Timothy C. Hain, MD, Chicago IL. Page last modified: July 26, 2020
More precisely, migraines are recurrent headaches separated by symptom-free intervals and accompanied by nausea and light sensitivity. Migraines are often accompanied by visual symptoms and are relieved by sleep; furthermore there is usually a throbbing quality. More often than not there is a family history of migraine.
Migraine is a "committee" diagnosis - -meaning that a group of people got together somewhere and decided what they were going to call a "migraine". Modern genetics has established that there are a large number of genes that contribute slightly to the probability that someone will be diagnosed with "migraine". In other words, we are not dealing with a "disease" - we have a group of symptoms that are subcategorized by a committee that was set up prior to the advent of modern genetics.
There are several other "committee" diagnoses relevant to vertigo -- including Meniere's disease. For "committee" diagnoses, "ruling out everything else" is usually required by the diagnostic process. Some disorders that we deal with commonly in dizzy persons, such as "cervical vertigo", don't have a committee, and being undefined, are less studied than disorders fortunate enough to have their own committee. On the other hand, if one needs a committee to define something, it means that the "somethings" very existence is subject to negotiation.
Accordingly, Formal criteria for migraine, have been offered by the IHS (international headache society) in a very long document. Because headaches are common and because the IHS criteria include nearly anything you can imagine, migraine is a very "inclusive" disorder. It is to be hoped that eventually a blood test or genetic test will be developed. So far, there seem to be many genes and it seems highly unlikely that a single blood test will identify the interaction between 40 genes.
If one attempts to distill out the most useful parts of the IHS document, migraine usually involves a combination of headache and sensory sensitivity.
Concerning sensitivity, there are various types of sensory inputs which may trigger migraine headache -- light (photophobia), sound (sonophobia), smells (such as perfume), and motion -- persons with migraine are often prone to become motion-sick. They are more sensitive to medications than most people, and they often have symptoms triggered by weather changes.
Psychological testing shows unsurprisingly that people with an active migraine headache don't think as well as usual (Meyer JS et al, 2000).
Although migraines are usually episodic, they can be chronic too. Chronic migraine is the most severe of all migraine syndromes, with headaches averaging grater than 15 days/month. Each year, about 2.5% of those with episodic migraine develop chronic migraine (Manack et al, 2011)
Although migraine headache is usually unilateral, opposite sides of the head are characteristically affected during different attacks. About 20% of Migraine's are preceded by an "aura" (see picture on the right), meaning visual symptoms, dizziness, numbness, or weakness. In fact, about 1% of the time, the aura may dominate the migraine, and there may be minimal or no headache ! Migraine auras usually last 5 to 60 min. Visual auras begin in central field of vision and move to the periphery. Another common visual aura is a scotoma (black spot). There is a genre of art called "migraine art", based on migraine aura. There are also sensory auras -- they often move from hand to arm to face and ipsilateral tongue. Presumably, any part of the brain might have an aura. The possibilities are endless !
Visual aura mimics include AVM, TIA, retinal disease, and some focal seizures.
There are numerous defined variants of migraine in the very long IHS document.
Migraine (i.e. the vaguely defined headache entity with 40 different genes) is associated with increased risk for both dementia and stroke.
According to several recent research articles looking at seniors with a "diagnosis" of migraine, , there is increased risk for dementia including , both “all cause”, and alzheimers, increasing the odds of diagnosis of dementia in older persons within 10 years by a factor of about between about 1.6 to 3 over the normal risk (Kostev, Bohlken, & Jacob, 2019; Morton, St John, & Tyas, 2019) Please note that we are not talking here about a study of 35 year old women -- this was a study of persons whose average age was 67.7.
Paraphrasing Kostev and associates recent study, " The main outcome of the study was the association between migraine headaches and the incidence of dementia within 10 years of the index date. RESULTS: This study included 7,454 individuals with or without migraine diagnoses. Mean age was 67.7 years (SD = 5.8 years), and 72.9% of patients were women. We observed a positive association between migraine diagnoses and all-cause dementia (hazard ratio [HR] = 1.43) as well as Alzheimer's disease (HR = 1.87). Sensitivity analyses further revealed that these associations were only significant in women (all-cause dementia: HR = 1.65; Alzheimer's disease: HR = 2.27). One wonders here how the risk scaled with the intensity of migraine symptoms.
Morton et al said: "A history of migraines was significantly associated with both all-cause dementia (odds ratio [OR]=2.97; 95% confidence interval [CI]=1.25-6.61) and AD (OR=4.22; 95% CI=1.59-10.42), even after adjustment for confounding and intervening variables. " Their study was of persons aged 65+. Again, one wonders here how risk scales with intensity. Ordinarily, there is little active migraine remaining in persons over the age of 65 (from epidemiological studies).
Treatment of migraine with SSRI’s seems to decrease risk of dementia. (Lee et al., 2017), so there are some mitigating factors.
Regarding stroke, a Migraine "diagnosis", also increases risk of a later diagnosis of stroke. An analysis from 2010 suggested that the relative risk ratio is about 2 fold. (Spector et al., 2010)
Quoting them directly, "The pooled adjusted effect estimates for studies that reported relative risks and hazard ratios, respectively, were 2.41 (95% CI, 1.81-3.20) and 1.52 (95% CI, 0.99-2.35). The overall pooled effect estimate was 2.04 (95% CI, 1.72-2.43). "
This study was a meta-analysis including a large number of studies within it. Of course, this conclusion depends greatly on what you call a "stroke", as in migraine there are often small white spots that are variously called strokes or not. We ourselves accept that there are more "white spots" in migraine, but at least in our clinical experience, we are not as sure about major "strokes". There is not much about the effect of migraine treatment on stroke risk, but it would seem reasonable to expect that it would reduce the risk. We suspect (but have no proof), that treatment with migraine with drugs that reduce vascular risk (such as beta-blockers or verapamil), woud also reduce the risk of stroke from migraine.
WHO GETS MIGRAINE ?
There have been some HUGE studies of migraine, perhaps because one can "diagnose" migraine from a survey. Supposedly, about 11 million Americans have significant problems with migraine headaches and about 10-15% of the population get migraines, at least occasionally (Stewart et al, 1994; Lipton et al, 2002; Burch et al, 2019). The main problem with this statistic is that there is no biomarker for migraine, and one must wonder what exactly these survey studies are finding -- perhaps they are showing that lots of people have bad headaches from time to time. As of 2017, it is obvious that "migraine" is not a single disorder, even though we have a substantial medical infrastructure built up around the idea that "migraine" is a disease. This is a common problem with "committee" diagnoses, that have no biomarker, and similar criticism can reasonably me leveled against the process of "diagnosing" many other illnesses (for example, most of psychiatry). We can only hope that as molecular genetics and biology progresses, we will be able to "split" up these conditions into more meaningful groups.
Well anyway, women are especially likely to get migraine (3:1 female:male ratio, 4:1 during childbearing years). The age group between 30 and 45 has the peak incidence, roughly 25%. Women probably get more migraine than men because of hormonal fluctuations, and the incidence appears to be increasing in women (Rozen et al, 1999). Women who have recently transitioned into menopause have more migraine.
Children have migraines too. See this page for more about migraine in children.
WHAT CAUSES MIGRAINE ?
Well, its messy - -current theories usually contain one or more of the following ideas:
Because migraine is so broad, there may actually be many causal variants. It seems unlikely that there could be so many "causes" in a single disorder. In fact, one might even question whether the term "migraine" is useful one at all, given the huge heterogeneity in symptoms and genetics.
The current consensus is that migraines are caused by abnormality in the brain which represents a combination of familial tendency, trigger factors such as stress, sleep disturbance, hormonal fluctuations, and certain foods. While in the past migraine was felt to be related to vasospasm, presently it is thought that the blood flow changes are not primary. Instead, it is felt that migraine is related to abnormal sensitivity to sensory inputs (Goadsby PJ, 2001). Nevertheless, there is reasonable evidence in the other direction -- for example, some forms of migraine are associated with a mutation in a gene that controls a potent vasoconstrictor (Tzourio et al, 2001), and there is also evidence for vascular muscle dysfunction in migraine (Napoli et al, 2009; Asghar et al, 2011). Really there is evidence for almost everything in migraine (sadly). This is probably because migraine is not a single disease, but rather a collection of symptoms named "migraine" by experts.
Our most effective medications for migraine headache (ergots, triptans) manipulate a blood chemical called serotonin. Medications that manipulate another neurotransmitter (dopamine), are also often effective treatments for migraine although they have side effects and abnormal dopamine is unlikely to the core central problem (Mascia et al, 1998)
Another neurotransmitter called CGRP -- calcitonin gene-related peptide also appears to be important to migraine. CGRP causes cranial vasodilation and facilitates pain reception. During a migraine attack, trigeminal activation results in release of CGRP. CGRP is increased in persons with chronic migraine. There are currently several drugs that target CGRP, but with relatively low response rates (think 1 in 5). There is also no reason to think that these drugs will help with light sensitivity, dizziness, sound sensitivity other than through placebo effects.
Migraine does not appear to be strongly related to allergy or immune disturbances in most instances. However, histamine released in allergy may trigger migraine.
HOW ARE MIGRAINES DIAGNOSED ?
The diagnosis is made through identification of characteristic features of the headache and examination. The features that one looks for are on the list of symptoms defined by the international headache society.
Patients with migraine typically are required to have sensory amplification symptoms --
In our practice, which largely contains persons with dizziness and headaches, we have also observed that many patients with "migraine" are very sensitive to changes in barometric pressure, and also that they are more sensitive to medication.
Migraine shares central allodynia with fibromyalgia, which interestingly enough, is often treated with a similar repertoire of centrally acting medications (e.g. SNRI's, gabapentin). Fibromyalgia is reported in 12.5-31% of patients with episodic migraine and about 36% of patients with chronic migraine, while 65% of patients with fibromyalgia have been reported to have migraine (Liu et al, 2015). Fibromyalgia, like migraine, is another symptom collection, and given recent genetic developments in migraine, perhaps the two are just variants determined by a particular subset of genetic and environmental risk factors.
We also see patients who have selective excessive sensory amplifications -- phonophobia, allodynia and motion intolerance being the main variants. There are also a few with smell intolerance. All of these central sensitivities also have other causes -- photophobia in particular was recently reviewed by Digre and Brennan (2012), who point out that dry eyes is a common cause too. If one pursues the idea that there are multiple genes that are associated with "migraine", one might also think that perhaps subsets of these genes also accounts for these limited sensory syndromes, lacking headache.
Interestingly, patients with "migraine" are "wired differently" and have thicker sensory cortex than patients without migraine (DaSilva et al, 2007). They are more "sensitive" to pain (Lipton et al, 2008). Patients with migraine have visual cortex that is more excitable than normal during migraine attacks (Denuelle et al, 2001).
As migraine is a "committee diagnosis", tests are used to exclude other entities rather than to make a positive diagnosis. No X-rays or blood tests are available to make the diagnosis, but in persons with severe and recent headaches, neurological findings, or complicated medical histories, the physician may recommend an MRI or CT scan (Silberstein, 2000). Imaging is generally not necessary if the examination is normal and the headache pattern is unchanged. Lumbar punctures are obtained when subarachnoid hemorrhage or meningitis is suspected.
We have collected together some of the papers on vestibular and hearing testing in migraine here. As a general comment, none of them are sensitive to migraine.
Migraine headaches are often misdiagnosed by patients themselves as sinus headaches. A study suggested that 88% of 2991 patients who had diagnosed themselves as having sinus headache, actually had migraine (Schreiber et al, 2004). What this study really said was that they didn't have sinus inflammation, as there is no methodology of positively confirming that one has migraine.
Migraine, being so broad, accepts headaches left over from other conditions that are defined with greater specificity. In other words, migraine is often a "catch all". Or a "wastebasket", if you prefer that term.
MRI scans may reveal white matter lesions in young persons with migraine. These can and often are confused with white matter lesions due to multiple sclerosis or white matter lesions that occur in older people (periventricular white matter lesions). This is not the case in the majority of people however. For more information, see this link: white-matter.
MR angiography can be used to diagnose vascular malformations. MR venography can be used for sagittal sinus thrombosis. MR also may detect low-pressure type headaches, tumors, as well as the Chiari malformation. Fortunately all of these are rare.
Persons with prolonged auras (> 60 min) should get coagulopathy testing--to look for lupus antibody, anticardiolipin antibody, protein's C and S, factor V Leiden. However, this is a low-yield endeavor, and headaches are not significantly associated with lupus (Mitsikostas et al, 2004).
It would seem likely to us that tests of high sensory reactivity, or inability to block out unwanted distractors, would be a reasonable method of screening for Migraine. Perhaps there is room for more research ?
The best strategy seems to be adjust the intensity of treatment to the severity of ones migraine condition (stratified care). It doesn't work as well to start cautiously, and then escalate if the first treatment doesn't work (Lipton et al, 2000). There are (at least) 6 different ways to manage Migraine. In all instances, it is best to keep a "log" of headaches, such as a diary, in order to determine whether or not a particular treatment is effective. In our neurology practice in Chicago, we use this form. It often works very well to simply print out a calendar on a page, and using a "traffic light" color scheme (green=good, red=bad), indicate days that one has symptoms.
1. Avoid trigger factors:
2. Simple non-drug treatment that should be tried first
3. Pain medications
For those who have more than 2 severe headaches/month and in patients with complicated migraine (migraine with stroke-like features), a daily medication may be worth while. These are generally highly effective (about 75% effective), but do require daily regular use. These drugs fall into three major classes: anticonvulsants, antidepressants, and antihypertensives. Examples are: Amitriptyline (Elavil), Corgard, Depakote, Inderal, Nardil, Verapamil (Calan, Isoptin). These drugs seem to work via several pathways: some are beta-blockers (e.g. Inderal, Corguard), some are calcium channel blockers (e.g. verapamil), some work in mysterious ways (e.g. Depakote, Nardil, amitriptyline). More information about these is in the next section.
5. Acute, specific medications (also called "abortive" medication):
For those who have severe but infrequent headache. Highly effective, especially Imitrex (sumatriptan). There are numerous others in the same family -- Zomig, Maxalt, and Axert, and Frova to name a few brand names. There are also older (cheaper) drugs, called ergots, that have similar effects but have more side effects -- examples of these types of drugs are: Cafergot, DHE, and Ergotamine. Drugs that are dopamine blockers include Metoclopramide and prochlorpromazine.
6. Nausea medications: For those with prominent nausea.
The "cyclic vomiting" variant of Migraine often calls for vigorous uses of these agents. These drugs are usually highly effective, but cause some major side effects. Examples are: Compazine, Phenergan, Reglan, Thorazine, Tigan. These drugs also have some utility as abortive agents, presumably because of their effects on the dopamine pathways. Droperidol is very effective in Migraine (Ashkenazi et al, 2003). but because of it's "black box warning" related to sudden death from cardiac side effects, we think it is best not used for a recurrent non life-threatening condition like migraine. Haloperidol is also very effective, and can be useful in occasional persons with severe vomiting.
As you can gather from the discussion above, "migraine" seems to be basically a collection of symptoms, that don't fit elsewhere, that usually include headache. The genetics suggests that the underpinning molecular biology is very broad. One would then expect that the treatment would be trial and error. That indeed does seem to be the case -- trial and error, ideally accompanied by good "clinical judgment", meaning someone who can recognize symptom clusters and predict the medications that are most likely to make an impact.
There are three main groups - - seizure medications (e.g. Topamax -- topirimate), blood pressure medications (e.g. verapamil or propranolol), and antidepressants (e.g. nortriptyline or venlafaxine). The author of this review usually starts patients with topamax, and proceeds on to try effexor, verapamil, propranolol and then ami or nor triptyline. It is very unusual that headache control is not attained with a single drug. When one "group" doesn't work, he may combine two or 3 groups simultaneously (anticonvulsant, blood-pressure agent, antidepressant).
Migraine abortive drugs
Imitrex (sumatriptan) is an example of a category of drugs called triptans. As a general comment, other than the generic sumatriptan, the triptans are expensive but very effective for headache. At this writing, this category includes sumatriptan, naratriptan, zolmitriptan, rizatriptan, almotriptan, frovatriptan and eletriptan. These drugs are all group-1 agents. These drugs are 5HT-1B and 1D (serotonin) receptor agonists. Some also affect 5HT-1F. Serotonin does a lot of things in the body and there are many receptors, presently ranging from 5HT1-7. Other drugs that affect serotonin are the SSRI and SNRI families of antidepressants and the "setron" family of antinauseants, such as ondansetron, which antagonize the 5HT-3 receptor. The FDA has advised caution when taking both triptans and SSRI/SNRI's together. We have never encountered problems, but nevertheless we suggest avoiding taking large amounts of triptans and SSRI/SNRI's together. This mainly is a difficulty when someone has two conditions - -major depression and migraine - -simultaneously.
Imitrex was first released as an injection given under the skin (subcutaneously) at time of headache. Imitrex is not recommended for patients < 18 years of age although several studies have been performed on adolescents with similar results to adults.
|(modified from Matthew and Loder, 2005)|
Presently triptans are also available as a pill (50-100 mg is best for sumatriptan), as a nasal spray and as a sublingual preparation (Maxalt and Zomig). The sublingual forms are generally preferred because of fast and effective action. Eletriptan, a recent addition, has a nasal form that is well suited to rapid treatment (Ashkenazi and Silberman, 2003), and both rizatriptan and eletriptan have a very rapid onset of action when taken orally.
A recent addition is frovatriptan (Frova) , which has a very long half-life (about 26 hours). Longer half-life drugs have less recurrences. The triptans are very effective for migraine (about 60-85% relief from active drug vs. 20-40% in placebo) but also very expensive (roughly $75/injection or squirt).
There are now some agents which cost a much less -- Sumatriptan, being generic, is not unusually expensive. Generic versions of the injectable sumatriptan are also reasonably priced. We can probably anticipate similar drops in the prices of Relpax and maxalt, once they emerge from patent protection. The triptans usually relieve migraine associated nausea as well as headache. These drugs are also reported somewhat effective for tension headache (Lipton et al, 2000).
Because of the high cost, some pharmacy programs limit quantities to about 8 uses/month. For the injection form, chest pain occurs in 3-5% of patients, usually from esophageal spasm. Myocardial infarctions, however, have rarely been reported after triptan use. For the nasal spray, a bitter taste in the mouth is the most common side effect (Ryan et al, 1997). Sumatriptan can be used up to twice/day. Very surprisingly, sumatriptan is not generally effective in children (Hamalainen, 1997), although some reports suggest that the nasal spray may work.
Triptans shouldn't be used by persons with heart trouble or blocked arteries (Maasen, Vandenbrink et al, 1998). Triptans should not be used within 24 hours of using an "ergot" type medication (see below) or Sansert (no longer available). Use in hemiplegic or basilar migraine (i.e. migraines with stroke like symptoms) is generally contraindicated but the evidence against using them is almost nonexistence. Nevertheless, In the author's practice, these drugs are also avoided in persons with MRI evidence of small infarcts, which is common in persons with severe migraine. Coadministration with MAO inhibitors is contraindicated (see preventive treatments). See table under "addictive medications" for guides for use. While triptans are generally effective in cluster headache, extreme caution is advised, because of the temptation to overuse this drug, and also the tendency for cluster to occur in persons with higher risks of heart attack.
Similar medications to sumatriptan (Imitrex) are zolmitriptan (Zomig), rizatriptan (Maxalt), and eletriptan (Relpax). Amerge (naratriptan) is a slower onset, more prolonged version. While triptans are not recommended for use in a preventive, "prophylactic" mode, nevertheless Amerge is sometimes used for several days in a row to prevent menstrual migraine. Frova is a very long half-life triptan. Axert (almotriptan) is also somewhat longer lasting. Relpax is one of the stronger ones (Sandrini, Farkkila et al. 2002; Farkkila, Olesen et al. 2003). Some pharmacy programs limit use of Zomig to 6 tablets/month of the 2.5 mg form, or 3 tablets of the 5 mg form, and Amerge to 9 tablets. It is generally felt that it is unreasonable to use triptans more often than every other day. Triptans can lead to rebound (withdrawal) headaches (Limmroth et al, 1998). Inappropriate use of medication including dependence is certainly possible and, because of it's high price, may impact health care spending significantly (Gaist et al, 1998)
An older similar medication, DHE (dihydroergotamine), also group 1, is less used because of greater incidence of side effects. DHE and other ergots are broader in their action than the triptans and this is probably the reason for their increased side effects vs. the triptans. Nevertheless, DHE is now been recent released as a nasal spray preparation (Migranal). DHE are contraindicated in patients with renal or hepatic insufficiency, coronary, cerebral or peripheral vascular disease, and uncontrolled hypertension. These agents are also contraindicated in women who are or may be become pregnant. Caution should be used with ergots in lactating women, in patients also receiving peripheral vasoconstrictors, 5-HT1 agonists, propranolol, nicotine and macrolide antibiotics (PCS, 1999).
Other ergots --
Ergot medications often undergo shortages -- these very old drugs are so often unavailable on a world scale. For example, ergotamine was mysteriously unavailable when Imitrex (sumatriptan) was first introduced.
Ergotamine tartrate (Wigraine) SL or suppository. This is obviously an "ergot" type medication too. Take at time of headache, repeat in 30 minutes if not effective. Wait 3 days before using again. Should NOT be used by persons with heart trouble or poor circulation. Coldness and tingling in the legs suggest a need to stop treatment. Otherwise, contraindications are similar to DHE (see above). Ergotamine is rarely used, being replaced by Imitrex.
As basic research suggests that these medications inhibit firing in trigeminal pain pathways via 5HT1B and D receptors (Goadsby and Hoskin, 1998), in theory at least, this group might also relieve other types of head pain such as sinus headache, tension headache, and toothache. Generally, it is not used for these purposes, and a good response to a "triptan" is considered a reasonable indication that the patient has a migraine.
Midrin (isometh/Dichlor/acetaminophen 65/100/325) was taken off the market as of 2011, evidently due to concern about one of its components, chloral hydrate. It reemerged thereafter, but with a change in formula to remove the chloral hydrate. Midrin was a group-2 drug, also considered an "ergot" type medication, although the pharmacology (see below) suggests that it is basically a combination of a vasoconstrictor, a sedative and analgesic. The instructions were: Take at time of headache, repeat in 30 minutes, up to maximum of 5. Wait 3 days before using again. The ingredients in Midrin also include dichlorophenazone (a combination of chloral hydrate and phenazone) and acetaminophen. Chloral hydrate, of course, is a sedative. Phenazone and acetaminophen are both mild analgesics. The rationale for combining them is obscure. Isometheptene is a sympathomimetic, that causes vasoconstriction. Similar to caffeine. Midrin was reclassified as a "schedule IV" drug due to the dichlorophenazone, which was considered a schedule IV drug in 2001. A Schedule IV drug is a low potential for abuse. Other countries, such as the UK, do not consider chloral hydrate to be a controlled substance. Chloral hydrate can be prescribed separately as well. Our thought is that Midrin was not causing any problems prior to 2011, and there must have been some other motivation to remove it from the market. Headache medications often vanish mysteriously from the market, especially when competitors are coming onto the market.
Naproxen (Alleve 220 mg) and Ibuprofen (Motrin 200 or 400 mg). These medications are "non-steroidals". There are many others -- there is no reason to suspect that they vary in any significant way other than side effects. Non-steroidals or "NSAID" medications are both useful for pain at the time of headache, and may also help on a daily basis. Main problems with them are stomach irritation and diarrhea. Don't take large amounts as these drugs can damage the kidney and liver too. Take with food. Aspirin may work as well. Cost for one of the older type NSAID's is about 0.20 $/day.
There are also other very strong nonsteroidals -- examples being Cambia (a powered variant of a common NSAID, volteran), Sprix (a nasal spray), and injectable nonsteroidals. When a very expensive NSAID works, usually one can find a generic equivalent.
Droperidol has been reported as an effective acute treatment for migraine (Silberstein et al, 2003). Because droperidol has been associated with occasional life-threatening side effects, however, we do not recommend it for this purpose. Haloperidol is also effective in aborting migraine, especially the basilar artery type migraine.
Steroids, according to Silberstein (2000), are not effective acute therapies for migraine. Nevertheless, they are often used in a short course to "break" a severe bout of migraine or cluster headache. This suggests they may be effective subacute treatment.
We do not recommend that you stop prescription medications without the permission of your doctor. However, certain medications can trigger headaches. Note that some medications in the same category that cause headache, may relieve headache. There are far too many medications that cause migraine to list here, however the most common ones are listed.
Migraine headaches often evolve or transform into daily headaches, and this pattern has been termed the "analgesic rebound headache". Common features are thought to be:
Some authors suggest that in addition there is tolerance to analgesic medication, and efficacy of usually effective medication is compromised by ongoing consumption of immediate relief medications. Treatment involves withdrawal of analgesic medication. This sometimes must be done during a hospital setting
A recent review by Bigal and Lipton (2008) suggested that risk of progression of acute migraine to chronic migraine, and associated rebound headache can be predicted from a combination of frequency of intake and the type of medication. They indicate that there is both an effect of medication inducing chronic headaches as well as pain inducing chronic medication use -- in other words, this is a "chicken and egg" problem. Of course, in general it is prudent to consider the possibilty that cause/effect is more conventionally arranged -- instead of medication->headache, why not headache->medication ? As the brain tumor gets bigger, perhaps the person is taking more pain medication ?
Bigal and Lipton suggest that drugs generally recocognized as addictive -- opiates and barbiturates -- should be carefully monitored and limited as much as practical. From their data, it would also seem prudent to limit and attempt to decrease the frequency the use of drugs not usually considered addictive -- namely triptans and antiinflammatory medications.
In our clinical practice, we will sometimes attempt a strategy of small amounts of several headache drug categories. This strategy is predicated on the hypothesis that habituation and addiction is less likely in persons who are not consistently on drugs of the same mechanism. If, all of these drugs work through the same mechanism to convert intermittent headache into chronic, this strategy would not work. More study is needed.
Other very reasonable approaches to the problem involve use of migraine prevention drugs rather than analgesics, and use of approaches such as "ice to the back of the neck", to avoid use of daily analgesics.