Predictors of Functional Recovery in Individuals Exposed to Late-Stage Exercise Programs after Total Knee Replacement: A Secondary Analysis of a Randomized Controlled Trial

Importance Total knee replacement (TKR) is common for advanced cartilage degeneration. Often, functional limitations persist, and late-stage exercise (2–4 months after TKR) may improve recovery. Objective The objective of this study was to identify predictors of late-stage functional recovery after TKR and their variations by exercise program. Design This study was a secondary analysis of a randomized controlled trial. Setting The settings were 1 physical therapy clinic and 4 community centers in Allegheny County, Pennsylvania. Participants The study included 199 individuals (mean age = 70 years; body mass index = 31 kg/m2; 60% women). Interventions Interventions were physical therapy exercises, community-based exercises, and usual care. Main Outcome and Measures The main outcome was functional recovery at 6 months after intervention. Functional recovery was defined as ≥50% improvement on the Western Ontario and McMaster Universities Osteoarthritis Index physical function subscale, at least somewhat better on the Global Rating of Change, and ≥ 20% improvement on ≥2 performance-based tests. Predictors were identified using logistic regression. Odds ratios (ORs) and 95% confidence intervals (CIs) were reported. Results In the entire cohort, 33% of participants achieved functional recovery, including 42% in the physical therapy group and 38% in the community group. Overall, better baseline health (odds ratio [OR] = 2.70 [95% CI = 1.34–5.47]), discharge to home (OR = 6.24 [95% CI = 1.94–20.09]), and higher Western Ontario and McMaster Universities Osteoarthritis Index pain subscale scores (OR = 1.22 [95% CI = 1.07–1.40]) predicted functional recovery. In the physical therapy group (n = 83), positive predictors were better health (OR = 4.40 [95% CI = 1.20–16.06]), discharge to home (OR = 11.67 [95% CI = 2.01–67.76]), regular use of nonsteroidal antiinflammatory drugs (OR = 3.76 [95% CI = 1.11–12.78]), and as-needed use of analgesics (OR = 10.53 [95% CI = 1.75–63.48]). Negative predictors, associated with a lower likelihood of recovery, were regular use of salicylates (OR = 0.22 [95% CI = 0.07–0.74]) and greater use of pain-coping strategies (OR = 0.51 [95% CI = 0.29–0.89]). No predictors were identified in the community group (n = 76). Conclusions Better baseline health, discharge to home after TKR, and higher pain levels predicted functional recovery. Physical therapy provided additional benefits for a subset of participants. Relevance This evidence helps clinicians set realistic functional recovery expectations and strategies to facilitate function late after TKR.