Role of surgical repair in FMR
The current role of surgical repair in FMR appears limited due to the conservative recommendations given by the 2017 U.S. and European guidelines and the position papers of professional societies,7,8,46 which are predominantly based on the results of 2 CTSN randomized studies.7,8,12,14Observational, non-randomized, and single-center experiences often lack robustness in study design with non-rigorous definitions of the degree of MR especially in patients with moderate and severe MR.48 Although RCTs are less susceptible to confounders and non-homogeneity of the samples, insufficient power and methodological or operational flaws might fail to generate evidence that can reliably be used to guide patient care.41
In the CTSN moderate IMR trial, the negative results after 2-year follow-up might be reasonably explained by the novel conceptual framework that all the enrolled patients were presented with proportionate MR (EROA of 0.2±0.1 cm2 and LV end-systolic volume index (LVESVI) of 54.8±24.9ml/m2).13,14 The contribution of CABG targeting ischemic ventricular muscles, not the restrictive annuloplasty targeting at mitral annulus, for alleviating MR is fundamental, and the results show that adding MR correction procedures to CABG showed no additional benefit in death and major adverse events.14 However, several flaws in this trial should be noted. Firstly, given the fact that 35.8% of patients in the CABG only group and 31.3% in the combined procedure group had no AMI, mixing patients with myocardial ischemia and infarction made the results inconsistent because pure IMR is reversible with revascularization and MR secondary to infarction-induced LV remodeling responds poorly to CABG;49 Secondly, the gradation of severity of IMR as moderate (EROA 0.20-0.39 cm2, vena contracta 0.3-0.69 cm and color Doppler jet area from 20% to 40% of the left atrial area) in this study now have been defined as severe in 2017 European guideline.7 As a result, these discrepancies cast shadow on the way the study was built and on its conclusions. Moreover, whether a survival benefit beyond 2 years in adding MV repair to CABG could be translated from better cardiac physical function, less IMR grade and higher reverse LV remodeling observed during 2-years follow-up is pending.14 Therefore, the impact of surgical MV repair in moderate IMR remains to be re-evaluated.
In the CTSN severe IMR trial, although a substantially recurrence of at least moderate MR developed in 58.8% of the patients in the repair group, the patients who did not have recurrent MR had greater LV reverse remodeling than patients who underwent replacement (LVESVI 42.7±26.4ml vs. 60.6±39.0ml) at 2 years.12 This outcome validates the well-proven principle that a “good repair” (with durable correction of MR) outperforms the best replacement in the setting of FMR.50 However, the high prevalence of recurrent MR in this cohort of patients contrasts with apparent durability of the same restrictive annuloplasty in patients with moderate MR (11.2%) studied by the same group of investigators.14 It is logically to attribute these differences to the larger LV in patients with severe MR. However, baseline LVESVI were only modestly greater in patients with severe MR (61.1±26.2 ml/m2)12 than in those with moderate MR (59.6±25.7 ml/m2).14 More importantly, among patients with severe MR, preoperative LV cavity size was not different between those who did or did not develop recurrent MR. Therefore, several authors have suggested the recurrence of MR was most likely related to a mismatch between LV end-systolic dimension (LVESD) and the mitral ring size. If the PMs remain laterally displaced relative to the mitral ring, tethering of the posterior leaflet can be exacerbated following restrictive annuloplasty.51,52 A post-hoc analysis by the CTSN authors indicated that a LVESD/ring size ratio >2 was associated with increased risk of persistent or recurrent IMR.51 Recent echocardiography-based studies have identified several valvular (e.g., coaptation depth >1.5 cm, posterior mitral leaflet angle >45°, distal anterior mitral leaflet angle >25°, systolic tenting area >2.5cm2) and ventricular (e.g., LVEDV >65mm, coaptation depth >1.5 cm, end-systolic interpapillary muscle distance >20 mm, and systolic specificity index >0.7) determinants failure of undersized annuloplasty in those with FMR.48,53 Given that these factors were not considered as inclusion or exclusion criteria for randomization, a crucial bias exists.50If these cases and those with a mismatch between the LV and mitral ring size were detected at the time of surgery, consideration should be given to an intervention directed to improving leaflet coaptation at valvular (e.g., leaflet augmentation plus true-size annuloplasty,[17,54] Figures 1 and 2) and subvalvular levels (e.g., papillary muscle approximation or relocation)[55,56]. Since the life-expectancy of patients undergoing replacement is endangered by the incremental risk of thromboembolic/hemorrhagic and prostheses-related events, well-designed RCTs without significant operational flaws are required to reconfirm the role of surgical MV repair in treating severe FMR.
In the era of fast-evolving transcatheter therapy, apart from the MitraClip, many other transcatheter devices designed to imitate surgical repairing techniques, including direct or indirect annuloplasty, have already been shown to be promising in reducing MR in experimental or ongoing clinical trials. Further studies will determine whether selected FMR patients (with or without different risk profile and predictors of repair failure) might benefit from tailored surgical or transcatheter repair.