ABSTRACT
ImportanceGrowing evidence suggests that coronavirus disease 2019 (COVID-19) is associated with neurological sequelae. However, the underlying pathophysiological mechanisms resulting in central nervous system (CNS) derogation remain unclear. ObjectiveTo identify severity-dependent immune mechanisms in the cerebrospinal fluid (CSF) and plasma of COVID-19 patients and their association with brain imaging alterations. DesignProspective cross-sectional cohort study. SettingThis study was performed from August 2020 to April 2021. Participants were enrolled in the outpatient clinics, hospital wards and intensive care units (ICU) of two clinical sites in Basel and Zurich, Switzerland. ParticipantsAge >18 years and a positive SARS-CoV-2 test result were inclusion criteria. Potentially matching individuals were identified (n=310), of which 269 declined to participate and 1 did not match inclusion criteria. Paired CSF and plasma samples, as well as brain images, were acquired. The COVID-19 cohort (n=40; mean [SD] age, 54 [20] years; 17 women (42%)) was prospectively assorted by neurological symptom severity (classes I, II and III). Age/sex-matched inflammatory (n=25) and healthy (n=25) CSF and plasma control samples were obtained. For volumetric brain analysis, a healthy age/sex-matched control cohort (n=36) was established. ExposuresLumbar puncture, blood sampling and cranial MRI and/or CT. Main outcomes and measuresProteomics, standard parameters and antibody profiling of paired CSF and plasma samples in COVID-19 patients and controls. Brain imaging and gray matter volumetric analysis in association with biomarker profiles. Follow-up after 10-months. ResultsCOVID-19 patients displayed a plasma cytokine storm but a non-inflammatory CSF profile. Class III patients displayed signs of blood-brain barrier (BBB) impairment and a polyclonal B cell response targeting self- and non-self antigens. Decreased regional brain volumes were present in COVID-19 patients and associated with specific CSF and plasma parameters. Conclusion and relevanceNeuro-COVID class III patients had a strong, peripheral immune response resulting in (1) BBB impairment (2) ingress of (auto-)antibodies, (3) microglia activation and neuronal damage signatures. Our data point towards several potentially actionable targets that may be addressed to prevent COVID-19-related neurological sequelae. Trial registrationThe trial (NCT04472013) was registered on clinicaltrials.gov. Key pointsO_ST_ABSQuestionC_ST_ABSDoes a severity-dependent pattern of immune mechanisms exist in the cerebrospinal fluid (CSF) and plasma of COVID-19 patients and are these associated with clinical and brain imaging findings? FindingsNeuro-COVID patients display a robust class III-specific peripheral immune response resulting in (1) blood-brain barrier (BBB) impairment, (2) ingress of (auto-)antibodies, (3) microglia activation and neuronal damage signatures. Integration of MRIs, brain volumetry and proteomics identified biomarkers associated with regional brain volume loss in severe Neuro-COVID. MeaningWe provide a multidimensional framework of mechanisms associated with severe Neuro-COVID and present possible targets to prevent COVID-19-related neurological sequelae.
ABSTRACT
Increasing evidence shows that the brain is a target of SARS-CoV-2. However, the consequences of the virus on the cortical regions of hospitalized patients are currently unknown. The purpose of this study was to assess brain cortical gray matter volume (GMV), thickness (Th), and surface area (SA) characteristics in SARS-CoV-2 hospitalized patients with a wide range of neurological symptoms and their association with clinical indicators of inflammatory processes. A total of 33 patients were selected from a prospective, multicenter, cross-sectional study during the ongoing pandemic (August 2020-April 2021) at Basel University Hospital. Retrospectively biobank healthy controls with the same image protocol served as controls group. For each anatomical T1w MPRAGE image, the Th and GMV segmentation were performed with the FreeSurfer-5.0. Cortical measures were compared between groups using a linear regression model. The covariates were age, gender, age*gender, MRI magnetic field strength, and total intracranial volume/mean Th/Total SA. The association between cortical features and laboratory variables was assessed using partial correlation adjusting for the same covariates. P-values were adjusted using false discovery rate (FDR). Our findings revealed a lower cortical gray matter volume in orbitofrontal and cingulate regions in patients compared to controls. The orbitofrontal grey matter volume was negatively associated with protein levels, CSF-blood/albumin ratio and CSF EN-RAGE level. CSF EN-RAGE and CSF/Blood-albumin ratio, which are neuroinflammatory biomarkers, were associated with cortical alterations in gray matter volume and thickness in frontal, orbitofrontal, and temporal regions. Our data suggest that viral-triggered inflammation leads to increased neurotoxic damage in some cortical areas.
