Preoperative anxiety and tourniquet use are associated with early postoperative pain trajectories in uncomplicated primary unilateral TKA: a retrospective cohort study

Early pain after total knee arthroplasty (TKA) can affect the first week of mobilization and rehabilitation. Preoperative anxiety and tourniquet exposure are clinically plausible correlates of early pain, but their associations with repeated pain measurements during uncomplicated primary unilateral TKA recovery have not been well described. We studied 236 patients who underwent primary unilateral TKA for knee osteoarthritis in a retrospective cohort. Preoperative anxiety was measured with the anxiety subscale of the Amsterdam Preoperative Anxiety and Information Scale (APAIS; range, 4–20). The primary outcome was post-exercise visual analogue scale (VAS) pain on postoperative days 1, 3, 5, and 7. We compared random-effect structures before fitting mixed-effects models and treated APAIS score as a continuous exposure. In the primary mixed-effects model, postoperative day was treated as a categorical variable to capture the observed nonlinearity trajectory. Trajectories of the Hospital for Special Surgery (HSS) scores, as a composite knee score, were analyzed in a secondary analysis. Restricted cubic spline (RCS) analyses were conducted as exploratory analyses to characterize nonlinearity dose–response relationships. Propensity score-weighted generalized estimating equation (GEE) models were used as sensitivity analyses to assess robustness to confounding. Of the 236 patients, 100 underwent TKA without a tourniquet and 136 underwent TKA with a tourniquet. In the categorical-time VAS model, adjusted VAS in the no-tourniquet group increased from postoperative day 1 (POD1) to POD3 (β = 0.660, 95% CI 0.467 to 0.854; P < 0.001), was similar on POD5 (β = 0.098, 95% CI -0.095 to 0.292; P = 0.320), and declined by POD7 (β = -1.115, 95% CI -1.309 to -0.922; P < 0.001). Tourniquet use was associated with higher adjusted VAS at each postoperative day: POD1 1.154 (95% CI 0.891 to 1.416), POD3 0.922 (0.660 to 1.185), POD5 1.144 (0.881 to 1.406), and POD7 1.696 (1.434 to 1.959); all P < 0.001. Each one-point increase in APAIS score was also associated with higher VAS (β = 0.147, 95% CI 0.107 to 0.187; P < 0.001). The APAIS-by-tourniquet interaction was not significant (β = 0.007, 95% CI -0.090 to 0.105; P = 0.886), and the three-way interaction terms were also not significant. In secondary analyses of the HSS score, tourniquet use was associated with lower scores across follow-up, whereas APAIS score showed no significant association. Among patients with uncomplicated recovery after primary unilateral TKA in this retrospective cohort, higher preoperative APAIS score and tourniquet use were associated with higher early post-exercise VAS pain. APAIS score did not appear to modify the tourniquet-pain association and was not associated with the HSS score, though this null finding should be interpreted cautiously given incomplete follow-up and missing subdomain data. These findings highlight the potential value of preoperative psychological assessment and day-specific pain trajectory evaluation in observational research; however, the retrospective design precludes causal interpretation, and these results should not be taken as evidence that anxiety management or tourniquet avoidance would improve outcomes. Not applicable.