Interobserver and intraobserver reliability of maximum canal compromise and spinal cord compression for evaluation of acute traumatic cervical spinal cord injury.

STUDY

DESIGN

Prospective, blinded validation study of an objective, quantitative measure to assess maximum canal compromise (MCC) and maximum spinal cord compression (MSCC) in individuals with acute cervical spinal cord injury (SCI).

OBJECTIVE

To examine the intraobserver and interobserver reliability of MCC and MSCC in individuals with acute traumatic cervical SCI.

SUMMARY OF BACKGROUND DATA: To date, few quantitative reliable radiologic methods for assessing the extent of spinal cord compression in the setting of acute SCI have been reported. MCC and MSCC, as assessed on mid-sagittal CT and T2-weighted MR images, respectively, appear to have potential clinical and prognostic value. To date, the validation of these assessment tools has been limited to a small number of observers at a single institution. However, to date no study has focused on the reliability of these radiologic parameters among a large cohort of spine surgeons from North America and abroad. This type of validation is critical to allow the broader use of these outcome measures in research studies and in clinical practice.

METHODS

Mid-sagittal MRI and CT images of cervical spine were selected from 10 individuals with acute traumatic cervical SCI. A total of 28 spine surgeons independently estimated CT MCC, T1-weighted MRI MCC, and T2-weighted MRI MSCC on two occasions using a calibrated ruler. In the first round of measurements, the observers estimated the radiologic parameters using only written instructions. The second measurement set was obtained after an interactive teaching session on the methodology. The order of the images was altered for the second set of measurements.

RESULTS

Analysis using parametric and nonparametric statistics indicated high intraobserver reliability for CT MCC, T1-weighted MRI MCC, and T2-weighted MSCC with interclass correlation coefficients (ICCs) of 0.92, 0.95, and 0.97, respectively. The interobserver reliability for all three radiologic parameters was considered moderate with ICCs ranging from 0.35 to 0.56.

CONCLUSION

Our results indicate that the intraobserver reliability for the MCC and MSCC was high. Although the interobserver reliability for all three radiologic parameters in the present study was below 0.75, the observed differences were small and largely accounted for by the limitations in the precision of the calibrated ruler. For cases with minimal cord compression, the measurement of canal stenosis (MCC) proved more accurate. In contrast, in cases with severe cord compression, the assessment of MSCC was more accurate. It is anticipated that the use of digital imaging technologies will further enhance the precision of these outcome measures.