Surgery in Acute Metastatic Spinal Cord Compression: Timing and Functional Outcome

Abstract

Background: Patients with metastatic spinal cord compression (MSCC) may experience long-term functional impairment. It has been established that surgical decompression improves neurological outcomes, but the effect of early surgery remains uncertain. Our objective was to evaluate the impact of early versus late surgery for acute MSCC due to spinal metastases (SM). Methods: We retrospectively reviewed a consecutive cohort of all patients undergoing surgery for SMs at our institution. We determined the prevalence of acute MSCC; the time between acute neurological deterioration as well as between admission and surgery (standard procedure: decompression and instrumentation); and neurological impairment graded by the ASIA scale upon presentation and discharge. Results: We screened 693 patients with surgery for spinal metastasis; 140 patients (21.7%) had acute MSCC, defined as neurological impairment corresponding to ASIA grade D or lower, acquired within 72 h before admission. Non-MSCC patients had surgery for SM-related cauda equina syndrome, radiculopathy and/or spinal instability. Most common locations of the SM in acute MSCC were the thoracic (77.9%) and cervical (10.7%) spine. Per standard of care, acute MSCC patients underwent surgery including decompression and instrumentation, and the median time from admission to surgery was 16 h (interquartile range 10–22 h). Within the group of patients with acute MSCC, those who underwent early surgery (i.e., before the median 16 h) had a significantly higher rate of ASIA improvement by at least one grade at discharge (26.5%) compared to those who had late surgery after 16 h (10.1%; p = 0.024). Except for a significantly higher sepsis rate in the late surgery group, complication rates did not differ between the late and early surgery subgroups. Conclusions: We report data on the largest cohort of patients with MSCC to date. Early surgery is pivotal in acute MSCC, substantially increasing the chance for neurological improvement without increasing complication rates. We found no significant impact when surgery was performed later than 24 h after admission. These findings will provide the framework for a much-needed prospective study. Until then, the treatment strategy should entail the earliest possible surgical intervention.