Neuroimaging Spectrum in COVID-19 Infection: A Single-Center Experience
We read with interest the article by Tiwari et al who reported on a retrospective
study of 180 severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected
patients who underwent neuroimaging in a single Indian center between March 2020 and
May 2021.[1] The most frequently detected neuroimaging abnormality was ischemic stroke (n = 77), followed by bleeding (n = 22), hypoxic lesions (n = 5), cerebellitis (n = 3), demyelination (n = 1), encephalitis (n = 1), acute hemorrhagic necrotizing encephalitis (n = 1), transverse myelitis (n = 1), cytotoxic lesion of the corpus callosum (n = 1), Guillain–Barre syndrome (n = 1), and microbleeds (n = 1).[1] The study is excellent but has limitations that should be discussed.
The first constraint is the inclusion criterion. According to the method part, patients
with polymerase chain reaction (PCR)-positive SARS-CoV-2 or coronavirus disease 2019
(COVID-19) pneumonia were included.[1] Accordingly, also patients with a negative PCR but COVID-19 pneumonia on imaging
were included. However, according to the exclusion criteria, patients with a negative
PCR or negative lung imaging were excluded. This discrepancy should be resolved. In
addition, we should know if COVID-19 pneumonia was diagnosed by X-ray or lung computed
tomography (CT) or both. In cases with negative PCR but imaging suggestive of pneumonia,
how were false-positive diagnoses ruled out?
A second limitation is that the majority of included patients had only cerebral CT
(CCT, n = 169) and only 28 had cerebral magnetic resonance imaging (cMRI).[1] Disadvantages of CCT are the low resolution and the fact that small or embolic ischemic
lesions can be missed. It is also unreported how many who underwent CCT or cMRI also
received contrast medium. Immune/infectious meningitis or encephalitis can easily
be overlooked if no contrast medium has been administered. According to the results,
an MRI was only performed if the CCT was not explanatory.[1] However, only 17 had both, CCT and cMRI. This discrepancy should be explained.
A third limitation is that comorbidities were not reported. In particular, we should
know how many had cardiovascular risk factors such as smoking, arterial hypertension,
diabetes, hyperlipidemia, or atrial fibrillation. How many of those with ischemic
stroke attributed to SARS-CoV-2 also had cardiovascular risk factors that could explain
the brain/nerve damage?
A fourth limitation is that in the 66 patients in whom neuroimaging provided no explanation,
no explanation was given for the neurological deficits We should therefore know the
clinical neurologic findings of these 66 patients with negative neuroimaging and what
other tests were performed to clarify the underlying pathology.
A fifth limitation is that the number of included patients was small.[1] This could explain why the number of imaging abnormalities was limited to 11.[1] However, the number of imaging abnormalities in SARS-CoV-2-infected patients is
much larger and also includes posterior reversible encephalopathy syndrome, infections/immune
encephalitis, infections/immune meningitis, acute disseminated encephalomyelitis,
acute hemorrhagic leukoencephalitis, multiple sclerosis, neuromyelitis optica spectrum
disorder, cerebral vasculitis, reversible cerebral vasoconstriction syndrome, giant
cell arteritis, ventriculitis, pontine myelinolysis, Wernicke encephalopathy, pseudotumor
cerebri, subarachnoid bleeding, hypophysitis, and rhabdomyolysis.[2]
There is a discrepancy regarding the study period between the methods (March 2020
to May 2021) and the results (May 2020 to May 2021).[1] This discrepancy should be resolved.
Overall, addressing these limitations would strengthen the conclusions and could improve
the status of the study.
