|
|
 
Morbidity and Mortality of a Failed Attempt at Mitral Valve Repair Converted to Replacement at the Same OperationNorthrup WF, DuBois KA, Kshettry VR. Journal of Heart Valve Surgery. 2003;12:700-76.Reviewer: Feroze Mahmood, MD
Background and Introduction:Over the years mitral valve repair (MVRpr) has been shown to provide a better postoperative outcome in patients with mitral regurgitation (MR) than valve replacement (MVR) and should be the preferred mode of correction of valve abnormality1. However, MVRpr is a technically complicated operation and requires additional skills and understanding of the mitral valve anatomy and pathology by the surgical team, and longer time on the cardiopulmonary bypass (CPB) and the possibility of going back on CPB if the repair fails.2 The authors, through a retrospective analysis of data, have tried to dispel the notion that a failed repair (FRpr), which is later converted to an MVR during the same operation, incurs additional morbidity or mortality. Methods:A database was examined for the time between January 1, 1979 to December 31, 1999, for all adult patients who underwent mitral valve operations (MVO), and only in-hospital period data was collected. All operations were performed using CPB. A repair attempt was considered failed, when the surgeon attempted to repair the valve and decided prior to left atrial closure that it was not going to work or when the repair was completed and subsequent transesophageal echocardiography (TEE) demonstrated 2+ or more MR. The majority of repairs involved annuloplasty ring insertions, although more complicated operations were also performed in a minority of the patients. Results:A total of 2017 MVO were performed, 1054 MVR (74.5%), 442 MVRpr (21.9%) and 71 FRpr (3.5%). Two surgeons performed >25% of all the MVO, >50% of all MVRpr and ~50% of all FRpr. These two surgeons also had a combined MVRpr rate more than three times that of the 15 other surgeons. >50% of all MVRpr and ~ 50% of all FRpr were carried out between 1996 and 1999. Rheumatic etiology of valve disease was more common in the group undergoing MVR and degenerative etiology of valve disease was more common in the groups undergoing MVrp and FRpr. All three groups were similar with respect to left ventricular ejection fraction (EF) and coexisting diabetes mellitus, renal failure and pulmonary disease. 55.7% of MVR patients were females, as compared to 39.4% of MVRpr patients and 38% of FRpr patients, and MVR patients were older than MVRpr and FRpr patients (64.0 + 12.4 years compared to 62.0 + 13.1 years and 59.3 + 12.7 years). NYHA functional class III/IV was more prevalent in MVR group (73.5%) as compared to MVRpr (54.2%) and FRpr (56.5%). Similarly, a non-elective status, congestive heart failure, pulmonary hypertension and cerebrovascular accident (CVA) were more prevalent in MVR than MVRpr and FRpr. Atrial fibrillation and previous cardiac surgery were more common in MVR than in either MVRpr or FRpr. MVR group had a higher prevalence of combined procedures and IABP use was higher in MVR and FRpr than MVRpr group. FRpr had a longer cross clamp and CPB time than MVRpr and MVR. MVR required fewer units of fresh frozen plasma in the first 24 hours than the other two groups. Postoperatively, the three groups were similar, except for longer ICU stay and ventilation and a longer overall length of stay for MVR group (LOS 13.8 ± 11.0 days for MVR vs. 10.9 ± 8.2 days for MVRpr and 10.2 ± 7.6 days for FRpr) and a higher reoperation rate for MVRpr for valve dysfunction (1.6% vs. 0.5% for MVR). Mortality was higher for the MVR group (7.2% as compared to 4.5% for MVRpr), but comparable to the FRpr group (7.0%). The mortality was increased when the procedure was combined with another cardiac surgery for all three groups. Comments:This paper described a retrospective analysis of data collected over two decades. Because there was no excess mortality or morbidity associated with failed attempts at MV replacement, the authors concluded that MVRpr can be safely undertaken when indicated, even with additional cardiac procedures. The study suffered from several limitations of a retrospective analysis. First, it was not clear whether the three study groups were truly comparable. The MVR group had a greater incidence of comorbidities and a greater frequency of emergent operations than the other groups. Although the data was collected from 1979, more than half of all MVRpr and almost one half of FRpr were carried out between 1996 and 1999, because of a marked increase in the number of MVO being performed in the last few years. This increase has been noted by numerous studies and has been attributed to multiple factors such as aging population, heredity, environmental factors and referral patterns.3, 4 Additionally, almost 50% of MVR and FRpr were done by two surgeons out of 17 in the group. Mitral valve surgery is a rapidly evolving subspecialty with a high degree of complexity, which requires specific training and experience.5 There could have been a historical bias and intersurgeon variability, which were not controlled for in this study. Furthermore, the authors have not clearly defined the criteria that they used for selection of patients for repair as opposed to replacement. Was it made preoperatively? Or was every patient a candidate for MVRpr until a decision was made based on the intraoperative TEE findings or direct visual examination and valve analysis? Similarly, it was also not clearly stated whether all the patients had intraoperative TEE for assessment of repair and whether and how frequently the diagnosis of failed repair was based on intraoperative TEE findings. A variety of repair techniques were used for repair but it was not mentioned which were more prone to failure, and if there was any specific personal preference on the part of the surgeons to perform a specific operation. Within these limitations, it was interesting to note that the FRpr was not shown to incur any increased morbidity or mortality when compared to MVR alone. The data possibly lends further reinforcement to the current trend that severe MR should be repaired whenever possible, especially given the better myocardial protection techniques available, allowing the surgeons to contemplate more and more complex repairs and longer time on the CPB. With the establishment of mitral valve repair as the standard operation for MR, intraoperative TEE has become extremely important in determining the feasibility of repair and assessing the success of repair. Intraoperative post-CPB TEE examination also has been shown to have an excellent correlation with late postoperative transthoracic echocardiography and thus can reliably predict late valve dysfunction.6 Functional anatomy of the mitral valve as delineated by TEE was strongly predictive of valve repairability and postoperative outcome.7 Intraoperative pre-CPB TEE demonstrates the mitral valve in a dynamic environment with real time loading conditions compared to a flaccid and paralyzed heart on CPB, thus making the valve analysis more accurate and aiding the surgical decision making.8 Reference:
Table of Contents:
© Society of Cardiovascular Anesthesiologists Questions or comments? Please send email to webmaster@scahq.org |