Brain compensation during visuospatial working memory in premanifest Huntington's disease

Abstract

Huntington's disease (HD) is a progressive autosomal dominant disorder characterised by impairment of movement, cognitive decline and psychiatric impairment. During the premanifest stages of HD (pre-HD), there are notable structural and functional brain changes detectable many decades prior to clinical diagnosis. Although caudate atrophy is one of the earliest and most prominent sites of neuropathology, there is also evidence of cortical atrophy, compromised white and grey matter integrity, cortical hypometabolism, and aberrant task-related functional magnetic resonance imaging (fMRI) activity. Pre-HD individuals typically show increased task-related fMRI activity suggested as a strategy to compensate for early brain anomalies. In this study, we used a quantitative model of compensation, known as the CRUNCH (Compensation-Related Utilization of Neural Circuits Hypothesis), to explicitly characterise compensation in pre-HD. The assumption of the model is that fMRI activity increases as task difficulty increases (compensatory effect). However, when individuals reach the ‘CRUNCH’ point, where task difficulty exceeds their neural resources, fMRI activity and performance declines. We acquired fMRI data (n = 15 pre-HD; n = 15 controls) during performance of an fMRI visuospatial working memory task with low, intermediate-1, intermediate-2, and high memory loads. Consistent with the CRUNCH prediction, pre-HD individuals showed decreased fMRI activity in left intraparietal sulcus at high memory load, compared to healthy controls who showed increased fMRI activity in left intraparietal sulcus at high memory load. Contrary to the other CRUNCH prediction, the pre-HD group did not show compensatory increase in fMRI activity at lower levels of memory loads in left intraparietal sulcus. Our findings provide partial support for the validity of CRUNCH in pre-HD.