• BACKGROUND
    • There appears to be substantial variability in outcomes > 2 years following total knee arthroplasty (TKA) that is masked by whole group analyses. The goal of the study was to identify trajectories of pain and function outcomes up to 5 to 8 years post-TKA and to identify baseline factors that are associated with different trajectories of recovery.
  • METHODS
    • Baseline, 6-month, and 12-month pain and function data were collected in a previous study investigating predictors of outcome following primary TKA (n = 286), along with a variety of baseline predictor variables. The present study obtained pain and function data at 5 to 8 years following TKA in the same cohort (n = 201). Latent class linear mixed models were used to identify different classes of pain and functional trajectories over time. The extent to which differences across latent classes were explained by baseline predictor variables was determined.
  • RESULTS
    • Three classes of pain and two classes of function trajectory were identified. While most patients (84% to 93%) followed a trajectory that showed an initial rapid gain following surgery that was sustained through 5 to 8 years, both pain and function included at least one trajectory class that showed a meaningful change after 12 months. No predictor variables were significantly associated with either the pain or function classes.
  • CONCLUSIONS
    • Most patients follow a traditional trajectory of recovery in knee pain and function over 5 to 8 years. However, alternative trajectories are observed in an important minority of patients such that knee pain and function at 12 months after surgery does not always reflect outcomes at 5 to 8 years.