Discussion
Variations in mitral valve surgery outcomes by geographic area and institution has been emphasized in multiple reports10-12. In this analysis, the outcomes of mitral valve surgery within the same institution were different according to the experience of the operator surgeon. Experienced surgeons were more likely to attempt repair, used different operative techniques and had higher repair rates. While surgeon experience was not an independent risk factor for post-repair MR recurrence, it was an independent risk factor for mortality (HR=2.66) irrespective of valve repair/replacement and the pre-operative characteristics of the patients.
As in the New York state data, we identified higher rates of both CHF and urgent/emergent surgery with less experienced surgeons. This suggests a possible trend in referrals where higher risk patients tend to be operated on by less experienced surgeons, who may be less inclined to perform MV repair under unfavorable conditions8.
Interestingly, the number of techniques used in the repair was not different between experienced and less experienced surgeons, but the management of the anterior leaflet using neochordae — a complex technique requiring a high level of experience — resulted in higher repair rates in the experienced group. Leaflet resection and annuloplasty rates were similar between both cohorts and also similar to the national average, but artificial neochordae rates differed. In the national study, 22.7% of mitral valve repairs had artificial cord implantation, compared to 41.5% of repairs with experienced surgeons and 18.7% with less experienced surgeons in this analysis13.
This analysis also addresses the effects of calcification on the mitral valve surgery outcomes. Extensive calcification extending to the leaflets was the most common reason for replacement, and mitral annular calcification (MAC) was the only independent risk factor for MR recurrence. MAC was reported in ~ 20% of patients undergoing mitral valve surgery14 and was an independent risk factor for valve replacement15. The extent of leaflet prolapse was not a risk factor for recurrence in this cohort, as opposed to previous literature where anterior leaflet and bi-leaflet involvement were associated with increased risk of recurrence16-18. Of note, these studies included patients who had surgeries decades ago, when the techniques of anterior leaflet repair were not well-established and adopted.
The recurrence rate over the study period in this cohort was 13%, of which 65% happened in the first year, which matches the existing literature16. The most common causes for recurrence were ring dehiscence and new lesions, which, at 36% each, largely matches previous studies19, 20. Surgeon experience was not a risk factor for MR recurrence after repair in this cohort as opposed to previous reports. This could be explained by the fact that less experienced surgeons were more likely to repair posterior leaflet prolapse and replace anterior and bi-leaflet prolapses.
Finally, patients operated on by experienced surgeons had better adjusted survival. This finding can’t be explained by the repair/replacement rate or the preoperative characteristics of the patients as it was shown to be an independent factor in the Cox hazard model. This was reported on a larger scale in the NY state data and requires further research8.