Changes in mitral annular morphology following transcatheter mitral valve repair. Clinical repercussion and importance of etiology

Introduction

Mitral regurgitation (MR) is the most prevalent valvular disease in the United States and the second most prevalent in Europe1,2. The transcatheter mitral valve repair (TMVR) treated with the MitraClip system (Abbott Vascular, Menlo Park, California, United States) imitates the edge-to-edge approach surgical technique proposed by Alferi to achieve an effective reduction of the degree of MR3,4. This technique is more widely used, particularly in patients of high or prohibitive surgical risk because it is less invasive and has shown good efficacy and safety results in the mid-term5-7.

This procedure is thought to be able to operate changes in the anatomy of the mitral annulus beyond the edge-to-edge approach of the valvular leaflets, but there is very little information on this regard. Some studies speak about a significant change of anteroposterior diameters in RM of functional etiology8, while others describe changes of diameter, in non-constant areas and in etiologydependent areas9.

The goal of this study is to analyze the morphological changes occurring in the mitral valve after the TMVR and its relation to the degree of reduction of MR in the short and mid-terms, and its association with the clinical goals.

Methods

This is an observational, prospective study conducted at Hospital Universitario Central de Asturias de Oviedo, Spain.

Inclusion de patients

Patients were included between October 2015 and October 2018. These were the inclusion criteria: patients with grade III-IV symptomatic mitral failure despite the optimal medical therapy considered of high surgical risk by the multidisciplinary team and who would adequately meet the anatomical criteria needed for the implant4,7. These patients were excluded: patients with prior mitral surgical annuloplasty due to the impossibility of measuring annular anatomic changes. A prior transesophageal ecochardiography was conducted in all patients. The etiology of MR was categorized into organic or degenerative, and functional. Patients with a mixed etiological profile in their MR were recategorized into one of the aforementioned groups based on their predominant component after 2 expert cardiologists studied the transesophageal echocardiography and achieved consensus. All patients received oral written information on the risks and benefits of the procedure, and they all signed a written informed consent according to the Declaration of Helsinki.

Description of the procedure

The TMVR was performed using the MitraClip system, which received the European certificate of conformity (CE mark) in March 2018. The implantation procedure has already been described in prior studies7. In sum, the intervention is conducted under general anesthesia and guided by a 3D transesophageal echocardiography and under the supervision of a MitraClip technical expert. More than one clip was implanted in cases where the reduction of the degree of MR was not of, at least, one grade, and as far as there was no significant residual mitral stenosis estimated through the average diastolic transmitral valve pressure gradient10.

Echocardiographic study

All patients underwent transesophageal echocardiographic studies in 2 and 3 dimensions before and right after the completion of the procedure that was conducted by an expert echocardiography expert using a state-of-the-art echocardiography machine model EPIQ 7 (Philips; Amsterdam, The Netherlands). The patient’s afterload hemodynamic condition was taken into consideration before and after the procedure.

In order to conduct the 3D study of the mitral annulus, 3D images were acquired (Zoom 3D, Philips; Amsterdam, The Netherlands) during the procedure that were later analyzed using the MVQ QLAB 10.0 mitral quantification software (Philips; Amsterdam, The Netherlands). Figure 1 shows an example of 3D reconstruction before and after the procedure.

The analysis of the leaflet grasping was estimated using the lengths of both leaflets before and after the procedure in the same plane of the implant of the device. The length before the clip was measured between the anchor site of the leaflet to the annulus and the leaflet free-edge and, the length after the clip was estimated between the anchor site of the leaflet to the annulus and the leaflet site immediately proximal to the part of the leaflet inside the device:

– Total grasping (mm): pre mitral leaflet length − post mitral leaflet length.

– Per cent grasping (%): ([pre mitral leaflet length − post mitral leaflet length]/pre mitral leaflet length) × 100.

Figure 1. 3D analysis before the clip (A) and after the clip (B) of mitral annulus in frontal view from the left ventricle. A, anterior; AL, anterolateral; Ao, aorta; P, posterior; PM, posteromedial.

Study variables

Echocardiographic variables

The technical success, the device success, and the procedure success were all defined according to the consensus document put together by the Mitral Valve Academic Research Consortium11. Both the etiology and severity of the MR were classified and assessed according to the clinical practice guidelines designed by the European Society of Cardiology12-14, being severity subdivided into four degrees in a similar way to what the EVEREST clinical trial did4,7.

Clinical variables

The patient’ functional capacity was assessed following the New York Heart Association classification. Admission to due heart failure was defined as patients coming back to their hospital floor or being assisted in the ER and having to stay and sleep over. The EuroSCORE II and the Surgeon Thoracic Score were estimated too. Follow-up event was defined as a hospitalization due to heart failure or all-cause mortality.

Study goals

The study goals were the assessment of the annular morphological changes, the recurrence of MR (at least grade III/IV) and a composite endpoint of rehospitalization due to heart failure and global mortality.

Statistical analysis

Qualitative variables were expressed as absolute number and percentage and quantitative variables as mean ± standard deviation. The Student t test for paired data was used to assess morphological changes before and after the procedure. The chi-square and Student t tests were used for different groups as methods to compare categorical and quantitative variables. Linear regression analyses were conducted to assess the predictors of annular quantitative modification, the binary logistics regression analysis was used for the study of MR recurrence, together with the survival analysis using the Kaplan-Meier method. A peak alpha error of 0.05 was assumed. All analyses were conducted using the Stata 14 software (Stata Statistical Software: Release 14. College Station, Texas: Stata- Corp LP).

Results

Fifty TMVR procedures were conducted between October 2015 and October 2018 in 48 patients: 48 MitraClip primary implants and two reinterventions due to the partial detachment of the posterior leaflet. The average age was 74.8 ± 7.2 years and 31.3% of the patients were females. Ten procedures (20.8%) were conducted in patients with organic MR and 38 (79.2%) in patients with functional MR. The baseline characteristics of the population based on the etiology of the MR and the echocardiographic data are shown on table 1 and table 2. An average 1.5 ± 0.5 clips per procedure were implanted. In 43 (86%) cases, the first-generation clip was used, while in 7 (14%) cases, the XTr clip was used. Technical success was 100% and the procedural success was close to 92% (46/50). The four unsuccessful cases were due to partial detachments, one failed reintervention, and persistent grade III/IV MR after the implant.

Table 1. Baseline characteristics of the population

Table 2. Echocardiographic characteristics

After the procedure, in the 3D analysis of the mitral annulus, there was a significant reduction of both annular diameters, the perimeter and both 2D and 3D areas(table 3). The comparative analysis based on etiology (table 4) found a greater reduction in the anteroposterior diameter in patients with functional MR compared to those with organic MR (13.2 ± 8.8 versus 8.6 ± 7.5 of per cent reduction, respectively; P = .05) and a greater tendency to a reduced area in the same sense (13.3 ± 12.4 versus a 7.2 ± 11.1 of per cent reduction, respectively; P = .01).

Table 3. Global annular changes

Table 4. Annular changes and mitral leaflet grasping based on the etiology of mitral regurgitation

When it comes to both leaflet-grasping it was observed that in patients with organic MR, a greater percentage of anterior leaflet tissue inside the device is approached (36.6 ± 11.5% in the organic MR versus 27.8 ± 11.4% in the functional MR; P = .02), while posterior leaflet grasping is similar in both subtypes (34 ± 8.1% in the organic MR versus 34.4 ± 10.6% in the functional MR; P = .04).

In the simple linear regression analysis conducted of predictor factors of reduction of the annular anteroposterior diameter we observed that the percentage of posterior leaflet grasping and the anteroposterior diameter before the implant were the only factors associated with a greater reduction. After adjusting for the etiology of the MR, the indexed ventricular volumes, the annular diameters before the implant, and the anterior leaflet grasping, this association between the posterior leaflet grasping and the reduction of anteroposterior annular diameter was still statistically significant (ß coefficient = 0.27; 95%CI, 0.05-0.48; P = .02).

After an average 454 days of follow-up (interquartile range, 195- 699), 7 out of the 48 patients (14.6%) and 8 out of the 50 procedures conducted (16%) showed grade III/IV MR. In the binary logistics regression analysis conducted for grade III-IV/IV MR predictors at the echocardiographic follow-up (table 5) it was observed that the percentage of grasping over the posterior leaflet was the only parameter statistically associated with a lower probability to develop significant MR (OR, 0.89; IC95%, 0.79-0.98).

Table 5. Binary logistics regression analysis. Predictors of grade III-IV/IV mitral regurgitation after transcatheter mitral valve repair

There was a 16% rate of rehospitalizations due to heart failure and a global mortality rate of 12.5% (table 6). The composite endpoint of all-cause mortality or rehospitalization due to hear failure occurred in 10 (20.8%) patients. In the regression analysis for the composite endpoint of mortality or rehospitalization due to heart failure we did not observe an association between the parameters of annular reduction or the leaflet grasping and the endpoint under study. The heart failure-free or all-cause mortality-free survival curve is shown on figure 2.

Table 6. Mortality causes

Figure 2. Mortality-free or readmission due to heart failure-free survival estimated using the Kaplan-Meier method.

Discussion

The main finding of our study is that after TMVR with MitraClip there are important anatomical changes when it comes to the reduction of anteroposterior and intercomissural diameters, the annular diameters and areas, measured both in 2D and 3D. It was observed that, except for the intercomissural diameter, the remaining annular measurements (anteroposterior diameter, perimeter, and area) were significantly enlarged in patients with functional MR compared to patients with organic MR.

Similar to other studies published8,9, sit has been observed a significant reduction of the anteroposterior diameter after the implant. However, unlike Remy et al.9, describe in patients with functional MR, there was a greater relative reduction of the anteroposterior diameter and a non-significant tendency to a greater reduction of these patients’ area Also, in our series, we saw a reduction of the intercomissural diameter, which may have to do with a significant and sudden reduction of the regurgitation volume and with left intra-articular pressure, rather than with a direct mechanical effect coming from the clip.

With respect to the repercussion of these anatomical changes in the significant clinical results during follow-up, we did not observe any statistically significant correlation between these changes and rehospitalizations due to heart failure or global mortality. There was, however, an inversely proportional correlation between the reduced anteroposterior diameter and the possibility of III/IV MR recurrence (OR, 0.95; 95%CI, 0.89- 1.05). This data has been published in former statistically significant studies15. It is believed that the lack of signification in our study when it comes to these goals, and the non-association between the magnitude of diameters before the implant and MR recurrence may be associated with the number of patients of the overall cohort and the low number of events during follow-up.

There was a greater per cent anterior leaflet grasping in patients with organic MR compared to those with functional MR. This data may be explained by the greater anteroposterior annular diameters of patients with MR of functional etiology and by their association with the tenting phenomenon or apical displacement from the coaptation site, thus making an angle of greater magnitude between the annulus and the leaflet and, therefore, more difficulties to encompass the anterior leaflet during the procedure. On the other hand, it was observed that the posterior leaflet grasping was similar in both groups, which is a particularly important aspect because larger grasping percentages are associated with a greater relative reduction of the anteroposterior diameter with coefficients close to 0.3, which implies that by achieving just a 10% more grasping of the posterior leaflet we would be achieving a 3% reduction in the anteroposterior annular diameter. The posterior grasping was also a protective element against the possibility of significant MR recurrence at follow-up. In this sense, it is believed that patients whose mitral annulus will not allow minimum leaflet coaptation at baseline or will cause excessive tension in the leaflets while grasping with the corresponding risk of tear and break are those patients that may benefit the most from an associated annuloplasty system.

The role that the new generation MitraClip XTr may play in the mitral annular changes of our cohort has not been studied due to the low number of implants of this last device. It would be interesting to publish in the future whether this new device causes changes of different magnitude compared to the previous device, and whether these changes have to do with significant clinical changes at follow-up.

Limitations

This is a single-center study with a modest number of patients (48) and procedures (50). The analysis of predictors of mortality and rehospitalizations due to heart failure may be affected by the small size of the sample and small number of events reported. Also, this is a relatively new technique at our center, meaning that the representation of patients who were followed in the long-term is scarce. Also, no long-term 3D analysis of the mitral annulus after the implant was conducted.

Conclusions

After TMVR with MitraClip there are morphological changes in the mitral annulus. The magnitude of these changes is different based on the MR etiology. The posterior leaflet grasping is the main factor that influences the appearance of changes and is also associated with a lower probability of significant MR recurrence at follow-up.

Funding

This study received no funding whatsoever.

References