Survivorship analysis is a useful way of expressing the long term results of joint replacement. With the increase of follow up period in total knee replacement, it has became possible to predict the survival rate of implants in total knee replacement. The survivorship of total knee replacement in 145 consecutive patients(209 knees) between 1982 and 1988 with follow-up period of up to 10 years was estimated by Dobbs' method(1980), and we analyzed, clinically and radiologically, 116 patients(177 knees) which were followed up more than 2 years(from 2 years to 10 years, average 6 years and 4 months). Also, we evaluated the knee function according to the functional knee scores of Hospital for Special Surgery. 1. The average age at operation was 55.6 years(from 19 years to 75 years). Male was 12 patients and female was 104 patients. 2. There were osteoarthritis in 83 knees(46.5%), rheumatoid arthritis in 71 knees(40.0%), secondary osteoarthritis in 18 knees(10.0%), tuberculous arthritis in 3 knees (2.0%), and Charcot joint in 2 knees(1.5%). 3. The Hospital for Special Surgery functional knee and pain scores was improved from average 43 points preoperatively to average 84 points postoperatively. The results were satisfactory in 162 knees(91.4%, excellent in 64.0% and good in 27.4%). 4. Revision surgery was performed in 15 knees of the 209 knees(7.2%), due to aseptic loosening in 9 knees(4.3%) and septic loosening in 6 knees(2.9%). 5. Survival rate of total knee replacement with reliability was 87.7%, 8 years after operation for 145 patients, 209 knees. In summary, total knee replacement is a relatively satisfying and predictable procedure which can obtain the relief of pain, restoration of function, and long term survivorship.