STUDY DESIGN: This was a radiologic validity and reliability study. OBJECTIVE: We aimed to assess the validity and reliability of measurements from standing lateral cervical radiography (XR), reconstructed midsagittal images of supine cervical computerized tomography (CT) scans, and supine magnetic resonance imaging (MRI) for evaluating C2-C7 distance (C27 SVA), C2-C7 Cobb angle (CL), T1 slope (T1S), thoracic inlet angle (TIA), and neck tilt (NT). SUMMARY OF BACKGROUND DATA: XR is the criterion standard imaging modality for measuring cervical sagittal alignment parameters. However, overlapping bony structures and soft tissue often make the upper end of the manubrium and cervicothoracic junction indiscernible. CT and MRI can overcome this limitation, but their reliability and validity have not been fully elucidated. METHODS: Fifty sets of three examinations from our database have been randomly selected. Three experienced spinal surgeons independently measured C27 SVA, CL, T1S, TIA, and NT. Paired t test and Pearson correlation were used to analyze the validity of CT and MRI in comparison with that of XR. Interobserver and intraobserver reliability were assessed by using the intraclass correlation coefficient (ICC). RESULTS: The paired t test confirmed the similarities of all CT (except C27 SVA) and MR parameters (except C27 SVA and CL) to the XR parameters (all P > 0.05). All parameters except C27 SVA were correlated (Pearson correlation coefficient, 0.523-0.913). Correlation was highest between CT and MRI and lowest between CT and XR. All three imaging modalities had excellent intraobserver and interobserver reliability (ICC range: 0.770-0.999). Intraobserver and interobserver reliability were highest with MRI and lowest with XR. CONCLUSION: CT and especially MRI were more reliable than XR for evaluating cervical sagittal alignment parameters. Considering the reliability and radiation exposure, MRI is a good alternative to XR for measuring cervical sagittal alignment parameters, especially thoracic inlet alignment. LEVEL OF EVIDENCE: 4.
STUDY DESIGN: This was a radiologic validity and reliability study. OBJECTIVE: We aimed to assess the validity and reliability of measurements from standing lateral cervical radiography (XR), reconstructed midsagittal images of supine cervical computerized tomography (CT) scans, and supine magnetic resonance imaging (MRI) for evaluating C2-C7 distance (C27 SVA), C2-C7 Cobb angle (CL), T1 slope (T1S), thoracic inlet angle (TIA), and neck tilt (NT). SUMMARY OF BACKGROUND DATA: XR is the criterion standard imaging modality for measuring cervical sagittal alignment parameters. However, overlapping bony structures and soft tissue often make the upper end of the manubrium and cervicothoracic junction indiscernible. CT and MRI can overcome this limitation, but their reliability and validity have not been fully elucidated. METHODS: Fifty sets of three examinations from our database have been randomly selected. Three experienced spinal surgeons independently measured C27 SVA, CL, T1S, TIA, and NT. Paired t test and Pearson correlation were used to analyze the validity of CT and MRI in comparison with that of XR. Interobserver and intraobserver reliability were assessed by using the intraclass correlation coefficient (ICC). RESULTS: The paired t test confirmed the similarities of all CT (except C27 SVA) and MR parameters (except C27 SVA and CL) to the XR parameters (all P > 0.05). All parameters except C27 SVA were correlated (Pearson correlation coefficient, 0.523-0.913). Correlation was highest between CT and MRI and lowest between CT and XR. All three imaging modalities had excellent intraobserver and interobserver reliability (ICC range: 0.770-0.999). Intraobserver and interobserver reliability were highest with MRI and lowest with XR. CONCLUSION: CT and especially MRI were more reliable than XR for evaluating cervical sagittal alignment parameters. Considering the reliability and radiation exposure, MRI is a good alternative to XR for measuring cervical sagittal alignment parameters, especially thoracic inlet alignment. LEVEL OF EVIDENCE: 4.