Comparison of different imaging modalities for the quantification of tricuspid valve geometry and regurgitation: A retrospective, single-center study

Abstract

Background and Aims: Tricuspid regurgitation (TR) is a frequent valvular heart disease with relevant adverse impact on patients' prognosis. Adequate TR imaging and evaluation is challenging. In this study, we aimed to compare different imaging modalities (echocardiography and multi-slice computed tomography) for the assessment of tricuspid valve (TV) function and geometry. Methods: We retrospectively investigated patients that presented to University Hospital Bonn, Germany, between September 2018 and March 2019, who underwent comprehensive echocardiography and multi-slice computed tomography (MSCT) to evaluate TR. MSCT was considered the reference approach for dimensional assessment of TV anatomy and echocardiography (transthoracic echocardiography + transesophageal echocardiography) for functional assessment of TV. We used Spearman's Rank order correlation, Bland-Altman analysis, and intra-class correlation to compare the different imaging modalities. Results: Forty patients (Mean Age ± SD: 77.5 ± 7.1 years; 35% female) with high grade TR (effective regurgitant orifice area, EROA: 0.49 ± 0.3 cm2, RegVol: 49.5 ± 13.4 mL) were included. There was a statistically significant but moderate correlation between 2D-TEE and MSCT for anteroposterior (AP) (r = 0.68, 95% confidence interval [CI]: 0.44-0.93, P =.05; intraclass correlation [ICC]: 0.77, P =.03) and sep-tolateral (SL) diameters (r = 0.71, 95% CI: 0.33-0.93, P =.03; ICC = 0.76, P =.05). MSCT and 3D-TEE showed a strong correlation for determination of TV annulus area (r = 0.94, 95% CI: 0.57-0.98, P =.002; ICC = 0.95, P =.4), perimeter (r = 0.9, 95% CI: 0.6-0.98, P =.002; ICC = 0.97, P =.3) and diameters (AP-Diameter: R = 0.73, 95% CI: 0.06-0.94, P =.03; ICC = 0.83, P =.09; SL-Diameter: R = 0.86, 95% CI: 0.47-0.97, P =.02; ICC = 0.95, P =.1). Only 3D-TEE allowed for direct measurement of planimetric EROA, which exhibited a significant difference from calculated EROA (0.49 ± 0.4 cm2, 0.67 ± 0.17 cm2, P = .05; r = 0.93, 95% CI: 0.5 to 0.99, P = .006). According to Bland-Altman analysis, we found a relevant agreement between MSCT and 3D-TEE only for TV area (bias: −1.95, 95% limits of agreement −3.6 to −0.1). Conclusion: Only 3D-TEE allowed for sufficient simultaneous functional and dimensional assessment of TR in our cohort.