Ninty-seven patients underwent prosthetic mitral valve replacement at Seoul National University hospital during the eight year period, from January 1, 1971, through September 20, 1978. Included in this group were patients who had concomitant aortic valve replacement (10 patients), tricuspid annuloplasty and valve replacement (12 patients), closure of atrial or ventricular septal defect (3 patients), and aortic valve replacement and tricuspid annuloplasty (1 patient). The ages of the patients at the time of operation ranged from 11 to 54 years, the mean being 32.0 years. Forty-eight patients were female and 49 male. Mitral valve replacement was carried out on 18 patients (18.6%) under 20 years of age. The operative findings of the mitral valve indicated rheumatic valvulitis in 94 patients, while in the remaining three the etiology was undetermined. Twenty-seven patients had predominant mitral stenosis, 29 predominant mitral regurgitation, and 27 mixed mitral lesions. Four patierts belonged to functional class II (NYHA), 45 to class III, and 34 to class IV. On physical and angiographic examinations, 13 patients had associated tricuspid regurgitation, 11 patients had aortic valvular disease, and 3 patients had atrial septal defect or ventricular septal defect. Forty-three patients had atrial fibrillation, 24 had normal sinus rhythm, and the remaining four had other supraventricular arrhythmia. Sixty-four patients were studied before operation by right heart catheterization. Severe pulmonary hypertension, indicated by a pulmonary arterial systolic pressure of 61 mmHg or greater, was present in 19 patients. In only one patients was the pulmonary arterial pressure normal. The average value for the systolic pressure in the 64 patients was 52 mmHg. One-hundred-five cardiac prosthetic valves were placed in 97 patients in the past 8 years. This series included 97 mitral, 11 aortic, and 7 tricuspid valve replacements. There were 18 perioperative deaths, an over-all mortality of 18.6 per cent. High perioperative mortality was seen in patients with associated aortic valvular diseases. Patients who were in class IV preoperatively had a higher motality (50 per cent) than those in class III (22 per cent). A dramatic decrease in the operative mortality from 100 per cent in 1971 to 5.7 per cent in this year was noted.