The steps towards understanding tricuspid regurgitation: one ‘careful’ step at a time and many more ‘careful’ steps needed

This commentary refers to ‘Secondary tricuspid regurgitation: incidence, types, and outcomes in atrial fibrillation vs. sinus rhythm’, by J.A. Naser et al., https://doi.org/10.1093/eurheartj/ehae346 and the discussion piece ‘What's the relationship between secondary mitral regurgitation and secondary tricuspid regurgitation in atrial fibrillation patients and how to treat them?’, by Y. Sun et al., https://doi.org/10.1093/eurheartj/ehae815. To the editor, We appreciate the interest from Sun and colleagues in our work on the incidence, types, and outcomes of secondary tricuspid regurgitation (STR) in atrial fibrillation (AF) and sinus rhythm.1 Sun et al. discuss how our study findings relate to their unpublished work which investigated changes in atrial secondary mitral regurgitation (A-SMR) and atrial secondary tricuspid regurgitation (A-STR) after totally thoracoscopic surgical ablation (TTA) of AF. They first found that the severity of A-SMR, but not A-STR, decreased after TTA. The finding is interesting but is difficult to interpret without full understanding of the criteria used to define A-SMR and A-STR and the characteristics of the study population. It appears that patients in their cohort potentially had pulmonary hypertension (PH) which suggests that tricuspid regurgitation (TR) was not purely due to the atrial secondary mechanism. Presuming that only patients with preserved ejection fraction were included, their A-STR population may have included significant aortic valve disease, heart failure with preserved ejection fraction (HFpEF), and pre-capillary PH. HFpEF can cause both A-STR and ventricular STR. Therefore, if these conditions remained inadequately treated, the lack of improvement in TR is not surprising. Although we showed1 that AF increases the risk of all types of STR due to right atrial enlargement and elevation in right ventricular systolic pressure, the role of rhythm control in regression of A-STR needs to be evaluated in future studies since previous studies included heterogenous TR.2 Furthermore, many pertinent details to the understanding of the data by Sun et al. remain unclear (e.g. time of follow-up, measurements to determine severity, duration of AF, success rate of TTA, and recurrence of AF). The difference between mitral regurgitation (MR) and TR reduction may also be related to the previous finding that STR develops following 40% dilation in the tricuspid annulus area, as opposed to SMR which develops with 75% dilation in the mitral annulus area.3 Thus, if both annuli are dilated to a similar extent, MR improvement can be achieved with a lower degree of annular area reduction as compared to TR. Sun et al. then discuss their second finding on the association between A-STR and A-SMR reduction. Only a minority of patients (<10%) had moderate/severe MR; most patients had mild/trivial MR that was reduced to trivial/no MR. Such low severity of MR would not be expected to cause PH, as suggested, but MR is likely a biomarker for coexistent left atrial myopathy and HFpEF.4 It is unclear whether this association is more pronounced among patients with successful rhythm control. A plausible explanation is that AF can worsen hemodynamics by loss of atrial systole, whereas restoration of sinus rhythm can improve hemodynamics including pulmonary pressures and thus TR. Another possible explanation is the reduction in AF burden may have helped in reverse right and left atrial remodelling and thus TR and MR. We wish to emphasize that the goal of our study1 was to evaluate the role of new-onset AF on the incidence and types of ≥moderate STR. Accordingly, structural heart disease at baseline (e.g. ≥moderate MR, other left-sided heart disease, PH), which is associated with AF, was an exclusion criterion, so that the impact of new-onset AF on incident TR can be relatively isolated. Importantly, patients developed all these conditions at follow-up, and they were considered in classifying incident TR types (i.e. were not excluded at that point). Patients with A-SMR without PH or diastolic dysfunction were indeed considered in the A-STR category, as mentioned in the methods.1 Since PH causes elliptical/spherical right ventricular remodelling and tethering of the tricuspid leaflets,5 PH was considered suggestive of ventricular STR according to the underlying process (e.g. left-sided valve disease, left ventricular disease, other PH). Finally, our study was not designed to evaluate the association between A-SMR and A-STR, the role of PH in TR, or the role of rhythm control in established significant TR; however, we agree that these are important questions that require dedicated evaluation in future studies designed for these purposes. Nothing to declare.