Prediction of Long-term Cognitive Functions after Minor Stroke, Using Functional Connectivity.

OBJECTIVE: To determine whether functional MRI connectivity can predict the long-term cognitive functions 36 months after minor stroke. METHODS: Seventy-two participants with first-ever stroke were included at baseline and followed up for 36 months. A ridge regression machine learning algorithm was developed and used to predict cognitive scores 36 months post-stroke on the basis of the functional networks measured using MRI at 6 months (referred to here as the post-stroke cognitive impairment (PSCI) network). The prediction accuracy was evaluated in four domains (memory, attention/executive, language and visuospatial functions) and compared with clinical data and other functional networks. The models' statistical significance was probed with permutation tests. The potential involvement of cortical atrophy was assessed 6 months post-stroke. A second, independent dataset (n=40) was used to validate the results and assess their generalizability. RESULTS: Based on the PSCI network, a machine learning model was able to predict memory, attention, visuospatial functions and language functions 36 months post-stroke (r2: 0.67, 0.73, 0.55 and 0.48, respectively). The PSCI-based model was at least as accurate as models based on other functional networks or clinical data. Specific patterns were demonstrated for the four cognitive domains, with involvement of the left superior frontal cortex for memory, attention and visuospatial functions. The cortical thickness 6 months post-stroke was not correlated with cognitive function 36 months post-stroke. The independent validation dataset gave similar results. CONCLUSIONS: A machine learning model based on the PSCI network can predict the long-term cognitive outcome after stroke.

Identification of a specific functional network altered in poststroke cognitive impairment.

OBJECTIVE: To study the association between poststroke cognitive impairment and defining a specific resting functional marker. METHODS: The resting-state functional connectivity 6 months after an ischemic stroke in 56 patients was investigated. Twenty-nine of the patients who had an impairment of one or several cognitive domains were compared to 27 without any cognitive deficit. We studied the whole-brain connectivity using 2 complementary approaches: graph theory to study the functional network organization and network-based statistics to explore connectivity between brain regions. We assessed the potential cortical atrophy using voxel-based morphometry analysis. RESULTS: The overall topological organization of the functional network was not altered in cognitively impaired stroke patients, who had the same mean node degree, average clustering coefficient, and global efficiency as cognitively healthy stroke patients. Network-based statistics analysis showed that poststroke cognitive impairment was associated with dysfunction of a whole-brain network composed of 167 regions and 178 connections, and functional disconnections between superior, middle, and inferior frontal gyri and the superior and inferior temporal gyri. These regions had connections that were specifically and positively correlated with cognitive domain scores. No intergroup differences in overall gray matter thickness and ischemic infarct topography were observed. To assess the effect of prestroke white matter hyperintensities on connectivity, we included the initial Fazekas scale in the regression model for a second network-based analysis. The resulting network was associated with the same key alterations but had fewer connections. CONCLUSIONS: The observed functional network alterations suggest that the appearance of a cognitive impairment following stroke may be associated with a particular functional alteration, shared specifically between cognitive domains.

Tau deletion promotes brain insulin resistance.

The molecular pathways underlying tau pathology-induced synaptic/cognitive deficits and neurodegeneration are poorly understood. One prevalent hypothesis is that hyperphosphorylation, misfolding, and fibrillization of tau impair synaptic plasticity and cause degeneration. However, tau pathology may also result in the loss of specific physiological tau functions, which are largely unknown but could contribute to neuronal dysfunction. In the present study, we uncovered a novel function of tau in its ability to regulate brain insulin signaling. We found that tau deletion leads to an impaired hippocampal response to insulin, caused by altered IRS-1 and PTEN (phosphatase and tensin homologue on chromosome 10) activities. Our data also demonstrate that tau knockout mice exhibit an impaired hypothalamic anorexigenic effect of insulin that is associated with energy metabolism alterations. Consistently, we found that tau haplotypes are associated with glycemic traits in humans. The present data have far-reaching clinical implications and raise the hypothesis that pathophysiological tau loss-of-function favors brain insulin resistance, which is instrumental for cognitive and metabolic impairments in Alzheimer's disease patients.