Introduction: Magnetization transfer ratio (MTR) histogram analysis can be used as a method for quantifying overall disease burden in MS. We studied correlations between MTR histogram and clinical parameters in MS subgroups. Contrary to earlier studies we placed special emphasis on the lower MTR range, to explore the effect of partial volume averaging effects with CSF. Methods: Seventy-nine patients with MS [26 primary progressive (PP), and 53 'relapse-onset', including 26 secondary progressive (SP)], and 23 healthy individuals were studied. MR imaging included 3 mm 2D gradient-echo images with and without an off-resonance MT pulse. According to the visually determined cut-off, histogram parameters were classified as parenchymal or CSF-related variables. Clinical measurements included the Expanded Disability Status Scale (EDSS) as a measure of global impairment/disability and the Paced Auditory Serial Addition Test (PASAT) as a measure of cognition. Results: SP MS patients differed from the other subgroups on many MTR variables, originating from both the lower and the higher MTR range. CSF-related low MTRs were clearly over-represented in SP patients, and showed a significant distinction between the SP and PP MS group. In the total group, as well as in the relapse-onset patients, significant correlations were found between MTR parameters and clinical parameters. No associations were found in the PP group. Conclusion: This explorative study suggests that MTR histogram analysis can distinguish between MS patients and controls, and best identifies the SP phenotype, partly as a result of increased CSF volume (atrophy). In addition, we show that MTR histogram analysis gives information about the level of impairment and disability in patients with a 'relapse-onset' course of MS, and therefore provides a useful tool to monitor the evolution of the disease in these patients.