Magnetic resonance imaging (MRI) monitoring of disease progression in multiple sclerosis is limited by the lack of correlation of abnormalities seen on T2‐weighted imaging, and disability. We studied the histopathology of multiple sclerosis lesions, as depicted by MRI, in a large postmortem sample, focusing on axonal loss. Tissue samples from 17 patients were selected immediately postmortem for histopathological analysis on the basis of T2‐weighted imaging, including normal appearing white matter and T1 hypointense lesions. In each region, we measured magnetization transfer ratios (MTR), T1 contrast ratio, myelin, and axonal density. T2 lesions (109 samples) were heterogeneous with regard to MRI appearance on T1 and MTR, whereas axonal density ranged from 0% (no residual axons) to 100% (normal axonal density). Of 64 T2 lesions, 17 were reactive (mild perivascular inflammation only), 21 active, 15 chronically active, and 11 chronically inactive. MTR and T1 contrast ratio correlated strongly with axonal density. Also in normal appearing white matter (24 samples), MTR correlated with axonal density. In conclusion, postmortem tissue sampling by using MRI revealed a range of pathology, illustrating the high sensitivity and low specificity of T2‐weighted imaging. T1 hypointensity and MTR were strongly associated with axonal density, emphasizing their role in monitoring progression in multiple sclerosis.

We investigated the histopathologic correlates of white matter changes in Alzheimer's disease (AD) patients (n = 6) and controls (n = 9) using postmortem MRI. White matter changes were rated on a 0 to 3 scale in 51 regions. Histopathologically, we subjectively rated the loss of myelinated axons in the deep and periventricular white matter, denudation of the ventricular ependyma, gliosis, width of the perivascular spaces, and leptomeningeal congophilic angiopathy; we measured structural changes in the walls of the blood vessels in the white matter in micrometers. The AD brains displayed significantly more white matter hyperintensities on MRI than controls. Histopathologically, the denudation of the ventricular ependyma and the gliosis were significantly more severe in AD than in controls, and there was a trend toward more loss of myelinated axons in the deep white matter in the AD brains (p = 0.07). The MRI abnormalities correlated with the loss of myelinated axons in the deep white matter (r' = 0.37; p < 0.01) and with the denudation of the ventricular lining (r' = 0.54; p < 0.01). We could not find any evidence for arteriolosclerosis, but the mean thickness of the adventitia of the arteries of the deep white matter in AD almost doubled the value in control brains (p = 0.0009). We conclude that white matter abnormalities in AD patients and controls consist of loss of myelinated axons, probably caused by arterial changes and breakdown of the ventricular lining. Since imaging/histopathologic correlation was similar in AD patients and controls, these changes probably represent some form of accelerated aging.

Later onset does occur in the disease of vanishing white matter, and both MRS and histopathology are compatible with a primary axonopathy rather than primary demyelination.

Demyelinating lesions of fiber bundles in and adjacent to the hypothalamus (i.e. the fornix. anterior commissure, internal capsule, and optic system) may be the basis for autonomic and endocrine alterations in multiple sclerosis (MS) patients. Therefore we investigated the presence and immunological activity of lesions in hypothalamic fiber bundles of 17 MS patients and 14 controls. In the MS group, 16 of 17 patients showed demyelinated lesions. The incidence of active lesions was high (60%) and outnumbered chronic inactive lesions in the internal capsule (p = 0.005). In 4 of 17 MS patients, axonal damage was observed and in 3 of 17 MS patients grey matter lesions were apparent. Duration of MS was inversely related to the active hypothalamic MS lesion score (r = -0.72, p = 0.001). Since comparison of hypothalamic lesions with MS lesions in other areas of the brain in the same patients (n = 7) showed a great similarity both as stage and appearance was concerned, this negative relation in all likelihood reflects the clinical consequences of high disease activity throughout the whole brain. In controls no demyelinating lesions were seen but in 11 control cases HLA expression was observed that was lower than that present in MS patients (p = 0.02). In the median eminence region that lacks a blood-brain barrier, all controls showed a strong HLA expression around the blood vessels. We conclude that systematic pathological investigation of the hypothalamus in MS patients reveals an unexpected high incidence of active lesions that may impact on hypothalamic functioning.