-based strategies

Methods

We measured basal RVD and LVD on presenting transthoracic echocardiograms (TTE) using the diastolic 4-chamber view on 127 PEs managed with surgical embolectomy (n=95, 75%), extracorporeal membrane oxygenation (n=23, 18%), or catheter-directed embolectomy (n=9, 7%) for intermediate- (64%) or high-risk (36%) PE (based on European Society of Cardiology criteria) between 2005 and 2022. The primary outcome was the composite of death (n=2) or survivors requiring CPR (n=10).

Results

A total of 127 patients were analyzed. Subjects were 57±14 years, 38% women, BMI 34±8. Mean RVD and LVD were 4.4±0.9 cm and 3.9±0.8 cm respectively. All presented with severe RV dysfunction on TTE and elevated cardiac biomarkers (Troponin-I or B-type natriuretic peptide). Mean RV:LV was 1.2±0.3 (range 0.7 to 2.8). Using logistic regression, higher RV:LV was associated with increased odds of death or CPR (odds ratio (OR) 15 [95% confidence interval (CI): 2.5, 82] per 1-unit increase, p=0.002. RV:LV >1.2 was the cutoff most associated with death or CPR, OR 7.2 [95% CI: 1.5, 34.5], p=0.01. Increasing RVD [OR 3.8 (95% CI: 1.1, 12.8), p=0.03] and decreasing LVD [OR 4.9 (95% CI: 1.3, 16.9), p=0.02] were independent predictors of death or CPR. RVD >5.0 cm (OR 5.9 [95% CI: 1,5, 23.2], p 0.01) and LVD <3.6 cm (OR 7.0 [95% CI: 1.7, 27.9], p=0.006) were the cutoff values most associated with the primary outcome. These cutoff values remained significant predictors even after adjustment for body surface area. Other parameters or RV size and function (diastolic area, systolic area and fractional area change) did not predict death or CPR. All of the subjects meeting the primary outcome had high-risk PE.

Discussion

In addition to RV:LV greater than 1.2, RVD and LVD were independently associated with death or CPR among high- or intermediate risk PEs. The independent contribution of declining LVD on PE mortality is a novel finding and highlights PE associatient LV pathophysiology (hyperdynamic and underfilled) antecedent to hemodynamic collapse. PE literature focuses on RV dilation as a predictor of PE mortality. LVD may represent a useful measure to risk stratify PE patients and predict hemodynamic decompensation.

Funding Acknowledgement

Type of funding sources: None.