One of the most important studies in the pandemic – studying the potential impact of Covid on the brain – was just published. The major findings of loss of gray matter, reduced brain size, and cognitive decline are concerning and need to be placed in context.
If you want to determine whether the SARS-CoV-2 virus can damage the brain, you would ideally have a MRI brain scan before and after the Covid infection and a matched (for age and sex, medical history) control group of people without infection who also had two sets of brain images. It just so happened that in the United Kingdom over tens of thousands of people enrolled in their UK Biobank had undergone a brain scan before the pandemic and a subset of these were brought back at an average three years later, with or without having had Covid. They also had basic cognitive testing – a connect-the-dots type of test – with their brain scans.
There were about 400 participants in each group, aged between 47 and 80, on average 59, at the time of their baseline scan. There was lack of diversity with 97% of both groups of white ethnicity. All but 15 people in the Covid group had mild to moderate Covid, not requiring hospitalization, and the results were not affected by deleting this small number people requiring hospitalization from the analysis. Importantly, in context, the study pertains to older white adults with mostly mild Covid.
As we age, there is typically a loss of gray matter of about 0.2% each year, which was seen in the control group. In contrast, the people with Covid in this study, at four months after their infection, had more gray matter loss than the control group, some up to tenfold more. Notably, the damage to the brain – the loss of gray matter – was chiefly in regions related to smell. Beyond the structural changes of the brain, there was a decline in the cognitive test among the infected group, taking longer to do the task. Separately, there was no difference in memory testing results between the two groups.
Now, what does all this mean? It’s a unique study design that is hard to replicate, but independent replication is important. We don’t have that yet, so we can’t consider the findings definitive, even though they are indeed concerning. Further uncertainties include the lack of knowledge about symptoms in the Covid group, such as loss of smell, and reliance for classification (Covid v no Covid) on different Covid tests, some of which do not have high accuracy. Were the two groups balanced? It was noted there was “a subtle pattern of lower cognitive abilities in the participants who went on to be infected”. While this would not influence the results of serial testing, or comparison with controls, it bespeaks some imbalances in features between the Covid and control groups. Even though it is the largest Covid brain imaging study, its size does not pre-empt multiple small differences between the groups, that cumulatively may have influenced the findings.
The mechanism for the structural brain changes is uncertain, but probably involves inflammation that originated from the nasal infection. Most studies of Covid and the brain support virus-induced inflammation, rather than direct infection of neurons, as the path to brain damage. The regions of the brain most affected in the Covid group are related to sense of smell, the limbic system, incriminating this nasal entry port. Whether the process is specific to this virus was examined in the current study by comparing Covid brain images with a small number of people with either influenza or pneumonia, and not seeing a similarity in pattern.
It is important to note that this was not a study of long Covid. The symptom of brain fog that is commonly reported in people suffering from long Covid and has been likened to cancer therapy “chemo brain” with brain inflammation chiefly involving white matter, is not related to the current report. The Covid group was assessed only one time after infection, about four months later, memory was not impaired, and details of symptoms were absent. The single assessment also raises the question as to whether the structural changes, and more likely the cognitive decline, may have some reversibility. While brain cells do not have high capability to regenerate, they have remarkable plasticity to form and reorganize synaptic connections – for preserving function. The flip side is also a possibility. Limbic system atrophy, the modest degree of which was seen in the Covid group, is one of the classic patterns of brain imaging of Alzheimer’s disease. For these reasons, subsequent brain imaging to determine potential recovery or progression, is essential.
Another question relates to whether the findings apply to younger adults and children. In the present study the rudimentary evidence of cognitive decline occurred mainly in people older than 70. Whether Covid may affect structural or functional brain changes in young people is yet to be established. That also brings up the cause-and-effect issue, since the evidence of brain impact must be considered as an association, since proof of Covid causality, although likely, is not absolute or certain.
In context, the study pertained to variants antedating Omicron, the ultra-transmissible strain that is thought to have potentially infected 40% or more of Americans and Europeans in recent months. It is important to note that loss of smell was considerably less with Omicron than Delta and prior variants, in some studies one-tenth as likely. This suggests that the liability for brain effects of Covid would be far less likely with Omicron. Each variant can have different affinity, known as tropism, to tissues and organs within the body; for reasons yet unexplained Omicron has less tropism to the olfactory bulb, the neurons at the base of the brain specialized in smell.
In two years, we’ve come a long way from considering severe Covid pneumonia as being the singular concern. While the list of uncertainties about Covid’s deleterious impact on the brain is long and unsettled, it is vital that we maintain a high regard for the potential liability and unpredictability of even mild infections.