Idiopathic normal-pressure hydrocephalus, cortical thinning, and the cerebrospinal fluid tap test

Abstract

When considering the underlying pathophysiological mechanisms involved in idiopathic normal pressure hydrocephalus (iNPH), white matter is often the main locus of investigation. However, when an axon in the brain is damaged, degeneration of the neuron can occur proximally (dying back) and Alzheimer's disease (AD), associated with cortical thinning, is a common pathologic comorbidity with iNPH. We investigated differences in cortical thickness between CSF tap test (CSFTT) responders and non-responders in iNPH patients and compared patterns of cortical thickness in iNPH patients with that of AD patients. Thirty-two iNPH patients (16 CSFTT responders and 16 CSFTT non-responders) and 16 AD patients were imaged with MRI, including 3-dimensional volumetric images for cortical thickness analysis across the entire brain. Among the iNPH patients, CSFTT non-responders, when compared to responders, had statistically significant cortical thinning in the left superior frontal gyrus at the level of a false discovery rate (FDR) p < 0.05, and tended to show widespread cortical thinning in most areas of the brain. Relative to the CSFTT responders, AD patients showed statistically significant cortical thinning in superior and medial frontal gyrus, left precentral gyrus, postcentral gyrus, paracentral lobule, precuneus, and superior parietal lobule after FDR correction (p < 0.05). However, comparing patterns of cortical thinning between AD patients and CSFTT non-responders revealed no statistically significant differences. Differences in cortical thickness correlated with CSFTT response for iNPH patients may indicate a possibility for considering patterns of cortical thinning in patients with ventriculomegaly as potential brain imaging markers for the prediction of CSFTT responders. And, our findings suggest that comorbid AD pathology might be related to the cortical thinning patterns found in CSFTT non-responders. Larger studies, using normal controls and combinations of other biomarkers associated with AD, would be necessary to evaluate these hypotheses.