Characterizing post-operative ICU admission after cervical spondylotic myelopathy surgery

Retrospective cohort OBJECTIVE: To characterize variables associated with post-operative intensive care unit (ICU) admission following cervical spondylotic myelopathy (CSM) surgery and determine if ICU stay is associated with increased length of stay (LOS).

Routine post-operative ICU stay for mean arterial pressure monitoring is an established practice after acute spinal cord injury. However, there is no formal recommendation for CSM patients post-operatively. Variability exists in ICU utilization after CSM surgery, with some surgeons routinely sending patients to the ICU while others do not.

Patients who underwent primary, single-approach CSM surgery at a single institution over a two-year span were divided into ICU and non-ICU cohorts based on immediate post-operative destination. Primary outcome was LOS. Secondary outcomes included discharge disposition, complications, clinical status, and reoperation within three years. Propensity score matching (PSM) created matched cohorts of ICU and non-ICU patients. Outcomes were assessed using Mann-Whitney, McNemar, or 2-sided t-tests.

452 patients were included in initial analysis. Variables that significantly predicted ICU disposition include age, increased number of stenotic levels, adapted spinal canal occupation ratio, Medicare insurance, management by orthopaedic service, posterior surgical approach, increased surgery duration, and increased estimated blood loss. After PSM to control these variables, non-ICU patients had shorter LOS (2.14 vs 4.14, p < 0.001) and higher rate of discharge to home (83.5 % vs 55.3 %, p = 0.0009) with no difference in neurologic status change or complications at 6 weeks.

In CSM patients, routine post-operative ICU admission is associated with longer LOS and discharge to rehabilitation facilities without decreasing post-operative complications compared to patients routinely admitted to the floor.