Inflammation plays a central part in the pathogenesis of multiple sclerosis. However, current surrogate magnetic resonance (MR) and immunological markers of inflammation are weakly associated and correlate poorly with clinical progression. Reasons for this are multiple and probably relate to the non-specific changes and insensitivity of current MR techniques, disease dynamics, anatomical factors, and the temporal profile and poorly defined complexities of the inflammatory reaction in multiple sclerosis. This paper provides an overview of the principles involved in the monitoring of inflammation in multiple sclerosis, discusses possible reasons for the weak correlation between MR and immunological markers of inflammation, and briefly reviews the studies correlating these modalities. In addition, the predictive values of MRI and CSF oligoclonal immunoglobulin are compared in determining future progression to clinically definite multiple sclerosis in patients presenting with clinically isolated syndromes compatible with demyelination.