Presurgical risk model for chronic postsurgical pain based on 6 clinical predictors: a prospective external validation
Antônio Montes, G Roca, Jordi Cantillo, Sergi Sabaté, for the GENDOLCAT Study Group
Abstract
No externally validated presurgical risk score for chronic postsurgical pain (CPSP) is currently available. We tested the generalizability of a six-factor risk model for CPSP developed from a prospective cohort of 2929 patients in 4 surgical settings. Seventeen centers enrolled 1225 patients scheduled for inguinal hernia repair, hysterectomy (vaginal or abdominal), or thoracotomy. The 6 clinical predictors were surgical procedure, younger age, physical health (Short Form Health Survey-12), mental health (Short Form Health Survey-12), preoperative pain in the surgical field, and preoperative pain in another area. Chronic postsurgical pain was confirmed by physical examination at 4 months. The model's discrimination (c-statistic), calibration, and diagnostic accuracy (sensitivity, specificity, and positive and negative likelihood ratios) were calculated to assess geographic and temporal transportability in the full cohort and 2 subsamples (historical and new centers). The full data set after exclusions and losses included 1088 patients; 20.6% had developed CPSP at 4 months. The c-statistics (95% confidence interval) were similar in the full validation sample and the 2 subsamples: 0.69 (0.65-0.73), 0.69 (0.63-0.74), and 0.68 (0.63-0.74), respectively. Calibration was good (slope b and intercept close to 1 and 0, respectively, and nonsignificance in the Hosmer-Lemeshow goodness-of-fit test). The validated model based on 6 clinical factors reliably identifies risk for CPSP risk in about 70% of patients undergoing the surgeries studied, allowing surgeons and anesthesiologists to plan and initiate risk-reduction strategies in routine practice and researchers to screen for risk when randomizing patients in trials.