Diagnostic abilities of magnetic resonance imaging in traumatic injury to the posterior ligamentous complex: the effect of years in training

Background context: The integrity of the posterior ligamentous complex (PLC) has been proposed to be an integral aspect in the treatment algorithm for spinal trauma. Magnetic resonance imaging (MRI) has been reported as the ideal tool to determine the integrity of the PLC. The ability to assess disruption of the PLC by reviewers of differing levels of training has not been described. In addition, the MRI sequence most suggestive of injury for each component of the PLC has not been clearly determined.

Purpose: This study was designed to determine the ability of reviewers with differing levels of training (fellowship-trained spine surgeon, fellowship-trained musculoskeletal radiologist, senior orthopedic surgery resident, and junior orthopedic surgery resident) to accurately interpret the results of MRI. The secondary purpose was to evaluate the MRI sequence that was most indicative of injury to the components of the PLC.

Study design: This is a prospective radiological study comparing reviewers of MRI to determine integrity of the PLC components using intraoperative notation as the gold standard for integrity.

Patient sample: Forty-five consecutive spinal trauma patients who underwent operative fixation after obtaining MRI.

Outcome measures: No patient outcome measures were used.

Methods: The sensitivity, specificity, and accuracy for each MRI reviewer in regard to MRI integrity were compared with the gold standard of intraoperative observation. In addition, the MRI sequence most suggestive of integrity of the PLC was noted by each reviewer for each component of the PLC.

Results: Forty-five patients (29 men and 16 women) with traumatic spine injuries were enrolled in the study. The sensitivity and accuracy of the surgeon were 0.83 (0.66, 0.92) and 0.81 (0.70, 0.88), respectively. The sensitivity and accuracy of the attending spine surgeon were not statistically significantly different from the other reviewers (p value=.2317 and .2582). However, the specificity of the surgeon was statistically significantly higher than that of the other reviewers (p=.0043). In the cervical, thoracic, and lumbar spine, the reviewers reached a 93% agreement that the sagittal short-tau inversion recovery (STIR) sequences were most helpful in visualizing injury to the supraspinous ligament (SSL), interspinous ligament (ISL), ligamentum flavum (LF), and the cervical facet capsules. The reviewers attained a 95% agreement that visualization of injury to the lumbar facet capsules is most optimal in the T2 axial sequences.

Conclusions: The interpretation of traumatic MRI is very sensitive and accurate regardless of years of training of the observer. The attending-level spine surgeon was statistically more specific in the evaluation of injury MRIs. The fluid-weighted STIR sagittal sequences are most useful in determining injury to the SSL, ISL, LF, and cervical facets capsules. Lumbar facet capsules are best evaluated with axial T2 MRI. The evaluation of the PLC on MRI can be accurately and efficiently interpreted by physicians at multiple levels of training, thus providing a key imaging modality in determining stability and need for stabilization.