Brain MRI correlates of magnetization transfer imaging metrics in patients with multiple sclerosis

Previous studies suggested that magnetization transfer ratio (MTR) histograms are highly correlated with other magnetic resonance imaging (MRI) measures and can be used as a reliable method for quantifying overall disease burden in multiple sclerosis (MS). However, the relative influence of burden and severity of macroscopic MS lesions and degree of brain atrophy on various MTR histogram parameters has not yet been fully elucidated. Aim of the present study was to investigate which MRI measure best predicts the values of MTR histogram parameters in MS patients. Forty-two MS patients underwent brain dual-echo. T1-weighted and magnetization transfer imaging (MTI) MRI scans. Hyperintense lesion load (LL) on proton density (PD)-weighted and hypointense LL on T1-weighted images were measured using a local thresholding technique. A measure of brain atrophy was derived from T1-weighted images by computing the volume of brain tissue segmented from a slab of five consecutive slices rostral to the velum interpositum. On MTI scans, MTR histogram analysis was performed for the whole brain and average lesion MTR was also calculated. PD-weighted LL, T1-weighted LL and brain volume were significantly correlated with several MTI-derived measures. When a multivariate analysis was performed, brain volume alone significantly predicted the values of all the MTR histogram-derived measures (P values ranged from 0.003 to 0.0002). The ratio between hypointense T1-weighted and hyperintense PD-weighted LL significantly predicted average lesion MTR (P<0.05). Our results confirm that MTR can be used as a reliable method to assess both the overall disease burden and the individual lesion intrinsic nature in MS patients. The significant influence of brain atrophy on MTR histogram parameters supports the concept that this method also provides information on the loss of brain parenchyma in MS.