Yan Li, Hong Hong, Jia-Hui Ma, Dong-Xin Wang
BACKGROUND: Heterogeneity of postoperative pain trajectories and their impact on clinical outcomes in older orthopedic patients remain insufficiently characterized. This study aimed to identify distinct postoperative pain trajectories, determine their potential predictors, and evaluate their impacts on postoperative outcomes. METHODS: This secondary analysis included 710 patients age 65 to 90 yr who underwent orthopedic surgery and received self-controlled analgesia. After surgery, pain intensity was assessed twice daily for 5 days using an 11-point numeric rating scale; occurrence of complications was followed up until 30 days; quality of life was evaluated at 30 days. Growth mixture modeling was used to identify pain trajectories according to pain intensity with movement. Multinomial logistic regression analysis was used to identify predictors of trajectory membership. Multivariable regression was used to evaluate associations between trajectory membership and postoperative outcomes. RESULTS: Four distinct pain trajectories were identified: the slow decrease pattern (57.9%), the rapid decrease pattern (23.4%), the increase-decrease pattern (10.8%), and the slow increase pattern (7.9%). Taking the slow decrease pattern as a reference, preoperative analgesic use was associated with higher risks of increase-decrease (adjusted odds ratio [OR], 3.76; 95% CI, 1.53 to 9.23; P = 0.004) and slow increase trajectories (adjusted OR, 3.89; 95% CI, 1.52 to 9.98; P = 0.005), whereas using dexmedetomidine in self-controlled analgesia was associated with lower risks of rapid decrease (adjusted OR, 0.47; 95% CI, 0.32 to 0.70; P < 0.001) and increase-decrease trajectories (adjusted OR, 0.15; 95% CI, 0.08 to 0.28; P < 0.001). When compared with the slow decrease pattern, patients with a rapid decrease trajectory required more opioids and developed more complications; patients with an increase-decrease trajectory required more opioids and had a lower physical domain quality of life score at 30 days; patients with a slow increase trajectory stayed longer in the hospital after surgery. CONCLUSIONS: Postoperative pain in older orthopedic patients followed four distinct trajectories that were influenced by baseline and perioperative factors. Specific pain trajectories were associated with adverse postoperative outcomes.