Three-dimensional right ventricular free-wall strain for identifying a higher Doppler-estimated PASP subgroup in high-altitude heart disease
Background High-altitude heart disease (HAHD) is characterized by pulmonary hypertension and progressive right ventricular dysfunction. Tricuspid annular plane systolic excursion (TAPSE) may have limited ability to characterize right ventricular impairment at higher pulmonary pressures. We aimed to determine whether three-dimensional speckle-tracking-derived right ventricular free-wall longitudinal strain, reported as an absolute magnitude (absolute RVFWLS), better identifies an operationally defined higher Doppler-estimated pulmonary artery systolic pressure (PASP) subgroup within HAHD than TAPSE, and to explore the relationships of PASP with right ventricular volumetric remodeling and strain. Methods This retrospective study included 100 participants who underwent standardized two-dimensional and three-dimensional echocardiography: 40 high-altitude controls and 60 patients with HAHD. Patients with HAHD were categorized using an operational analytical threshold into higher-PASP (PASP ≥60 mmHg) and lower-PASP (PASP <60 mmHg) subgroups. Single-index receiver operating characteristic (ROC) curves were compared using paired DeLong tests. A logistic model evaluated absolute RVFWLS beyond TAPSE, age, and sex. Exploratory regression-based mediation-model frameworks with 5,000 bootstrap iterations examined whether right ventricular end-systolic volume (RV-ESV) statistically accounted for the associations of PASP with three-dimensional right ventricular ejection fraction (3D-EF) and absolute RVFWLS. Results Of the 60 patients with HAHD, 26 met the operational higher-PASP threshold. Absolute RVFWLS showed better discrimination than TAPSE, with areas under the ROC curve of 0.886 and 0.647, respectively (difference, 0.238; P = 0.002). Adding absolute RVFWLS to TAPSE, age, and sex improved model fit (likelihood-ratio P < 0.001); the odds ratio per 1-percentage-point decrement in absolute RVFWLS was 1.81 (95% confidence interval, 1.29–2.54). RV-ESV statistically accounted for 45.0% (95% confidence interval, 34.8%–67.2%) of the PASP–3D-EF association, whereas it did not materially account for the PASP–absolute RVFWLS association (−1.6%; 95% confidence interval, −8.4% to 5.7%). Septal longitudinal strain was also reduced in HAHD. Conclusion Three-dimensional absolute RVFWLS showed better discrimination than TAPSE for identifying an operationally defined higher-PASP subgroup within HAHD. The association between PASP and absolute RVFWLS was less closely related to RV-ESV than the PASP–3D-EF association. These exploratory cross-sectional findings require prospective validation against invasive hemodynamic measurements and clinical outcomes.