Factors that can predict pain with walking, 12 months after total knee arthroplasty

Background and purpose — Functional limitations after total knee arthroplasty (TKA) are common. In this longitudinal study, we wanted to identify subgroups of patients with distinct trajec- tories of pain-related interference with walking during the fi rst year after TKA and to determine which demographic, clinical, symptom-related, and psychological characteristics were associ- ated with being part of this subgroup. Patients and methods — Patients scheduled for primary TKA for osteoarthritis (n = 202) completed questionnaires that evalu- ated perception of pain, fatigue, anxiety, depression, and illness on the day before surgery. Clinical characteristics were obtained from the medical records. Interference of pain with walking was assessed preoperatively, on postoperative day 4, and at 6 weeks, 3 months, and 12 months after TKA. Results — Using growth mixture modeling, 2 subgroups of patients were identifi ed with distinct trajectories of pain-related interference with walking over time. Patients in the Continu- ous Improvement class (n = 157, 78%) had lower preoperative interference scores and reported a gradual decline in pain- related interference with walking over the fi rst 12 months after TKA. Patients in the Recurrent Interference class (n = 45, 22%) reported a high degree of preoperative pain-related interference with walking, initial improvement during the fi rst 3 months after TKA, and then a gradual increase—returning to preoperative levels at 12 months. Patients in the Recurrent Interference class had higher preoperative pain, fatigue, and depression scores, and poorer perception of illness than the Continuous Improvement class. Interpretation — 1 in 5 patients did not improve in pain-related interference with walking at 12 months after TKA. Future studies should test the effi cacy of interventions designed to modify preop- erative characteristics.