What Causes Migraine and What Can We Do About It?
We saw in the previous chapter that migraine is an all-encompassing brain and body event involving many of the pathways, brain structures, neurotransmitters and hormones that we have seen in other types of headaches. However, migraine is the dark symphony of all of these components working together. With so many factors involved, it stands to reason that migraine may be caused in many ways. Our understanding of who migraine affects and how it creates such malaise has evolved since the dawn of thought. If we look through the history, we will be able to build a picture of what kind of person was thought to suffer, and what it was about what they did that brought on such malaise.
Given that we have evidence that migraine was being experienced by humans millennia ago, you might think that we have a rich resource by which to work out what it is that makes us so unfortunate as to have our lives touched by this spectre. The demons of ancient times seemed to be indiscriminate; there was no thought given to the individual characteristics of those who were struck down by migraine. By contrast, literature dating from the 19th century, which is often dismissed or ignored by modern scientific and clinical literature, was prolific in its descriptions of the migraine attack, providing a vividness that there seems to be no space to provide nowadays. No one wants to reinvent the wheel, but we should also not ignore what that wheel looks like.
Ancient thought to modern times
In the past, there have been many theories about the causes of migraine and descriptions of the symptoms in literature. For example, the causes outlined by Hippocrates,1 the first to describe the migraine aura around 400 BCE, took two forms, the first of which was ‘humoral’. The humours in Hippocrates’ theory were the four bodily fluids: blood, phlegm, yellow bile and black bile, all of which needed to be kept in balance for good health. Illness was caused when these fluids became out of balance, sometimes requiring the reduction in the body of a humour through bloodletting or purging.
Several hundred years later, in the 1st century CE, the celebrated Greek physician Aretæus of Cappadocia put forward an alternative theory: ‘The cause of these symptoms is coldness with dryness.’ Sorted. So all we need to do to avoid migraine is stay warm and wet! If only it were that simple.
In the 6th century CE, Alexander Trallianus turned back to the humours and suggested that migraine may be caused by an excess of the yellow and black bile, collectively called the bilious humours. Anything that caused the movement of yellow and black bile to the stomach would result in tummy upset. Equally, constipation was seen as a preamble to migraine. Treatment was based on remedies that would purge these humours through the action of emetics that make you vomit, laxatives for defecation and bloodletting to, well, let blood.
This humoral theory wasn’t totally wrong: bile is indeed released by the liver and the gall bladder to break down fatty foods. However, Trallianus’s theory was wrong in that it erroneously categorised those who experienced migraine as those who ate a decadent, high-fat diet. Migraineurs were therefore encouraged to become sparse in their dietary habits, eschewing high-fat and -protein foods such as butter, meat pies, hot buttered toast, ales... To some, this meant they had to deny themselves anything good in their lives. Puritanicals such as the British Quaker John Fothergill of the 18th century, for instance, saw this as both salvation from his lifelong migraine but also validation of his abstinent ways and teachings.
The second cause of migraine was purported by Hippocrates to be ‘sympathetic’ in nature. This idea grew as humoral theory took hold in 400 BCE but instead of it being related to bodily fluids, it centred on a particular organ of the body, such as the stomach, the bowel or the uterus. From here, there would be some form of unconscious communication throughout the body, spreading the malaise. The Greeks called it ‘sympathy’ (the root of which can mean either ‘feeling’ or ‘disease’) and the Romans called it ‘consensus’ (an ‘agreement of the senses’).
The interesting thing about this theory 2 is that, early though it was, it connected the head, or at least the consequences that were found there, to something that happened in the body. This went against the grain of the prevailing common philosophy, which itself has had a lasting legacy. Not only did Aristotle, shortly after Hippocrates’ time, argue cogently (yet erroneously) that there was no way the brain had anything to do with the mind, but it also still wasn’t accepted even as late as the 17th century that mind and body could be linked. Instead, dualism was the flavour of the day – the philosophical position supported by the Frenchman René Descartes in the early 1600s that our behaviour is controlled by two entities: the mind (and it was still up in the air whether or not this had anything to do with the brain) and the body.
Based on the sympathetic theory, many of the symptoms of migraine, such as nausea or disturbed appetite that was often coincident with menstruation, could be explained. English physician Thomas Willis, who was active in the mid-1600s, advanced our understanding by marrying keen clinical observation with detailed anatomical investigation. 3 The language with which Thomas Willis described the migraineous episode is at once descriptive, metaphorical and gripping. Speaking of one of his patients (‘a most noble Lady’), who was ‘extremely punished with this Disease’; he described her ‘Distemper’ (at this time taken to mean ‘bad temper’) as ‘having pitched its tents near the confines of the Brain’ and ‘had so long besieged its regal tower, yet it had not taken it; for the sick Lady ... found the chief faculties of her soul sound enough.’ If I were to write a scientific paper with this kind of language today, I’d be butt-bounced before it even went to review; it would never be allowed. What a loss in the name of progress. Thomas had, however, correctly identified that even though a person could be completely incapacitated by a migraine, they are fundamentally non-fatal.
We find ourselves now dealing with theories that might point to the migraine’s cause. Is it a psychological issue? Is it something to do with a lack of constitutional strength? Is it behavioural, maybe because you like a bacon butty? Were you born that way? Thomas, for once, put it very succinctly and yet didn’t answer the question either...
‘...An evil or weak constitution of the parts ... sometimes innate and hereditary ... an irritation in some distant member or viscera ... changes of season atmospheric states, the great aspects of the sun and moon, violent passions and errors in diet.’
That pretty much covers everything, doesn’t it, as well as conflating predisposing and behavioural conditions. Let’s break this down a bit.
Is it me?
What does the migrainous constitution look like? It’s not like we can identify somebody who is prone to migraine walking down the street, although we might spot somebody if they were in the throes of one. Here, we rely on observational correlations. Which of the following clinical characteristics is true?
1 Migraine is much more prevalent in people under 1.6m (5ft 3in) tall.
2 A luxuriant head of hair indicates a migraineur.
3 Migraine is more widespread in women.
4 If you have an inverted nipple, chances are you get migraine.
5 Migraine is commonly seen in red-headed women.
True or not, all have been reported in the scientific literature. This illuminates the age-old problem we have in academic enquiry: the difference between correlation and causation. Correlating the greater prevalence of migraine among the bountiful of follicle does not mean that good hair growth causes it. On the other hand, we do know that women suffer more from migraine than men, but this is because of the hormonal link, which we will discuss later; there is a cause here related to female physiology.
Grace Touraine and George Draper, psychiatrists from New York, sought to define a classic migraine type in their writing in 1934 and settled on a person with some bulbous features about their skull, high intelligence but an unsettled emotional make-up. The hormone link gave the (mostly male) clinicians of the 1900s licence to restrict their comments, outside of their own experiences with migraine of course, almost exclusively to the female of the species and in particular with respect to their form and purported intellect. In 1959, American physician Walter Alvarez wrote about ‘Some Characteristics of the Migrainous Woman’ observing that such women dressed their small, trim bodies with firm breasts well, were eager of mind and were socially very attractive. They also seemed to age well. Who wouldn’t want to suffer from migraine if it meant we got to be the kind of woman Walter saw in his clinic?
Discussion of the psychological underpinnings was similarly conflated between observational correlations and causal attribution. Herman Selinsky, another American physician, writing in 1939, accepted that the headache of migraine was related to vascular changes, but he wanted to understand common psychological factors in his group of 200 migraine patients (with a ratio of four females to every male in the sample) with a view to developing a therapy in this domain. Herman argued that the migraine episode could be seen as a mechanism of escape for the patient from a bad situation, that it could be used to evoke sympathy and not the anger that the situation that brought about the migraine might more appropriately deserve. In his observations, Herman thought that the harassed housewife was very prone to migraine, particularly if they had intellectual leanings that could not be expressed through ‘the bearing and raising of children under adverse economic conditions, and the constant drudgery of housework’. They just had no ethical, moral or social way to release the pressure.
This might have made sense in its time but doesn’t really get us to the real cause of migraine, rather it tries to explain the function of migraine – what it was for, what it served to do. The best we can deduce from this kind of theory is that these situations may be coincident with migraine, but the population statistics of the time don’t stand up to the argument that every harassed housewife got migraine. What is it about the ones who do that made their stress manifest in this way? And was stress the only cause? The people whom Grace Touraine, George Draper, Herman Selinsky and others met in the 1930s tended to be:
‘....often of an intellectual type with a tendency to worry, rigidity of attitude, driving ambition, exaggerated sense of responsibility, marked sensitiveness to critics and meticulous attitude towards work or responsibility. Furthermore in women, there is usually impairment of ability to achieve sexual gratification.’ 4
In this framework, the migraine was seen as a psycho-physiological outcome and such leaps of logic made it difficult to define the underlying cause, and sometimes masked it. Having inverted nipples or red hair doesn’t cause migraine. Is there some unifying factor that can tie this together with a possible propensity for high intellect and the sexual dysfunction? Is it possible to separate clinical observations from their time and the social pressures that existed within it? Could it really be true that only intelligent people experienced migraine?
It took until 1971 and William Estlin Water’s epidemiological studies to debunk this myth. (The word epidemiology comes from the Greek epidemia, meaning the prevalence of disease. Epidemiological studies look at the incidence, spread and control of diseases and health disorders.) In his preferred sample, the population of South Wales, he didn’t find any evidence that those who suffered from migraine were more intelligent than those who didn’t, but he did see a hint that those who were of higher social standing and more intelligent were more likely to consult a doctor. After all, not everybody would have had access to a doctor before the 1950s, certainly in the UK, and so this skewed the view of John Fothergill and others throughout history that migraine only existed in the decadently fed, posh and clever domain. It turns out, migraine does not respect these factors at all.
The point of peas
A more biological approach is required and our eye will first fall on heritability as predicted by Thomas Willis back in 1654. The prevalence of traits, and of course illness, in families was noted in ancient times but it wasn’t until the late 19th century that Gregor Mendel formalised the science of what we now know as the field of genetics (from the Greek genesis, meaning ‘origin’).
Gregor was an Augustinian Friar who lived in Moravia (now in the Czech Republic), and over the course of seven feverish years, starting in 1856, he noted what happened when he bred and cross-bred pea plants that had different characteristics, such as their height, pea shape and colour, and many others. He realised that there were invisible ‘factors’ that determined what trait was passed on, and in what proportion in the offspring. For example, when a green pea plant was crossed with a yellow pea plant, only yellow pea plants resulted in the first generation. This is an example of a ‘dominant trait’: yellow always wins. But curiously, in the next generation, there are three yellow to one green pea plant, and so the green isn’t lost completely, it’s a trait that is lurking in the background. Gregor called this a ‘recessive trait’. He didn’t know what these invisible factors were, the ‘units of inheritance’ as he called them.
He published his results in 1866, which was very lucky because then it was in the public domain. If he hadn’t published, the knowledge would have been lost since all of Gregor’s papers were burned on his death in 1884 to end a tax dispute he had had with the local government in his capacity as abbot of the monastery. 5
These units of inheritance, or ‘genes’ as we now know them, were more clearly defined 30 years later when Mendel’s work was rediscovered. Sitting on the chromosomes, or genetic material in every nucleus of every cell, the genes determine what proteins are made. These proteins are the building blocks of every material that can be found in our body. As we touched on in Chapter 5, environment also has a role to play in this. For instance, if you have two genetically identical plants and you raise one with ample water and the other without, one will grow much taller than the other.
Another example is Dolly, the first cloned sheep. Born in July 1996, Dolly looked different to the ewe that donated the original cell containing the genetic material. Keith Campbell, Ian Wilmut and their team at the Roslin Institute in Scotland tried using lots of different body cells with their full complement of 46 chromosomes but they went with the udder one in the end. (See what I did there...) It was because this cell originally came from a mammary gland that the cloned sheep was given the name Dolly... 6
Dolly died when she was just over six-and-a-half years old, having had six lambs of her own in the regular way (Bonnie, twins Sally and Rosie, and triplets Lucy, Darcy and Cotton). That’s a bit of a short life for a Finn Dorset sheep, which is what she was; we might expect them to live until 11 or 12 years old. That said, it’s fair to think that because Dolly’s genetic, or first, ‘mum’ – a white-faced Finn Dorset – was six years old when she ‘donated’ her mammary cell that actually Dolly was six years old on the day she was born; there are a couple of age-related characteristics of her DNA that support this. However, Keith and Ian think that the balance of evidence falls elsewhere. Dolly had arthritis from the age of four, a condition that is age-related for sure, and unusual for a four-year-old sheep, but what killed her was a type of lung cancer called pulmonary adenocarcinoma. This was caused by a bug called Jaagsiekte 7 sheep retrovirus, which is quite common in sheep, particularly those kept indoors – as Dolly was for her own security. A retrovirus is a nasty bug because it gets into your cell nucleus and changes the DNA, replicating itself as the cell divides. Apparently, other sheep in her flock died of the same thing, as it is contagious.
Dolly’s ‘mum’ was the ewe that had donated the egg cell. All of the genetic material of that cell was removed; because it was a sex cell, it would only have contained 23 chromosomes (see here). And because it was a sex cell, it was the type of cell (called a stem cell) that is built to divide in the unlimited way necessary to create all the cells required to build a body and everything inside it. So, the DNA of Dolly’s genetic ‘mum’ was put into an egg cell from Dolly’s ovum ‘mum’ and the whole kit and caboodle was implanted into the uterus of Dolly’s birth-giving, or third, ‘mum’. Both the ovum ‘mum’ and the birth-giving ‘mum’ were Scottish Blackface sheep. When Dolly was born, the only lamb out of 277 attempts, she was a white-faced Finn Dorset, which provided evidence that it is the genetic material that determines breed. And yet the environment she lived in – a pampered indoor sheep palace, totally different to the living conditions of any of her mums – meant that she displayed various differences in how those genes were expressed.
Subfields of genetics
This interaction of environment with genetics is a subfield called ‘population genetics’ and can be an interesting way to view heritability of traits and particularly disease across evolution.
Epigenetics, on the other hand, looks at how genes and their activity might be altered, but without messing around with the DNA (as the retrovirus does) that codes for a particular protein. These factors sit on top of the heritable traits that Gregor Mendel identified and can be passed on down to offspring as well.
The study of whether we inherit migraine came a little bit late to the game. It made sense to anybody who thought about it that migraine is a family trait – it is incredibly common to find multiple people in one family who complain of it. The mechanism of this was unclear though. In the 1960s and 1970s various theories abounded: that it was linked to a recessive gene, or a dominant gene, or lots of genes, or (my favourite) that there wasn’t a clear genetic picture at all. The problem was, all of these studies relied on self-report of family trees of pain, and of course such a process is fraught with error due to misdiagnosis of other family members’ malaise and also a bias that existed because nobody wanted to be left out.
William Waters, who worked for the Medical Research Council’s epidemiology unit in South Wales, sought to separate all of these biases. He concluded, in 1971, that genetically determined heritability wasn’t as big a deal as those before him believed. For him, environment was a much bigger factor; you weren’t born with migraine, you acquired it somehow.
One gene to rule the world?
Perhaps again we need to turn to a finer-grained biological study. With all of our advances in our knowledge of genetics, mapping the genome, working out what specific genes do and playing about with their activity, we should have linked ‘the migraine’ to a specific ‘migraine gene’ by now, right? Well, instead of there being one gene to rule the migraine world, science has identified a number of candidate genes, changes to each of which raises the risk of migraine. Many of these studies have been done with people who experience migraine with a clearly defined aura for reasons of clearer diagnosis of the migraine headache (not to be confused with any other type). This means that there is the possibility that migraine without aura has a different genetic profile, but as we saw in the previous chapter, the same brain changes may be going on in somebody without aura, it’s just that they don’t experience them.
We know with some certainty that there is a particular form of migraine called hemiplegic migraine that is passed down through families. Patients experience extreme muscle weakness or pins and needles down half of their face or head or even their body (hence hemiplegia from the original Greek: hemi or ‘half’, plege or ‘stroke, plague or wound’) that can actually border on or tip into paralysis, similar to symptoms of a stroke. However scary, the effects are temporary.
Hemiplegic, or familial, migraine is linked to an autosomal (not linked to the sex chromosomes and so both the mother and father can carry it) dominant gene, meaning that you only need one parent to have the altered gene to pass it on. Even though you might have the gene, though, it isn’t necessarily expressed; in some cases no effects are seen at all. We still need to find out why this is so. What are the environmental triggers that turn this gene on? Are they linked to the regular migraine triggers we know about and will talk about shortly? There is a serious game of ‘join the dots’ to be done here.
When I was starting university, genetics research was the great hope both for understanding humanity and treating the diseases that affected them. Spotting genes that were responsible for cancer, neurological diseases or heart failure, for example, became a boon to the prediction of disease onset opening up the possibility that we could mitigate them somehow. The Human Genome Project was a Herculean international effort to map all of the genes that are found in the human body. It only started in October 1990 but a mere 13 years later, two years ahead of schedule and under budget, the entire human genetic code was elucidated. It’s amazing what we can do when we work together.
Up until now, for the most part genetic studies have focused on candidate genes that often were proposed as possibilities due to the coincidence of migraine with other disorders such as epilepsy, for example. Indeed, four candidates have been proposed to explain this shared genetic basis:
1 CACNA1A on chromosome 19 (makes calcium channels)
2 ATP1A2 on chromosome 1 (makes the sodium/potassium pump)
3 SCN1A on chromosome 2 (makes sodium channels)
4 PRRT2 on chromosome 16 (involved in neurotransmitter release at the synaptic cleft)
All of these are involved in the release of neurotransmitters and/or the balance of ions between the inside of the nerve cell and outside, the importance of which we saw in the previous chapter. If these abnormal genes are not creating the correct proteins, it will lead to problems with the excitability of these cells and how the signal is passed on from one to the next. As we know, these are problems in various degrees in both epilepsy and migraine.
By spotting these genes and following their inheritance down family trees, coupled with how the condition is manifested in these individuals, we can eventually get some kind of handle on the mechanism of how each altered gene leads to behavioural effects we can actually see. If we can understand the role and the power of these genes, then we can work to normalise them. We’re not quite there yet, but in theory we can create a normal gene in the lab and change the patient’s DNA so that it will forevermore express the normal gene as opposed to the altered one. This can be done using the trickster retroviruses to ferry the correct form of the gene into the cells of body, like the scorpion crossing the river on the back of a frog.
This so-called ‘candidate gene’ approach has a couple of methodological drawbacks, however. It is incredibly time-consuming because families expressing the altered gene and an associated condition have to be identified and followed, often over generations. Also, while precise for the families it follows, it is not holistic. There may be other genes outside of the candidate chosen that are just as important, that we may miss or, worse, what you are looking at may just have been found by chance and doesn’t generalise to other people at all. The sample size is typically very small and the environmental influence is varied to the point where it is difficult to truly know what’s going on. Lastly, while candidate genes are identified and followed, it is rare to have each of them followed in the same familial study to look at interactions between their actions. 8
Another way to define the genetic basis of a disease is to carry out genome-wide association studies (GWAS). Instead of focusing on one or a few genes at a time, looking for problems, we can now mix what we know about the genome with population genetics. What’s interesting about this approach is that it doesn’t start with a hypothesis – a theory that you are looking to either prove or disprove. The candidate gene approach does this: you might start off by saying ‘I think the PRRT2 gene is involved in migraine. Can I prove this?’ as opposed to the GWAS, which would merely ask a question like ‘what are the genes associated with migraine?’ Everything gets reported in this latter case – both the associated genes and the unassociated ones – whereas in the candidate gene approach, only when a hypothesis is proved does publication usually result. This is why I am such a fan of a question-led approach; anything can happen, nothing is off the table, and it is up to you as a scientist to design a good experiment and interpret the results in some kind of intelligent framework based on what we already know, or one that looks beyond to how we might understand something in the future.
It turns out that there are more than 40 genes that can be implicated in migraine. The inheritance of these genes is much higher in hemiplegic migraine than in other forms, and also seems to be higher in those reporting migraine with aura as opposed to without. Three of the four candidate genes for hemiplegic migraine have been confirmed using GWAS, with PRRT2 losing its credibility as a driving factor in the process. Interestingly, even outside of this most clearly hereditable form of migraine, there is a genetic basis for specific clinical features of other migraine that can be passed down. A strong family history of migraine incidence is associated with a lower age of onset, more frequent migraine episodes and migraine with aura. GWAS studies looking at populations of people who experience migraine without aura also shows robust genetic associations, but there are subtle differences with aura populations probably leading to their threshold of aura consciousness being different. Overall, these two migraine subtypes are more alike than different.
What’s the point?
So, migraine is in our genome, with at least one form being distinctly genetically heritable (remember, familial heritability takes the environment into account also). But this makes no sense. Migraine is a figurative pain in the ass and a literal pain almost everywhere else. What use is it to us as human beings in the 21st century?
The human body is adept at letting redundant features go throughout the course of evolution. We have no use for the appendix any more, since we no longer have the kind of diet it was useful for, since it helped us digest the foods we no longer eat so much of. We also have little use for wisdom teeth, those huge brutes that break through our gums in our teenage years (as if we didn’t have enough going on), again because our diet has changed. ‘OK,’ evolution said, ‘I take your point’; both appendices and wisdom teeth are being selected out of the genome as we speak. At least one in 100,000 people are born without an appendix, and they are only spotted because of surgery or scans for something else, they never knew, and so the number could be much higher. Up to 35 per cent of people will never cut wisdom teeth.
Other adaptations include the fact that kids are being born nowadays with better musculature and control over their thumb movements, to deal with the requirements of video games and swiping apps. These are the offspring of the Nintendo generation, who began to adapt to such movements, and because it was advantageous to do so this trait was passed down to the next generation. (I am of the Atari generation; I’m not sure any good came from that aside from a love of Pong.)
The bottom line of evolution is this: adaptations (either accidental ones like mutations, or experiential ones) that help our species survive and make life easier and/or more efficient will eventually be passed down. Altered genes or mutations that are not helpful will not. So why do we still have migraine?
We might make the same argument for depression, the most common affective mood disorder out there. Our moods are complex but are generally driven by two main aspects of thinking behaviour; we all use varying levels of a problem-solving approach and a more reflective brooding approach. A greater propensity to the brooding approach is coincident with the onset of depressive symptoms. Given the genetic basis of personality type and thinking style, one might think that given the risk of a brooding style to the onset of depression we should have selected that out of our genome already, because there is nothing helpful about depression, right? Wrong. I recently supervised a PhD student called Yan Birch, who was puzzled by this from his undergrad days. He did a series of experiments that showed that a brooding, thinking style was really important for certain thinking tasks. Sure, a more problem-solving approach was important in some other tasks, but brooding allows us to bring extra neural resources to the table, allowing us to dwell on problems we find it hard to break down.
Of course, at some point, this can become maladaptive – we use up a lot of our neurotransmitters like serotonin in the process of the work our brain is doing and if we are not replenishing them as we go through our behaviours and interactions with others then we can fall into an emotional hole with increasingly negative thoughts. However, the critical point is that we need this kind of thinking to solve problems, so it would be detrimental to us to have selected our ability to do it out of our genome.
So, is there any evidence that we need migraine? For example, it could be that migraine has a beneficial side effect – as is the case with the mutation that causes sickle cell anaemia, which also gives you protection from malaria – although it appears that there’s nothing so striking going on. We could also look at this from a behavioural evolutionary perspective, as Herman Selinsky did, back in New York in the 1930s, in effect saying that migraine is required to save the harassed housewife. (A pretty extreme measure I would have thought, even in 1939.) However, his theory comes from too high up in the behavioural tree for me to be happy for it to explain why migraine exists. The vasodilation that happens after the constriction in the migraine pathway could be seen as protective and this causes at least part of the headache, but the migraine experience is so much more. Why are our neurons going bananas in the first place? What’s the value of that?
The crimes of fashion against humanity
We have some tantalising evidence – and the first factor lies in the visual system. People who experience migraine have a much more excitable visual cortex than others. If other parts of the brain weren’t involved in processing the visual picture and we just had this area, called V1, we would see the world as a group of lines of every orientation there is. This is because the neurons that are found here respond only to lines, with each cell only firing off an action potential if there is a line of the right orientation appearing in exactly the place in space that neuron is tuned to. You have millions of these neurons, and they all add together successively over the rest of the visual system (with such imaginative names as V2, V3, V4 and V5) to build up the picture and detect edges and movement and overall contrast.
The role of V1 is straightforward. We test this in the lab by giving people very simple visual search tasks to do. For example, you might have to pick out a / from a load of \ forms for a really low-level task that would just occupy V1, or you might have to find a particular car from other pictures of cars, which would require higher-level processing. Back in 1995, Shirley Wray and her colleagues from Harvard showed that migraineurs, who rather crucially were not experiencing a headache at the time, were much faster than control people at the V1 task, but that both groups were as fast as each other at the higher-level task. In the early 2000s, Ed Chronicle from the University of Lancaster and his team showed that this advantage was linked to hyperexcitability of V1, which is similar in those who experience aura and those who don’t. This not only explains the extreme photophobia migraineurs experience during the headache but it also seems that even outside of the migraine episode these neurons are on a hair trigger for activation.
Opticians are already well aware of how much more sensitive a migraineur’s peripheral vision is. We can prove this in the lab by sending magnetic pulses into your brain to generate the phosphenes or spots of light we talked about in the previous chapter. It takes much less stimulation to generate a phosphene in somebody who gets migraine than other people. There are lots of things that affect the reactivity of your brain to stimulation like this (as I know after spending a gazillion hours in the lab, many stimulating myself for scientific kicks [this sounds bad, the insertion of the word ‘scientific’ didn’t help at all]) but the lower threshold for migraineurs is there no matter what. In addition to behavioural tests (e.g. the visual search one described above) and transcranial magnetic stimulation, you can also detect this difference using a visceral response. Flash up a picture of an Indian rug featuring lines going in lots of different directions and the response from your audience at the regional Migraine Society meeting will tell you all you need to know. In addition to not making any friends, I learned some new curse words from that audience. Tough crowd.
This can happen much closer to home and is often something people don’t think about. Venetian blinds? Invention of the devil. If you have an excitable visual cortex, just imagine the torture you are putting it through if you are surrounded by venetian blinds in your house, activating all of the neurons selective for horizontal lines all at once. These bladed blinds were actually invented in Persia, but were introduced to Europe by the Venetians in 1760. The French resisted the credit belonging to Venice though and still call them Persian blinds (Les Persiennes). As a migraine sufferer myself, I have a different (but impolite) name for them; in fact, if hell is a place, it is shaded by venetian blinds. By the end of the 19th century they were everywhere, in homes, churches and courthouses, even being present to witness the signing of the American Declaration of Independence in 1774. Immortalised in art by American Impressionist Edmund Charles Tarbell, his picture, descriptively named ‘The Venetian Blind’ (1898) depicts the back of a scantily clad woman in a recumbent pose in front of a venetian blind. I’m not surprised; she probably had a killer migraine. Perhaps actually the woman was Venetian and she was blinded by the light in her malaised state. Or perhaps I am reading too much into it…
Yet we are surrounded by lines so much so in fact that we really don’t notice them most of the time. But sometimes, they will rankle. About a year ago I had a really bad migraine that I just couldn’t seem to shift – it kept coming back over the course of two weeks. I had my eyes tested. I even went to the dentist to make sure I didn’t have anything going on in my teeth. It turned out that my wife had been shopping and had taken to wearing lots of stripy tops, some of them with tightly packed black-and-white lines. Well, you may as well have given me arsenic. I forbade her to wear them in my presence again and my headache went away. I am used to sharing meetings with men in stripy shirts but I am not contractually obliged to give them the whole of my visual attention, and so that doesn’t affect me so much. Marriage, on the other hand, is different. 9
It’s not just lines that the hyperactive visual cortex is sensitive to; it can also be flashing lights and strip fluorescent lights that are slightly out of phase with each other. These are commonly found in clothing stores – another reason why I don’t like to shop. Migraineurs can detect a much more subtle flicker in strip lighting than the rest of the population.
The hyperactive visual cortex can be linked to abnormal neurotransmitter balance, in particular a lack of inhibitory control. And the migraine features, namely the wave of excitability that starts off the cortical spreading depolarisation followed by the depression, are triggered by a certain spatial frequency, or how close together the lines are. That’s why not all patterns of lines around us set it off, and it’s fair to say that every migraineur might have a different spatial frequency that triggers them. Activate every single V1 neuron by presenting lines of every orientation packed together in a neat rectangle, though, and you will trigger all sufferers.
But back to evolution. The value of having a very sensitive primary visual cortex lies in your ability to see very small contrasts and other differences in the visual scene. This will have been of value in our hunter-gatherer past, for picking out movement of potential predators or prey in tall grass or trees. Also, because your V1 is very sensitive, you will be able to maximise the signals coming from your retina, making your vision more acute in dim light. By this view, migraineurs are evolutionarily more advanced. Said no migraineur, ever.
Other badges in biology
In other news, there is a lesser incidence of alcoholism in people who experience migraine, but this may be due to the individual protecting themselves from alcohol-induced headache and not really show an evolutionary advantage at all.
On the other hand, Knut Hagen followed 70,000 people in Norway over the course of 10 years and found that those with a diagnosis of Type 1 diabetes had a much lower incidence of migraine. There may be a genetic link here, or a behavioural one in that Type 1 diabetics tend to be more controlled about what they eat. In this study, no link was found between Type 2 diabetes and migraine but a French 10-year study did find that women were up to 30 per cent less likely to develop Type 2 diabetes if they got migraines. Although these migraines were self-reported and therefore sensitive to misdiagnoses, this number is too high to ignore; migraine seems to be protective against developing Type 2 diabetes. Again, this might be because those who suffer from headaches are much more careful around foods thought to be migraine triggers, thus perhaps having healthier eating habits, and so lessening the chance of developing diabetes. Or perhaps if we turn this on its head, there may be something about elevated blood sugar that stops headaches from happening and the chocolate (and carbs) that we crave in the prodrome phase may be self-medication with sugar and is not particularly serotonin-related after all.
Alternatively, there may be a link with proteins that may be found in abundance in migraine but not diabetes. One such candidate is calcitonin gene-related peptide (CGRP). This is released in great quantities as part of the cortical spreading depression we came across in the last chapter, but is actually produced to a lesser degree following the onset of diabetes. It’s a protein that is important in the inflammatory response in the tissues of the body and part of this is the proper regulation of fuel to the brain and other tissues of the body. This may be the link between migraine and diabetes. Too much CRGP leads to pain, too little leads to high blood sugar. And one precludes the other.
Knowing this means we can target CRGP as a new treatment for migraine. If we block the receptors on which CRGP works, we can cut off the inflammatory response and so the pain signals. This is the culmination of work Peter Goadsby from UCL and his colleagues have been doing since the early 1990s. Called Erenumab, it is injected below the skin and improved migraine incidence in 30 per cent of patients in the test group, but only 14 per cent in the control group. The actual improvement seen over the placebo effect is a 16 per cent decrease in incidence. It offers another mode of treatment for those in whom acute treatment such as sumatriptan doesn’t work. (You might remember sumatriptan as being a good treatment for cluster headache. It is a serotonin agonist acting just like serotonin in the brain and also constricts the vasodilation that causes the pain in migraine.)
We’ve talked before about how we might start to try to self-medicate with chocolate and sex to boost our serotonin levels and so if not head off (sorry) a headache at least treat ourselves in some other way. But it turns out that we do this already, all of the time! Tim Houle looked at sexual desirability scores in a sample of Chicagoans who suffered from either tension headache or migraine. He found that men were 24 per cent more interested in sex than women (not hugely surprising) but that women who experience migraine are right up there with the average man for sexual desire. Overall, migraineurs desired sex 20 per cent more than those who suffered from tension headache. Serotonin is therefore a big factor in the migraine mix (but not tension headache), explaining why taking a serotonin mimic works for some migraineurs, as well as people with cluster headache.
In fact, migraineurs tend to have lower serotonin levels than the rest of the population; genome-wide association studies reveal that migraine shares the same genetic variant risks as depression. There is a higher co-incidence of migraine with depression and the underlying pathway, at least in part, relates to this serotonin imbalance. Selective serotonin-reuptake inhibitors that make serotonin hang around in the synapse for longer may help raise and stabilise serotonin levels in migraine as they do in depression. However, behavioural mechanisms such as social connectedness, fun, satisfaction and love all work to sustain them over time.
The menstrual migraine
For women, there is a further clear hormonal pathway to trigger migraine. Three times more women than men suffer from migraine and of these women, 70 per cent experience menstrual migraine. It’s the fluctuation of hormones that is the culprit here. The menstrual cycle is roughly 28 days long and has four distinct but overlapping phases. Day one is counted as the first day of menstruation when all hormones are at their lowest levels. The follicular phase starts on the first day of your period and lasts right up to ovulation. In this phase, the hypothalamus (the puppet master of the endocrine system) kicks the pituitary gland into gear to release follicle-stimulating hormone (FSH), which does exactly what it says. About 20 follicles, or cyst-like nodules, each carrying an immature ovum or egg, form on the surface of the ovary. The developing follicles release oestrogen to coincide with the end of menstruation, which encourages a thickening of the lining of the uterus in preparation for receipt of a mature egg. The hypothalamus detects the increase in oestrogen and in reply it prompts the pituitary to release luteinising hormone (LH) – from the Latin luteum for ‘egg yolk’ – and another burst of FSH. The levels of all three – oestrogen, LH and FSH – peak just before ovulation, when a mature egg 10 bursts out of one of the follicles and is swept up by the oviduct (or Fallopian tube, named after Gabriello Fallopio, the Italian anatomist who described them in the 16th century), which transports the egg to the uterus.
While oestrogen, LH and FSH levels are falling following ovulation, the burst follicle gradually seals itself off, transforming into a structure called the corpus luteum. The corpus luteum releases progesterone and a little bit of oestrogen. Both of these maintain the thickened lining of the uterus, waiting to see if a fertilised egg will implant in its folds. If it does, then the corpus luteum sticks around and keeps up the release of progesterone and oestrogen. But if not, it just withers away by day 12, just like the unused follicles at ovulation do. Progesterone and oestrogen are no longer released by the dead corpus luteum and so concentrations fall, meaning the lining of the uterus falls away by day 28, starting the menstrual cycle all over again. Menstrual migraines are mostly linked to the last few days or the first few days of the menstrual cycle, a time when hormone levels are fluctuating from high to low and back again.
More evidence that fluctuating oestrogen levels are implicated in the incidence of migraine comes from Simone Ferrero, a gynaecologist from the University of Genoa. She and her team found that women with endometriosis are more than twice as likely to suffer migraine as women without it. Endometriosis is when the lining of the uterus, the endometrium, forms outside of the uterus itself and sticks to organs and tissues in the abdominal cavity; it can be incredibly painful and debilitating and a greater propensity for migraine is all the poor patient needs. Some 13.5 per cent of the women in the endometriosis group experienced migraine with aura, in comparison with 1.2 per cent of the control group who did not have endometriosis.
Coupled with the fact that endometriosis presents with high oestrogen levels, it may be that there is a causal link between oestrogens and cortical spreading depolarisation and the threshold by which individuals are conscious of its perceptual effects. Low oestrogen seems to be linked to migraine without aura and so the role of oestrogen may lie in how we actually experience the aura. Oestrogens can certainly play with neuronal excitability in lots of ways, and they interact with the blood vessels of the brain. In the case of endometriosis, though, there are added factors causing pain, such as the release of prostaglandins and nitric oxide, which are both part of the inflammatory response and direct stimulators of the trigeminal nerve.
There is some evidence that migraine is related to lower levels of gonadal hormones such as oestrogen, progesterone and testosterone in both men and women, but of course, these have a more cyclical role to play in women, increasing the prevalence of their effect. Since all of these hormones are ultimately controlled by the hypothalamus, and if we put this together with the hypothalamic symptoms of the prodrome phase we discovered in Chapter 6, there is of course the chance that hypothalamic dysfunction or, more specifically in this case, underactivity of the hypothalamic–pituitary–gonadal axis could be the root of all of this evil.
The use of external hormones might ease symptoms here as they work to stabilise the hormone concentrations. Because they stop the fluctuation of hormones necessary for the normal conditions whereby an egg is released by setting oestrogen and progesterone to higher-than-normal levels, it removes the possibility of conception. You might know these drugs better as oral contraceptives. The initial dose of oestrogen might lead to migraine to start with (due to the link of high oestrogen with migraine aura) but this tends to stabilise over time.
As we know, the fluctuation of oestrogen from low to high and back again causes various psychoactive effects; it can affect our mood and even our cognitive function. German psychologist Markus Hausmann and his team have investigated various cognitive tasks across the menstrual cycle and have concluded that these changes in hormonal concentration affect the brain such that more of the brain gets involved and functions that would usually require only one side of the brain now include both. Remember Ed Chronicle’s view that the hyperactivity in visual cortex in migraine is due to a lack of normal inhibition between the two hemispheres? Put this together with how hormones change the balance of activity across the hemispheres and we begin to get a sense of why our hormonal milieu might set the scene for the migraine episode. Markus has essentially found that these hormone-induced changes in brain activity in women do mean that they find it hard to engage the spatial perception required to parallel park. But don’t get carried away, fellas, the effect is very limited to a specific and relatively small window in the menstrual cycle. Quit your generalising.
The heart of the matter
There is another, perhaps surprising bodily function that may cause migraine, and it lies in your heart. This is a pretty vital organ, which takes all the deoxygenated blood from your body into its right chambers and redirects it straight to the lungs to become oxygenated again. The left side of the heart gathers all of this lovely oxygenated blood and redirects it around your body. It’s a pump, and it can beat faster or slower depending on the demands of your body for blood to various areas, and that’s governed by the autonomic system.
Before you were born and you were in utero, even though your lungs were developing, you didn’t breathe through them (if you did, you would have got a lungful of amniotic fluid and that wouldn’t be good). So even though your circulatory system was developing to prepare you to breathe normally when you were born, there wasn’t really any point in the blood being diverted to the lungs because there was no oxygen there to collect. You got all your oxygen through the umbilical cord from the placenta; this is where the transfer of oxygen occurs, instead of the lungs. This problem is taken care of quite easily in the developing foetus by having an open passage between the top chambers (called the atria) of the heart, which are usually separated by an impenetrable wall called the septum. Blood comes in from the body in the regular way on the right side but then instead of being redirected to the lungs, it passes through the septum of the atria using a gap called the foramen ovale (literally an oval hole). The left side of the heart then redistributes this blood around the foetal body.
After birth, the foramen ovale is sealed shut in 75–80 per cent of people. But in some people it is left open or ‘patent’, leading to the condition’s name: patent foramen ovale. It can be left completely open, requiring surgery at birth, or it can be incompletely closed, by a flap of tissue in the atrial septum. This doesn’t always cause symptoms and you might never know that it’s there, but when pressure is created in the chest by coughing or sneezing the flap can open. This means that at that moment, blood can flow in either direction between the right and left atria. The problem isn’t so much to do with oxygenated or deoxygenated blood ending up in the wrong place if the flap only opens rarely. The issue for the brain’s purposes is what else the blood contains. If blood passes from the right atrium to the left atrium and is distributed around the body, then it means it hasn’t passed through the lungs yet. This is important, because as well as oxygenating the blood the lungs act as an important filtration system for the circulating blood, removing bits of debris, such as small blood clots, for example. Although the kidneys do this too, it is the primary function of the spleen, making it an incredibly important (and my second-favourite) organ in the body, even if it is the first thing to be removed in medical dramas on television.
Essentially, there is a direct route for unfiltered blood to get to the brain. Once there, the cerebrovascular system’s arteries (that carry oxygenated blood) branch off into smaller arterioles and tiny capillaries where any of this debris can get stuck, and if it does it stops the blood flow to that part of the brain, directly affecting its activity. In a worst-case scenario, these neurons die due to the lack of oxygen and nutrients, and this is the basis of both transient ischemic attacks (where blood flow in the brain is temporarily blocked) and also its scary big brother, ischemic stroke (where the blockage is longer-lasting and causes brain damage).
You can see how this phenomenon might link to migraine. A small piece of debris blocking a blood vessel in the brain will not only alter the activity there but will also set up rebounds both with respect to compensatory overactivation of the neurons in the brain (setting up the wave of excitation underlying cortical spreading depression) and rebound surrounding inflammation and vasodilation to cope with the ischemic area. However, it took until 2005 to correlate the coincidence of patent foramen ovale with people who experience migraine. Markus Schwerzmann and his team from Bern, Switzerland, compared 93 migraine patients with 93 controls. They found that 47 per cent of the migraine group had a patent foramen ovale in comparison with 17 per cent of controls, and while the control ovales tended to be small, the migraine sufferers were more likely to have mid-sized or large gaps with a right-to-left flow between the atria. None of these people know that they even had this condition – many don’t until they suffer a stroke or other medical emergency.
So, now we can look for patent foramen ovale as a possible anatomical cause for migraine. Once this is known, treatment becomes more straightforward. Beta blockers, which interact with the autonomic control of your heart – meaning your heart beats more slowly and with less force – have shown some success. Blood thinners can help, although I always think these are poorly named. The likes of aspirin or warfarin don’t actually make your blood thinner or break up clots, but they can stop you forming new ones and slow the growth of the ones you already have. Anti-coagulant is a much more representative moniker.
Warfarin and its uses
Warfarin, which we’ve been using for more than 60 years, stops the formation of vitamin K-dependant clotting factors in your liver. It was first discovered after herds of cows died in North America and Canada from a bleeding disorder that happened either spontaneously or through nicks and scrapes. In 1930, Lee Rodrick from North Dakota realised that this had to do with an anti-coagulant that was made when sweet clover went bad, and that all the cows who had died had been given mouldy silage made from clover.
Ten years later, Karl Link and his student Harold Campbell from the University of Wisconsin – Madison isolated the chemical that was causing the complete breakdown of the clotting system and called it 4-hydroxy coumarin. Ten years after that, the reality that coumarin could be used as a biological weapon against rats sank in and warfarin was first developed, getting its name as a hybrid of the funders of the work (the Wisconsin Alumni Research Foundation, WARF) and the chemical it comes from (coum-arin). By 1954, its medical value to humans was defined, thankfully in much lower doses than are contained in the rat poison. It remains the most widely used anti-coagulant in the world, irrespective of the side effects of bleeding that can occur.
It is, however, possible now to heal the incompletely sealed flap that may be causing all the problems. In 2007, cardiologist Michael Mullen and his team at London’s Royal Brompton Hospital developed a bioabsorbable (something that can be absorbed by living tissue) patch that acts as a temporary plug, allowing the body’s healing response to cover it over and replace it with normal healthy tissue. This only takes 30 days to do in the body – all it needed was a bridge. It’s an improvement on the previous grafting procedure, which often led to inflammation problems, because it was permanently present and seen as a foreign object by the body. The best thing of all is that the patch is carried into a body through a catheter, a flexible tube that wends its way through the vascular system from the groin, where it is inserted. The operator can see where they are going as the catheter also contains a tiny camera that beams back live pictures from inside your body. I’m guessing you can tell how fascinated I am by this. When the catheter gets to the right place in the heart, it deposits the patch and your immune system does the rest!
Why migraine, why now?
The reasons why people get migraine are myriad. So far, we know it can be baked into our biology through genetic factors and that these may (or may not) affect how excitable our visual cortex is, how we produce sex hormones, how many inflammatory proteins like CGRF we have floating around our systems and the anatomical development of our heart. But what about the role of environmental factors like tiredness, stress or diet? Well, these aren’t so different to triggers for other headaches. The difference lies in how you and your brain, as a migraineur, react to them.
Migraine is a neurovascular headache. Triggers can affect the neural activity of the brain directly (as stripy lines do) or may have an impact on the vasculature that indirectly affects brain activity, both generating the specific migraine experience. The fact that there is such a thing as a special ‘migraineous brain’ determines why everybody doesn’t get them.
Treat yourself right
We saw in the previous chapter that there has been a tremendous amount of misunderstanding about so-called ‘trigger foods’; foods that our hypothalamus encourages us to eat in the prodrome phase don’t actually trigger the migraine, it is just our brain yanking our chain. You have an urge to eat chocolate, eat the chocolate! But are we throwing the baby out with the bathwater? Are there some foods or dietary habits that really can cause the migraine experience? When I talk to migraineurs, they often mention that missing meals or fasting can invariably lead to a migraine. This stands to reason and is not always restricted to migraine headache; it happens in tension headache, too. This is because vasodilation will occur to maximise the delivery of glucose to the brain if concentrations are low, but of course, as we now know, this will have specific effects in the migraineous brain.
Dana Turner, who works in North Carolina with Tim Houle (who, you might remember, was interested in how interested migraineurs were in having sex) and others, asked 34 migraineurs who experienced at least two headaches a month with headaches present between 4 and 14 days in the month to keep a diary for six weeks. When they looked in detail at the diary days that followed a non-headache day (to reduce any changes in behaviour that might have been induced by the headache) they found that night-time snacking resulted in a 40 per cent drop in the odds of experiencing a headache as opposed to having no food at night. Eating a late dinner resulted in a 21 per cent decrease in the chances of a headache developing with respect to having no food, but the difference here wasn’t statistically significance (and so could have equally been due to chance).
The snacking finding was significant though. Putting this finding together with what we know about migraineurs having more inflammation-causing calcitonin gene-related peptide (CGRP) floating around their systems, we might suggest that utilising that CGRP in its fuel regulation role gives it less chance to get involved in its inflammatory activities. This fits well with the protective effect of migraine for diabetes (people who get migraines are much less likely to develop diabetes); high blood sugar might stop headaches, low blood sugar causes them.
I spend a lot of time debunking the chocolate myth with migraineurs, but they will often cite other triggers such as cheese, Chinese food and processed foods. Tyramine is a unifying factor here as it is present in many ripened and aged cheeses such as Camembert and Brie but also soy sauce, miso, cured meats and fish.
Tyramine is a simple neurotransmitter called a monoamine – just like serotonin and dopamine are – that regulates our blood pressure by causing vasoconstriction. Too much of it, or indeed too little of the monoamine oxidase that breaks it down and removes the excess from our bodies (which leads to too much tyramine in our systems) can lead to vascular changes. Given how sensitive our brains are to vascular changes, and migraineous brains in particular, this might explain why these foods cause headaches.
What’s more, monoamine oxidase inhibitors (MOIs), which are often prescribed for depression since they stop the breakdown of mood-affecting neurotransmitters such as serotonin and dopamine, will also inhibit the breakdown of excess tyramine. For this reason, eating a meal that’s high in tyramine while taking MOIs can lead to serious hypertension (high blood pressure) because of overconstriction of the blood vessels.
The MSG argument
A further link with Chinese food is the presence of monosodium glutamate (MSG), a substance that is found in lots of other cooking and products too, and naturally in mushrooms, seaweed, tomatoes and soy, and also Parmesan cheese, among other things. The additive form was first developed in 1908 by a Japanese chemist called Kikunae Ikeda at Tokyo Imperial University, who was fascinated by the factors that gave his food flavour. He noticed that if kombu, a type of kelp, was added to the broth, it made his soup taste delicious. Some further investigation led him to identify the fifth human taste, umami, and that what those taste receptors are detecting is glutamate, a building block of proteins. As Kikunae pointed out, we have no doubt developed a taste for glutamate because it indicates the presence of vital proteins we should be ingesting to keep us alive.
Glutamate itself doesn’t have the umami flavouring but it activates the glutamate receptors in the taste buds in the mouth, which our brain detects as a savoury meaty flavour. In effect, it adds punch to somewhat tasteless dishes. However, it took 100g (4oz) of dried kelp to isolate 1g of glutamate through a very long and convoluted process, so for his next trick, Kikunae set about trying to make this easier for the purposes of home cooking. He needed something with the physical characteristics of salt or sugar to add to stock bases and the like. With this template in mind, he looked for chemicals to buddy up with the glutamate that would be granular and robust to moisture and humidity but be soluble in water. The isolated glutamate by itself looks like little brown crystals and are very powerful; distribution within the bond of another chemical would lower the concentration and make it much easier to work with. Sodium was the ideal candidate and the resultant bond between one molecule of sodium to every glutamate molecule became the salt-like monosodium glutamate. Kikunae knew he’d cracked it and called the new seasoning Ajinomoto (味の素) or ‘essence of flavour’. It duly went into mass production in 1909 using a more efficient method than the kelp procedure (it now involves wheat and soybeans) and today, the Ajinomoto Company, Inc. employs more than 32,000 people in 35 countries.
MSG is down as a migraine trigger in the current version of International Classification of Headache Disorders. The problem is, the evidence is ropey beyond feedback from migraine sufferers, and the mechanism by which it might cause headaches isn’t very clear at all.
The controversy started in 1968 when Robert Ho Man Kwok wrote a short letter to the New England Journal of Medicine that was subsequently published. At the time, he was a medical doctor working as a senior research investigator at the National Biomedical Research Foundation in Maryland. He explained that he had regular symptoms following his visits to Chinese restaurants since his arrival in America, experiencing ‘numbness in the back of the neck, gradually radiating to both arms and the back, general weakness, and palpitation.’ He discounted the high salt content as well as the cooking wine and soy sauce (which was weird, considering the fact that the latter contains both tyramine and glutamate) as triggers, saying he cooked with them at home and they didn’t have any effect. He suggested that perhaps it was the monosodium glutamate that is used liberally as seasoning in Chinese restaurants, particularly cuisine from the north of China, and called other doctors to arms in the investigation of this theory. Because he titled his letter ‘Chinese Restaurant Syndrome’ he not only coined a new pejorative phrase (it was later changed to ‘MSG symptom complex’) and validated the malaise but he also sparked off years of study and anti-MSG sentiment, forcing Chinese restaurants to advertise that they don’t use MSG in their cooking. How did a speculative letter gain such traction in America and across the world so fast?
Published by the Massachusetts Medical Society, the New England Journal of Medicine is the oldest and certainly one of the most prestigious medical journals in the world. Its weekly editions contain articles and state-of-play reviews in addition to a letters section that continues to garner lively debate. Its impact factor, which is a measure of how respected it is through how much other people cite its papers, is over 79 points, dwarfing the mere 53 The Lancet scores (which is pretty huge too). It’s A Big Deal and the journal has historically seen itself as the arbiter of the acceptable – indicating what the scientific community should be interested in. There was a tradition at the time of comic syndrome letters appearing alongside more strait-laced ones in the letters pages of the New England Journal of Medicine, often expounding on the phenomenon of common problems with overly scientific, pretentious and clinical language, seemingly just for fun but presumably it floated the boat of the Letters Editor. There was even a letter on Cryogenic Cephalalgia, which all of us reading this book now know as brain freeze, and others on French Vanilla Frostbite, Space Invaders Wrist or Credit-Carditis. This would seem to be how clinicians at the time got their kicks, but in re-reading them, some of them strike me as primers to nascent but possible societal conditions.
In response to Kwok’s letter there were many replies, as many completely endorsing the syndrome as there were pooh-poohing it. Then the media, including as respected an outlet as The New York Times, got hold of the controversy and the snowball rolled on from there. All argument on the matter soon superseded the role of MSG, appropriating the main thrust of the problem to Chinese food, which was entirely stupid as MSG is found in everything from nature to crisps.
One respondent, Herbert Schaumburg, a pharmacologist, actually was true to his word and went on to intensively investigate the effect of MSG on human health, designing his tests in collaboration with a neurologist called Robert Byck. They chose to inject MSG in large quantities into 13 people, and this, quite understandably, didn’t go so well since MSG is ordinarily ingested and not injected. They did also give MSG orally to participants and noticed symptoms such as burning, facial pressure, chest pain and headache (bear in mind, the original letter never mentioned headache explicitly), although people had vastly different sensitivities to MSG. A big problem with the experiment was that it wasn’t blinded (the subjects knew what was being tested), which is a bit like having a biased juror when you are being tried for a crime; in this case, MSG was in the dock. There were also some very strange experiments that involved injecting huge amounts of MSG into baby mice and monkeys, because that’s representative of the human experience, not, showing that the mice and monkeys grew up with serious impairments. Well, yes, as glutamate is the most ubiquitous excitatory neurotransmitter in the nervous system, this is hardly surprising.
By 1970, lots of studies were appearing debunking the earlier human studies, but the horse had bolted: it was in the public domain and the public were worried. So much so in fact that in the early 1970s Ralph Nader, a consumer activist and much later a US presidential candidate (in the Bush/Gore/Nader election in 2000), lobbied Congress to ban its use in baby food. This is highly ironic given that glutamate is present in breast milk. Good luck with that one, Ralph.
Since then, loads of studies using placebos have shown that MSG causes no different effects to a placebo substance. But the public were still not convinced. In 1995, the Food and Drug Administration (FDA) in America, decided to put the controversy to bed, asking the Federation of American Societies of Experimental Biology to investigate it thoroughly. It turns out, if you eat six times the normal amount of MSG, on an empty stomach, you might experience something akin to what Herbert Schaumburg found in his human studies, but the number of people affected by this approach was very small. What’s more, the glutamate in MSG does not cross the blood brain barrier and so can’t have psychoactive effects or set off cortical spreading depolarisation in migraine, so its effects wouldn’t be particular to migraine anyway (which was not mentioned in Ho-Man Kwok’s original letter!). So there you go: MSG does not cause migraines, as so many people seem to believe.
Ham-fisted approach
Another, more theoretically sound dietary candidate takes the form of nitrates. Nitrates are naturally occurring chemical compounds that contain nitrogen and oxygen. Found in green leafy vegetables and also carrots and celery (perhaps in slightly lesser concentrations in organic food that has not been exposed to nitrogen-based fertilisers), they are powerful antibacterial agents. It is for this reason that they are added to processed foods such as bacon, sausages, cooked meats – indeed any meat that has been smoked, salted or cured. You might remember that nitric oxide is an important inflammatory substance in the body and it has the power to induce dilation of the blood vessels it is working on. Usually, this is an important mechanism in the cardiovascular system; it keeps blood flow to the heart regular and optimises it for the amount of work it has to do. People with angina, or narrowing of the blood vessels that feed the heart, will often be prescribed a Glyceryl Tri-Nitrate (GTN) spray to be deployed under their tongue when they experience symptoms. The spray serves as a fast way of introducing nitric oxide into the body, causing widespread vasodilation releasing the pressure on the cardiovascular system. However, vasodilation also happens in the cerebrovascular system, often leading to headache following the use of the GTN spray. But we also saw how spiralling levels of nitric oxide induced by muscle hardness can lead to tension headache by tugging on the trigeminal nerve so annoyingly. The ingestion of nitrates, therefore, would seem to be not good for anyone. But is there the chance that migraineurs are more sensitive to ingested nitrates?
For this we need to think about our digestive system, and specifically the mouth. Here, we find lots of substances: saliva, or salivary amylase, an enzyme that is the first to start breaking down the food we ingest; and a whole host of natural flora – bacteria that do the same job. Some of these bacteria work to reduce nitrates into nitrites and then nitric oxide for quick absorption in the body (which is why the GTN spray under the tongue is a clever delivery system).
As part of the Great American Gut Project Cohort in 2016, Antonio Gonzalez, working in Rob Knight’s team in the University of California, San Diego, found that there were more bacteria devoted to breaking down nitrates found in foods in the mouth of a migraineur than in those who don’t experience migraines. It remains to be seen if this translates into higher concentrations of nitric oxide in the system that may lead to the cerebrovascular effects underlying the typical migraine, or indeed if this difference is causative of migraine, or just an effect. However, it is a rather ingenious way of getting to the root of a problem, and offers another window on why nitrates affect migraineurs in particular. Perhaps this is another protective mechanism of migraine, representing a way to conserve cardiovascular health? It’s hard to see it that way when you are suffering from a headache.
One way to combat this is to use an antibacterial mouthwash, which decreases the concentration of the nitrate-reducing bacteria in your mouth. Vikas Kapil and his team from Queen Mary University in London showed in 2013 that this measure alone has the power to increase blood pressure, due to a 25 per cent decrease in nitrites (the precursor to nitric oxide) in the system. Or you could just stop eating bacon sandwiches.
Bar brawls
Let’s end this trigger talk with a trip to the bar. We know that alcohol has various effects on our brain and the cerebrovascular system and that the resultant dehydration will cause a nasty headache. But is it a headache trigger? Alessandro Panconesci has trawled the literature for evidence of how this is reported. It turns out migraineurs don’t report alcohol as a trigger any more than people who get tension headaches, and it’s the same for men and women. Only 10 per cent of migraineurs report alcohol as a frequent trigger, but that might be because the rest of the migraine population avoid alcohol – a suggestion that seems to be backed up by consumption data.
Red wine is reported to be the villain of the piece although there is controversy because other studies implicate white wine and other drinks to a greater degree. Part of this is regional; people from the UK think that red wine is the worst; people from France and Italy think white wine and champagne are headache inducing (from my experience, I’m with them, but I do like a bit of bubbly). Perhaps this is more of a cultural narrative situation, then: because everybody in our social group says so, it must be true. Let’s have a pint of science and discuss it.
There are a few suspects in the dock. Sulphites, tyramine, histamine and flavonoids in different forms of alcohol all stand accused. What the prosecutor needs to do is determine which of these have the power to induce headache (separate to the hangover or dehydration headache we talked about in Chapter 1) and if any of them are specific to migraine.
Rumour has it that sulphites are the biggest bad in wine. Sulphur dioxide is used in wine making as a preservative, since it’s an efficient antimicrobial and an antioxidant. White wine contains a much higher concentration of sulphites than red wine, and dessert or sweet wines have higher concentrations again. There is a link between sulphites and histamine release, which in certain sensitive individuals could lead to breathlessness (more common in asthmatics), and this is caused by the sulphite’s ability to release histamine, which narrows the bronchial tubes in your lungs, and is also on our trigger list for headache. Sulphites might also boost serotonin levels, so while our mood may improve while drinking, the serotonin is causing vasoconstriction. This may cause the neurovascular effects seen in migraine, but certainly causes rebound vasodilation, which will pull on the vessels’ sensory nerve endings. Other than that, the evidence is scant; sulphites get off on a technicality: other foods (e.g. crisps, raisins, dried fruits, juices) contain 10 times the amount of sulphur dioxide than wine does but not that many people complain about those.
We’ve encountered tyramine in Chinese food and other places already, a repeat offender we might say, but the concentrations of tyramine in alcoholic beverages (and there is little variability between them) is very small and much less than that which is used by clinical scientists to prove tyramine has a vasoactive effect. So, tyramine in alcohol is released without charge. Tyramine in everything else, though, is still under suspicion.
Another repeat offender, histamine, doesn’t get off so lightly, but mainly because it is a perpetrator of inflammatory responses with all alcoholic drinks and is egged on by the sulphites. It hangs in the red wine gang, having a greater concentration in your Malbec rather than your Muscadet. Even though we know it can provoke headache, and migraine in those who are prone, we still don’t know if it is the driving factor of how alcohol triggers them.
Histamine’s big brother in the red wine gang is the flavonoid. Flavonoids are naturally occurring in many plants and act as antioxidants; in alcohol they contribute to the colour, taste and mouthfeel of the drink. They are present in red wine in concentrations 23 times greater than those found in white wine, and include catechins (also found in tea and cocoa); anthocyanins, which give the wine its colour; and tannins, which contribute to the taste. The problem is, these particular flavonoids, which together represent 30 per cent of the flavonoid concentration in alcoholic beverages (of which there is more in red wine), are potent inhibitors of PST (phenolsulphotransferase)-P, which breaks down phenols, high concentrations of which can be toxic in the body, causing an immune inflammatory response.
Phenols in nature are powerful antiseptics and are the main ingredient in carbolic soap (my dad’s favourite faux swear word). Nobody makes a habit of eating carbolic soap, but we do ingest phenols in the most common painkiller (among other drugs): paracetamol. Not being able to break phenols down means they hang around for longer and can have a harmful effect; inhibiting PST-P will lower the toxic threshold of the drug.
These effects of red wine and other alcoholic beverages are not, however, specific to migraine; they will trigger malaise in everyone. One point to note, though, is that in 1995 Mark Sandler from Queen Mary University in London found that there is a deficit of PST-P in the migraineurs he tested, meaning that alcohol may be a double whammy for them. In other words, not only do they have a deficit anyway, but alcohol will also lower it further. What’s more, many alcoholic drinks have vasodilatory effects caused by the release of nitric oxide from the blood vessel walls and nerve endings through the action of the ethanol and histamine. And finally, ethanol itself promotes the release of CGRP, a prolific vasodilator and an agent that is already high in concentration in migraineurs.
Overall, though, the jury is out, still. For every study that says red wine is the baddy, others will say white. Wine has been linked to headache since the earliest medical encyclopaedias of Celcus at the beginning of the Common Era through to those of Paul of Aegina in the 7th century CE. There is therefore a historical precedence for this narrative. For every vasoconstriction mechanism, there is a vasodilatory one. And it’s really difficult to show if it is what is in the beverage aside from the ethanol content that is acting as the trigger. It is also hard to be specific to migraine – all of these factors might cause any vascular headaches – but it is the special features of the migraneous brain that makes some people experience migraine from alcohol, though as Alessandro Panconesci found, only 10 per cent of migraineurs report alcohol as a specific trigger. The International Classification of Headache Disorders Criteria are strict; they say that migraine cannot be diagnosed when there are other possible causes for headache, including substance-induced toxicity. However, as we have seen, such toxicity may trigger migraine in the ‘migraneous’, but you won’t be diagnosed as a migraineur based on your experience of an intoxication headache alone. Finally, we should also be mindful that these experiments involving different types of alcohol are very difficult to do. After the third glass of plonk, who really cares anymore?
Figure it out for yourself
Putting all of this together it is clear that there is a plethora of triggers for migraine. Some are major, such as patent foramen ovale, and some are relatively avoidable, such as nitrates. Some of them may sound familiar to you, some may be worth investigating; the individual differences here are vast and there is no one-size-fits-all solution. Prevention is better than cure of course, although there have been great strides in this field with the arrival of sumatriptan and Erenumab, both born of our clearer understanding of what is happening in the brain and body preceding and during a migraine episode and what made that happen; what is special about the migraneous brain. But you? You are inimitable. You have to identify your own triggers and you are uniquely placed to understand them. There are factors out there that wouldn’t bother a regular person but make the migraineur much more susceptible to the induction of a headache. Working out which ones you can control and which ones you can’t will allow you to wrestle the initiative back from this dreaded event.