37. Early versus late decompression for lumbar spinal nerve injury: a propensity score matched analysis

BACKGROUND CONTEXT

Traumatic lumbar spinal injury often necessitates surgical decompression of the thecal sac, nerve roots, or peripheral nerves. While there is some evidence in the literature to suggest a benefit to early surgery within 24 hours, there has yet to be a consensus and society recommendations for the timing of decompressive surgery for lumbar spine injuries.

PURPOSE

To evaluate the effect of early versus late decompressive surgery on inpatient outcomes at a nationwide level in the United States.

STUDY DESIGN/SETTING

Retrospective cohort database study.

PATIENT SAMPLE

Patients from the American College of Surgeons National Trauma Data Bank (NTDB) from 2017-2021.

OUTCOME MEASURES

The primary outcome measures are all-cause mortality and overall hospital length of stay (LOS). Secondary outcome measures entail hospital complications such as pressure ulcers and acute kidney injury (AKI) and discharge disposition such as routine discharge to home and discharge to skilled nursing.

METHODS

The NTDB was queried from 2017-2021 for all patients with a lumbar spinal cord or nerve injury matching the ICD-10-CM code S34. Patients younger than 18 years, who did not undergo surgical decompression, or who were missing outcome data were excluded. Patients were divided in the early surgery group if they underwent decompression within 24 hours and in the late surgery group if they underwent decompression at or after 24 hours. Propensity score matching was performed using the k-nearest neighbors algorithm based on patient age, sex, race, ethnicity, comorbidities, Glasgow Coma Scale, and insurance type. Equal post-match balance was evaluated using a standard mean difference threshold of 0.1. Early and late patients were compared using Student's t-tests and Pearson's chi-square tests.

RESULTS

A total of 1499 patients matching the inclusion and exclusion criteria were identified, of which 905 had early surgery and 591 had late surgery. Following propensity score matching, 591 matching patients in the late surgery group were identified. Post-match, the early surgery group had a lower mortality rate (0.17% vs 1.69%, p<0.01) and shorter overall length of stay (2.47 vs 3.79 days, p<0.01), as well as lower rates of unplanned intubation (1.02% vs 2.88%, p=0.02), AKI (0.17% vs 1.35%, p=0.02, stroke (0% vs 0.68%, p=0.045), pressure ulcer (0.68% vs 2.2%, p=0.03), unplanned intensive care unit admission (1.02% vs 4.06%, p<0.01), and ventilator-associated pneumonia (0.34% vs 1.69%, p=0.02) compared to the late surgery group. Additionally, the early surgery group was more likely to be discharged to inpatient rehabilitation (53.64% vs 40.61%, p<0.01) but less likely to be discharged routinely to home (26.73% vs 34.52%, p<0.01) or a skilled nursing facility (4.74% vs 7.61%, p=0.04).

CONCLUSIONS

Early surgery within 24 hours for traumatic lumbar spinal cord and nerve injury is associated with lower mortality, shorter LOS, and fewer short-term complications. Additionally, it is associated with a greater rate of discharge to rehabilitation, suggesting a greater potential for longer-term functional recovery. This study is one of the largest on these topics and provides additional evidence towards the benefit of early decompression. Supplementing the existing literature from Fehlings, Badhiwala, and others, this work can aid in the development of additional guidelines for the timing of decompressive surgery for lumbar spinal nerve injuries.

FDA Device/Drug Status

This abstract does not discuss or include any applicable devices or drugs.