The review found that MRI can be used to rule out cervical spine injury in blunt trauma patients. The review methods were poorly reported, analysis was weak and available data were limited. The conclusions of the review reflect the data presented, but should be viewed with caution given the paucity of these data.

To assess the performance of magnetic resonance imaging (MRI) in ruling out cervical spine injury in patients with suspicious or potentially missed ("unevaluatable") blunt trauma.

MEDLINE, CINAHL and the Cochrane CENTRAL Register were searched to August 2005 for English language publications; search terms were reported. The bibliographies of relevant review articles and clinical guidelines were searched for additional studies.

Prospective or retrospective studies with a minimum of 30 trauma patients with clinically suspicious or potentially missed ("unevaluatable") blunt trauma cervical spines were eligible for inclusion; patients with obvious spinal cord injury were excluded. Included studies were required to conduct MRI within 72 hours of injury and to report sufficient data for the construction of 2 x 2 contingency tables comparing the results of MRI with the reference standard of clinical follow-up. A true positive (TP) MRI was defined as an abnormal MRI scan that resulted in clinical management (immobilization or surgical stabilization), a false positive (FP) MRI was defined as an abnormal MRI where no further management was instituted and physical examination at follow-up or flexion radiographs were normal, a true negative (TN) MRI was defined as a normal MRI which allowed discontinuation of immobilization and where follow-up examination was normal, and a false negative (FN) MRI was defined as a normal MRI with follow-up evidence of injury. A three member panel assessed abstracts and full papers for inclusion.

The authors do not state that they assessed study validity. They describe the studies included, based upon the criteria stated above, as 'level I' studies.

The authors did not state how data were extracted for the review, or how many reviewers were involved in data extraction.

Simple pooled estimates of sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were obtained using fixed and random-effects models and weighting by inverse variance. Between study heterogeneity was assessed using the Q statistic and random effects models were used where significant heterogeneity was present.

Five studies, with a total of 464 participants were included in the review.

The pooled estimate of sensitivity was 97.2% (95% CI: 89.5, 99.3); no statistically significant between study heterogeneity was identified (p = 0.756). The pooled estimate for specificity was 98.5% (95% CI: 91.8, 99.7); significant between study heterogeneity was present (p = 0.023).

There were 97 TP MRI scans (21%) that had been missed by three-view plain radiograph series or CT scans.

A normal MRI scan can be used to rule out cervical spine injury and that MRI detects injuries missed by other imaging techniques. However, the number of false positive MRIs cannot be accurately determined.

The authors presented a clear research question, which was defined by appropriate inclusion criteria. A number of sources were used to identify potentially relevant studies, but the restriction of searches to English language publications may have resulted in a loss of data. Description of the review process was limited and it was unclear whether measures were take to limit error and bias. Though the authors stated that they included only 'level I' studies, the criteria used to define these were limited and omitted many factors important to the methodological quality of a diagnostic accuracy study. Available data were sparse (there were no false negative MRIs and only 3 false positives) and may have been better presented as a narrative summary; the generation of simple pooled estimates was uninformative. The authors' conclusion that MRI can be used to rule out cervical spine injury is a reasonable interpretation of the data presented, but must be subject to considerable uncertainty given the limited data upon which it is based.

Practice: The authors state that a normal MRI scan can be used to rule out cervical spine injury. However, the logistic problems of MRI scans have to be considered when using MRI for screening patients.

Research: The authors state that further prospective research is needed to evaluate the relationship of MRI findings with significance of soft tissue injury and reduce the amount of over treatment consequent upon abnormal MRI findings.

Not stated

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.