An Interview with Michael VanElzakker Ph.d
Van Elzakker’s hypothesis could explain several of the mysteries associated with chronic fatigue syndrome
Michael VanElzakker’s Vagus Nerve Infection Hypothesis (VNIH) for Chronic Fatigue Syndrome may be the most intriguing hypothesis to come along in the last twenty years. If it’s correct it could explain several mysteries including how a virus might trigger the disease and then seemingly disappear, why the Lipkin study failed to find an active infection, why cytokine studies have trouble finding evidence of the ‘never-ending cold’, why when antivirals work they often take so long to do so.
It’s raised a lot of interest. Now Michael VanElzakker ‘sits down’ and answers some questions about it.
What is your background? When did you get interested in this subject and start developing this hypothesis?
I am a neuroscientist. I mostly focus on posttraumatic stress disorder (PTSD), which I consider to be very different from CFS and a separate arm of my research interests (although there are many interesting overlaps between the views of CFS and PTSD within our culture).
A sick friend prompted his interest in this disorder
However, I got my bachelor’s and master’s at the University of Colorado – Boulder, at a time when the Psychology & Neuroscience department there was very focused on psychoneuroimmunology-related phenomena. Some of the many research programs there related to neuropathic pain, cytokines, and the vagus nerve. All it took was for someone who was thinking about CFS to be exposed to these different literatures and to start fitting them together.
I started thinking about CFS because I have a sick friend. She got sick back when CFS was viewed even more skeptically than it is today – I remember one MD referring to it as “yuppie flu.” I knew that my friend was not malingering – why would she be? She had to put her life on hold. It was pretty devastating.
The focus on the vagus nerve is simply because that organ is responsible for symptoms during “normal illness” that strongly overlap with CFS symptoms.
Testing the Theory
Apparently imaging techniques are not able yet to find localized infections in the vagus nerve. How far are they from being able to that? Is there interest in developing those kinds of techniques?
You reported that PET scans have shown ‘promise’ in identifying activation of the microglia – a prominent part of your theory. (The VNIT proposes chronic microglial activation causes the vagus nerve to send signals to the brain that result in flu-like symptoms.) It seems that we could settle the VNIH theory right now if PET scans showed clusters of infected areas around the vagus nerve. How effective are PET scans at doing something like this now?
I’ll answer these two questions together. As far as the effectiveness of PET scans finding local vagus infection: We don’t know. That’s part of what my group is trying to work on. There are a lot of technical problems that will require pilot testing.
Most PET scans are done of the brain. VE would like to do one further down in the vagus nerve area
One of the big problems is that PET scans cost thousands of dollars per hour. It’s difficult to convince funding agencies to give us money to pilot test a method so that we can even begin to ask questions about a hypothesis that may or may not be accurate. But I’ve got some really good people on my team and we’re working on it.
There’s interest because of this hypothesis, but most imaging of the vagus nerve thus far has been at the level of the brain, trying to understand the mechanism of vagus nerve stimulation for epilepsy or depression. We’re trying to image it farther down.
If someone made an animal model of the hypothesis, that would help raise interest. I laid out a protocol in the paper for creating an animal model; I hope somebody with a rat lab takes the idea and runs with it. I’m not jealously guarding these ideas, I put them in the paper in the hopes that other groups would work on them too.
You suggested that future CFS research use radiolabeled antibodies to localize clusters of speciﬁc virus types. This is done to find tumors. Is it possible to radiolabel antibodies so that they pick up clusters of infection in the body?
Yes, it is. But there are several problems to this with regards to the VNIH of CFS. One is that antibodies are specific, but CFS could be caused by a number of different pathogens. So, we could flood someone’s body with radiolabeled antibodies against HHV-6, but maybe in that specific case, their symptoms are caused by HHV-4 (Epstein–Barr).
Another problem is that some of the pathogens that might be most likely to cause CFS are found in the vast majority of humans. So, radiolableled antibodies against HHV-1 would find a signal in most people, but only cause CFS if the viruses are in a vagus (para)ganglion. And the vagus nerve is so highly branched, that could be all over the trunk. Another problem is that antibodies cannot always get inside ganglia, which are immunoprivileged. But despite all of that, I still think it’s a research program that is worth pursuing.
Herpesviruses apparently love to set up house in the sensory ganglia, but you suggest that antiviral drugs might have trouble getting to them and destroying them there. Why is that?
Antivirals may take much longer to work in ME/CFS because they have difficulty accessing the vagus nerve where the viruses are present
Similarly to how the central nervous system has a blood-brain barrier (BBB), peripheral nervous system ganglia are immunoprivileged. According to the hypothesis, the frequent failure of drug therapy and also one’s own immune system to eliminate infections within vagal ganglia and paraganglia is just like how some drug doses and antibodies do not penetrate the BBB.
You noted that behavioral/stress management therapies such as CBT are moderately effective in about 30% of people with chronic fatigue syndrome, and that CBT resulted in lower viral loads and improved immune functioning in HIV. Why would this be?
Stress causes a cytokine response. So, if someone who doesn’t like public speaking is giving a presentation, their immune system is generating a cytokine response. If such a person even thinks about giving a presentation, they are likely to generate a cytokine response.
People with CFS have an authentic reason to be concerned about any activity that requires physical exertion, and it’s called post-exertional malaise: worsened symptoms after exertion.
According to the vagus nerve infection hypothesis (VNIH), there is a physiological reason behind post-exertional malaise: Exercise provokes muscle tissue to produce the proinflammatory cytokine IL-6, which would then exacerbate the ongoing local cytokine response within vagus nerve ganglia or paraganglia. That’s the hypothesized mechanism behind post-exertional malaise.
The CBT practitioners in the infamous PACE study were focused on avoidance/fear of activity because they began with the assumption that CFS is psychological. They think the fear of activity itself is the cause of CFS; I’d say that fear of activity is justified, but like all fear, it can become dysfunctional. For the vast majority of their patients, CBT did not help. The three out of ten that found some slight improvement may have used CBT to figure out exactly what level of activity they should be worried about. So, the moderate improvements they reported in a minority of patients were probably related to stress reduction.
In patients with HIV, reducing something like stress that taxes the immune system is bound to help a little bit.
I understand that this is a really charged topic among CFS advocates, and there is a lot of misinformation out there. Just to be clear, cognitive-behavioral therapy (CBT) does not get at the root cause of CFS. CBT offers coping strategies and is not a cure. But I can’t think of a single medical condition that isn’t exacerbated by stress. CFS is no different. Having a chronic illness is stressful and it makes one’s life complicated and there’s a grieving process. CBT is for those parts of the illness. It’s intended to help people solve problems and to challenge dysfunctional patterns. If you’re seeing a CBT practitioner who views CFS as a psychologically-based illness and is approaching your CBT that way, fire them. Find someone else.
While CBT can help people with serious and chronic medical problems, it should be used as an adjunct and not a primary treatment. It would be crazy, for example, for a doctor to prescribe CBT instead of chemotherapy for cancer. But chemotherapy is a known, empirically tested treatment for cancer. CFS doesn’t have such a thing yet.
Behavioral practices like CBT that reduce stress can be helpful in immune mediated diseases such as HIV and ME/CFS, but are adjunct, not primary therapies
Without a cure, the next best thing is to focus on quality of life. I am very much focused on finding an explanation for CFS, which would then lead to a cure. I have hypothesized that CFS is a neurological illness triggered by a foreign pathogen infecting the vagus nerve. But the fact is that stress has profound impact on immune system function. CBT for CFS patients can reduce stress, which is one mechanism of action to improve symptoms.
I should also say – CBT sometimes gets conflated with graded exercise therapy as well. Some studies have combined these two techniques but they are not the same thing. In the paper I gave an example of a treatment regimen that included graded exercise therapy.
Again, to be clear, if the VNIH is correct and some combination of glial inhibitor, antivirals and vagus nerve stimulation can be used to quell symptoms, then and only then does it make sense to begin graded exercise therapy. At that point, the root cause of CFS symptoms has been dealt with, and the next priority is to deal with muscle deconditioning which is not an insignificant factor in ongoing symptoms.
I absolutely do not condone forcing still-sick patients to exercise if it’s making their symptoms worse. This should be obvious.
The heart rate variability evidence suggests the parasympathetic nervous system (vagus nerve) is under-activated in ME/CFS while the sympathetic nervous system is over-activated. The SNS activation, in fact, may be due to the PNS’s inability to rein it in. The increased heart rate at rest, for instance, could be due to be due to the inability of the vagus nerve to slow it down. In your theory, though, the vagus nerve appears to be activated by the localized infection. I’m a bit confused.
Is an infection contributing to the autonomic nervous system problems in ME/CFS?
This has to do with the fact that the vagus nerve is a mixed cranial nerve, meaning it has both sensory (afferent, or towards the brain) and motor (efferent, or away from the brain) divisions. Its parasympathetic influence over the body results from efferent activity; its function in detecting peripheral infection and triggering sickness behavior results from afferent activity.
However, terms like over-active and under-active are a bit too simplified – what matters is that the nerve is able to respond and signal appropriately, to be able to create a functional signal-to-noise ratio.
Researchers have been looking for cytokines in ME/CFS for decades. Sometimes they find them, sometimes they don’t. When they do find them sometimes research groups find similar cytokines and sometimes they don’t. The one constant is that they keep looking. You mentioned that lung infections are also not associated with increased cytokine levels in the blood. Are there many other infections like this?
Well, to be more accurate, it’s not necessarily that lung infections won’t show a cytokine response in the blood. It’s more that we cannot be certain to find a cytokine response from a local infection – that is, any local infection. If a lung infection were severe enough, you might find cytokines in the blood. Cytokine studies are quite prone to false negatives, and it’s a mistake to infer from a negative cytokine blood test that there is no cytokine response happening anywhere in the body.
In studies that look for cytokines in blood, there are 3 relevant questions:
- Is there any cytokine response in the first place?
- Did that cytokine response diffuse into peripheral blood?
- Did the method of detection work?
VE notes the difficulties present in finding a cytokine response in an infectious disease
The question we’re interested in is #1, however it’s a big assumption that the answers to #2 and #3 are “yes” when we infer from a negative blood test that there is no cytokine response.
Those of us who think that CFS is not psychologically-based tend to think there’s an immune dysfunction of some sort. People have been looking for cytokines because they are an obvious potential link between the immune system and CFS symptoms, but a lot of studies have ignored how cytokines work.
If a research group is unfamiliar with the cytokine literature they may have also made some easy mistakes in the cytokine assay – the actual lab methods for looking for these proteins.
For example, cytokines are relatively labile, meaning unstable. If someone who didn’t know any better stored their blood samples in a refrigerator instead of a -80° freezer, you can bet they did not find cytokines. If blood samples went through freeze-thaw cycles, the cytokines will also start to denature (break down). There are definitely a lot of really good researchers who have looked into cytokines, but the literature can get muddied pretty easily by bad studies. And because the symptoms are systemic, most people have been thinking in systemic terms (i.e., not thinking about a localized infection causing CFS).
In general, I’m skeptical of any attempts to find a “cytokine profile” for CFS or any other infectious disease. That doesn’t mean it can’t be done, but it’s difficult. Cytokine responses are very complex, they interact with each other and they change in daily and hourly rhythms. You could study one individual and not find a “cytokine profile” unless you took several samples a day for many days.
Response to the Hypothesis and the Future
This is a really intriguing theory. Kristin Loomis of the HHV-6 Foundation was excited by it. Have you gotten much response from it?
I’ve actually been really pleasantly surprised by the response. I’ve had the idea for quite a long time, and spent a lot of time and effort trying to set up a collaboration with a rat lab, to create an animal model. To make a very long and frustrating story short, nothing worked out.
The response to VE’s hypothesis has been very positive; he is working on putting a study together to test his hypothesis
I’ve been telling anybody who would listed about the hypothesis. It’s not like doors were getting slammed in my face, but most people simply didn’t have a background in the different literatures that the hypothesis ties together.
It wasn’t until recently that I discovered this unique journal, Medical Hypotheses, so I made some time to write up the idea. It gave me a forum to really give people the background they needed to understand the idea, and allows people to check the citations themselves. Based on past experience, I thought I’d have to keep cold-calling researchers to push the hypothesis. But it really took off right away.
I put it up online for free, and it’s been downloaded over 1000 times there; I don’t know how many people have downloaded it from the publisher through a university or hospital subscription. I’ve heard from researchers from 5 continents. Somebody translated the paper into Dutch and put it up online.
The week the paper came out, Anthony Komaroff contacted me, we’ve been in contact since. He finds the hypothesis to be “provocative and plausible” and shares my hope that functional imaging can help to shed some light on it. I’ve been in contact with a lot of other well-known CFS researchers, and I think the idea is at least changing the way that some people think about the problem.
Van Elzakker will be at the IACFSME conference with a poster presentation of his hypothesis
I also know that the paper is already being taught at some universities and medical schools, so hopefully it will at least get young scientists to start thinking about CFS. I hope people start to think about new CFS findings through the lens of this hypothesis because in my experience, it explains a lot of phenomenology.
Even if the hypothesis doesn’t turn out to be accurate, or is only partially accurate, I hope that it gets us closer to effective treatments that are actually attacking the root causes of CFS symptoms and not just helping people cope with them.
Some reports suggest you’re engaged in a pilot study. Can you comment on that?
On the record, I’d just say that I’ve put together a really great team to pursue the VNIH and that Dr. Komaroff is part of it. There are a lot of technical issues but we’re hoping to use functional imaging to gain enough preliminary data that we can pursue it further.