Introduction

Ischemic mitral regurgitation (iMR) results due to remodeling and/or distortion of the left ventricle following ischemic heart disease that can impair leaflet coaptation, papillary muscle function, and overall mitral valve geometry and competence [1, 2]. iMR has been shown to impact and predict long-term mortality, and surgical revascularization alone is often insufficient [2–5]. Current consensus guidelines suggest revascularization and surgical mitral valve intervention at the same time has a benefit, however, the type of mitral valve intervention (repair or replacement) is still a topic of discussion [6–8]. Surgical mitral valve intervention in the setting of ischemic cardiomyopathy is intended to prevent left ventricular (LV) volume overload and worsening LV remodeling and dysfunction [9–11]. Despite this theoretical benefit, surgical correction of moderate to severe mitral regurgitation in the setting of ischemia has not been shown to improve survival in multiple studies [1, 12–15]. When considering surgical intervention for iMR, the CTSNet moderate and severe iMR trials are often referenced, which concluded that patients who underwent mitral valve repair (MVr) had higher rates of postoperative moderate to severe MR when compared to mitral valve replacement (MVR) [5, 16]. This contradicted previous reports of patients having improved survival and fewer complications following MVr [1, 17–20]. In the present study, we looked at our single-center outcomes for patients who underwent mitral valve repair or replacement for ischemic mitral regurgitation.

Methods

This study was reviewed by the Institutional Review Board (IRB) and was exempt from approval due to its retrospective nature (45 CFR 46.101(b)). Informed consent was waived for this study (45 CFR 46.116 (D)). Clinical trial registration was also not applicable for this study.

Results

Study population

A total of 111 patients underwent mitral valve surgery for ischemic mitral regurgitation by a single surgeon at our center between January 2010 and June 2018. The median age of the cohort was 66 years, and 76% of the cohort was male. Overall, 62 (56%) had a mitral valve replacement and 49 (44%) had a mitral valve repair. The baseline characteristics of both groups were similar (Table 1) with the exception that the mitral repair group had a lower preoperative median left ventricular ejection fraction (EF), (MVr 40% vs. MVR 55%, p < 0.01) (S1 Data). The MVr and MVR groups had similar rates of moderate and severe mitral insufficiency, both groups with predominantly severe MR (69% and 74%). The MVr group notably had a larger preoperative median left ventricular internal diameter at end diastole (LVIDd) (5.6cm vs. 4.6 cm, p < .01).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic and perioperative data on patient groups.

Data reported as median with interquartile range.

https://doi.org/10.1371/journal.pone.0307449.t001

Primary outcomes

Of the 111 patients reviewed, 93 achieved 1-year follow-up (42 (86%) of MVr patients and 51 (82%) of MVR patients). Of these patients, 72 patients (31 (74%) MVr and 41 (80%) MVR patients) had ECHO information available at this visit. At 1-year, more patients in the MVr group had greater than trace mitral regurgitation (71% vs. 39%), with 65% of MVr patients having mild MR. The number of patients with moderate or severe mitral regurgitation was similar between the MVr and MVR groups (6% vs. 12%, p = 0.69). At 2-years, the MVr and MVR groups had no difference in moderate or severe mitral regurgitation (7% vs. 13%, p = 0.68)(see Table 2). One patient in the MVR group required reoperation 5-years postoperatively for recurrent symptomatic severe MR, which was replaced with a mechanical mitral valve along with concomitant CABG to the RCA.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Perioperative and ECHO-based outcomes of patients who underwent either mitral valve repair or replacement.

https://doi.org/10.1371/journal.pone.0307449.t002

Discussion

In this study, we describe our experience and outcomes using both mitral valve repair and replacement strategies for ischemic mitral regurgitation. Overall, our patient populations in both groups were similar aside from an expected shorter duration of CPB and cross-clamp time in the repair group. The CTSNet severe iMR trial compared these two surgical techniques in patients with severe MR [21]. Patients who underwent MVr for ischemic MR had a higher rate of moderate or severe MR (32.6% vs. 2.3%, p < .001 at 1-year, 58.8% vs. 3.8%, p < .001 at 2-years) when compared to those who underwent MVR. Additionally, there were more heart failure related adverse events and readmissions related to cardiac disease.

The CTSNet moderate iMR trial compared outcomes from MVr with concomitant coronary artery bypass grafting (CABG) to CABG alone. These results suggested an overall similar survival with decreased rates of moderate to severe MR, but with increased early neurological events and supraventricular arrythmias in the CABG + MVr group [22]. Following this and other various studies, AATS guidelines suggest MVR with preservation of the posterior and anterior chords, or MVr with an undersized complete rigid annuloplasty ring are reasonable to consider for severe iMR, and MVr may be considered in the setting of other indicated cardiac surgery (e.g. CABG) in patients with moderate iMR.

Compared to the CTSNet data from moderate and severe ischemic MR trials, our overall incidence of moderate to severe postoperative MR at 1- and 2-years after MVr is relatively low. While the present study included both patients with preoperative moderate and severe mitral regurgitation, the CTSNet severe iMR trial only looked at patients with severe MR. Despite this difference, the rate of postoperative moderate to severe MR in our MVr patient population (6% at 1-year, 7% at 2-years) was low compared to the CTSNet data (32.6% at 1-year, 58.8% at 2-years). This suggests that an appropriately selected mitral valve repair has good hemodynamic durability when performed well by an experienced surgeon. When compared to patients who underwent MVR in the present study, our MVr patients had a lower preoperative EF and larger LVIDd, possibly suggesting more advanced LV remodeling and worsened baseline disease, however, further analysis would be needed to determine this significance. It is unclear whether this difference in preoperative LV dilation impacted our results. A more dilated LV may benefit from undersized annuloplasty band and reduction in annular dimension, leading to preserved mitral valve function rather than an oversized mitral valve replacement. On the contrary, a more dilated LV may suggest more advanced disease and dysfunction, leading to worsened outcomes with repair or replacement given the unclear promise of reverse LV remodeling. Reverse LV remodeling is a theoretical benefit to MVr or MVR in the setting of iMR. In available studies, however, despite improved rates of moderate to severe mitral regurgitation, the improvement in reverse remodeling varies [22, 23].

Five patients who underwent MVR developed clinically significant mitral regurgitation in follow-up. In review of these five patients, 2 of the patients developed severe mitral regurgitation, 1 secondary to infective endocarditis with paravalvular abscess at 3 years after surgery, and 1 who initially was lost to follow-up, however, presented 6 years later and was found to have severe MR in the setting of new onset atrial fibrillation. The 3 additional patients developed moderate mitral regurgitation. 1 patient developed infective endocarditis, 1 developed complete heart block 3.5 years after being lost to follow-up, during which they were found to have moderate MR, and the final patient was additionally lost to follow-up and found to have central moderate MR after eventually following up with his cardiologist 6 years after surgery.

Furthermore, retrospective studies and various meta-analyses have shown a mix of outcomes after MVr when compared with MVR, suggesting that these outcomes are variable [24–27]. Overall, it is still unclear when MVr is a more appropriate option than MVR, however, a well-performed MVr in an appropriately selected patient can have non-inferior outcomes to MVR, and the criteria for choosing one surgical intervention over the other deserve further discussion and prospective studies.

Limitations

This study contains the limitations inherent in a retrospective review. While follow-up between groups was similar, the lack of follow-up ECHO data in 35% of the study population limits the overall scope of this study. The low number of patients studied similarly limits the ability to broadly draw conclusions, however, it does offer insight into surgeon-specific experience in performing these procedures. The single-center, single-surgeon nature of this study allows for consistency and expertise in technique and decision-making, to yield results that may not be able to be extrapolated to the greater population. Each case was discussed with the multidisciplinary heart team prior to operation, however, the decision to perform MVr vs. MVR was made intraoperatively at the discretion of the experienced surgeon, and not by a multidisciplinary team. While this may limit the standardization of principles described here, it also suggests that surgeon expertise may play a role in performing a quality repair. Additionally, the Carpentier-McCarthy-Adams IMR ETlogix (Edwards LifeSciences, California, USA) annuloplasty ring was used in all patients who underwent MVr. This may limit the overall conclusions one may draw about mitral valve repair with other devices. This study was performed in patients who experienced mitral regurgitation in the setting of ischemic heart disease and cannot be extrapolated to patients with primary mitral regurgitation.

Conclusion

Patients with moderate to severe ischemic mitral regurgitation may benefit from mitral valve surgery at the time of coronary revascularization. MVr and MVR have similar hemodynamic outcomes at two years in our population with an overall low incidence of clinically significant MR and equivalent survival. Surgeon experience, regional factors, and careful preoperative and intraoperative decision making may impact the success of both methods of mitral valve surgery. Mitral valve repair can be a durable option in appropriately selected patients and should be considered first in patients with ischemic mitral regurgitation.

Supporting information