Right ventricular systolic strain in patients with pulmonary hypertension: clinical feasibility, reproducibility, and correlation with ejection fraction.

Abstract

Right ventricular (RV) systolic function is the major determinant of prognosis in patients with pulmonary hypertension (PH) with quantitative assessment by speckle-tracking strain echocardiography emerging as a viable candidate measure. We evaluated a prospective cohort of 231 patients with known or suspected PH referred for clinical echocardiography. All underwent measurement of RV free-wall systolic strain by sonographer staff. Digital images were recorded for blinded offline assessment by an expert echocardiographer. Reproducibility was assessed using the analysis methods of Bland-Altman and the Cohen's-Kappa coefficient. RV strain was feasible in 213 (92%). The average RV systolic pressure was 59 ± 22mmHg. RV systolic strain correlated with functional class, NT-proBNP, and the degree of RV enlargement. The average free-wall systolic strain was -20 ± 7% (range -2 to -37%). The RV strain measures (clinical practice versus blinded expert) had an excellent correlation with a normal distribution (R2 0.87, p < 0.0001). By Bland-Altman analysis, the mean difference in measurement was -1.7% (95% CI -1.4 to -2.1) with a correlation of 0.93, p value of < 0.0001. The reproducibility of RV strain for clinically relevant thresholds was also excellent (Kappa coefficients 0.68-0.83). There was no effect on the variability of strain measures across body mass, pulmonary pressures, or rhythm. RV strain correlated with RV diastolic volumes and ejection fraction with RV free wall strain being the best echo predictor for a reduction in ejection fraction. Here RV systolic strain was found to be highly feasible and reproducible in clinical practice with excellent levels of agreement for clinically relevant thresholds.