21. Early vs delayed surgery for incomplete cervical spinal cord injury with pre-existing cervical stenosis: the OSCIS randomized clinical trial

Abstract

<h3>BACKGROUND CONTEXT</h3> The timing of surgery for acute traumatic cervical spinal cord injury (SCI) has long been an issue of intense debate. The superiority of early surgical decompression for incomplete cervical SCI without concomitant bone injuries has not been established. <h3>PURPOSE</h3> The aim of this study was to determine whether early surgical decompression within 24 hours results in better motor recovery than delayed surgery following at least two weeks of conservative treatment. <h3>STUDY DESIGN/SETTING</h3> Multicenter randomized controlled trial conducted at 43 tertiary referral hospitals in Japan. <h3>PATIENT SAMPLE</h3> Patients aged 20 to 79 years with motor-incomplete cervical SCI with pre-existing canal stenosis (American Spinal Injury Association [ASIA] Impairment Scale C; without fracture or dislocation) were included. The presence of canal stenosis was verified based on baseline imaging studies, including MRI and CT. <h3>OUTCOME MEASURES</h3> The primary endpoints were improvement in the mean ASIA motor score, total score of the spinal cord independence measure (SCIM), and the proportion of patients able to walk independently at one year post-injury. <h3>METHODS</h3> Participants were allocated (11) by computer-based randomization to urgent surgery within 24 hours after admission or delayed surgery (later than two weeks post-injury). <h3>RESULTS</h3> Among 72 randomized patients, 70 patients (mean [SD] age, 65.1 [9.4] years; 5 women and 65 men) were included in the full analysis population at one year post-injury (37 patients assigned to early surgery and 33 to delayed surgery). We found a consistent, but insignificant, trend favoring early surgery more than delayed surgery (mean change in ASIA motor score, 53.7 vs 48.5, difference, 5.2; 95% confidence interval, -4.2 to 14.5; p=0.27; mean SCIM total score, 77.9 vs 71.3, p=0.34; the proportion of ambulatory patients, 70% vs 61.5%; p=0.51). A mixed-design analysis of variance revealed a significant difference in the ASIA motor scores between the groups (p=0.03). The early surgery group showed greater motor scores than the delayed surgery group at 2, 3 and 6 months post-injury. Complications were common in both groups (death, 3 patients vs 3 patients). <h3>CONCLUSIONS</h3> Among motor-incomplete cervical SCI patients, early surgery produced similar motor regain at one year post-injury as delayed surgery, but showed accelerated recovery within the first six months. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.