Neuro-COVID-19: What are we talking about?

In a very short time after the onset of the COVID-19 infection, a “flood” of reports regarding SARS-Cov-2 related neurological manifestations were published. However, careful analysis shows that robust data establishing this relationship are lacking.

While more than three and a half million individuals are affected worldwide to date, most arguments of COVID-19-related neurological manifestations are supported by case reports or small series [1], [2], [3]. Also, reported frequencies could be over or underestimated because patients were usually examined by non-neurologists, which represents a limiting factor to an accurate description of these neurological presentations.

Another limit is that most neurological manifestations are listed as symptoms or signs and not as specific neurological syndromes. For example, headache is included in the prevalence calculation in some large series [4], albeit, it is not always a marker of neurological involvement, since it is commonplace during all viral infections and febrile states. Furthermore, laboratory investigations (neuroimaging, CSF analysis, electromyography, muscle biopsy…) could have been of great support to the observed neurological manifestations and their mechanism, but were frequently not conducted.

On the other hand, a review of the available data shows that the implication of the SARS-Cov-2 virus as a direct cause of neurological complications is not established yet. First, unlike other viruses such as the herpes or rabies virus, which neurotropism is clearly established, the neuroinvasive potential of the SARS-Cov-2 virus is currently hypothetical in humans. Indeed, the arguments are extrapolated from animal studies, using other coronaviruses as exemplars [5]. In support to this, to date, only 2 cases of meningitis or encephalitis with evidence of viral SARS-Cov-2 detection in CSF are reported [6], [7].

Other interesting symptoms like the loss of smell or taste are a frequent complaints, but their mechanism (mucosal or nervous fibers involvement) is still debated [8]. To date, the virus has not been detected in nervous fibers and the olfactory bulbs appear normal on imaging [9], which argues against a nervous implication.

So far, a dozen of COVID-19-related Guillain–Barré syndromes (GBS) have been reported, which seems to be marginal compared to the high prevalence of infected individuals [10]. While GBS occurs in the subsequent month following an infection, no signal of an increased frequency is seen after more than 4 months of this pandemic onset in China or other countries early affected. When as for other viruses, such as the Zika virus, thousands of GBS cases were triggered by this infectious agent [11]. Concerning muscle injuries, the causality link to SARS-Cov-2 virus was suggested on Hyper-CKemia in a group of Intensive Care Unit (ICU) patients, receiving several drugs that may interfere with this finding [4]. A single case of myositis was reported, but this diagnosis was made using a muscle MRI without a histological confirmation [12].

Finally, the majority of the reported CNS features and damage, like altered mental status, seizures, increased cerebrovascular events… are typically seen in severely affected patient in ICU [4], [13]. These findings may reflect more broadly critical illness condition, not specific to a COVID-19 infection (microinfarcts during hypercoagulability state, sepsis, multiorgan failure, hypoxic encephalopathy, drug toxicities and other inherent morbidities due to prolonged stay on intensive care…).

All these arguments emphasize that, given the high rate of COVID-19 infection, we must be cautious to infer a causal linkage, because coincidental occurence of other neurological condition is also possible. Several months following the pandemic onset, with millions of affected patients and a few hundred thousand deaths worldwide, the real impact of the SARS-Cov-2 virus on the nervous system remains unclear.

However, the potentially long-term risks related to this virus, if its neurotropism is confirmed, must be taken into account. A possible latent state could give rise to a recurrent encephalitis, or be implicated in the pathogenesis of several CNS disorders, especially the autoimmune ones.

Disclosure of interest

The authors declare that they have no competing interest.