ObjectiveTo compare the clinical outcomes of mitral valvuloplasty (MVP) and mitral valve replacement (MVR) for infective endocarditis, and to investigate the effect of MVP under different surgical risks. MethodsA retrospective study was done on the patients with mitral infective endocarditis, who underwent surgical treatment in our department from January 2018 to March 2022. According to the procedures, the patients were divided into a MVP group and a MVR group. Propensity score matching method was applied with a ratio of 1:1 to eliminate the biases. The early and midterm outcomes were compared between the two groups after matching. According to the European System for Cardiac Operative Risk Evaluation Ⅱ(EuroSCORE-Ⅱ), the effect of MVP was compared between high and low risk patients. ResultsA total of 195 patients were collected. There were 141 patients in the MVP group (120 males, 85.1%) and 54 patients in the MVR group (41 males, 75.9%). The mean follow-up time was (34.0±16.1) months. Patients in the MVP group were younger [(42.7±14.6) years vs. (56.8±13.0) years, P<0.001] and had better preoperative conditions. The patients in the MVP group had a shorter ICU stay [3.0 (2.0, 5.0) d vs. 4.0 (3.0, 8.0) d, P=0.004], and lower incidences of low cardiac output syndrome (0.7% vs. 9.3%, P=0.007), in-hospital mortality (0.0% vs. 3.7%, P=0.023), and follow-up mortality (4.3% vs. 15.4%, P=0.007). However, after 1:1 propensity score matching, there were no statistical differences in the baseline data or postoperative and follow-up adverse events between the two groups (P>0.05). Also, there was no statistical difference in the mortality of high-risk patients between MVP and MVR group (P>0.05). There was no statistical difference in the reoperation or recurrent severe mitral regurgitation between high and low-risk patients in the MVP group (P>0.05). Conclusion MVP is feasible for treating mitral lesions caused by infective endocarditis with good early and midterm outcomes. For patients with severer preoperative conditions, if the leaflet damage is not severe, MVP may be a viable option, but validation with larger sample sizes is needed.