Inject an agent that stimulates the immune system (usually something called lipopolysaccharide) into a rat and the animal will exhibit a constellation of behavioral symptoms called “sickness behavior.” The symptoms include reduced locomotion, decreased appetite (anorexia), sleepiness, and decreased engagement in usual activities like grooming. In other words, the rat seems to be depressed. Scientists can demonstrate that the immune system activation that causes sickness behavior in rodents also produces physical activation of immune markers in the brain.
Humans who get infected with various viruses and bacteria can relate: when we are sick we generally have decreased interest in things (called “anhedonia” by psychiatrists and psychologists), decreased energy, sleep disturbances, reduced appetite, and a sad feeling. Is this because stimulation of the immune system by these infectious agents is having a physical effect on our brains or is it a psychological reaction to feeling sick that is not accompanied by any discernible physical brain changes?
This question, which is very difficult to answer, has become especially relevant during the current coronavirus pandemic. The syndrome of Post-Acute Sequelae of SARS-CoV-2 Infection (PASC), better known as Long Covid, has sparked intense interest throughout media and the internet and is the subject of intensive research by basic and clinical scientists throughout the world. Many of the manifestations of Long Covid are neurological and psychological in nature, including headache, “brain fog,” confusion, loss of the senses of taste and smell, depression, and anxiety. This raises the question of how the virus that causes Covid-19 affects the brain. Are the neurological and psychological symptoms of Long Covid related to physical changes in any part of the central nervous system?
Virus Not Detected in Human Brain
Some viruses are neurotropic, meaning they can infect nervous system tissue directly. Examples are the measles, polio, and rabies viruses. So far, however, scientists have not found evidence that the virus that causes Covid-19 (SARS-CoV-2) is neurotropic in humans, although the virus has been detected in the brains of mice and non-human primates. Autopsy studies have failed to produce any signs of direct infection of neurons or other cells in the brains of people who have died from Covid-19. That means that it is unlikely that ongoing viral brain infection is the cause of Long Covid’s psychological and cognitive symptoms.
One theory that has been prominently advanced and studied to explain Long Covid brain-related symptoms is the possibility that an ongoing immunological process is affecting the brain. Given the well-established effect in rodents of inflammation elsewhere in the body on both immune function in the brain and on behavior, this is certainly possible. While the evidence for this mounting, it is still far from definitive.
The Difficulties of Studying the Human Brain
Studying the human brain is a difficult undertaking because it is encased behind something called the blood-brain barrier, a protective layer of cells and blood vessels that keep many things circulating in the bloodstream from entering the brain and many things in the brain from exiting into the blood. This means that blood tests generally do not reveal much about what is happening in the brain. And unlike other organs of the body, we usually cannot directly sample brain tissue in living people.
Blood studies have shown that Covid-19 can be associated with the production of autoantibodies, antibodies that mistakenly recognize tissues of the body as if they were foreign invaders. These autoantibodies have been shown to attack central nervous system tissue in the laboratory and to persist in the blood of people for weeks and even months after recovering from acute Covid-19 infection, suggesting that this could be one mechanism by which abnormal immune system activation affects the brain. It is not known yet, however, whether in fact Covid-19 stimulated autoantibodies do cross the blood-brain barrier in humans and adversely affect the brain.
Studies using formal neuropsychological testing have shown cognitive abnormalities months after recovery from the acute phase of Covid-19 infection. Brain imaging studies can further tell us many things about brain structure and function. Using the technology of positron emission tomography (PET) scanning, for example, scientists reported decreased metabolism in parts of the brain in patients with severe cognitive impairment in the days following recovery from Covid-19. Six months after recovery from Covid-19, the same investigators found improvement in both cognition and brain metabolism in the study participants, but there were still residual abnormalities in both domains. These are important studies, but they do not provide us with an understanding of the mechanism by which SARS-CoV-2 causes disturbances in brain function.
Animal studies can clarify the effects of the Covid-19 virus on the brain. Studies in a mouse model show that the virus can cause inflammation in the brain without actually entering the central nervous system, with activation of the brain’s immune cells, called microglia. The same pattern of microglia activation has been observed in autopsied human brains. Once again, however, it is unclear to what extent this is a direct result of previous Covid-19 infection or how directly correlated it is with Long Covid symptoms.
Examining the Cerebrospinal Fluid
That leaves examination of the cerebrospinal fluid (CSF), the fluid that surrounds the brain and runs down alongside the spinal cord. Samples of CSF can be obtained by performing a lumbar puncture, also known as a spinal tap, a standard procedure when brain infections like meningitis are suspected. Although lumbar puncture is less painful than people often anticipate and generally quite safe, it is not something done casually and therefore research studies involving examination of CSF samples often involve relatively small numbers of patients. Furthermore, even though the CSF originates in the brain, what is in it doesn’t always represent exactly what is occurring within the brain tissue itself.
Studies in which CSF has been sampled from patients with Long Covid are now available. They show elevation of a number of markers of abnormal immune system activation. As noted above, however, these studies involve small numbers of participants and different studies have found different immune systems markers to be activated. Although suggestive, the data do not yet nail down the exact role of immune system activation in Long Covid.
It is also important to remember that rates of anxiety disorders, depression, and posttraumatic stress disorder have increased during the pandemic. When symptoms of these conditions emerge in people who have recovered from Covid-19 it is difficult to know whether it is the psychosocial burden of living under pandemic conditions that is responsible, the physical effects of ongoing neuroinflammation, or some combination of the two. Because of the stigma surrounding mental illness, it is sometimes difficult to tell patients with Long Covid that one of their problems is depression and that treatment with evidence-based psychotherapy and/or antidepressant medication might be helpful. Here, however, trying to make the distinction between what is “physical” and what is “psychological” may be misleading. There is a long literature on the relationship of depression to immune system abnormalities and overactivation. Hence, it would be unsurprising to find that both psychosocial and immunological factors are involved in Long Covid.
Evidence is mounting that immune system activation is a factor in the cognitive, emotional, and neurological symptoms associated with Long Covid, but this is far from settled science. As a recent viewpoint article in Science noted “The pathophysiological mechanisms are not well understood, although evidence primarily implicates immune dysfunction, including non-specific neuroinflammation and antineural autoimmune dysregulation.” The search for what causes Long Covid and its brain-related symptoms thus continues.