Value of echocardiography using knowledge-based reconstruction in determining right ventricular volumes in pulmonary sarcoidosis: comparison with cardiac magnetic resonance imaging

Abstract

Abstract Background Right ventricular (RV) dysfunction in sarcoidosis is associated with adverse outcomes. Assessment of RV function by conventional transthoracic echocardiography (TTE) is challenging due to the complex RV geometry. Knowledge-based reconstruction (KBR) combines TTE measurements with three-dimensional coordinates to determine RV volumes. Purpose The aim of this study was to investigate the accuracy of TTE-KBR compared to the gold standard cardiac magnetic resonance imaging (CMR) in determining RV dimensions in pulmonary sarcoidosis. Methods Pulmonary sarcoidosis patients prospectively received same-day TTE and TTE-KBR. If performed, CMR within three months after TTE-KBR was used as reference standard. Outcome parameters included RV end-diastolic volume (RVEDV), end-systolic volume (RVESV), stroke volume (RVSV) and ejection fraction (RVEF). Results 282 patients underwent same day TTE and TTE-KBR. In total, 122 patients received a CMR within 90 days of TTE and were included. TTE-KBR measured RVEDV and RVESV showed strong correlation with CMR measurements (R=0.73, R=0.76), while RVSV and RVEF correlated weakly (R=0.46, R=0.46). Bland-Altman analyses (mean bias ±95% limits of agreement), showed good agreement for RVEDV (ΔRVEDVKBR-CMR, 5.67±55.4mL), while RVESV, RVSV and RVEF showed poor agreement (ΔRVESVKBR-CMR, 21.6±34.1mL; ΔRVSVKBR-CMR, −16.1±42.9mL; ΔRVEFKBR-CMR, −12.9±16.4%). Image quality, time to CMR and learning curve showed no impact. Conclusions TTE-KBR is convenient and shows good agreement with CMR for RVEDV. However, there is poor agreement for RVESV, RVSV and RVEF. The use of TTE-KBR does not seem to provide additional value in the determination of RV dimensions in pulmonary sarcoidosis patients. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): ZonMW (The Netherlands Organisation for Health Research and Development) Figure 1. Correlation plotsFigure 2. Bland-Altman plots