Prediction of Postoperative Ambulatory Function Using Nomogram in Patients with Metastatic Epidural Spinal Cord Compression
Abstract
Background/Objectives: This study was conducted to develop and internally validate a nomogram that predicts postoperative ambulatory function in patients with metastatic epidural spinal cord compression (MESCC) undergoing decompressive surgery. Methods: Of 352 patients who underwent surgery for metastatic spinal cancer at a single tertiary center between 2014 and 2021, 97 with MESCC (Bilsky grade 2–3) and preoperative neurological deficits (ASIA grade D or worse) were included in this retrospective study. Binary logistic regression was used to identify candidate predictors of 3-month postoperative ambulatory status in the full cohort (p < 0.10 in univariable analysis), which were then entered into a multivariable logistic regression model and incorporated into a nomogram using the rms package in R. Model discrimination and calibration were internally validated using Harrell’s bootstrap optimism-correction method with 1000 resamples. Results: Postoperative ambulation was regained in 24 patients (24.7%). Four variables met the prespecified threshold (p < 0.10) for inclusion in the nomogram: preoperative motor grade (OR 2.48; 95% CI 1.41–4.34; p = 0.002), symptom onset–to–operation interval (OR 0.81; 95% CI 0.62–1.04; p = 0.098), preoperative KPS (OR 1.04; 95% CI 1.00–1.07; p = 0.070), and age (OR 0.97; 95% CI 0.93–1.01; p = 0.087); in the multivariable model, only preoperative motor grade and symptom onset-to-operation interval were independently associated with ambulatory recovery (p < 0.05). The nomogram achieved an apparent AUC of 0.827 (optimism-corrected, 0.800) and acceptable-to-good calibration (optimism-corrected calibration slope, 0.848; intercept, −0.088). At the Youden-optimal threshold, apparent (optimism-corrected) sensitivity, specificity, and F1 score were 79.2% (74.5%), 78.1% (76.7%), and 64.4% (60.6%), respectively. Conclusions: The nomogram demonstrated acceptable discrimination and good calibration for predicting postoperative ambulation in patients with MESCC. This tool may support individualized surgical decision-making and patient counseling. External validation is required before routine clinical use.