DISCUSSION:
The major findings of the present study are as follows:
  1. Intraprocedural MG was found to be a strong predictor for 12-month mortality (cut-off value: 4.5 mmHg) and adverse functional outcomes (cut-off value: 3.9 mmHg).
  2. We found the worst functional outcomes assessed by NYHA functional class and walk distance as well as the highest 12-months mortality in patients with persistent MG ≥ 4.5 mmHg at discharge independently from the MR aetiology followed by in patients with an in-hospital decrease in MG to values below 4.5 mmHg at discharge.
  3. LAI (<1.11) was found to be the strongest predictor for unfavourable intraprocedural MG (>4.5 mmHg) followed by baseline MG, the number of implanted clips, and central clip localisation.
Elevated MG following the MitraClip procedure is considered to be associated with adverse outcomes (20). Therefore, the current guidelines recommend avoiding an intraprocedural transmitral pressure gradient above five mmHg (11). On the other hand, some contradicting studies currently show no relevant association between MG and outcomes after the MitraClip procedure, particularly in patients with functional MR (6), (7). Therefore, the effect of MG on outcomes after MitraClip is ambiguous.
Intraprocedural assessment of mitral inflow patterns and hemodynamics might be challenging owing to altered mitral geometry, such as the double- or multi-orifice MVs after transcatheter MV repair. Additionally, intraprocedurally assessed MG might be altered by ”abnormal” hemodynamic conditions through general anaesthesia or inotropes. Despite difficulties in MG assessment by using continuous-wave Doppler in the setting of catheter-based MV repair, it was proven to be superior over planimetric evaluation of MVA for stenosis assessment, in a study with 38 patients by Biaggi and coworkers. They found peri-interventional planimetry to be unacceptably time-consuming and also to underestimate the MVA due to technical and acquisition-related issues (18). Accordingly, we used the Doppler-based assessment of MG in the present study.
A linear relationship between intraprocedural MG >5 mmHg and decreased functional capacity, as assessed by using the NYHA functional classification, was demonstrated by Thaden et al. in a cohort with 90% DMR (n=112) and preserved LV function (mean LV-EF: 50%). Correlatively, MG at baseline, the number of implanted clips, and central clip implantation were shown to be independent predictors for high intraprocedural MG in our study (21).
Neuss et al. found an invasively measured MG >5 mmHg and an echocardiographically estimated MG >4.4 mmHg to be independent predictors for all-cause mortality in 268 patients (mean age: 75 years) with moderately reduced LV-function (mean LV-EF: 39%) and moderate-to-severe or severe MR, who underwent the MitraClip procedure (19). Despite some differences in clinical and echocardiographic characteristics, such as older patients with better LV-function and a lower FMR rate, their findings support the validity of our results.
Utsunomiya et al. found a moderate correlation between MG and MVA after the MitraClip procedure in 97 patients with pre-existing pulmonary hypertension. Confirmatively, they discovered that intraprocedural MG predicts adverse outcomes in a cohort of comparable patients (20).
In a monocentric study including 51 elderly patients (mean age: 75 years), Boerlage-van Dijk et al. demonstrated that intraprocedural assessment of MG systematically underestimates the value compared to real life. Of note, the authors found no correlation between higher intraprocedural MG and increased heart failure symptoms at FU (9). In contrast, we found higher NYHA functional classes and lower six-minute walk distances at FU, as well as higher one-year mortality in patients with intraprocedural MG ≥4.5 mmHg, which might be due to the fact that the majority of patients included in the cited study had FMR (74%), and mostly suffer from chronic heart failure. This might hamper discerning persisting advanced heart failure symptoms and symptoms due to elevated MG at FU - high competing risk. Our cohort comprised a balanced number of MR etiologies (DMR: 40%, FMR: 42.8%, mixed: 17.2%), which might be a reason for the divergent finding.
An intraprocedural MG >4.4 mmHg was shown by Patzelt et al. to be predictive for a combined endpoint consisting of all-cause mortality, redo procedure, and LVAD implantation after MitraClip only in patients with DMR, but not with FMR. Moreover, higher intraprocedural MG was correlated with lower functional capacity at FU in the same study. The authors found the patient’s age to be the strongest independent predictor for the combined endpoint followed by residual MR >II and intraprocedural MG (7). We also found that baseline MG and central clip implantation are relevant predictors for unfavourable intraprocedural MG, although we included more patients with FMR (42.5%) and in more advanced stages of heart failure. We additionally found a negative influence of elevated intraprocedural MG on clinical outcomes regardless of MR aetiology.
Itabashi et al. showed that increased dimensions of the MV and the LV might be accountable for a somewhat lower intraprocedural MG in a comparable cohort after one-clip implantation (22). Contrary to our findings, patients with FMR showed a tendency for developing higher intraprocedural MG in this study, which might be due to smaller annular dimensions in their study cohort (AP diameter: 32mm vs 38mm).
Of note, MR geometry has a direct influence on the development of elevated MG. We found LAI to be the strongest predictor for unfavourable MG after clip deployment. This new parameter, which reflects on the length of the leaflets in relation to annular dimensions and offers an adequate geometrical assessment of the MV, was not assessed in all of the studies cited but seemed to be associated with residual MR, MG, and outcomes after the MitraClip procedure (14).
Apart from residual MR as a well-known prognostic parameter, postinterventional MG appears to be an independent predictor for clinical outcomes despite in-hospital decrease to values below 4.5 mmHg. It shows a dynamic postinterventional process and is influenced by various haemodynamic parameters such as blood pressure, heart frequency, volume condition of the patient, sedation or anaesthesia, haemoglobin, inotropes. Therefore, its sporadic assessment may lead to under-or overestimations, which may lead to misinterpretation intrainterventionally. Understanding of MG dynamics and its predictors is desirable to get more favourable outcomes after a successful interventional MR reduction compared to just residual MR based decision-making. The definite pathomechanism of this clinical entity stays still unexplained as an encouraging reason for further prospective multicentric studies. Considering that higher postprocedural MG is associated with worse outcome, forthcoming procedural and device/system-related improvements are desirable.