The Utility of Cervical Spine MRI in Non-Examinable or NEXUS-Positive Patients With Suspected Blunt Cervical Spine Trauma

Abstract

Background

While numerous studies regarding the need for magnetic resonance imaging (MRI) for evaluation of suspected blunt cervical spine injury (BCSI) exist, the resulting recommendations are often conflicting and are less reliably applicable to non-examinable or National Emergency X-radiography Utilization Study (NEXUS)-positive patients. This study sought to identify the utility of MRI in characterizing BCSI in such patients who had already undergone computed tomography (CT) imaging of the cervical spine.

Methods

Records from 402 unique patients presenting to a Level 1 trauma center were analyzed. Incidence of positive MRI in the setting of negative CT, unstable BCSI on either modality, need for surgical intervention, time in a cervical collar, and hospital readmission rates were calculated.

Results

Non-examinable or NEXUS-positive patients with BCSI identified on both CT and MRI were less likely to have a stable BCSI compared to CT-positive alone (53% vs 88%, P = 0.001). Out of 189 CT-negative patients, 53 (28%) were found to have BCSI on MRI, with 13 (6.8% overall) requiring operative intervention. Out of 100 BCSIs read as “stable” on CT, 28 (23.1%) were deemed “unstable” on subsequent MRI. Patients with negative MRI findings spent less time in a cervical collar than patients with positive findings (median 2 days vs 57 days, P < 0.001) and had lower 180-day readmission rates (12 patients [7.5%] vs 35 [15%], P = 0.031).

Conclusion

While CT remains vital for diagnosing BCSI, non-examinable or NEXUS-positive patients with negative CT should undergo confirmatory MRI prior to cervical collar removal.