Craniometric parameters for the evaluation of platybasia and basilar invagination on magnetic resonance imaging: a reproducibility study

OBJECTIVE: The present study aims to perform a reproducibility study of the clivus-canal angle (CCA), Welcker’s basal angle (WBA), and the distance from the odontoid process to Chamberlain’s line (DOCL) on magnetic resonance imaging (MRI). MATERIALS AND METHODS: Two medical students and two radiologists respectively evaluated 100 and 50 consecutive MRI scans of adult skulls, selected randomly. Each examiner, working independently and blinded to the previous results, performed readings for each patient on two different occasions. Measurements were performed in T1-weighted sequences acquired in the midsagittal plane. The levels of intraobserver reproducibility and interobserver agreement were evaluated by calculating the intraclass correlation coefficients (ICCs) and the corresponding 95% confidence intervals. RESULTS: The mean values obtained by the examiners were 150º for the CCA, 130º for the WBA, and 2.5 mm for the DOCL. The ICC for interobserver agreement was 0.980, 0.935, and 0.967, for the CCA, WBA, and DOCL, respectively, for the students, compared with 0.977, 0.941, and 0.982, respectively, for the radiologists, and 0.980, 0.992, and 0.990, respectively, for all of the examiners together. In the analysis of intraobserver agreement, the ICC ranged from 0.929 to 0.959 for the CCA, from 0.918 to 0.964 for the WBA, and from 0.918 to 0.981 for the DOCL. CONCLUSION: The measurement of the CCA, WBA, and DOCL appears to show excellent intraobserver reproducibility and interobserver agreement on MRI.