One thousand five hundred ninety-eight patients who underwent cardiac valve replacement were reviewed. One hundred fifty-two patients (10 percent) required a second valve replacement. The indications for repeat valve replacement were prosthetic valve dysfunction in 53 patients (35 percent), development of a new valvular lesion in 46 patients (30 percent), simple closure of a perivalvular leak in 14 patients (9 percent), change of the valve poppet in 13 patients (8 percent), severe hemolysis or emboli in 21 patients (14 percent), and prosthetic valve endocarditis in 5 patients (3 percent). The mean preoperative New York Health Association functional class improved from 3 to 1.5 in the nonreoperated patients and from 2.9 to 1.8 in patients who underwent a second valve replacement. Similarly, the mean cardiac index improved from 2.5 to 2.9 and from 2.5 to 2.8 in nonreoperated and reoperated patients, respectively. The operative mortality rate was 14 percent in the nonreoperated patients and 16 percent at second operation in the reoperated patients. Using actuarial techniques, the risk of repeat valve replacement was 1 to 4 percent per year. Long-term survival was compared between groups. Using actuarial techniques, the estimated survival rates at 1, 5, and 10 years were 89 percent, 69 percent, and 52 percent, respectively in nonreoperated patients and 87 percent, 60 percent, and 37 percent in reoperated patients. This study has documented the excellent improvement in functional and hemodynamic state after second cardiac valve replacement. The operative mortality and long-term survival rates were similar to those of the nonreoperated patients. Patients having repeat cardiac valve replacement can expect good improvement in length and quality of life.