Variability of right ventricular linear measurements of size and function using a standardized echocardiographic protocol

Abstract Funding Acknowledgements Type of funding sources: None. Background Right ventricular size and function have a prognostic value not only in right heart conditions (pulmonary hypertension, congenital heart disease), but also in left ventricular (LV) disease states. The right ventricle (RV) has a unique shape, for which a simple geometrical model is not achievable as is the case for the LV. In clinical practice simple measurements are used for the RV, since these are considered to be most reliable and reproducible: linear cavity dimensions, tricuspid annular plane systolic excursion (TAPSE) and tissue Doppler systolic velocity s’ at the tricuspid annulus. The latest RV quantitation guidelines made important modifications to RV measurements, in particular, by shifting from the apical 4-chamber view to the RV focused apical view. Very scarce data exist on the variability of these measurements. Purpose In this study we aim to analyze the intra and inter-observer variability of the linear measurements of RV size and function using a standardized methodology. Methods A mixed cohort, randomly selected from the echocardiographic database, comprising 50 patients, was retrospectively analyzed. We measured linear diameters (basal = 1, longitudinal = 2, mid = 3) in apical RV-focused view, TAPSE and s’. Based on computer simulations, we propose a step-by-step approach: tracing the longitudinal diameter as perpendicular to the mid of the basal and generally parallel with the interventricular septum, and the mid as perpendicular to the middle of the longitudinal (Figure). Two observers measured independently the same images at baseline, three (inter-observer) and six months (intra-observer). Results The longitudinal diameter had the lowest relative bias (3% inter and 4% intra-observer), followed by the basal (7% and 2%), and mid (6% and 6%). The limits of agreement (LOA) were small for s’ (-3 to 1 intra and -1 to 1mm inter-observer), TAPSE (-6 to 5 and -3 to 3mm), and larger for diameters (1=-5 to 10 and -8 to 6; 2=-13 to 17 and -11 to 6; 3=-11 to 7 and -9 to 4mm). ICCs for individual measurements were very good (1= 0.94 intra and 0.94 inter; 2 = 0.78 and 0.94; 3: 0.72 and 0.56; TAPSE: 0.86 and 0.94; s’=0.94 and 0.98), p < 0.001 for all (Table). Conclusion A systematic approach to linear RV-focused apical view measurements may lead to reproducible results. It is essential that size measurements be performed in the RV focused view. Our study shows best consistency and reproducibility for the basal diameter and the linear functional parameters. Echocardiographic laboratories may benefit from implementing a consistent analysis protocol and assessing its reproducibility. Abstract Figure: linear RV measurements Abstract Table: variability results