Background: We compared and analyzed differences between repair vs. replacement of mitral valves on severe rheumatic mitral stenosis by looking at mid-term clinical outcomes. Methods: Patients with severe rheumatic mitral stenosis (mitral valve area <= 1.5 cm(2), with or without mitral valve regurgitation) from January 2011 to September 2017 were divided into two groups: a mitral valve repair group (MVP) and a mitral valve replacement group (MVR). After propensity score matching between the two groups, we compared changes in post-operation clinical outcomes. We also monitored changes in left ventricular longitudinal and circumferential strain in successfully matched patients (20 pairs) by echocardiography speckle tracing. Results: A total of 921 patients were recruited (221 in MVP and 700 in MVR). After a propensity score matching, 216 cases were selected with 108 patients in each group. With a follow-up period of 3 months to 7.1 years, the incidence of heart failure was observed to be significantly higher in the MVR group than in the MVP group (P<0.05). Echocardiographic speckle tracking imaging analysis showed that left ventricular longitudinal strain decreased in the MVR group, suggesting that a change of cardiac structure may affect the cardiac function. Conclusions: Selecting suitable patients for mitral valve repair is feasible even for patients with severe rheumatic mitral stenosis. In our study, compared with prosthetic valve replacement in these patients, valve repair still significantly reduced the valve-related complications rates and improved the quality of life during the follow-up.