Objective: The objective of this study is to analyze time to surgery as both a continuous and discrete variable to determine its association with outcomes in cauda equina syndrome (CES).
Methods: Patients at a single center whose medical record allowed precise calculation of time to surgery were included. CES was defined as at least four of the following: bladder dysfunction, saddle anesthesia, lower extremity weakness, lower extremity sensory disturbance, bowel dysfunction, or acute lower back or leg pain. Time to surgery was analyzed as a continuous variable using logistic and ordered logistic regression, and as a discrete variable by comparing patients treated before and after set thresholds.
Results: Forty-five patients were identified. Analysis of time as a continuous variable did not reveal any significant association with outcomes. A parsimonious model with adjustment for age, sex, race, acute onset of CES, saddle anesthesia, motor deficit, and bowel dysfunction at presentation was used to analyze the continuous influence of time to surgery on bladder dysfunction and an aggregate outcome of symptoms. Neither time to surgery nor any of the covariates were significantly associated with either outcome. Discrete analysis of outcomes across thresholds of 12, 24, 36, 48, 60, and 72 hours did not reveal prognostic time points.
Conclusion: In this single-center CES series, time to surgery did not have a convincing continuous or discrete relationship with outcome. Future prospective studies are needed to determine the best timing for surgery in patients with CES.
Keywords: 24 hour; 48 hour; Cauda equina; Cauda equina syndrome; Spine; Time to surgery.
Copyright © 2016 Elsevier Inc. All rights reserved.