Not every really painful headache is a migraine, in much the same way that a really bad cold is not the ’flu. Unless you’re a man of course (see here – the theoretical implication of your bigger preoptic nucleus aside, gentlemen). Influenza is an entirely different animal, as those of you who have actually experienced it will know. Migraine, too, has a specific symptom set that means it is recognised now as something totally separate from other classes of headache. Over the next two chapters, we will investigate these symptoms to understand what is going on to create them and how our body reacts to them.
Migraine falls into two-and-a-half categories. You can have ‘migraine with aura’, which is called ‘classic migraine’, or ‘migraine without aura’, which is ‘common migraine’. I have experienced both on occasion myself. You can also have ‘ocular migraine’, which is the aura bit without the pain part – you might just jump straight to feeling groggy, which is why I call it half a category. However, as we shall see, the four component stages that make up a migraine are somewhat shared across these types.
The most interesting aspect of migraine is that it is an experience that goes beyond the headache component. It is not just a headache, it is a phenomenological event that comprises four distinct phases:
1 The prodrome phase
2 The aura
3 The pain phase
4 The postdrome phase.
It is important to understand each because this tells us something new about what’s going on in our brain to cause them, and also opens windows on how we might be able to solve them.
Something wicked this way comes – the prodrome phase
I have spoken to many migraine sufferers in my time and have learned a lot. However, the most fundamental thing I have discovered is that migraineurs (as we refer to migraine sufferers in the biz) are quite bad at spotting the first stage of migraine: the ‘prodrome’ phase. As a case in point, I met a 25-year-old man called David and his mum at a headache event once and witnessed a rather heated exchange about this very thing. David’s mum swore that she could spot his migraine before he could. ‘With every respect, Mum,’ and you just know the tone of what’s coming next is going to be laced with disrespect, ‘I think I, of all people, would know when I am getting a headache, after all.’ I interjected (unhelpfully for David) that the last person to know is often the migraineur themselves. Indeed, there is hard, albeit recent, evidence to support this. Ana Gago-Veiga from the Headache Unit in Sanitaria Hospital, Madrid, Spain, worked out that only around a third of patients she surveyed could be classified as good predictors – in that they could spot an impending migraine more than 50 per cent of the time. Even then, it was only because their prodrome symptoms were quite obvious, including pronounced yawning, drowsiness, food craving, adversity to light, increased thirst or blurred vision. Having explained all of this to David, and his rather rapt mum, he sheepishly reported that he had never put these symptoms together with his migraine before. ‘I had,’ his mum announced, rather too triumphantly, I thought. But it just goes to show: mums really are always right.
As we now know, it takes a certain amount of self-awareness to spot the first stage, the prodrome phase. In this premonitory (something that predicts something bad is about to happen) phase you may be displaying a bunch of behaviours related to the changes that are happening in your brain, and these can occur a couple of days or hours before the migraine starts properly. You might yawn more than usual, be less alert – or even drowsy – have cravings or be hungrier than usual. Walking into shops with really bright mirrored lighting might unsettle you to the point of distraction. These symptoms should not be underestimated.
For the past 10 years, Peter Goadsby from King’s College London and some others have been talking about the value of tracking these symptoms to their underlying biology. It’s the changes that happen in your brain that sometimes causes the behaviour that we think of as being ‘triggers’ for migraine. We have to unpack this; ultimately, I am interested in why we have these symptoms. Are these behaviours in some way generated to redress some neurochemical balance that is out of kilter in our brain? How can we manage this to stop the migraine from going any further? Here’s what we know so far.
Yawning chasm
Let’s start with yawning. Yawning is an interesting behaviour as it is both physiological and psychosocial. It’s very easy to ‘catch’ a yawn – there have even been studies that have linked exactly how easy you find this to how empathic you are. Well, I am yawning my head off writing this so take from that what you will, and if you feel the urge to yawn right now, go ahead, I won’t judge. The reason why we yawn together is because it engenders group alertness; yawning introduces a big dose of oxygen to the body, a fair amount of which reaches your brain, helping you feel more refreshed (which is why we also yawn when we are tired). Evolutionarily, this may have been important when the group was off hunting a woolly mammoth or something; nowadays, military parachutists often report having a jolly good group yawn before they jump out of the plane. (Although between you and me I think yawning might be further down the list of the many functions my body would perform all by itself in this instance.)
It isn’t all about the shot of oxygen, though. In 2007, father-and-son duo Andrew and Gordon Gallup from New York State University put cool packs on people’s foreheads as they watched videos of people yawning. They discovered that people yawned much less frequently when the cool packs were there, so this indicates that yawning helps cool our brains too; if our head is already cool, we don’t catch a yawn as easily. This cooling may help us feel more alert, independent of the extra oxygen. And perhaps the cooling is interacting with a neural pathway that is active in yawning, stifling it for us.
Yawning is an unconscious or automatic behaviour that originates in the brainstem, where lots of species-specific behaviours, such as grooming, originate. So, yawning may be triggered as a thermoregulatory response (we’re too hot) or to introduce more oxygen into the brain, which might lead us to think that in the pre-migraine state, the brain is inflamed in some way that increased its local temperature and/or that it is somewhat oxygen depleted. Yawning is our way of self-medicating to reduce that inflammation and increase the oxygen levels.
This relates to neurochemistry in a very specific way through the action of dopamine. Dopamine is one of the main excitatory neurotransmitters in the brain, important in alertness but also movement and how rewarded we feel having behaved in ways that keep us alive (like eating and drinking, and sex, although that is not classed as a behaviour that keeps us alive). Dopamine neurons trigger the act of yawning by causing the hypothalamus to act on the brainstem to carry out the act of yawning itself. In this way, dopamine induces a behaviour to improve our alertness. We know that chemicals that act like dopamine in the brain can cause yawning, and that people with dopamine deficiencies, such as Parkinson’s disease patients, yawn less. Because of this, the prevailing wisdom is that higher levels of dopamine cause the symptoms seen in migraine. But the story is more nuanced than that, with the migraineur’s sensitivity to dopamine being the key issue; migraineurs seem to be particularly sensitive to dopamine concentrations.
There is another view, however. Building on the knowledge that people prone to migraine are hypersensitive to dopamine, Piero Barbanti, the head of the Italian Headache Society, and his colleagues from San Raffaele Hospital in Rome, Italy, have developed another somewhat complementary idea. Piero has a theory that lower amounts of dopamine, not more, kick off the prodrome symptoms of migraine. I can buy this theory because lower dopamine would have a depressive effect on the brain and so yawning could be instigated to wake the brain up. What happens next, Piero explains, is that the dopamine levels are regulated by the hypothalamus, which takes emergency steps to keep all levels of hormones and neurotransmitters within strict confines. The quick boost of dopamine it administers into the system causes the nausea that is then seen in the next stage of migraine. In experiments where subjects are given a dopamine agonist (that acts just like dopamine in the body) a very small amount induces yawning in migraineurs, whereas those who don’t get migraine have to take a lot more to have that happen. With increasing dosage, a control person is merely yawning, whereas people who get migraines are by now throwing up all over the place because they are so much more sensitive to it.
How do we reconcile these two theories? The common point here is the migraineurs’ sensitivity to dopamine, whether the concentration of dopamine needed to cause the symptoms of the prodrome phase is low or high. You might ask why we care about what causes the migraine symptoms at a neurobiological level, but understanding this might be a way of stopping migraines from happening in the first place. For instance, we can link dopamine to how we feel and we can manipulate it through our behaviours. It makes you think of all the times when you might have unwittingly done something to head off a migraine by boosting your dopamine levels, by giving yourself a rewarding experience like eating something you fancy or having an orgasm. Unfortunately, we’ll never know what these behaviours were. This is one of the things we know we don’t know – it’s fairly impossible to get evidence for what we did in order not to experience an event.
It is a bit of a conundrum though, because dopamine is a good mediator of pain, blocking the signal from the trigeminal pain pathway to the brain. But of course the problem is not caused by dopamine alone; dopamine works hand in hand with serotonin – serotonin improves the action of dopamine – with receptors for both often located together. So another argument is that dopamine may not be doing its job properly because serotonin is low, and any regulatory boost the hypothalamus tries to implement is just too late to stop the pain cascade.
Craving a cure?
It is not just yawning that is a striking symptom in the prodrome phase. Changes in appetite are, too. Some people experience a loss of appetite before a headache hits, whereas others crave certain foods, generally of the sweet variety. As we know, appetite is also controlled by the hypothalamus, and a strong hormonal suspect in the latest migraine research is neuropeptide Y, which is a kind of neurotransmitter just like dopamine is. It turns out that orexin (a hormone we met in Chapter 5) also directly interacts with neuropeptide Y (NPY) and so it is implicated as well. Orexin makes us crave specific foods, and when we put this together with our knowledge of prodrome migraine behaviour, it explains why people turn to cheese or chocolate or sometimes high-carbohydrate meals in this phase. That doesn’t mean the chocolate is a trigger of your headache though; the changes in your brain chemistry are the trigger. We have to be really careful about this: laying off chocolate will not prevent headaches. The changes of activity in your hypothalamus that precede the headache are directly influencing your dietary choices.
Can these cravings tell us anything about what our bodies are lacking? Is our brain pulling our strings to self-medicate us? Migraineurs are generally attracted to sweet foods that will quickly break down into sugar in the body. This might mean that a period of poor diet or irregular eating patterns has led to a hormone imbalance in the hypothalamus, which it is now encouraging you to replenish. It could also be a way to take in energy to help you cope with what is to come (the migraine). Either way, our hypothalamus keeps us in such tight homeostatic control that it is hard to believe that the specific craving that we have has no correlation with what is wrong, and what is causing the headache.
Perhaps understanding what is happening with our hormone levels will help us? Our normal appetite control is a complicated dance between the action of orexin on NPY neurons and vice versa, in addition to the circulating levels of the hormone leptin that is released from fat cells to inhibit hunger. If the level of one of these hormone levels falls, then that affects the action of the others. For example, high leptin leads to a decrease in NPY through the action of orexin, thus discouraging further feeding, and low leptin does the opposite. But the orexin–NPY axis can work outside of the influence of leptin (remember how easy it is to override how full you feel when you absolutely have to have dessert?).
Alas, the link between levels of these hormones and the incidence of migraine is befuddled across the scientific literature. Migraine incidence has been linked to low levels of NPY at least in younger patients during the prodrome phase. Levels jumped as the pain phase took over. Orexin levels seem to be low in people who experience episodic migraines (which would theoretically lead to a loss of appetite) and high in those who have a chronic variety (those who get cravings).
More evidence about the role of orexin comes from those who experience the sleep disturbance narcolepsy, or uncontrolled falling asleep. This is caused by a loss of orexin neurons in the hypothalamus and such patients report at least a doubling of migraine prevalence in comparison with non-narcoleptics, boosting the evidence that low orexin is a driving factor in migraine. But as with cluster headache, the critical factor may not be the actual concentrations of orexin but rather our sensitivity to it.
How this might manifest itself in real life can explain the cravings. We know that orexin itself can cause specific cravings, but if NPY isn’t there to control the appetite, then the craving will be unleashed. As the pain phase of the headache takes hold, the levels of NPY increase because just like dopamine, it is a powerful painkiller in the trigeminal pathway and also causes constriction in the vasculature of the head. And just like dopamine, this higher concentration boost happens too late to stop the pain from starting.
The clearest things we can say are that these hormones are implicated somehow in the disordered physiology, or pathophysiology, of the migraine and have the power to manipulate our behaviour. Directly affecting the action of these chemicals medically is a possible pathway to intervention.
A chemical conundrum
So we now know that dopamine, serotonin and neuropeptide Y all lead to that feeling that something is not quite right through their actions, and can make us agitated and cranky. This discomfort or dissonance with what is going on around you is no doubt linked to dips in serotonin levels as well as oxytocin. Even outside of headache, our levels of these hormones peak and trough throughout our lives and are quite dependant on what life has bestowed upon us, or indeed hit us with. Serotonin in our story is consistent with chocolate craving; packed with tryptophan, which is converted into serotonin in our bodies, chocolatey goodness is bound to prop up our concentrations.
Oxytocin, meanwhile, is the bonding hormone, released in large amounts in females after giving birth – and postnatal women report a much lower incidence of migraine. Individuals of both sexes experience a boost in oxytocin in the early days of a love affair, too. It helps people become somewhat addicted to each other, making them feel uncomfortable when they are apart from their amour. This explains why new affairs eschew the involvement of others that might compete with the attentions of the lovers, hence all previous friends are dropped like hot potatoes. It doesn’t last though; levels normalise, there is time once again for friends, and being with your partner is not as big a deal. Sorry for reducing such a wonderful human experience to the specific action of a hormone – of course it is just a little more complicated, but here is my point: if your oxytocin dips, how does that make you feel?
Migraine sufferers in the prodrome phase feel disconnected; I have heard many use the phrase ‘I just feel needy’. This feeling should never be ignored as it is entirely related to low oxytocin. Hugs to the rescue! Even having some time alone with a loved one’s undivided attention or telling a loved one how you feel about them, and why, helps to boost their oxytocin levels (e.g. you clean the bathroom, you make nice risotto, your smile fills me up … feel free to improvise) and yours will get a nice bump too. Sex helps, but only with partners with whom there is an emotional connection (although serotonin and dopamine don’t mind so much about that). This all matters because there are loads of oxytocin receptors on the trigeminal pathway neurons and if oxytocin binds with them, the trigeminal neuron can’t pass on its signal. In other words, if the oxytocin receptors are not fed, then the pain signals pass at full power to the brain. Low oxytocin is therefore a big problem for the migraineur. Choose a hugger as your partner; they don’t call it [neuro]chemistry for nuthin’!
Do you see what I see? The aura
After the prodrome phase we have a bit of a divergence. Some people go straight to the pain phase and this is called ‘common migraine’, but some have another step: the migraine aura, turning their experience into what we refer to as ‘classic migraine’. I have spoken to many people about their experience of the migraine aura, but actually, of the migraine population, only 20 per cent have experienced an aura. They may have experienced one or two in their lifetime or they may experience them every time they have an episode of migraine. And then there are those who experience this stage without going on to the pain phase of migraine – theirs is called ‘ocular migraine’. This name flags up the most common form of aura, which is a visual disturbance.
Ben was a 22-year-old PhD student when he experienced his first aura. It was a Sunday morning, and he had to go into the university to do some writing for his thesis. He didn’t really want to because he had felt ‘fuggy’ for the previous couple of days, but he hadn’t got as much done as he had wanted to that week so he was giving it another push to reach his target. It was a sunny, fresh spring day, the best time for clearing his head. Ben remembers smiling to himself as he put the key in his office door; he was thinking this might be the first beer garden afternoon of the year if he could get finished up by three. And then the door disappeared.
I asked him if this came as a shock. He looked at me as if I had 16 heads. ‘Well, of course; that didn’t usually happen.’ After a couple of moments just staring at where the door should be, Ben realised the whole door hadn’t disappeared, just the part with the door handle and the key, and that this was in his peripheral vision as he was looking straight ahead at the name plate mounted on the door. He cautiously moved his eyes around and realised that the ‘hole’, which scientists and clinicians call a ‘scotoma’, was in the same place in his peripheral vision no matter where his eyes were pointed, and he couldn’t see whatever fell into the hole.
What’s interesting about Ben’s report is the sudden onset, and it went away just as quickly – by the time he had managed to sit down at his desk to gather his thoughts, it was gone. He experienced it for five minutes, tops. Most people who report aura can detect the onset as a gradual change in their perceptions. It usually lasts between 5 and 20 minutes and is undetectable after an hour. Just like Ben, most people are completely freaked out by their first experience and whether or not it is followed up by a headache, it is well worth talking to a clinician about it, and certainly if the aura lasts for more than an hour.
If we put all of the ways people experience aura together, a pattern soon emerges. In their simplest forms, visual auras are composed of phosphenes – our perception of spots of light that aren’t really there. You can induce your own phosphenes easily by pushing on your eyeball gently, close your eyes first! That spot of light is caused because you have activated the retina mechanically through the pressure you’ve put on the eye. Because the retina has been activated, the brain ‘perceives’ light, even though it is illusory. You can induce phosphenes by activating the part of the brain that decodes the electrical signals coming from the retina, too. In the 1930s, Wilder Penfield from McGill University would send electrical currents into the open brains of conscious patients he was operating on in order to see what different parts did. If the bit he stimulated was important in arm movements, the patient’s arm would move. If it detected light in a specific part of the visual field, the patient would see a phosphene there. Wilder managed to make many amazingly useful maps of the function of different parts of the brain in this way. You might think this sounds barbaric, but even in the 1930s it was possible to operate on conscious patients under just local anaesthetic. Indeed, it was preferable, as it guided neurosurgeons to areas of vital importance for life, such as speech areas, or movement regions, so that they could be spared if possible. It is a practice still continued today, although scientific investigation is less invasive now. By sending magnetic pulses into your brain, I can activate the neurons that make your body move, or that make you see a phosphene. Called transcranial magnetic stimulation, it is a great tool to gently and reversibly switch on bits of your brain so that we can know exactly what it is doing, and when.
Scientifically induced phosphenes can be obvious. They are like flashes of light against a black background, or more subtle changes in quality of the resolution of your vision in an area of your visual field. In a migraine aura, there is more structure. When you are perceiving a phosphene, it obscures what you should really be seeing at that point in your visual field and so you are essentially blind in that area. American psychologist Karl Lashley experienced aura himself, and was interested in what was happening in his brain to cause these perceptions.
What Karl realised in 1941 was that while your earliest perception may be a small spot (easily overlooked by the visual system) it seems to expand, and the shape of the scotoma never changes, it just gets bigger. He linked this to how electrical activity might be spreading over the area of the visual system in the brain; the earliest area (V1, also known as primary visual cortex and also striate cortex because of how it looks under a microscope) is important in perceiving lines, hence the perception of zigzags made of straight lines. More advanced areas in the visual system catch light from much bigger areas in the visual field and so stimulation there will cause a much bigger scotoma. If the electricity is passed in a non-random way, the scotoma will hold its shape and merely appear to be expanding. The edges of the scotoma, or the ‘fortifications’ as they are called, do not expand, but are duplicated to fit the bigger shape and there seems to be a scintillation across the levels of boundaries, ‘like the illusion of movement of a revolving screw’. When he compared notes this seemed to be common for all observers. Karl was onto something. By plotting the rate of enlargement of his scotoma, the rate of the scintillation (about 10 waves per second) and putting that together with what he knew about how the visual cortex propagates the electrical signal, he concluded that the two were related, and at least in his case, this wave of excitation in his brain must be travelling at a rate of 3mm (0.12in) per minute.
These ‘ripple-waves in the cerebral pond’ had already been suggested in 1904 by a British neurologist Sir William Gowers, but without precise neurological insight at the time, he proposed them more as a metaphor to understand the phenomenon. Now, Karl had added some observational heft to the story. But at the time, the neurological community was consumed by the idea that each bit of the brain did one thing, or ‘functional localisation’ as we still call it, and wasn’t interested in hearing how waves of activity could pass between one area and another resulting in complex perceptions and behavioural effects. It took until 1990 and Gregory Barkley’s team in Wayne State University, Detroit, Michigan, to definitively visualise these waves in the brain of patients experiencing migraine aura using a technique called magnetoencephalography (MEG) (see here), which detects the magnetic fields emanating from your head, an indicator of the electrical activity that is going on inside your skull.
Other types of aura
Yet aura is not just a visual issue. The next most common type is sensory aura, which takes the form of a creeping tingling in your skin, sometimes over the head and sometimes in the limbs. It can make your skin hypersensitive and even present as a dull response in your limbs to movement, or paralysis. As with visual disturbance, it’s always important to go to a medical professional the first time you experience this to rule out other causes such as stroke. Somatosensory tingling during aura is not to be mixed up with Autonomous Sensory Meridian Response (ASMR), which is also described as a tingling over the head and neck after triggers such as repetitive movements or whispering.
Although it is relatively new to scientific enquiry, there is already a multimillion-pound YouTube cottage industry trying to induce that pleasurable feeling some people describe as a brain orgasm. Because of this, there has been much interest in trying to induce it as a happiness boost in depression and other affective disorders. Most research to date focuses on pleasant inducement of the ASMR effect, although there have been some reports of ASMR occurring when people view particularly violent scenes. It certainly needs more investigation, if not just to understand any commonalities that may exist between ASMR and somatosensory aura in migraine. For instance, Nick Davis from Manchester Metropolitan University and his team have shown that background music stops ASMR from happening, so that distraction from another modality (in this case hearing) stops a somatosensory effect. Can we mine this knowledge to minimise the effect of migraine aura?
It is even possible to have other sensory auras including olfactory (smelling something that isn’t there or extra sensitivity to smell), auditory (hearing dripping taps, loss of hearing, tinnitus or extra sensitivity to sound) and taste (particularly manifesting as a metallic taste in your mouth). Interestingly, these are also symptoms often reported in geriatric patients suffering from urinary tract (and particularly kidney) infections that have caused severe dehydration.
Aura can therefore manifest itself in any of our sensory systems but we have left one out. Everybody thinks about the five senses: seeing, hearing, taste, smell, touch. But touch is a very specific term; it would be better to use a more umbrella term such as somatosensation, which encompasses both touch and also our awareness of where our body is in space. It is this ‘proprioception’ that can also be disturbed by aura, leading to a certain clumsiness. Some of us just have this all the time unfortunately, and have the broken nose to prove it.
We now know that there is a clearly defined pattern of activity in the brain that occurs during the aura that causes these sensory effects, independent of outside stimuli. Work to define the migraine aura intensified in the 1990s after Sir William Gower and Karl Lashley’s suggested ‘ripple-waves’ were seen using MEG.
It turns out that what you are observing during aura is a wave of excitation across your cortex. In this wave, many neurons become active in a coordinated way and this electrical activity travels as a surge. This is what causes the sensory disturbance, as brain cells are active and usually their activation is because you are sensing something; so you see things that aren’t really there, feel touch when you are not being touched, etc. However, crucially, this wave of excitation is followed by a wave of depression called ‘cortical spreading depression’, where your neurons go completely asleep or more properly, become inactive. This is the big picture of how the abnormal pattern of electrical activity affects your behaviour in migraine and kicks off the pain. But what is happening to create this activity?
This is important, because some of the causes of migraine interact directly with a particular phenomenon at the level of the cell that is happening in all of us, all of the time! I’m referring to what makes you tick – quite literally, because that’s how this phenomenon sounds to our ears when amplified in the lab! If you listen to the electrical activity in a single neuron it sounds like static noise over a radio. Slow it right down and you can hear every crackle, every tick. Each one of those ticks is called an action potential, a little pulse of electricity passing down that neuron, a signal, a nerve impulse, each one lasting about a millisecond. It is well named because it truly represents your potential for action. For you to be able to think, walk, watch television, read this book or do anything at all, you need millions of these ticks in just the right places in your brain at just the right times. And it’s all going on without you even knowing it’s there.
It’s really worth looking at how action potentials are made so that we can find out how these waves of excitation kick off the pain we experience in migraine.
How do you tick?
You might remember from school that electricity is all about the movement of charged particles. In the body, these are ions like sodium (with its chemical name Na+) and potassium (K+). When the neuron is not doing anything and resting, there are many more negative particles on the inside of the neuron than the outside. These nerve cells are really particular about what comes in and goes out and have little doors made out of protein in their membrane that only open for a particular thing, like a gate that only lets sodium in and out, for example. Whether these gates (called ion channels) are open or closed depends on how negative the inside of the cell is. If there is a little spark of electricity, that might have come from another neuron (we’ll talk about this soon), or a feather brushing your arm or some other sensory stimulus, these gates open and ions like sodium move in. Sodium is a positive ion and it wants to rush into the neuron because there isn’t much there already. Also, the inside of the neuron is very negative, so because opposites attract the positive sodium ions are drawn there. So, very rapidly the inside of the cell gets even more positive than the outside was, and this flip in polarity is the first part of the action potential and is called depolarisation. For the sparkier among you, it is about a 110mv difference (from -70mv to +40mv).
Now we have to get the neuron back to normal resting state again because until it is back to resting state, it can’t generate another signal. Because the inside is more positive at this point, the gates for sodium shut. It’s trapped! But when the inside is this positive, the gates for potassium open. Potassium is also a positive ion and is usually much more concentrated inside the cell than outside the cell. It leaves the neuron, fast, because the outside is more negative now and opposites attract, and there is not much potassium out there so it wants to populate it. In a fraction of a blink, the upswing of the pulse has been turned back down again. The action potential is done and this part of the cell is repolarised, back to having the inside more negative again. It even has an extra dip in negativity to finish with a flourish before it gets to the normal resting state. The neuron sets about kicking all that sodium out again by a kind of ion prisoner exchange, using a protein in the membrane that pumps out three sodium ions for every two potassium ions it lets in.
Anything that blocks the way ions move into or out of the cell through the ion channels, or plays with the concentration of ions, will mess with this entire process, and there will be consequences. For example, the wave of excitation in migraine will cause potassium to get trapped on the outside of the cell, and this activates pain receptors in the blood vessels – not good for us migraine sufferers. More on this later too.
FIGURE 3 Two Neurons ‘Talking’
The role of axons
The action potential is kicked off in the cell body of a neuron, in an area called the axon hillock just at the start of the axon (or in less technical terms, ‘the stringy bit’) that carries the signal to the next neuron. If we think about one neuron in particular, there will be many neurons connecting to it trying to pass on their signals. It is the axon hillock’s job to decide if it has had enough excitatory input to pass on the signal. If it has, then it generates an action potential. If it hasn’t then it just keeps quiet.
Some other things are worth saying that are important for our cortical spreading depression story. Action potentials are all-or-nothing events. They all look exactly the same. You can’t add them together. So how do you know if there is a particularly forceful stimulus for example, or a really bright light as opposed to a dim one? Well, magnitude is transmitted by frequency. A little touch will elicit a few action potentials per unit time, but a proper punch will cause loads of action potentials. You also can’t have two action potentials in the same patch of neurons at the same time because the ions are in the wrong place to generate one. And action potentials travel in one direction only, away from the cell body. The extra dip in negativity of the inside of the cell when potassium has flooded out takes care of that detail.
The action potential travels down the axon by regenerating itself at every point down the neuron. Support cells called glial cells wrap themselves around most neurons, forming a myelin sheath that acts like the insulation we see on electrical cable. The action potential can then jump between gaps between the glial cells and only has to regenerate itself every now and again. This makes nerve conduction faster and more efficient. We can track rates of myelination (the creation of the myelin sheath) in different parts of the brain to the acquisition of new skills in babies and young people. 1
The case of Golgi vs Cajal
Because we are living human beings and don’t catch excitability from some neuroscientist in a lab coat with an electrode stuck into the Petri dish in which we are living (although there is a movie in that), we have to think about how it happens in real life, where neurons don’t live in isolation. Let’s think about two neurons. How does activity in the first get passed on to the second? This fundamental question caused one of the biggest and certainly most infamous tiffs in scientific history. It started when the Italian Camillo Golgi developed a way to see the structure of a neuron and what was in it under the rudimentary microscopes that were around in the mid- to late-1800s. He would slice up brain tissue as thinly as he could, shine light through the material and magnify the image through a number of convex lenses. However, brain tissue is translucent, and so he had to develop a silver substance to wash the brain tissue in. The silver particles would stain the individual neurons black and show very fine details of what they looked like and where they went, which led Golgi to describe them as an interconnected series of ‘tubes’.
Soon after, the Spanish neuroanatomist Santiago Ramón Y Cajal refined Golgi’s technique and discovered that neurons are discrete entities in and of themselves. He also linked age to what the neurons he could see looked like, with their form becoming much more complex with maturity.
Golgi was furious, and diametrically opposed to Cajal’s views; Golgi believed that the nerve cells he could see acted much like the blood vessels in the body, whereas Cajal correctly identified that they were separate, with gaps in between them that had their own functions, not just acting like transit stations in a network. In recognition of their work, both Cajal and Golgi won the Nobel Prize in Physiology or Medicine in 1906. Awkward.
The drama didn’t end there. Both anatomists used their acceptance speeches in Sweden to attack the other’s view, leading to quite fractious scenes and a definite sense of unease in the auditorium. Probably just as well they hadn’t won the Nobel Peace Prize, although there was no way that was in either of their futures. Of the two, Cajal was the more magnanimous, acknowledging that if it hadn’t been for Golgi, he could not have made his own discovery. Golgi was bound by his education and the prevailing views at the time, whereas he, Cajal, had the kind of mind that could look beyond current wisdom to interpret what they could both see in a different, and ultimately more prescient, way.
Bridging the gap
The gap between nerve cells is called the synapse and provides a bit of a problem for our conduction of our action potential; the electrical signal can’t jump this kind of fissure. What seems like a complicated system, however, allows for a really discrete way to control excitability in the brain. Instead of nerve conduction being entirely electrical, the synapse introduces a chemical component. The arrival of the action potential at the end of the neuron, called the terminal bouton, opens up calcium ion channels in the terminal or presynaptic (before the synapse) membrane. Calcium comes into the neuron and binds with little pouches or vesicles full of neurotransmitters such as glutamate or the dopamine and serotonin that we have already met. The vesicles bind with the presynaptic membrane and the contents are dumped out into the synapse. It is the neurotransmitters that lock onto specific receptors in the membrane of the next neuron (postsynaptic membrane) to open ion channels specific for either excitatory (positive ions such as sodium) or inhibitory (negative ions such as chloride) inputs. The axon hillock then decides whether or not there is enough net excitation to start off an action potential in the second (or postsynaptic) neuron. If there isn’t enough, then an action potential simply isn’t generated.
Wave action
In the traditionally called ‘cortical spreading depression’, the first step is a wave of excitation, so it has now been renamed ‘cortical spreading depolarisation’. This is a huge and co-ordinated excitation that radiates from a point in the cortex, most regularly being in the visual cortex (this is situated at the very back of your brain and is the first place the electrical signal that is made in your eyes in response to light goes). Literally every cell is depolarised and active at the same time, it’s a brainstorm. It is also very like the brain activity seen in epilepsy.
The link between epilepsy and migraines
Epilepsy, from the Greek epi- or ‘upon’ and -lepis’ meaning ‘taking hold of’ or ‘grasping’, is diagnosed after two such seizures. The abnormal brain activity in epilepsy usually begins in a small area of cortex that is damaged or malfunctioning in some way. Very often, it begins in the temporal lobe, the area of the brain that processes what objects look like as well as hearing, memory and speech comprehension (in the left hemisphere). This activity in the epileptic brain can cause hallucinations, just as with migraine aura – the experience of seeing something or hearing something that isn’t there. The difference between what happens in epilepsy and migraine aura is that in migraine, following the initial activation of neurons, all neural activity stops behind the leading wave; no more action potentials, no more activity at all and so neural activity is completely depressed and, crucially, because of this, there is no epileptic seizure. In epilepsy, the brainstorm continues wave after wave.
Perhaps not surprisingly, there is a genetic link between the two; families with two or more close relatives with epilepsy have been found to also have a prevalence of migraine with aura. The two conditions are genetically similar, but while both are rooted in hyperexcitability, there is a fundamental difference in the receptors that underlie the effect, modulating the behavioural consequences.
The wave of excitability explains the symptoms of the aura, as it slowly passes over the cortex at a rate of around 3mm (0.12in) per minute. It is possible for people to recognise more than one aura; it’s not always the case that they only experience a single type, although they do tend to occur sequentially. We can visualise it this way. Let’s say that the excitability starts in your occipital cortex at the back of your brain and radiates out from there. Since your occipital cortex is concerned with vision, activity here will cause you to see an aura pattern as described before. When it hits the temporal cortex at the bottom of your brain there may be some auditory or perhaps memory disturbance. The wave will hit the parietal lobe at pretty much the same time, so it is possible to have two symptoms simultaneously, and this is where you might get some tingling, usually on the opposite side to where the wave is passing through your somatosensory cortex.
Wilder Penfield mapped this out using his electrophysiology technique on awake patients who were having brain surgery. He realised that the more sensitive an area of the body, such as the lips or fingertips, the more neurons were devoted to the processing of sensory information from there. Each area of the body has a place in the somatosensory cortex, which lives just before the middle of the brain moving from back to front. If you point to the centre of your head (humans are strangely very accurate when pointing to the centre of their heads!), your somatosensory cortex lies about a finger space behind that and extends down the side of your brain, with sections devoted to all the parts of your body. So as the wave passes by it activates all of these regions.
The wave usually stops at the central ‘sulcus’ (the big crack that denotes the start of frontal lobe), but if it does pass it immediately encounters the primary motor cortex, which is organised in much the same way as the somatosensory cortex. Again, though, Wilder found that the size of a region devoted to moving part of the body was related to the dexterity we need there, so the area for moving the fingers is much bigger than that for moving the thigh, even though our hands are so much smaller than our thighs. Further forward, our wave finds planning, thinking and emotion centres, which means that as it passes over the frontal cortex, you might feel a bit shaky, clumsy, foggy and generally out of sorts. If the wave gets as far as the very front of your brain your sense of smell and taste will be affected, too. For the wave to traverse the entire brain would take 50 minutes at 3mm (0.12in) per minute for the average 15cm (6in) brain, but up to 30 minutes for the usual half a brain the wave extends to.
However, if the wave doesn’t get as far as the frontal cortex then why are these symptoms that come from there so prevalent during the aura? The answer lies in anatomical and functional connectivity between the back of the brain and the front.
I once ran an experiment with a great colleague Alison Lane and a group of other fine people in which we decreased the activity in the back of the brain using electrical stimulation and looked to see what was happening in the whole brain using functional MRI. It turns out that even though we hadn’t done anything in the front of the brain, there were a load of regions there where there was less activity than usual. It stands to reason that the wave of excitation and following wave of depression will be having remote effects much further forward than where the wave stops, never mind what is happening under the cortex in the hypothalamus and other areas. But what is driving the complete depression of activity that follows the excitation?
Own goal
It’s a little bit like the last minute in a football match when you are 1–0 behind. You only need to draw to win the league but if you lose, you get nothing. Your team throws all of its players forwards, including the goalkeeper. But disaster, your number 11 has a bad touch, loses possession and the other team goes on the break. All of your players are out of position and can’t defend against the inevitable goal. If we think of your players as ions, and link this to what happens in the action potential, we’ve got our explanation.
Our action potentials are fast and furious. Remember that the number of action potentials denote the strength of the response because action potentials are all-or-nothing events. But these continual action potentials put all of the ions that cause the action potential in the wrong place after a while. Sodium rushes in and potassium rushes out, but with continual action potentials the sodium-potassium pump can’t redress the balance to a proper resting state where there would be more potassium inside the cell than outside of the cell and more sodium outside than inside. The glial cells can’t clear the potassium quickly enough. Since the lowest glial-to-neuron ratio in the brain is found in the visual cortex, this may be why this is the most usual, and first, aura that is experienced. Ultimately, sodium gets trapped on the inside and potassium on the outside, making it impossible to generate an action potential at all! Everything stops.
However, the effects of lots of potassium being in the extracellular space where it doesn’t belong are wider than just shutting down the ability to generate action potentials. Potassium acts directly on the tiny branches of the arteries, called arterioles, which then lead to the even tinier capillaries that transfer oxygen and nutrients to the brain tissue as part of the blood–brain barrier. All of the potassium that is floating around the extracellular space acts on the smooth muscle that makes up the wall of the arterioles to constrict the muscle, thereby slowing down the blood flow to that region and reducing the amount of oxygen to the brain tissue, prolonging the depression of neural activity. Because the blood flow is decreased, the high levels of potassium can’t be removed efficiently so it hangs around for even longer. As the concentration rises even further, the vasoconstriction gets even more pronounced, perpetuating the cycle that results in the prolonged depression of local activity and ‘ischemia’ (from the Greek ishaimos or ‘stopping blood’).
This neuronal silence lasts for a few minutes but the recovery to normal activity takes up to 30 minutes. The regional blood flow changes, and the ‘hypovolemia’ (low blood volume) it causes seems to track the neurological changes seen during cortical spreading depolarisation and the following depression. In 1981 this was named ‘spreading oligaemia’ 2 by Jes Oleson, a Danish neurologist, one of the first to suggest cortical spreading depression as the mechanism underlying migraine with aura. This was helped in no small part by his decision to set up the Copenhagen Acute Headache Clinic, a place where people in the throes of a headache could come to get some help if they happened to be in Copenhagen, but who also provided valuable data to allow Jes and his colleagues to test their theories. What Jes found was that this hypovolemia spread at the same rate as the neuronal excitability wave, and would continue for 30–60 minutes depending on the patient. Jes observed two other issues:
1 Cortical blood flow continued to be reduced after the aura and during the initial phase of the headache, with patchy increased flow seen up to 2–6 hours later.
2 The side of the headache usually corresponded to the side of the vascular change.
This is excellent evidence to point to potassium as our culprit, because potassium acts directly on the trigeminal pain receptors, or ‘noiciceptors’, of the arterioles. Therefore, the pain that is generated in the headache phase of the migraine that has been preceded by aura comes from something that happened in the brain and is not purely a vascular issue (as in other headaches). With blood flow reductions of about 20–25 per cent, it’s not thought that this is powerful enough to underlie the focal effects of the migraine pain, but it does serve to trap the potassium in situ to do its worst on your noiciceptors and can become critically focused in one place. And what happens when you activate the pain receptors? Your brain thinks you are under threat and arranges a hefty inflammatory response to sort it out, including things we’re familiar with by now – prostaglandins, mast cells full of histamine and nitric oxide, all of which will try to induce vasodilation to regulate the blood flow again.
Other changes outside of the cortex are happening, too; the hypothalamus is activated, because it is certainly involved in the autonomic response to threat, and dopamine and neuropeptide Y concentrations increase. The brainstem is cooking because this is where all of the cell bodies of the sensory pain neurons are grouped and activity here is tightly coupled with the hypothalamus through the production of certain proteins that is prompted by the cortical spreading depolarisation and depression. Neuropeptides such as Substance P and nitric oxide as well as calcitonin gene-related peptide (CGRP, more of which here) are released, which perpetuate the inflammatory response in the meninges or the membranes covering the brain and cause vasodilation there. Also, the middle meningeal artery is dilated via the ‘trigemino-parasympathetic reflex’ – sensory activation of the trigeminal nerve automatically leads to dilation of this blood vessel that feeds much of the back of the brain and brings blood to the frontal regions, too.
This all happens even without the pain signals being perceived by you, in much the same way that if somebody taps just below your kneecap, your leg extends by reflex to protect the tendon there because the tap makes it think it has been stretched.
The aura and what is happening in the brain to underlie these sensory disturbances very clearly explains what then causes the classical migraine pain in a particular place on your head. But hold on, you can have the very same migraine pain without experiencing the aura. Does that mean that common migraine is a purely vascular headache with no underlying brain changes driving the conditions that create the pain (that comes from the blood vessels, as with all other headaches)? Well, Peter Goadsby from King’s College London believes that all migraine headaches start with these activity and chemical changes in the brain. He argues that cortical spreading depolarisation happening is actually happening in the brain of every migraineur, it’s just they might not experience it as a sensory disturbance. Andrea Harriott from the National Yang-Ming University in Taipei puts it beautifully. It is possible, she says, that some migraine patients without a perceived aura still have the electrical and chemical changes going on in their brains, but that their ‘cortical ineloquence stops it from being felt.’ It also explains why people like Ben – for whom the door disappeared – has only experienced three auras in his life, although he has had countless migraines in his 44 years on the planet.
What a pain
Migraine pain is particular. True to its original moniker – hemi (‘half’) and krania (‘skull’), morphing into the word ‘migraine’ through Late Latin and then French – the pain is felt in one side of the head. It might always be the same side, or even the same spot, but it doesn’t have to be. Either way, the pain doesn’t usually switch sides, though eventually it may be felt all over the head as the pain phase progresses.
The ubiquitous throbbing pain that migraineurs describe is often centred in the forehead towards the temple but actually can be felt anywhere, but with the very back of the head being the least reported pain zone. (In fact, headaches here perversely often indicate eye strain as this is where most of the early visual processing is done.) The throbbing often appears to follow the heartbeat and indicates the extreme vasodilation that is occurring in the brain.
The demon within
The pain of migraine has been described by cultures for more than 5000 years, going back to the Mesopotamian poems of Sumer and Babylonia, when it was given equal status with the other physiological effects of photophobia (sensitivity to light), nausea, vomiting and a tension in facial features. It’s not hard to imagine how this taking over of one’s physiological state five millennia ago would have been seen at the time as some form of possession. The finger of blame was pointed at demons, in particular in Mesopotamia and the demon class Asakku, 3 which were held responsible for everything from epilepsy and headache (interesting, now knowing the genetic link they have!) to infectious diseases but that manifested themselves as shivers and chills, jaundice and exhaustion. Migraine was treated with incantations but also trephination, or making a hole in the skull to let the poisoned blood and, of course, the demon out. Archaeology tells us that this was a widespread medical practice from Neolithic times, wherever early humans found themselves, although the justification for it may have changed over time.
Given that migraine has been around for so long, it is rather fascinating that it hasn’t been selected out of our genome by evolution if it serves no valuable purpose; there must be some reason why we need it like a hole in our head. We’ll come back to this later.
But why do we experience it at least initially in one place? The answer to this lies in our old friend, referred pain. The nerve fibres that carry the pain signals from both the meninges and skin and muscles are all lumped together in one tract, and also can come from a wide area of the brain. It is the same principle whereby pain signals from the heart – perhaps if the heart muscle is not getting enough oxygen – is not perceived by you as a ‘pain in your heart’ but rather a ‘pain in your left arm or jaw’. It’s because the sensory nerves from your heart are bound up with the pathway from your arm and jaw, and the brain can’t distinguish between them. So in migraine, you feel the pain in one focal place and where that is depends on your anatomy.
The migraine pain signals in the headache phase are created in three ways. The first is the extension of the blood vessel walls themselves, indicating danger. The second thing to think about is that the substances that are causing the inflammation are making the peripheral branches of the trigeminal nerve (and remember, this extends all over the head) much more sensitive to pain than usual. The third problem is that the spinal trigeminal nucleus of the brainstem is in overdrive because those peripheral receptors are going nuts. These activations are passed through connections with the great sorting house of signal known as the thalamus, 4 which sits above the hypothalamus (and is in fact how the hypothalamus gets its name, borrowing from the Greek hypo for ‘under’), which has anatomical links all over the brain and so explains many of the other symptoms of migraine. Activations from here to the motor cortex make you unable to control your muscles, including, most commonly, those of the face. Attempting to move makes pain even worse because you are trying to make a part of your brain that is already buzzing do something else. You might be clumsy and unable to focus because your parietal cortex is being stimulated (meaning you can’t use it for your aims of planning to get from A to B). Somatosensory cortex activation will of course make you feel the pain in your head but it will also make you tingly or numb or even hugely sensitive to touch, to the point where you feel it as pain anywhere in your body. If your temporal cortex goes off then the same will happen with sound. You might have memory loss for a while, too, in addition to being extremely sensitive to odours. The gold standard of course is photophobia, where overactivation of the visual cortex means that any more activation by light sends it into hyperdrive, further perpetuating the pain response. This completes the cacophony of cortical catastrophe (now there’s a tongue twister).
The structures that live under the cortex are vulnerable too. For example, the trigeminal nucleus of the brainstem is connected to various areas of the hypothalamus that control the endocrine hormone system as well as the autonomic nervous system of our whole body. So this explains the systemic response of nausea, vomiting and appetite changes. Really interestingly, there is a pathway that extends to the zona incerta, which sits below the thalamus, too. We talked about this region in Chapter 4 because it is an area that we are beginning to understand as being involved in your subjective experience of pain, i.e. it is part of the system that determines your pain threshold. Standing on a Lego brick in your bare feet in the dark is the most unimaginable pain known to humans (OK, we are talking about me here) save for childbirth. But some people brush it off because they have a high threshold for activation of their zona incerta. Your levels of endogenous opioids, or your body’s natural painkillers, are also important in this dance our brain does to decide how bad pain is. But the zona incerta is also involved in chronic pain.
Sameth Eldabe, a world-renowned pain consultant, tells me that people who had spinal surgery in the past because of a trapped nerve, misplaced disk or similar sometimes revisit their clinician years later because their pain symptoms have returned. After tests, the doctor may have to tell them that there is no physical cause for their pain – the spine is fine, nothing is impeded. This is chronic pain. The doctor can only help people cope with the pain. But guess what? Spontaneous activation of the zona incerta is seen at the arrival of the perceived pain. So, we can separate out the sensory component of pain (the somatosensory cortex) and the affective component (the zona incerta). If, as Rodrigo Noseda, a lifetime migraine researcher from Harvard Medical School, says, there is a direct activation of the zona incerta in the migraine, there is an instant mechanism by which any pain will be dialled up in your perception. What a mess! This phase can take from 4 to 72 hours to clear but of course depending on your physiological state before the attack, it could take longer. And it doesn’t end there.
It ain’t over till it’s over – the postdrome phase
Without treatment, this phase of the migraine has to take its course. The brain’s natural way of blocking pain – including boosting dopamine, serotonin and natural endorphins (which act like opioid painkillers such as morphine) – will eventually manage to get control and the trigeminal pathways in the brain will calm down, which in turn will stop the nausea through the hypothalamic pathway. And now we enter the postdrome phase. Energy is low, you are mentally foggy. Two plus two might make four but who the heck cares? It is a literal headache hangover, and you had zero fun getting there. Having said that, some people feel euphoric in this phase, which is due to a boost in both the endorphins needed to kill the pain but also the dopamine that has also been released to help block the pain signals. 5
This ‘out-of-whack’ or ‘zombie-like’ feeling is present after the peak pain has passed. Even though the pain is dissipating, the wider hypothalamic functions still need sorting out. Your hormones are all over the place. The length of time your postdrome phase takes is dependent on how disturbed these hormones are and given that it is a very interdependent system and controls most of your bodily functions, it’s not hard to see why it has such a widespread effect. Pyari Bose who was doing his PhD with Peter Goadsby at King’s College London recently used fMRI (see here) to show that there is a distinct decrease in blood flow around the brain in the postdrome phase, which is hardly surprising in light of the vasoconstrictor brakes the autonomic nervous system had to apply to all those dilated blood vessels! Also, it is unwise to ignore it. Symptoms such as this are trying to tell you something. Mental tiredness means your neurotransmitters need replenishment or rebalancing; rest and sleep help with that. Both mental and physical exhaustion can be helped by introducing a sensible eating pattern with adequate hydration.
Give your body time; it has undergone a pretty spectacular brain event that’s affected everything, it has to recover. If you don’t, and your hormones and neurotransmitters remain haywire, a migraine will happen again, and sooner than you would wish.