Аннотация
A BSTRACT Background: Postoperative pain after total hip arthroplasty (THA) is influenced by biological, surgical, and psychosocial factors. Pain catastrophizing, kinesiophobia, and central sensitization are clinically relevant constructs in pain psychology, but their independent association with postoperative pain outcomes remains uncertain, particularly in small open datasets. Objective: To conduct an exploratory secondary analysis of an open clinical dataset and evaluate whether preoperative psychological factors are associated with postoperative pain at rest and during walking after THA. Materials and Methods: This study used publicly available patient-level data from Harvard Dataverse. The source dataset included 92 adult patients undergoing THA; the final analytical sample included 91 patients after listwise deletion. Sex was coded as 1 = male and 2 = female. Preoperative psychological variables included the Pain Catastrophizing Scale (PCS), Tampa Scale for Kinesiophobia-J (TSK-J), and Central Sensitization Inventory (CSI). Co-primary postoperative outcomes were Visual Analog Scale (VAS) pain at rest and VAS pain during walking. Spearman correlations were used for bivariate analysis because postoperative VAS outcomes were nonnormally distributed. Benjamini-Hochberg false discovery rate (FDR) correction was applied across the six bivariate tests. Secondary adjusted linear regression models used HC3 robust standard errors. Logistic regression analyses were considered exploratory, including a sensitivity model using postoperative walking pain ≥10 mm. Results: Before multiplicity correction, PCS was weakly associated with postoperative pain at rest (ρ = 0.214, P = 0.0416), while TSK-J (ρ = 0.210, P = 0.0454) and CSI (ρ = 0.279, P = 0.0074) were weakly associated with postoperative walking pain. After FDR correction, only CSI remained significantly associated with postoperative walking pain (FDR-adjusted P = 0.0444). In adjusted linear regression, none of the psychological variables were statistically significant independent correlates of postoperative VAS pain. In exploratory logistic regression for any postoperative walking pain, older age was associated with lower odds of pain (odds ratio = 0.499, 95% confidence interval 0.284–0.879, P = 0.016). However, in the sensitivity model using postoperative walking pain ≥10 mm, no psychological variable or age was statistically significant. Conclusion: In this exploratory secondary analysis, preoperative psychological variables showed weak bivariate associations with postoperative pain outcomes after THA. Only the CSI-walking pain association remained significant after FDR correction, and psychological variables were not independently associated with postoperative pain in adjusted or sensitivity models. Findings should be interpreted as hypothesis-generating rather than confirmatory.