Comparison of qualitative and quantitative analyses of age-related changes in clivus bone marrow on MR Imaging

Magnetic resonance (MR) imaging is capable of detecting a wide spectrum of diseases involving clivus bone marrow. To evaluate any pathology in this region, the normal MR signal intensity patterns according to age should be known. Assessment of clivus bone marrow is usually and more efficiently made visually (qualitatively) on routine MR imaging To compare the qualitative and quantitative analyses of age-related changes in clivus bone marrow on MR imaging, midsagittal T1-weighted MR images of the clivus bone marrow were prospectively evaluated in 201 subjects. MR signal intensity patterns of clivus bone marrow were qualitatively graded from I to III according to the proportions of low and high signal intensity areas within the clivus (Grade I: predominantly hypointense, Grade III: predominantly hyperintense). Signal intensity measurements were also made from the clivus, pons, and the cerebrospinal fluid within the fourth ventricle. Grade I pattern was observed in 92% of the subjects in the first decade, whereas Grade III pattern was not seen in this decade; Grade I pattern was absent in all but one of the subjects over the age of 50. Mean values of clivus/pons and clivus/CSF signal intensity measurements gradually increased with age in both males and females. Comparison of these ratios in each gender showed statistically significant higher values for males (P < 0.05). Comparison of the qualitative and quantitative results showed that calculated mean values of intensity ratios for each grade were different from each other; however, there was an overlap of the ranges of signal intensity ratios for Grades II and III. Visual evaluation of the clivus bone marrow MR signal is validated with the quantitative assessment results. Statistically significant higher intensity ratios in males may reflect a difference between gender in the bone mineral content under the influence of sex hormones. (C) 2003 Wiley-Liss, Inc.