Search publications

Reset filters Search by keyword

No publications found.

 

Developing Topics

Authors: Silva NCBSStein RGGu YHsu CLTam RCSalluzzi MMcCreary CRAlkeridy WALam KMacKay ALKolind SCossette BGriffith LEHogan DBMcMillan JMRaina PSmith EELiu-Ambrose T


Affiliations

1 Concordia University, Montréal, QC, Canada.
2 University of British Columbia, Vancouver, BC, Canada.
3 The Hong Kong Polytechnic University, Hong Kong, Hong Kong.
4 University of Calgary, Calgary, AB, Canada.
5 King Saud University, Riaydh, Saudi Arabia.
6 University of Sherbrooke, Sherbrooke, QC, Canada.
7 McMaster University, Hamilton, ON, Canada.

Description

Background: Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is the causative agent of COVID-19 and has infected >700 million persons worldwide. Individuals infected with SARS-CoV-2 are at risk for cognitive decline and at higher risk of dementia compared with those diagnosed with other respiratory tract infections. Data from animal models suggest that SARS-CoV-2 infection triggers an overaggressive neuroinflammatory response resulting in myelin loss. Whether SARS-CoV-2 is associated with myelin loss in older individuals remains unknown.

Methods: We investigated the impact of SARS-CoV-2 on myelin in older individuals from the Canadian Longitudinal Study on Aging COVID-19 Brain Health Study who underwent brain MRIs. We included SARS-CoV-2 confirmed cases at baseline (2021-2022) via positive serological testing or health care provider diagnosis. Non-infected controls had negative serological testing and reported no COVID-19 diagnosis. Myelin data were acquired via myelin water imaging using a 3D MRI gradient and spin echo sequence for T2 measurement. Myelin content was extracted from 16 regions-of-interest within the cerebral white matter. 3D T1-weighted scans were acquired for registrations and to estimate intracranial volume. T2- and PD-weighted scans were acquired for segmentation of white matter lesions. We performed cross-sectional comparisons via analysis of covariance. Exploratory analyses were conducted to assess the association of SARS-CoV-2-related symptom incidence and severity with myelin content by group. All models were adjusted for age, age2, sex, ethnicity, white matter lesion burden, intracranial volume, and study site.

Results: We included 352 community-dwelling individuals (SARS-CoV-cases, n= 64; controls, n=288). Their mean [SD] age was 65.26 (8.35) years, and 50.3% were female. There were no differences between SARS-CoV-2 cases and controls on myelin content across all regions-of-interest. Cases showed greater incidence (p <0.001) and severity (p <0.001) of symptoms compared with controls (Figure 1). Exploratory analysis revealed significant interactions between symptom incidence and severity with group after correcting for multiple comparisons (Table 1, p corrected < 0.05). Post hoc analysis showed that symptom incidence and severity were inversely associated with myelin in SARS-CoV-2 cases but not in controls across multiple regions-of-interest (Figure 2).

Conclusions: Myelin loss may occur in older individuals who experienced greater incidence and severity of SARS-CoV-2 infection symptoms.


Links

PubMed: https://pubmed.ncbi.nlm.nih.gov/41434542/

DOI: 10.1002/alz70861_108935