Intraoperative Bulbocavernosus Reflex Monitoring for Predicting Bowel Function After Resection of Conus Medullaris and Cauda Equina Tumors.

Abstract

Study DesignRetrospective cohort study.ObjectiveTo evaluate the predictive value of intraoperative bulbocavernosus reflex (BCR) monitoring for postoperative bowel dysfunction (PBD) following conus medullaris and cauda equina tumor (CMCET) resection.MethodsA total of 118 patients were analyzed. Neurogenic Bowel Dysfunction Scores >6 at 1 month and 6 months postoperatively defined short- and long-term PBD. The amplitude reduction ratio (ARR) was calculated from bilateral BCR waveforms using the worst-side ARR for analysis. Multivariate logistic regression and ROC curve analysis determined independent predictors and optimal thresholds. Stratified 10-fold cross-validation confirmed threshold stability.ResultsPatients with short-term or long-term PBD had significantly higher ARRs than those without (<i>P</i> < 0.001 for both comparisons, Mann-Whitney U test). Furthermore, ARR emerged as the only independent predictor for either short-term (OR 1.287, 95% CI 1.018-1.627, <i>P</i> = 0.035) or long-term PBD (OR 1.153, 95% CI 1.055-1.261, <i>P</i> = 0.002). The threshold values of ARR for predicting short-term and long-term PBD were 59.6% (AUC = 0.991, 95% CI 0.974-1.000, <i>P</i> < 0.001) and 63.6% (AUC = 0.988, 95% CI 0.971-1.000, <i>P</i> < 0.001), respectively. Cross-validation confirmed these thresholds, with mean AUC values of 0.992 and 0.993 for short-term and long-term PBD, respectively.ConclusionsThe ARR of the intraoperative BCR waveform showed a high predictive value for PBD in patients with CMCET. We propose a 60% ARR reduction on either side of the BCR waveforms as an exploratory, hypothesis-generating early warning criterion for BCR monitoring during CMCET surgery, pending external validation in independent prospective cohorts before clinical implementation.