Chapter 5 . 1 Imaging the tip of the iceberg : visualisation of cortical lesions in multiple sclerosis

Background: Cortical lesions (CLs) occur frequently in multiple sclerosis (MS), but only few CLs are observed on conventional magnetic resonance imaging (MRI). Why some CLs are visible and others are not, is currently unknown. Here, we investigated whether CLs that are visible on conventional MRI differ from MRI-invisible CLs in terms of underlying histopathology and quantitative MRI (qMRI) measures. Methods: A total of 16 brain slices from 10 chronic MS patients were analysed histopathologically and with conventional and qMRI. A region-of-interest approach was used to compare MRI-visible CLs to MRI-invisible CLs. Results: Although under-powering cannot be completely excluded in this study, MRIvisible CLs did not seem to differ from MRI-invisible CLs in terms of histopathology or qMRI measures. They were, however, significantly larger than their invisible counterparts (mean 13.3 ± 1.7 mm2 versus 6.9 ± 1.3mm2; P = 0.001). Furthermore, the number of MRIvisible lesions correlated with the overall number of cortical lesions in the brain slice (r = 0.96, P < 0.01) and with the overall percentage of demyelination (r=0.78, P < 0.01) per hemispheric brain slice. Conclusion: MRI visibility of CLs is determined by lesion size, and not by any distinctive underlying pathology. Visible CLs are associated with a higher total cortical lesion load, which suggests that when CLs in MS patients become detectable on MRI, they merely represent ‘the tip of the pathological iceberg’.