ABSTRACT
The coronavirus SARS-CoV-2 causes COVID-19, a predominantly respiratory disease that has been reported to be associated with numerous neurological signs, symptoms and syndromes. More than 20 published studies have used RT-PCR methods to determine viral SARS-CoV-2 genomic presence in postmortem brain tissue and the overall impression is that viral brain invasion is relatively uncommon and occurs in low copy numbers, supporting indirect mechanisms as the cause of most neurological phenomena. Hypoxic-ischemic brain injury and stroke are one such possible indirect mechanism, as acute ischemia or stroke concurrence with COVID-19 has been reported as being 0.5% to 20%. Immunohistochemical stains for {beta}-amyloid precursor protein (APP) have been suggested to be a "signature" change of hypoxic leukoencephalopathy or COVID-19 brain disease, although prior reports have not had a non-COVID-19 control group. We therefore compared the prevalence and intensity of white matter APP staining in the brains of subjects dying with and without COVID-19. Clinical and neuropathological results, including semi-quantitative assessment of the density of white matter APP staining, were compared between 20 COVID-19 cases and 20 pre-COVID-19 autopsy cases, including 10 cases with autopsy-proven non-COVID-19 pneumonia and 10 cases without pneumonia. Positive APP white matter staining in at least one of the two brain regions (precentral gyrus and cingulate gyrus) studied was not significantly more common in COVID-19 vs controls (14/20 vs 12/20). Comparing density scores from both brain regions combined, the mean scores for COVID-19 cases were higher than those for controls of both types together but not significantly different when restricting to controls with pneumonia. Among control cases, cases with pneumonia had significantly higher scores. The presence or absence of a major neuropathologically-defined neurodegenerative disorder did not significantly affect the APP scores. The major finding is that while APP white matter staining cannot be regarded as a specific marker of COVID-19, as it does not occur with significantly greater probability in in COVID-19 brains as compared to non-COVID-19 brains, it is possible that white matter APP staining, representing acute or subacute axonal damage, may be a common occurrence in the perimortem period, and that it may be more intense in subjects dying with pneumonia, regardless of cause.
ABSTRACT
The coronavirus SARS-CoV-2 (SCV2) causes acute respiratory distress, termed COVID-19 disease, with substantial morbidity and mortality. As SCV2 is related to previously-studied coronaviruses that have been shown to have the capability for brain invasion, it seems likely that SCV2 may be able to do so as well. To date, although there have been many clinical and autopsy-based reports that describe a broad range of SCV2-associated neurological conditions, it is unclear what fraction of these have been due to direct CNS invasion versus indirect effects caused by systemic reactions to critical illness. Still critically lacking is a comprehensive tissue-based survey of the CNS presence and specific neuropathology of SCV2 in humans. We conducted an extensive neuroanatomical survey of RT-PCR-detected SCV2 in 16 brain regions from 20 subjects who died of COVID-19 disease. Targeted areas were those with cranial nerve nuclei, including the olfactory bulb, medullary dorsal motor nucleus of the vagus nerve and the pontine trigeminal nerve nuclei, as well as areas possibly exposed to hematogenous entry, including the choroid plexus, leptomeninges, median eminence of the hypothalamus and area postrema of the medulla. Subjects ranged in age from 38 to 97 (mean 77) with 9 females and 11 males. Most subjects had typical age-related neuropathological findings. Two subjects had severe neuropathology, one with a large acute cerebral infarction and one with hemorrhagic encephalitis, that was unequivocally related to their COVID-19 disease while most of the 18 other subjects had non-specific histopathology including focal {beta}-amyloid precursor protein white matter immunoreactivity and sparse perivascular mononuclear cell cuffing. Four subjects (20%) had SCV2 RNA in one or more brain regions including the olfactory bulb, amygdala, entorhinal area, temporal and frontal neocortex, dorsal medulla and leptomeninges. The subject with encephalitis was SCV2-positive in a histopathologically-affected area, the entorhinal cortex, while the subject with the large acute cerebral infarct was SCV2-negative in all brain regions. Like other human coronaviruses, SCV2 can inflict acute neuropathology in susceptible patients. Much remains to be understood, including what viral and host factors influence SCV2 brain invasion and whether it is cleared from the brain subsequent to the acute illness.
