Mitral regurgitation (MR) is a prevalent valvular heart disease in western countries (1) and, when left untreated, leads to dilation and dysfunction of the left ventricle, as well as to heart failure decompensation and a poorer long-term prognosis (2). Open heart surgery has been the treatment of choice in MR patients for decades. However, many patients are deemed unsuitable for cardiac surgery based on comorbidities and surgical risk (3). Hence, less invasive percutaneous techniques have appeared in order to overcome this unmet clinical need. Catheter-based therapies mimicking surgical mitral valve repair face some limitations like their applicability on difficult anatomies (e.g., calcified mitral apparatus, short and restrictive posterior mitral leaflet) and the recurrence of significant MR overtime (4). Transcatheter mitral valve replacement (TMVR) constitutes a less invasive approach compared to open heart surgery, and it may help overcome some of the limitations associated with transcatheter edge-to-edge repair (TEER). Several dedicated TMVR devices for the mitral position are under clinical evaluation (5), and a growing body of evidence for this therapy is currently available.

In this systematic review and meta-analysis, we aimed to delineate pooled clinical outcomes observed for this technique using current TMVR systems, and to provide data on MR reduction and patientś clinical improvement after the intervention.

A comprehensive systematic review of published studies reporting on patients with significant MR who received TMVR was performed following the specified on the Preferred Reported Items for Systematic Reviews and Meta-Analysis (PRISMA) statement (6).

A computer-based search was performed on PubMed and EMBASE databases in order to identify relevant studies, and manual search of references from the selected studies (backward snowballing) was also used. Reviews, meta-analyses, and editorials were also reviewed in order to identify other potential entries.

The following keywords were used: “transcatheter mitral valve replacement” and “transcatheter mitral valve implantation.” The last access to databases was on October 31st 2022.

Eligible studies were those of original design, which were published in English, and that reported on clinical outcomes after TMVR with the use of dedicated devices for the mitral position. When the same patient population was used for several reports, the study with the largest sample size was finally selected. Studies including patients who underwent simultaneous TMVR and other valve interventions were excluded, as well as single case reports and congress communications. Studies informing on the use of transcatheter valves intended for the aortic position in patients with severe mitral annular calcification, a prior mitral annuloplasty, or a failed surgical bioprosthetic valve were also excluded.

A standardized data abstraction sheet was used for data extraction. Clinical characteristics, in-hospital, 30-days and mid-term outcomes were collected as reported by authors. Two investigators (A.A and I.P.) performed the literature search, study selection, and data acquisition. Any discrepancies were solved by a third investigator (P.A.).

The main findings of this systematic review and meta-analysis can be summarized as follows: (i) among 347 patients comprising the experience with current TMVR devices, there was a high technical success rate (95.4%), and LVOT obstruction and stroke rates were relatively low (2.6% for both); (ii) 30-day mortality was 8.4%, and the rate for early significant bleeding is still of concern (15.6%); (iii) there was a statistically significant reduction in ≥ grade 3+ MR and in the number of patients exhibiting poor functional class (NYHA 3 or 4) after the intervention; and (iii) both quality of live and exercise capacity improved significantly after mitral valve replacement.

There has been a gradual refinement in valve technology since the beginning of the TMVR experience. It consists of ameliorated device deliverability, a reduction in the size of vascular access sheaths, and the adoption of straightforward and more reproducible implantation techniques. This has led to improved clinical outcomes, with a steady reduction in the rates of unsuccessful device implants, early bleeding, stroke, and mortality. As a matter of fact, technical success rates were below 80% for some systems that are no longer under clinical use/evaluation (21), whereas technical success rates were above 95% in our updated systematic review.

Stroke has probably been the most dreadful complication in the transcatheter structural field. Our pooled data show that currently used TMVR devices associate relatively low rates of cerebrovascular complications (2.4%), and these rates range between 0% and 3% for the most recently published studies in the field (12, 15). These promising results highlight the reduction in stroke events when compared to the very early clinical experience in TMVR (22). However, procedural safety still lies way behind in comparison with other percutaneous mitral valve interventions such as TEER, where stroke rates have been reported to be systematically below 1% (23).

Major (and life-threatening) bleeding remains one of the primary shortcomings of TMVR, with pooled rates above 15% in our analysis. This limitation could be partially explained by the learning curve associated with the use of some devices. Notably, many studies had sample sizes below 20 patients; hence, they illustrate the early stages of clinical use for several systems. Besides, the predominant utilization of the transapical route may have played a major role in these findings, as pericardial and thoracotomy-related bleedings are often of concern. Further device and delivery system iterations are needed to increase procedural safety. Besides, wider adoption of TMVR and growing operator experience will help overcome this issue.

In our study, most of the implantations were performed transapically (315/347, 90.78%), which may be mainly explained by the obtention of the CE mark by the Tendyne device system (Abbott, Menlo Park, California). It should be noted that the growing use of the transfemoral approach for some new devices may help overcome the detrimental effects associated with the transapical access, such as pericardial bleeding and myocardial injury. Additionally, the shift from transapical to transfemoral TMVR procedures could be associated with a more straightforward patient recovery translating into a shorter hospital stay. The appearance of new devices purposely designed for transfemoral-transeptal access (e.g., Cephea, Evoque, M3), and the iterations allowing for transfemoral implantations in systems which were firstly developed exclusively for transapical procedures (e.g., Intrepid) are of upmost importance to improve TMVR safety and to facilitate its expansion.

Valve performance and sustained MR abolition are probably the most significant advantages of TMVR over other transcatheter techniques used in MR patients. Accordingly, we have shown a significant MR reduction after the procedure when dichotomizing moderate-severe vs. less than moderate MR, and the number of patients with significant MR despite TMVR is minimal. In addition, these results seem to be maintained over time, as no significant MR was observed at mid-term follow-up echocardiography. The 2-year results of the largest TMVR single-arm trial have been reported, and MR at 2-year follow-up ranged between none-trace (93.2%) and mild (6.8%) (24), with no cases of significant mitral stenosis. Future studies and larger follow-ups are necessary to comprehensively assess valve durability and its performance over time, although these early results are encouraging. It should be noted that these results apply exclusively to patients with severe MR. Some studies have included patients with a certain degree of mitral stenosis, especially those reporting on TMVR in mitral annular calcification (9, 15, 18). However, there are no specific reports for patients with significant mitral stenosis without regurgitation, and the scarce number of patients with concomitant stenosis does not allow for a separate subanalysis. TMVR performance in this setting (isolated mitral stenosis) remains widely unknown.

MR does not exclusively impact life expectancy but also the quality of life and functional capacity. Our study demonstrates a rapid improvement in functional class, quality of life, and exercise capacity early after TMVR, probably based on an effective MR reduction with no need for a sternotomy or major surgery. These findings should be taken into perspective, as patients ultimately treated with TMVR exhibited several comorbidities and advanced age. Therefore, both qualitative and quantitative ameliorations in quality of life are of great clinical relevance.

Despite the findings in our systematic review, some issues remain to be addressed. For instance, screening failure based on anatomical features such as the risk for LVOT obstruction and mitral annular sizing was not systematically reported among the selected studies, but it may have been relatively high. This issue could hamper TMVR applicability in routine clinical practice. Additionally, whether to perform mitral valve repair or replacement when TEER is doable from a technical standpoint is controversial. TMVR outperforms TEER in terms of MR reduction, although it implies a higher stroke and bleeding risk. Besides, long-term complications associated with prosthetic valves, such as infective endocarditis or valve thrombosis (scarcely evaluated in current TMVR studies, ought to the short clinical follow-ups) should be considered. On the other hand, TMVR carries an almost negligible risk for high residual mitral gradients. As previously reported (25), TMVR may be a good alternative when TEER implies a prohibitive increase in diastolic mitral gradient. Ongoing trials comparing TMVR vs. TEER will shed more light on this matter (NCT03433274). For the time being, and given the amount of clinical evidence available for TEER, it seems reasonable to undertake an upfront percutaneous repairment strategy when feasible.

Some limitations merit acknowledgement in our study. First, the nature of the included studies was observational for the vast majority, and no control group was used for the trials included. Therefore, TMVR performance vs. medical therapy or other invasive techniques should be further evaluated. Publication bias is a potential limitation as for any meta-analysis. However, Eggeŕs regression analysis and funnel-plots did not suggest its presence (Supplementary Material Table S1 and Supplementary Material Figures 1–3). Furthermore, the lack of substantial clinical experience with some devices might have significantly impacted pooled outcomes, as well as the predominant use of the Tendyne device among the patients included. Future systematic analysis as TMVR experience grows over the following years will be paramount to monitor the field's evolution.

In conclusion, TMVR with currently available systems yields a significant MR reduction and an early improvement in functional class and quality of life, but with the trade-off of relatively high major bleeding rates.