Transcatheter tricuspid valve repair in patients with pulmonary hypertension

This commentary refers to ‘Clinical characteristics, diagnosis, and risk stratification of pulmonary hypertension in severe tricuspid regurgitation and implications for transcatheter tricuspid valve repair’, by P. Lurz et al., on page 2785. We have been intrigued by some aspects of report by Lurz et al. on the negative impact of pulmonary hypertension on outcome in patients undergoing transcatheter tricuspid valve repair reported by Lurz et al.1 The authors defined pulmonary hypertension by a systolic pulmonary artery pressure (sPAP) >50 mmHg, which corresponds to a mean PAP (mPAP) of 32 mmHg. This is much higher than the internationally agreed definition by a mPAP >25 mmHg, recently decreased to 20 mmHg.2 The 50 mmHg cut-off value should have been derived from a ROC curve since the multivariable analysis demonstrated sPAP to be an independent predictor of outcome. Invasive and non-invasive measurements of PAP were compared using correlation calculations. Agreed statistics better rely on a Bland and Altman3 analysis to derive measures of accuracy (bias) and precision (limits of agreement). Absence of difference on the means of sPAP shown in Table 1 suggested accuracy, in contrast with a regression line which deviated from the line of identity shown in Figure 2 suggesting pressure-dependent underestimation. The under-estimation of sPAP by Doppler transtricuspid gradients is expected in severe tricuspid insufficiency because of increased right atrial pressure but also tricuspid valve deformation which limits the application of the simplified form of the Bernouilli equation. Brecker et al.4 discussed this problem with high fidelity pressure measurements and showed a systematic under-estimation of 0.8 × sPAP − 0.2 mmHg. It was −11% in the sPAP under-estimation subgroup in the study by Lurz et al. The report did not discuss the tricuspid annular plane systolic excursion (TAPSE)/sPAP ratio, a valid estimate of right ventricular function adaptation to afterload of known potent predictive capability in various types of pulmonary hypertension on heart failure with a cut-off value of 0.36.5 It was an independent predictor in the study by Lurz et al., but no cut-off value was ROC curve-determined. We, therefore, propose a perhaps more robust mPAP/(TAPSE/sPAP) decision tree to assist the difficult decision of tricuspid valve repair in pulmonary hypertensive patients (Figure 1). Decision-making algorithm. mPAP, mean pulmonary artery pressure (at right heart catheterization); PVR, pulmonary vascular resistance; RHC, right heart catheterization; sPAP, systolic pulmonary artery pressure (at echocardiography); TAPSE, tricuspid annulus plane systolic excursion. Decision-making algorithm. mPAP, mean pulmonary artery pressure (at right heart catheterization); PVR, pulmonary vascular resistance; RHC, right heart catheterization; sPAP, systolic pulmonary artery pressure (at echocardiography); TAPSE, tricuspid annulus plane systolic excursion. Conflict of interest: none declared.