Objective: Evaluate the operative results of mitral valve repair (MVV) and mitral valve replacement (MVR) performed through standard and smaller incisions.
Methods: From January 1997 through December 2002, 821 consecutive patients underwent mitral valve operation. Of these procedures, 475 were MVV and 346 were MVR. A logistic regression model was developed to identify the risk factors for early mortality and to evaluate the effect of replacement versus repair and standard versus small incision.
Results: Replacement patients were older, more likely New York Heart Association (NYHA) class III or IV, more likely female, and had more frequent previous median sternotomy and stroke (all P <.05). The mitral diagnoses in the 2 groups were markedly different. Prolapse and ischemia dominated the repairs, whereas calcific and rheumatic diagnoses required replacement. There were 667 concomitant procedures performed on these patients, most commonly coronary artery bypass graft (229), aortic valve replacement (170), maze (79), and tricuspid valve (TV) repair/replacement (73). Thirty-three patients (4.0%) died in the postoperative period, 2.3% after repair and 6.4% after replacement ( P <.01). Endocarditis (4/17), calcific disease (7/73), and ischemic disease (9/121) accounted for 26% of patients and 60% of deaths. Multivariate regression analysis identified NYHA class, emergent status, concomitant TV operation, and history of renal failure, but not repair versus replacement, as independent risk factors predicting mortality. We estimated that 356 of the 821 patients (43%) were candidates for small-incision operations, the others were excluded by the need for concomitant procedure or other cause. A total of 205/356 (57%) actually underwent small-incision operations, all with central cannulation and standard techniques. From 1997-1999, 32% of eligible patients were so treated, but from 2000-2002, with increasing surgeon experience, this percentage rose significantly to 71% ( P <.01). Eligible patients who underwent small-incision operation were younger and had lower NYHA classifications, lower preoperative creatinine, and shorter length of stay (all P <.01) than those who had standard incisions. Cross-clamp time, perfusion time, and mortality rate were not significantly different.
Conclusions: The mortality rate for MV operations is concentrated among a few diagnoses. In some patients surgery may be approached safely through smaller incisions without introducing new elements of operative risk.