Feature-tracking computed tomography-derived right ventricular strain in patients with severe tricuspid regurgitation evaluated for transcatheter direct annuloplasty

Background Multi-detector computed tomography (MDCT) is key in the evaluation for potential tricuspid valve intervention with transcatheter direct annuloplasty in patients with symptomatic severe tricuspid regurgitation (TR). Novel software permits assessment of right ventricular (RV) global longitudinal strain (GLS) and free wall longitudinal strain (FWLS) from MDCT data. Purpose To investigate the feasibility of feature-tracking MDCT-derived RV GLS and RV FWLS specifically in patients with severe TR undergoing transcatheter direct tricuspid annuloplasty, and evaluate its agreement with echocardiographic measures of RV GLS and RV FWLS. Methods Patients from an ongoing clinical registry of patients with symptomatic severe TR, evaluated with MDCT for transcatheter direct tricuspid annuloplasty were included in this analysis. Feature-tracking MDCT-derived RV GLS and RV FWLS were compared with RV GLS and RV FWLS assessed by speckle-tracking echocardiography (Figure 1). Results In total, 26 patients (78 ± 6 years, 73% females) were included in the current analysis. The majority of the patients had symptoms or signs of heart failure (81% had New York Heart Association functional class III or IV) and all patients had atrial fibrillation. By echocardiographic assessment, half of the population had severe TR, while 5 patients (19%) were diagnosed with massive TR and 8 patients (31%) with torrential TR. On average, RV systolic function was preserved with a mean RV fractional area change of 37 ± 9% and mean tricuspid annular plane systolic excursion of 16 ± 3mm. Speckle-tracking echocardiography showed a mean RV GLS of -17.8 ± 6.1% and mean RV FWLS of -20.1 ± 6.4%. By feature-tracking MDCT analysis, RV GLS and RV FWLS were respectively -17.7 ± 4.4% and -21.7 ± 4.8%. The correlations between feature-tracking MDCT-derived and traditional echocardiographic measured RV parameters are shown in Figure 2. Speckle-tracking echocardiography showed moderate agreement compared to MDCT-derived strain measures (r = 0.638 and r = 0.598, respectively for RV GLS and RV FWLS). The Bland-Altman plot also showed a relatively small bias between both methods, 0.10% and -1.64%, for RV GLS and RV FWLS respectively, and the range of agreement included >95% of the differences between both measurements. Conclusions Assessment of RV GLS and RV FWLS on dynamic feature-tracking MDCT data is feasible in patients evaluated for transcatheter direct tricuspid annuloplasty because of severe TR and, and results show a moderate correlation with speckle-tracking echocardiography.