Predictors of Chronic Postsurgical Pain in Elderly Patients Undergoing Hip Arthroplasty: A Multi-Center Retrospective Cohort Study.

Background: Identifying patients at high risk of developing chronic postsurgical pain (CPSP) is of extreme importance in order to help launch appropriate therapeutic strategies and intensive initiation of pain management.

Aim: In this study, we aimed to conduct a multi-center retrospective cohort study to establish a prognostic model and a nomogram to predict the risks of CPSP in elderly patients who underwent hip arthroplasty at 6 months postoperatively.

Methods: We collected data from 736 patients aged >65 years who had undergone hip arthroplasty from October 1, 2016 to September, 30, 2018 at multiple tertiary referral centers in Guangzhou, China. All data were randomly stratified into a training set and a testing set at a ratio of 8:2. Data were analyzed via multiple logistic regression analysis with receiver operating characteristic (ROC) curves and areas under the curve. This model was further validated by estimating calibration and discrimination. A nomogram was ultimately developed.

Results: A total of 736 eligible patients were enrolled, 27.20% of whom developed CPSP within 6 months postoperatively. Preoperative pain in the surgical area (OR=2.456, 95% CI:1.814-3.327, P<0.001), preoperative depression state (OR=1.256, 95% CI:1.146-1.378, P<0.001), surgical type (OR=7.138, 95% CI:3.548-14.364, P<0.001), acute postoperative numerical rating scale score (OR=5.537, 95% CI:3.607-8.499, P<0.001) and analgesic type (patient-controlled epidural analgesia: OR=0.129, 95% CI:0.055-0.299, P<0.001; patient-controlled intravenous analgesia: OR=0.033, 95% CI:0.011-0.097, P<0.001) were identified as independent significant factors associated with CPSP. A prognostic model was established and further validated. An ROC curve confirmed the predictive ability of this model with a high sensitivity value of 92.12% (95% CI:86.90-95.74) and specificity value of 91.72% (95% CI:88.77-94.11). A nomogram was developed to simplify the use of the predictive model in clinical practice.

Conclusion: This prognostic model could be of great value in clinical practice, serving as the basis for early personalized analgesic management of elderly patients undergoing hip arthroplasty.