Post-stroke cognitive impairment (PSCI) is underdiagnosed and undertreated as
much as other stroke complications, such as motor weakness. The pathophysiology
of PSCI is very complex and has unrevealed aspects in high cortical function. Beyond
the c...
Post-stroke cognitive impairment (PSCI) is underdiagnosed and undertreated as
much as other stroke complications, such as motor weakness. The pathophysiology
of PSCI is very complex and has unrevealed aspects in high cortical function. Beyond
the clinical and conventional neuroimaging marker studies of PSCI, network analysis
is required for resolving the local lesion effects throughout the brain.
Patients with acute ischemic stroke within one week of symptom onset were
enrolled. The participants underwent electroencephalography (EEG) and brain
magnetic resonance imaging (MRI) and were assessed for cognitive function using
the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment
(MoCA). Acute lesion characteristics and chronic lesions following the STRIVE
criteria were evaluated with brain MRI.
Neuropsychological tests were evaluated at baseline during admission and 3
months after the stroke. The change scores between them were analyzed. We divided
the patients into a cognitively impaired group and a cognitively normal group based
on the 3-month MMSE Z-score -1.0. EEG was analyzed at network levels using
imaginary coherence (iCOH) analysis. Based on the iCOH values, network features
indicating segregation and integration were calculated
A total of 87 patients among the recruited 99 patients were included in the final
analysis. Their mean age was 65.4 years (SD 11.7) and 53 (60.9%) were men. Left
hemispheric acute lesions were more common (n=51, 60.7%). The median score of
the MMSE and the 3-month MMSE, the MOCA and the 3-month MoCA were 25 and
27, and 20 and 23 points, respectively. As part of the MoCA domain, the executive
index, the attention index, and the memory index score were significantly changed.
There were 49 cognitively normal groups and 38 cognitively impaired groups.
Weighted modularity was significantly lower in the cognitively impaired group
compared with the cognitively normal group. The weighted modularity was
significantly correlated with the post-stroke 3-month MMSE scores after adjustment
for age, sex, education, and initial stroke severity.
In this study, we investigated the surrogates of PSCI by using EEG. The modularity
of the cognitively impaired group was significantly decreased. We demonstrated that
the EEG during the acute stroke period could provide information about the features
of the brain network with stroke, among which the baseline modularity was a
reflective indicator of global cognitive dysfunction (MMSE) 3 months after a stroke.