Although several reports have described excellent early and long-term results of minimally invasive mitral valve repair [5], few studies have analyzed MIMVR cases in detail. The present analysis of a consecutive series of patients who underwent an MIMVR procedure found it to be safe and associated with excellent postoperative outcomes, while good procedural times, shorter hospital and ICU stays, and acceptable mid-term results were also revealed. As demonstrated by complete echocardiographic follow-up results, no paravalvular leakage developed in any of the patients, including during the late phase. Specifically, overall in-hospital mortality and postoperative neurological events were nearly 0%, with the latter lower than the range of rates for mortality and neurological events recently presented by Modi and colleagues [6] in a meta-analysis of MIMVR cases (4.9% in 5 cohorts with a total of 979 patients, 2.3% in 3 cohorts with a total of 778 patients).
In the present study, since there were no cerebral vascular accident incidents, preoperative contrast CT results were analyzed to detect aorta condition, i.e., the presence of arteriosclerotic disease or parietal thrombus, and determine the perfusion program for each case. Details of the perfusion protocol were previously reported by Nakamura et al. [7]. When significant arteriosclerotic disease was found in the entire aorta and iliac artery, right axillary artery cannulation was employed to establish antegrade perfusion for CPB.
Although early good results and quick recovery following MIMVR have been noted in prior reports [8], there are concerns regarding its long-term efficacy, which has not been adequately elucidated. We found 2 reports that included long-term follow-up findings after MIMVR procedures performed at high-volume centers. The longest functional and echocardiographic follow-up study was conducted by Glauber and colleagues, who analyzed 476 patients who underwent MIMVR via an RT approach between 2003 and 2013 (mean age 67±11 years, incision length 5-7 cm, in-hospital mortality 3.2%, neurological complications 2.6%, reoperation for bleeding 6.4%) [9]. In those cases, survival after replacement at 1, 5, and 10 years was found to be 91.0±1.4%, 81.3 ± 2.5%, and 76.2 ±3.4% respectively, while the rates for freedom from reoperation after replacement at 1, 5, and 10 years were 98.6±0.6%, 94.5±1.6%, and 83.9±5.5%, respectively. In 2003, Casselman et al. reported early and long-term results in a review of a series of 80 patients who underwent endoscopic MVR with an RT procedure (incision length 4 cm, mean follow-up period 19.6±17.3 months, completed in all) [10]. Although 2 (2.5%) of their patients experienced new onset endocarditis during the follow-up period and 1 underwent a reoperation with a median sternotomy, they noted excellent freedom from late phase death (4 years) in 92.0±3.6% as well as freedom from anticoagulation-related complications (4 years) in 97.0±1.9%. In the present cohort who underwent MIMVR (mean age 70.7±11.3 years, mean incision length 5-6 cm), 5-year survival, freedom from a mitral valve-related reoperation, and freedom from cerebrovascular events and anticoagulation-related complications showed favorable rates of 76.5%, 100%, and 81.7%, respectively. In addition, those rates were also good after first-time MIMVR at 83%, 100%, and 93%, respectively. As compared to those two other recent reports, our mid-term survival results were relatively better, while the freedom from reoperation was 0% at the 5-year follow-up examination, and there were no device-related complications or cases of paravalvular leakage.
Reoperative mitral valve surgery through a median sternotomy can be particularly challenging due to dense adhesions and is known to carry a substantial risk of injury to vascular structures, which have been found to occur in 7–9% of cases and are associated with increased mortality [11]. A valid alternative that could be employed to avoid risks associated with a redo median sternotomy approach is a right anterolateral mini-thoracotomy approach [12]. Unfortunately, only 5 patients who underwent a conventional sternotomy were available as a control group for the present study. Our strategy consists of selecting MIMVR whenever possible even for redo-mitral valve procedure cases, as that procedure has been our the first-line choice since 2014. However, that introduces a critical selection bias including cases of redo-mitral valve surgery between MIMVR and conventional sternotomy groups for analysis of procedures performed at our institution. Thus, we sought to compare the present results with those previously reported for cases of redo-MIMVR or conventional full sternotomy redo-mitral valve surgery. Studies regarding the feasibility of redo-MIMVR via an RT procedure have been presented by Ricci et al. [13], Sharony et al. [14], and Thomson et al. [15], with excellent results reported in each (Table 5). The results obtained in the present cases are favorable as compared with those, though remaining issues related to our method include longer prolonged CPB and cross-clamp times as compared to those prior findings. On the other hand, there was only one in-hospital mortality among the present cohort. Furthermore, the rates for freedom from stroke events and reoperation for bleeding were nearly 0%. In addition, 3 other retrospective observational studies all demonstrated the superiority of an RT approach as compared to a median sternotomy for reoperative mitral valve surgery, with excellent results noted in those cases (Table 5) [16-18]. Pooled analysis showed a mini-thoracotomy mitral valve surgery procedure as a safe alternative to a standard sternotomy, with reduced mortality rates, length of hospital stay, and reoperations for bleeding, along with a comparable risk of stroke.
There is a paucity of literature showing outcomes of MIMVS in older patients. Iribarne et al. demonstrated that it can be performed safely in patients at least 75 years old [19]. Furthermore, they noted that even though a minimally invasive approach was associated with slightly longer CPB and cross-clamp times than a conventional sternotomy, there were no significant differences regarding postoperative morbidity or mortality. Importantly, the mean and median durations of hospitalization were 3 and 1 day shorter, respectively, for patients who underwent an MIMVS procedure, findings with important implications for resource use. Our strategy consists of selecting MIMVR whenever possible even for elderly cases, thus we were not able to compare with a conventional sternotomy approach. In the present cohort, elderly patients (≤80 years old) who underwent MIMVR showed favorable rates for 5-year survival, freedom from a mitral valve-related reoperation, and freedom from cerebrovascular events and anticoagulation-related complications of 64%, 100%, and 100%, respectively. On the other hand, the cause of cardiac-related death could only be determined in 1 of 3 cases (septic shock after 2 months, fatal arrythmia after 3 months, unknown after 13 months). Nevertheless, the present results demonstrated the superiority of MIMVR in elderly patients the same as with the young group.
Factors with potential effects on the outcome of patients who undergo MIMVR are numerous and many confounding variables exist, though only dependent variables related to in-hospital mortality and prolonged hospital stay (>30 days) were examined in the present study. Univariate analysis indicated infectious endocarditis as a predictor, which might be related to the influence of post-operative antibiotic administration. Interestingly, previous cardiac surgery, concomitant operation, and elderly status were not significant. A multivariate logistic regression model of 409,904 valve procedures performed between 1994 and 2003 and cataloged in the Society of Thoracic Surgeons database demonstrated that the third most important preoperative variable influencing operative mortality is a reoperation (OR 1.61, P <0.001) [20]. However, the present data suggest that an RT approach after a previous median sternotomy is not an independent predictor of mortality (P=0.26). This important finding indicates that this technique, in least in consideration of early outcome, should be the first choice for reoperative mitral valve surgery in patients who do not need concomitant aortic valve replacement or coronary revascularization. That is in concordance with previously reported data demonstrating equivalent or lower mortality rates, and less morbidity associated with an RT approach as compared to a reoperative sternotomy [16, 21, 22]. Nevertheless, findings regarding mid- and long-term outcomes are needed.
For treating mitral valve disease or degenerated bioprosthetic valve failure, transcatheter mitral valve replacement (TMVR) therapy has emerged as an encouraging option. Early reports have noted utilization of TAVR devices for degenerated surgical bioprosthetic valves (transcatheter mitral valve-in-valve implantation with balloon-expandable valves) (23) or annuloplasty rings (transcatheter mitral valve-in-ring implantation with balloon-expandable valves) (24), with reasonable results shown. Additionally, utilization of a TAVR valve in the mitral position has been described for treating patients with significant mitral annular calcification (MAC), though early results have demonstrated high rates of mortality and significant complications (25). Overall, survival (30-day all-cause mortality) after these various procedures has been greatest in patients who underwent a mitral valve-in-valve procedure (6.2%), followed by valve-in-ring (9.9%), with the worst results reported for valve-in-MAC (34.5%) (26), and those findings raise questions regarding whether TMVR is an entirely appropriate alternative for surgical mitral valve surgery. Results presented thus far show that MIMVR is more effective. Additional long-term studies with larger numbers of enrolled patients will be necessary to better assess the efficacy of TMVR and MIMVR.
This study has some limitations. It was conducted in a retrospective manner and lacked a control group for appropriate comparisons. Thus, it was not possible to compare the present case series with a control group, since MIMVR has been our standard approach since 2014 and patients typically demand less invasive procedures. Even though this was a follow-up study conducted over 5 years, the number of patients was small and only 12 were found to be at risk at the 60-month follow-up examination. A well-designed study with an appropriate sample size will be required in the future to validate the advantages of MIMVR.
To conclude, we found that MIMVR can be safely performed with encouraging short and mid-term outcomes, including elderly patients, with very low rates of conversion and mortality as compared to a conventional sternotomy procedure. In addition, a mitral valve re-operation can be safely and effectively performed through a smaller right thoracotomy. Finally, MIMVR was found useful for elderly patients. Based on these findings, we consider MIMVR to be an attractive alternative to conventional MVR.