Résumé : BACKGROUND: The definition and incidence of patient-prosthesis mismatch (PPM) in the mitral position are unclear. OBJECTIVES: To determine the impact of PPM on late survival and functional status after mitral valve replacement with a mechanical valve. METHODS: Between 1992 and 2005, 714 patients (mean [± SD] age 60±10 years) underwent valve replacement with either St Jude (St Jude Medical Inc, USA) (n=295) or Carbomedics (Sulzer Carbomedics Inc, USA) (n=419) valves. There were 52 concomitant procedures (50 tricuspid annuloplasties, 25 foramen oval closures and 20 radiofrequency mazes). The mean clinical follow-up period was 4.4±3.3 years. The severity of PPM was established with cut-off values for an indexed effective orifice area (EOAi) of lower than 1.2 cm 2/m 2, lower than 1.3 cm 2/m 2 and lower than 1.4 cm 2/m 2. Parametric and nonparametric tests were used to determine predictors of outcome. RESULTS: The prevalence of PPM was 3.7%, 10.1% and 23.5% when considering values of lower than 1.2 cm 2/m 2, lower than 1.3 cm 2/m 2 and lower than 1.4 cm 2/m 2, respectively. When considering functional improvement, patients with an EOAi of 1.4 cm2/m2 or greater had a better outcome than those with an EOAi of lower than 1.4 cm 2/m 2 (OR 1.98; P=0.03). When building a Cox-proportional hazard model, PPM with an EOAi of less than 1.3 cm 2/m 2 was an independent predictive factor for midterm survival (HR 2.24, P=0.007). Other factors affecting survival were age (HR 1.039), preoperative New York Heart Association class (HR 1.96) and body surface area (HR 0.31). CONCLUSIONS: In a large cohort of patients undergoing mitral valve replacement with mechanical prostheses, PPM defined as an EOAi of lower than 1.3 cm 2/m 2 significantly decreased midterm survival. This level of PPM was observed in 10.2% of patients. Patients with an EOAi of 1.4 cm 2/m 2 or greater had greater improvement of their functional status. ©2010 Pulsus Group Inc. All rights reserved.