OBJECT
The purpose of this study was to characterize changes in American Spinal Injury Association motor score (AMS) in the 1st week after traumatic central cord syndrome (CCS) to identify predictors of improved early outcome in patients treated with early versus delayed surgical intervention.
METHODS
All patients presenting to a regional spinal cord injury center between January 2004 and June 2009 were queried for those with a diagnosis of CCS. Patients treated conservatively were excluded. A prospectively maintained spinal cord injury database was used to track AMS throughout each patient’s hospitalization. Hospital records provided information regarding demographics, presenting neurological examination, imaging findings, comorbidities, timing and nature of surgical procedures, and length of stay (LOS) in the hospital and intensive care unit (ICU). Patients were separated into those who underwent early surgery, within 1 day of presentation (early group), and those who underwent surgery on a delayed basis (delayed group). Differences between groups were analyzed using the Student t-test and chi-square test. Predictors of outcome were identified using correlation analysis and multiple linear regression.
RESULTS
Of 426 patients in the database, 80 (18.8%) were diagnosed as having CCS, and 68 of them ultimately underwent surgical decompression. Nineteen (28%) of 68 patients underwent surgery within 1 day of presentation (early group) while the remaining 49 patients (72%) underwent surgery on a delayed basis (delayed group). The mean age in the early group was significantly younger than that of the delayed group (52 vs 59 years, p = 0.049). Other characteristics were similar between groups including sex, proportion of patients with cord edema on MRI (44% early vs 55% delayed, p = 0.47), and proportion of patients with cervical fracture (26% early vs 28% delayed, p = 0.98). Patients in the early group presented with an AMS of 62.5 versus 70.0 for the delayed group (p = 0.36). No difference in the change in AMS was seen at 7 days between the early group (-2.9 points) and the delayed group (-4.2 points) (p = 0.34). Additionally, the number of patients who had early improvement was similar between the early and delayed groups (50% vs 48%, respectively, p = 0.94). Neither time in the ICU (3.4 vs 3.4 days, p = 0.84) nor the overall LOS (10.5 vs 12.5 days, p = 0.59) was different in the early versus delayed groups, respectively. Correlation analysis and multiple linear regression demonstrated early surgery was not associated with change in AMS or AMS at Day 7. Age was identified as the only significant predictor of change in AMS and had a negative effect (coefficient = -0.34, p = 0.025).
CONCLUSIONS
Early treatment of patients with CCS remains controversial. Although some long-term neurological recovery is expected in patients with a CCS, surgeons and patients should not expect early neurological improvement with or without early operative intervention.