Background Mitral valve replacement (MVR) has a high mortality and mor
bidity. It has been suggested that preservation of the subvalvular app
aratus and more optimal timing of surgery might improve outcome. Metho
ds and Results We performed a retrospective study of 612 consecutive p
atients who underwent mitral Valve repair or replacement: 226 patients
had repair, 68 had replacement with subvalvular preservation (MVR/SVP
), and 318 had replacement without subvalvular preservation (MVR/NoSVP
). Baseline characteristics were most unfavorable in the repair group
with respect to age (P=.002) and in the repair and MVR/SVP groups with
respect to NYHA functional class and left ventricular function (P=.04
4). Thirty-day mortality was lower in the repair (1.8%, P=.046) and MV
R/SVP (1.5%, P=NS) groups than the MVR/NoSVP group (5.0%). Overall sur
vival at 7 years was better in the repair (71.2+/-5.6%, P=.022) and MV
R/SVP (66.2+/-12.4%, P=.017) groups than the MVR/NoSVP group (63.5+/-3
.4%). Myocardial failure caused 66 of 107 complication-related deaths.
Multivariate analysis confirmed independent beneficial effects of rep
air on 30-day mortality (odds ratio, 0.27, P<.05) and of repair and MV
R/SVP on overall mortality (hazard ratios, 0.43, P<.001 and 0.40, P<.0
5, respectively) and complication-related death (hazard ratios, 0.38,
P<.001 and 0.35, P<.05, respectively). Preoperative NYHA class III or
IV symptoms and left ventricular impairment were independent risk fact
ors for death and myocardial failure. Conclusions Mitral valve repair
is superior to replacement. If repair is not feasible, the subvalvular
apparatus should be preserved. Early surgery before the development o
f severe symptoms and demonstrable left ventricular impairment is also
needed to optimize outcome.