Early Surgical Decompression and Pulmonary Embolism Risk in Traumatic Cervical Spinal Cord Injury: A Propensity Score-Matched National Analysis.

Abstract

Study DesignPropensity score-matched retrospective cohort study.ObjectivesTo determine whether early cervical decompression is associated with in-hospital pulmonary embolism (PE) after traumatic cervical spinal cord injury (SCI).MethodsAdults with traumatic cervical SCI undergoing coded cervical decompression after direct admission were identified in the National Trauma Data Bank, 2019 to 2024. Early decompression was within 24 hours; delayed decompression was after 24 hours. Patients were matched 1:1 using 30 covariates. The primary outcome was detected in-hospital PE. Robustness analyses included sensitivity, subgroup, landmark, negative-control, competing-risk, clustered-error, bootstrap, and McNemar analyses.ResultsAmong 11,530 decompression-coded patients, matching yielded 4,480 balanced pairs. Early decompression was not associated with detected PE (OR 0.97, 95% CI 0.70 to 1.35; p = 0.868; absolute risk difference -0.04%, 95% CI -0.58% to 0.49%). Robustness analyses supported this null finding, including a null negative-control outcome (OR 1.10, 95% CI 0.75 to 1.62). No subgroup interaction was significant after multiplicity correction. In patients with Injury Severity Score ≥25, PE risk was similar (OR 0.88, 95% CI 0.56 to 1.39). Early decompression was associated with higher in-hospital mortality overall (OR 1.54, 95% CI 1.30 to 1.82), creating potential competing-risk bias.ConclusionsEarly cervical decompression after traumatic cervical SCI was not associated with increased detected in-hospital PE. PE risk alone should not justify delaying otherwise indicated decompression, but differential mortality and residual confounding require cautious interpretation and prospective confirmation.