Correlation of Echocardiographic Right Ventricular Functional Parameters with CMR-Derived Right Ventricular Ejection Fraction in Pulmonary Hypertension: A Systematic Review and Meta-Analysis

Abstract

Background: Pulmonary hypertension (PH) is associated with increased pulmonary arterial pressure and impairment of right ventricular (RV) function. Reliable characterization of RV performance is important for prognosis, risk assessment, and therapeutic planning. Cardiac magnetic resonance (CMR) is considered the reference technique for quantifying RV volumes and function, but its expense and limited availability can restrict routine implementation. Echocardiography is more widely accessible; however, the extent to which individual echocardiographic indices correspond with CMR-derived RV ejection fraction (RVEF) remains incompletely defined. Objective: Our objective was to determine the correlation of RV fractional area change (RV-FAC), RV free-wall longitudinal strain (RV-FWLS), and RV global longitudinal strain (RV-GLS) with CMR-derived RVEF in individuals with PH. Methods: PubMed, Embase, ClinicalTrials.gov, Scopus, and the Cochrane Library were searched, yielding 404 records; six studies fulfilled the eligibility criteria. Studies were eligible when they enrolled patients with PH and provided extractable correlations between at least one prespecified echocardiographic measure and CMR-derived RVEF. Correlation coefficients were converted with Fisher’s z transformation, combined using common-effect and random-effects approaches, and then transformed back to correlation coefficients. Pooled mean RV-FAC was evaluated as a separate outcome. Between-study heterogeneity was quantified with I2 and τ2, with random-effects estimates emphasized for interpretation. Results: Each evaluated echocardiographic measure was positively correlated with CMR-derived RVEF. RV-FWLS had the largest pooled correlation and the greatest consistency (r = 0.79, 95% CI: 0.73–0.83; I2 = 0.0%). The pooled correlation for RV-GLS was 0.75 (95% CI: 0.60–0.85; I2 = 80.3%), while that for RV-FAC was 0.72 (95% CI: 0.61–0.80; I2 = 61.8%). The random-effects pooled mean RV-FAC was 31.10% (95% CI: 26.77–35.43), with substantial heterogeneity (I2 = 90.9%). Conclusions: RV-FWLS showed the highest numerical and most consistent association with CMR-derived RVEF, suggesting that it may serve as a useful non-invasive adjunct for RV assessment in PH, especially where CMR is not readily available. Larger studies using harmonized acquisition and analysis methods are required for confirmation.