A 5 year prospective study of patient-relevant outcomes after total knee replacement.

OBJECTIVE: To prospectively describe self-reported outcomes up to 5 years after total knee replacement (TKR) in Osteoarthritis (OA) and to study which patient-relevant factors may predict outcomes for pain and physical function (PF). METHODS: 102 consecutive patients with knee OA, 63 women and 39 men, mean age 71 (51-86) assigned for TKR at the Department of Orthopaedics at Lund University Hospital were included in the study. The self-administered questionnaires Knee injury and Osteoarthritis Outcome Score (KOOS) and SF-36 were mailed preoperatively and 6 months, 12 months and at 5 years postoperatively. RESULTS: Response rate at 5 years was 86%. At 6 months significant improvement was seen in all KOOS and SF-36 scores (P<0.001). The percentage of patients performing more demanding functions related to sports and recreation increased postoperatively. The best postoperative result was reported at the 1 year follow-up. Compared to the 1 year follow-up, a significant (P</=0.01) decline was seen at 5 years in the KOOS subscale activity of daily living (ADL) function (82-73) and the SF-36 subscale bodily pain (72-63), PF (61-51) and vitality (69-59). Patients who scored in the lowest quartile preoperatively in the KOOS subscales pain and ADL made the greatest improvements to 1 year (18-82, 22-80) but also declined the most from 12 months to 5 years (82-72, 80-66). Being 10 years older pre-operatively predicted 5-7 points worse scores in KOOS pain and KOOS symptoms at 1 and 5 years. When adjusted for age, sex and comorbid conditions, pre-operative SF-36 scores did not predict postoperative KOOS pain or PF scores. CONCLUSION: Compared to preoperatively, a significant improvement was still seen 5 years postoperatively. However, the best result was reported at 1 year, indicating a decline from 1 to 5 years after TKR. To fully evaluate the results of TKR with regard to pain and PF, follow-ups longer than 2 years are needed, and items of more demanding PFs should be included. Older age to some extent predicted more postoperative pain and other symptoms, however, no predictors of postoperative PF were found, indicating the difficulty of determining preoperatively who will benefit more or less from the procedure