Adult patients who received computed tomography (CT) alone or CT-magnetic resonance imaging (MRI) for the evaluation of cervical spine injury.Objective.
To evaluate the utility of CT-MRI in the diagnosis of cervical spine injury using propensity-matched techniques.Summary of Background Data.
The optimal evaluation (CT alone vs. CT and MRI) for patients with suspected cervical spine injury in the setting of blunt trauma remains controversial.Methods.
The primary outcome was the identification of a cervical spine injury, with decision for surgery and change in management considered secondarily. A propensity score was developed based on the likelihood of receiving evaluation with CT-MRI, and this score was used to balance the cohorts and develop two groups of patients around whom there was a degree of clinical equipoise in terms of the imaging protocol. Logistic regression was used to evaluate for significant differences in injury detection in patients evaluated with CT alone as compared to those receiving CT-MRI.Results.
Between 2007 and 2014, 8060 patients were evaluated using CT and 693 with CT-MRI. Following propensity-score matching, each cohort contained 668 patients. There were no significant differences between the two groups in baseline characteristics. The odds of identifying a cervical spine injury were significantly higher in the CT-MRI group, even after adjusting for prior injury recognition on CT (odds ratios 2.6; 95% confidence interval 1.7–4.0; P < 0.001). However, only 53/668 patients (8%) in the CT-MRI group had injuries identified on MRI not previously recognized by CT. Only a minority of these patients (n = 5/668, 1%) necessitated surgical intervention.Conclusion.
In this propensity-matched cohort, the addition of MRI to CT alone identified missed injuries at a rate of 8%. Only a minority of these were serious enough to warrant surgery. This speaks against the standard addition of MRI to CT-alone protocols in cervical spine evaluation after trauma.Conclusion.
Level of Evidence: 3