Grey matter analysis in non-severe COVID-19 points out limbic-related cortex and substantia nigra

Background & Objective: It is unknown whether occult neurological damage exists in mild COVID-19 patients. Obtaining direct histopathological evidence is difficult and often inappropriate. Radiological tools provide important clues regarding this issue. We aimed to investigate any overt or subtle changes in brain magnetic resonance (MR) scans of patients who recovered from non-severe COVID-19 at subacute stage. Methods: Cortical thicknesses/areas were measured in the olfactory bulb (OB), gyri recti, amygdalae, hippocampi, entorhinal cortices, perirhinal cortices, insulae, and substantia nigrae (SN) and compared with controls. Gross findings have also been reported. We assessed the correlations between radiological and clinical parameters. Results: Twenty percent of the patients had abnormal MR scans (mild ventriculomegaly, a hyperintense lesion, a lacune and hydrocephalus) although their relevance to COVID-19 was unknown. We found increased cortical thickness in bilateral OB, left entorhinal cortex, right perirhinal cortex, bilateral insulae, and bilateral gyri recti (p<0.05 for all). Right OB was thinner in patients with anosmia (p=0.015) and ageusia (p=0.004). Left perirhinal cortex and left gyrus rectus were thicker in patients with vertigo (p=0.040; p=0.032 respectively). Sleep disturbance was associated to increased thickness in left perirhinal cortex (p=0.033). Patients with brain fog had smaller SN bilaterally (right p=0.028 and left p=0.011). Patients with anxiety symptoms after COVID-19 had increased right hippocampal area(p=0.023). Neutrophil-to-lymphocyte ratio was correlated to the thickness of right perirhinal cortex (r=-0.57, p=0.02) while CRP, time since COVID-19 and age were not. Conclusion: These changes in limbic areas, insula and SN necessitate close monitoring of patients for autonomic complications, and secondary neurodegenerative processes.