Surgery Open Science · Oct 19, 2026
Preoperative predictors of early patient-reported recovery trajectories after major surgery: a secondary analysis of a randomized controlled trial
By van der Velde M, de Graaf JM, Geleijn E et al.
OBJECTIVE: To examine recovery trajectories of Physical Functioning (PF) and the Ability to Participate in Social Roles and Activities (APS) after major surgery, and identify preoperative predictors. BACKGROUND: Recovery after major surgery is typically evaluated at isolated time points, limiting insight into recovery dynamics. Understanding recovery trajectories and their determinants may inform prognostic model development. METHODS: This exploratory secondary analysis used data from a multicenter randomized controlled trial including adults undergoing major elective surgery. Patient-Reported Outcomes Measurement Information System (PROMIS) Computer Adaptive Test PF and APS were assessed preoperatively and at 1, 3, 6, and 12 weeks post discharge. Recovery trajectories were modeled using linear mixed effects models with random intercepts and slopes. Preoperative characteristics were examined using interaction terms for time and time². RESULTS: Among 369 patients, PF and APS followed curvilinear patterns. For PF, older age (β = -0.04, 95% CI [-0.06 to -0.02]), unintentional weight loss and higher comorbidity were associated with slower recovery. Higher baseline PF predicted with faster recovery. Differences exceeding the Minimal Important Change (MIC) were observed for age and baseline PF.For APS, older age was also associated with slower recovery (β = -0.04, 95% CI [-0.08 to -0.004]), and intermediate education (vs low) with differences in recovery curvature. Higher baseline APS predicted higher postoperative scores. MIC-exceeding differences were observed for baseline APS and age. CONCLUSIONS: Early postoperative recovery is curvilinear. Multiple preoperative factors were associated with recovery trajectories, supporting their role as candidate predictors for future trajectory-based prognostic models. TRIAL REGISTRATION: This study was approved by the Medical Ethical Committee of Amsterdam University Medical Center, location VUmc, under registration number NL61503.029.18.
doi.org/10.1016/j.sopen.2026.10.002Open accessPubMed 42990824Populations