Pulmonary pressure-to-longitudinal strain ratio by echocardiography: a rapid surrogate to magnetic resonance for right ventricular failure assessment

Abstract Funding Acknowledgements Type of funding sources: Public hospital(s). Main funding source(s): United Christian Hospital Ruttonjee and Tang Siu Kin Hospitals Background Better risk stratification in pulmonary hypertension (PH) by echocardiography (echo) to detect ventricular vascular uncoupling may act as gate-keeper for downstream management, such as MRI and expensive therapies. Studies showed modest correlation found between RV peak global longitudinal strain (RVGLS), which is afterload dependent, and right ventricular ejection fraction (RVEF) Purpose To test the accuracy and optimal cut-off of echo derived mean PA pressure-to-RVGLS ratio against MRI detected severe RV dysfunction (defined as RVEF< 35%), RV dilatation (defined as RVEDVi >87ml), and correlate native T1-values (nT1) Method Strain analyses by echo and volumetric assessment by 1.5 tesla MRI were performed in all patients. Contoured MRI short axis images provided RVEF. In a subgroup of pulmonary arterial hypertension (PAH), right heart catheterization and MRI non-contrast native T1 mapping were performed (Figure 1). Using previous study data, to identify a difference of 1.8 pressure-to-strain ratio between mild and severe PH with a variance  of 2.2 , power of 80% and a significance level of 0.05, a total of 11 participants per group were needed Result Thirty-one PH patients (13 female, age 60 ± 14y, 13 had PAH) were recruited prospectively. Strong correlation was demonstrated between the mean PA pressure-to-RVGLS ratio to MRI derived RVEF (r = 0.80, p < 0.01), and to catheterization derived pulmonary vascular resistance and indexed cardiac output (r = -0.80, p= 0.001; r= -0.75, p = 0.003 respectively). The cut-off value of -2.5 had best accuracy in ROC analyses (Table 1) In PAH patients, this ratio correlated with global nT1 at basal short-axis level (r= -0.91, p = 0.004), but not at the mid short-axis level. Their basal posterior interventricular insertion regions had significantly higher nT1 than those of age-matched normal controls at the same region on the same scanner (1256 ± 217 ms vs. 932 ± 25 ms, p = 0.04) Conclusion In terms of detection of severe right ventricular dysfunction by echocardiography, mean PA pressure-to-RVGLS ratio performed better than RVGLS alone, and a ratio cutoff of -2.5 predicts MRI determined ventricular vascular uncoupling in pulmonary hypertension Table 1 Echo detect MRI AUC standard error 95% CI sens (%) spec (%) p Mean PA pressure -to-RVGLS ratio RVEF < 35% 0.86 0.073 0.71-1.00 72 83 0.007 RVEDVi > 87ml 0.81 0.081 0.65-0.97 83 70 0.004 RVGLS RVEF < 35% 0.76 0.100 0.57-0.96 60 83 0.048 RVEDVi > 87ml 0.73 0.090 0.55-0.91 67 70 0.032 PA pulmonary artery; RVGLS: RV global longitudinal strain; RVEDVi: indexed RV end-diastolic volume Abstract Figure 1