Exploring the relationship between MRI changes and cognitive/neuropsychiatric complaints in a cohort of long COVID‐19 patients: A cross‐sectional and longitudinal study

Abstract

AbstractBackgroundPost‐acute Covid‐19 syndrome (PACS) frequently refers to cognitive complaints. It is not yet clear whether there is an association between cognitive symptoms with brain changes or neuropsychiatric symptoms. Our aims are 1) to study cross‐sectional and longitudinal MRI brain measures in a cohort of PACS and 2) their association with cognitive performance and mood disturbances.MethodWe performed a prospective single‐center study with 3T‐T1w MRI of 49 PACS patients at a cross‐sectional level. These participants had confirmed SARS‐CoV2 infection, ≥ 8 weeks after symptoms onset and cognitive complaints. All participants completed a comprehensive neuropsychological battery (NPS) and questionnaires assessing depression, anxiety, and subjective cognitive complaints (SCD). We obtained global MRI measures (e.g; gray and white matter volumes and mean cortical thickness) with FreeSurfer. We measured correlations of global MRI measures with SCD, memory and executive function outcomes, anxiety, and depression. All analyses were corrected for multiple comparisons. 44 PACS subjects had a 6‐month follow‐up MRI: in these, we performed longitudinal analyses with Generalized Linear Mixed‐Effects Models to study changes between visits in global MRI measures.ResultDemographics are shown in Table 1. We did not find any correlation between clinical outcomes (SCD, anxiety, depression) and MRI findings. Visual memory (Rey figure’s recall) and cognitive interference inhibition and processing speed ((Stroop’s color‐word condition) were positively correlated with global gray and white matter volume measures (Figure 1). We did not identify changes in global MRI measures at 6 months in PACS participants.ConclusionIn PACS, worse visual memory and executive function, but not other clinical outcomes, were associated with lower global structural MRI indexes. We did not observe global longitudinal changes in MRI.