why migraine and diabetes are associated

Why Glucose Testing for Diabetes is Inaccurate

8271 views

Do you have type 2 diabetes (T2D)? How about prediabetes? OK… how about insulin resistance (IR)? No? Are you sure? I am not sure at all! Not, based on the many cases I evaluate daily. Here I present you with one case study that is not from the US or Canada. The reason why I picked one that is not from North America is because if you live in North America, more often than not your T2D or IR is not discovered by conventional blood glucose testing. This article will demonstrate the problem.

Why You Don’t Know that You Have T2D or IR

For simplicity, I will use the term T2D to represent all levels of metabolic syndrome: type 2 diabetes, prediabetes, and insulin resistance. According to the CDC, the total diagnosed T2D in age groups of 18 to 65 was around 12% of the US population (table 1) in 2017, though in 2015 the diabetes and prediabetes population was estimated at 56.1% (table 3), meaning the undiagnosed with T2D is 34% of the population, nearly three times that of the diagnosed. Currently, there 132,000 children under the age 18 assumed to have T2D but remain undiagnosed. Of the US population, about 5% of the people (of all ages) are estimated to have type 1 diabetes, which is either a genetic condition, an autoimmune disease, or in some cases may be viral—this is not yet well understood. The 5% is about double that of 2015.  It is interesting to note that type 1 diabetes is also on the rise both for children and also for people with LADA (latent onset type 1 diabetes). In this paper, I only focus on T2D, what causes T2D and the reasons that so many of us remain undiagnosed. In the case study presented here, the patient permitted me to use her blood test results.

Undiagnosed T2D: A Case Study

I am fortunate to be able to work with thousands of migraine sufferers, many of whom bring their friends and family to my migraine groups even though they may not have migraines. This case study is about such a person, whom I will call now Nicole (not her real name).

Nicole joined my migraine group because she has a relative with migraines and she had her own health conditions. I have a questionnaire that each new member fills out, and the answers are sent to me by FB private messaging. Nicole went to her doctor and received her blood test. It was an oral glucose tolerance test (OGTT), in which they first test a fasting measure, then she drinks 75 gr glucose, and 1 hour and 2 hours after finishing the glucose drink her blood is tested again. Here are her blood glucose test results:

Since she is from a county where they use metric, on her test you see metric blood glucose results. Here are the results in US standard:

  • Fasting: 105 mg/dl
  • 1 hour after OGTT: 172 mg/dl
  • 2 hours after OGTT: 125 mg/dl

Note that all three are unremarkable though her fasting glucose is higher than the desired <99 mg/dl (<5.5 mmol/L) but it is not flagged—the Dawn Phenomenon may increase fasting blood glucose and so slight elevations are often ignored, as they were here. Both her 1-hour and 2-hour tests show normal.

A key important point to make here is that if you live in the US or Canada, this is where your OGTT ends, you are patted on the shoulders “job well done” and sent home. This is the core of the problem. When trying to find out if someone has T2D or IR, one needs to check insulin. The glucose measure is used as a proxy but is it really a good proxy? Not at all. I explain later—let’s continue to two other measures Nicole received, none of which would have been performed in North America, but which are the most critical in understanding whether one has T2D.

The Urine Test

For a metabolically healthy person, glucose is never found in the urine. Glucose is an important energy source for some organs (not all), and in prehistoric times glucose was not easy to come by. As a result, a healthy body converts all excess glucose (what cannot be used immediately) by the liver into glycogen or fat to be stored for the future as energy. Glucose can be found in the urine of those with a compromised metabolic system, because they have a sluggish glucose absorption as a result of IR, and they urinate out the glucose that was not possible to absorb and convert into stored energy.

In the case of Nicole, while her blood glucose levels tested normal, the first sign of trouble shows in the 1-hour mark when there is glucose in her urine.

Note the 1-hour reading with 50 mg/dl glucose in the urine, while the accepted lab range is <15 mg/dl. Her urine contains over 3 times normal glucose amount. This is a major sign of inefficient glucose processing and of serious metabolic disease.

I need to remind the reader again: urine test is not provided with the OGTT in North America. I believe that this is a big mistake.

The Insulin Test

In the US, in order to get your fasting insulin test, you need to beg your doctor. Many healthcare facilities will refuse to offer it and insurance will not cover the associated costs in most places. This is the biggest mistake of them all. If we are looking for insulin resistance, why are we not checking insulin? Let’s see how Nicole did with her insulin test.

Look at how high her insulin is. While the lab range noted here is 6-24.9 ulU/ml, in reality the preferred range is 4-11, normal lab range is 2-22, where 5 uIU/ml is ideal. Her fasting insulin is over 5 times what would be considered as normal in the US and it is above normal of the acceptable by this lab as well. The one-hour postprandial (PP)—meaning after glucose solution—insulin should have been <90 uIU/ml max (hers is 165.3, which is nearly twice the maximum). The two-hour insulin PP should be <60 and Nicole’s remains high at 158, about 2.5 times the maximum (the lab range of >60 is already metabolically ill). As you see, with time passing, her insulin remains increasingly higher than normal. This is a very slow reduction. The high absolute number is not the only thing that matters. How long insulin stays high matters as well.

Diabetes In-Situ

Dr. Joseph Kraft, in his book “Diabetes Epidemic & You” discusses how easily patients with T2D (he called this diabetes in situ) passed OGTT in great numbers with flying colors as if they had no metabolic health concerns at all. In his table, 40% of those with diabetes in situ had fasting blood glucose <110 mg/dl1 (page 6)–just like Nicole. A perfect diagram was designed by Ivor Cummins, which places the 4 types of insulin responses on one graph for easy viewing.

Kraft Curves in Situ by Ivor Cummins
Kraft Curves in situ by Ivor Cummins

See the original of this graph here.

Looking at Nicole’s lab results and superimposing her fasting and two hours of insulin reading on this graph:

Nichole via Kraft Curve overlaying Kraft in Situ by Ivor Cummins
Nicole via Kraft Curve overlaying Kraft in situ by Ivor Cummins

What you see is that she fits right somewhere along Pattern 4, meaning full blown type 2 diabetes. Of course, her doctor prescribed for her medication immediately, but she decided to follow the dietary approach to T2D reversal instead of the medicinal approach—more on this at the end.

Why Do We Test Blood Glucose Instead of Insulin?

Nicole’s blood glucose test shows absolutely no T2D and barely even the slightest possibility of IR with a slightly higher fasting glucose than usual. Why do we use blood glucose as a proxy for insulin? While the reasoning defies my logic, I can speculate.

  • It is cheap to measure blood glucose. Blood glucose meters are sold everywhere very inexpensively so lab tests are only needed for an OGTT.
  • It is easy to read what a blood glucose test shows.
  • People can easily measure their blood glucose at home several times a day, before meals, after meals, etc.
  • Insulin is harder to test and is more expensive.
  • Many medical practitioners don’t understand insulin.

The following quote explains it all:

Measuring insulin levels is an unnecessary health care expenditure, added Dr. Arslanian, the Richard L. Day Endowed Professor of Pediatrics at the University of Pittsburgh. “We’re already talking about how expensive health care costs are in the United States. Why do that when your eyes can tell you — or the body mass index can tell you. If you’re obese, the insulin level will be higher. You treat the obesity and the insulin comes down. You don’t treat the insulin.” (see here)

Chicken or the Egg?

In most medical literature, metabolic disorders, such as T2D and IR, start with obesity. However, obesity is the outcome of IR and T2D and not the cause. To understand this, doctors would have to recall something important from their rote memory from many years ago when they started med school. The role of insulin is to take excess glucose out of the blood and pile it into the liver, whose job it is to store it as fat. Glucose cannot be stored as glucose for several reasons, of which one reason is that excess glucose in the blood is toxic. Storage of glucose presents a space issue as well: a glucose molecule is C6H12O6. If you recognize the formula of water (H2O), you can see that a single glucose molecule incorporates 6 water molecules. And water is heavy. Storing glucose would mean storing a lot of water as well, unfortunately not part of the human physiology.

Insulin takes glucose to the liver, where the liver extracts the water and discards it, converts the remaining glucose-remnant into fat, and deposits all into adipocytes (fat cells). Part of the reason for excess urination by T2D patients is this constant removal of the extra water (polyuria) that is attached to glucose and excess thirst (polydipsia), because the water homeostasis is disrupted.

Thus, long before T2D shows up on any medical radar, your body is building up a hefty fat storage. Where that fat is stored, also matters and is largely genetic. Some people are genetically predisposed to have much of this fat deposited as ectopic fat (inside organs). They may look thin but their organs, such as liver (non-alcoholic fatty liver disease) or pancreas are full of fat. These people are called TOFI (Thin Outside Fat Inside) and just by looking at them, there is no way to assess their metabolic status.

On the flip side, there are many people who are genetically blessed and though are obese, have no metabolic disease at all. As a result, making a metabolic disease judgement based on “looks” is deceiving. The official criteria for the chance of receiving a T2D screening test is as follows:

Criteria for Testing for Diabetes or Prediabetes in Asymptomatic Adults

(see table 2 here)

  • Testing should be considered in overweight or obese (BMI ≥25 kg/m2 or ≥23 kg/m2 in Asian Americans) adults who have one or more of the following risk factors:
    • First-degree relative with diabetes
    • High-risk race/ethnicity (e.g., African American, Latino, Native American, Asian American, Pacific Islander)
    • History of CVD
    • Hypertension (≥140/90 mmHg or on therapy for hypertension)
    • HDL cholesterol level <35 mg/dL (0.90 mmol/L) and/or a triglyceride level >250 mg/dL (2.82 mmol/L)
    • Women with polycystic ovary syndrome
    • Physical inactivity
    • Other clinical conditions associated with insulin resistance (e.g., severe obesity, acanthosis nigricans)
    • Patients with prediabetes (A1C ≥5.7% [39 mmol/mol], IGT, or IFG) should be tested yearly.

Note that the list of criteria is based on the pre-qualification of a higher than accepted BMI (body mass index). As noted earlier, there are TOFI who have low BMI and have T2D and then there are the obese with large BMI who have no T2D. The criteria is biased based on “obesity causes T2D” rather than the truth, which is “T2D causes obesity”.

Migraine and T2D

I work with thousands of migraineurs and find that nearly all (at least of those who so far tested) have IR and some T2D, yet nearly all also have normal or below normal BMI, low blood pressure, normal triglycerides, great HDL cholesterol levels, may not have anyone in the family with T2D, have no CVD (coronary vascular disease), may not have polycystic ovary syndrome (PCOS), and are physically active (many are athletes). In other words, the migraineurs I work with would not meet any of the above criteria set as the guideline by which to screen for T2D.

Migraine sufferers are not the only ones that are misdiagnosed, since nearly all young people don’t fit the model for testing criteria. However, even if testing is recommended, they only test blood glucose. So, when is T2D diagnosed? Often decades after the condition has already started. T2D often has no symptoms until it is in advanced stage. T2D is preventable and can be put to remission by eating a low carbohydrate diet. T2D is discoverable by the insulin test. If your doctor doesn’t want to prescribe a fasting insulin blood test, my recommendation is: change doctor. If all else fails, in the US you can get your insulin blood test via private services for a fee.

Case Study Conclusion

Nicole is a very lucky person that the country she lives in tested her for urine and insulin in addition to glucose. This likely saved her from years of agonizing pain, loss of vision, and loss of limbs. It saved her from a most horrific death of slowly rotting from the inside from excess glucose. She is currently on a very low carbohydrate diet (ketogenic) and recovering steadily.

Sources

  1. Kraft, J., R;. Diabetes Epidemic & You. revision 1 edn,  (Trafford, 2011).

We Need Your Help

More people than ever are reading Hormones Matter, a testament to the need for independent voices in health and medicine. We are not funded and accept limited advertising. Unlike many health sites, we don’t force you to purchase a subscription. We believe health information should be open to all. If you read Hormones Matter, like it, please help support it. Contribute now.

Yes, I would like to support Hormones Matter. 

Migraine as a Metabolic Disease

10540 views

Migraine is assumed to be a disabling neurovascular disease with various manifestations that affect the vasculature as well as the physiology of the brain. Migraine is categorized either as episodic headache (<15 pain days a month) or chronic (>15 pain days a month). It is also understood that migraine is associated with hyper sensory sensitivities and that about 15% of migraineurs suffer aura migraines. Migraine is strongly associated with metabolic disease1-9. Based on many studies, migraine leads to metabolic disease—either because of migraine prevention medicines or because of dietary lifestyle. Genetic factors also strongly underlie why migraineurs end up with metabolic syndrome. Before we investigate what these genetic factors are and how a migraine sufferer can avoid metabolic syndrome,  we first need to understand what migraine is.

So, What is Migraine?

The current scientific understanding of migraine ignores many key factors because most doctors only see migraineurs when they are in pain or have aura, and most research is, therefore, only focused on pain or aura. However, every single migraineur experiences migraines without pain—silent migraines. Silent migraines are usually associated with aura migraines but cyclic vomiting syndrome, irritable bowel syndrome (IBS), and restless legs syndrome (RLS) can also be present. These other syndromes are often precursors of the migraine-brain development (particularly in children) and they come without migraine-pain. If one can have pain-free migraines, it means the pain in migraines is only a symptom. A symptom of an underlying condition that my research and publications have been identifying and describing.

I have written many blogs about migraine and its genetic connections, see here and here and have published a book that details all genetic, neurovascular, and metabolic connections to migraine, which you can find here. In these publications, I discuss how even 2-year-olds can have migraines, how cyclical vomiting, IBS, RLS, and anxiety often precede the presentation of migraine pains by years. Because of this, migraine can appear as “not migraine” for many years and these seemingly independent symptoms may end up being treated separately. In my research, working with thousands of migraineurs over the years in my FB migraine group, I have found that nearly all of them have some form of insulin resistance (IR), which may be exhibited by hyperglycemia (HG) or reactive hypoglycemia (RH), all of which nearly always remain undiagnosed for a very long time. This is particularly important, since contrary to non-migraineurs with metabolic syndrome, migraineurs sport low blood pressure and are often very thin—meaning several commonly used key factors for the classification of metabolic syndrome are not present, based on which medical professionals would even consider testing for metabolic syndrome.

What is Metabolic Syndrome?

If three of the following five categories are met (based on the new revision of the ranges; see original article here), the person is said to have metabolic syndrome.

  1. Waist circumference > 40 in (males) or >35 in (females)
  2. Fasting glucose ≥ 100 mg/dl
  3. Triglycerides ≥ 150 mg/dl
  4. HDL cholesterol < 40 mg/dl (males) or <50 mg/dl (females)
  5. Systolic blood pressure >130 mmHg or diastolic blood pressure > 85 mmHg

I repeat here the same in bold striked out the ones that migraineurs don’t typically have:

  1. Waist circumference > 40 in (males) or >35 in (females)
  2. Fasting glucose ≥ 100 mg/dl (true for a few)
  3. Triglycerides ≥ 150 mg/dl (true for a very few)
  4. HDL cholesterol < 40 mg/dl (males) or <50 mg/dl (females)
  5. Systolic blood pressure >130 mmHg or diastolic blood pressure > 85 mmHg<.strike>

As you see, migraineurs don’t meet three (and often none) of the above to qualify for a metabolic syndrome examination. Yet, when I ask for an at-home 5-hour fasting blood glucose test, checking for blood sugar fasted, pre-breakfast, 30 minutes after breakfast, and then every 30 minutes after that for 5 hours, I find some very complex IR and RH cases. Below find two common examples:

Hyperglycemia
Hyperglycemia

Hyperglycemia (high glucose)—note that the “normal” may appear low to you but my migraineurs are on a low carbohydrate diet where the maximum blood sugar spike should not exceed 20 points. The above hyperglycemia graph shows both an overly large blood glucose spike and a significant delay in the spike, suggesting serious insulin resistance.

Reactive Hypoglycemia Example
Reactive Hypoglycemia Example

Reactive hypoglycemia is represented by blood glucose level that drops significantly after a meal.

Note that both of the above graphs come from people on a low carbohydrate diets, where breakfast is usually eggs with bacon with perhaps some raspberries with yogurt or a mix of nuts with cottage cheese. So the spike at the 2-hour mark usually represents the conversion of egg whites’ protein to glucose—hence, the spike at 120.00 minutes for the RH and a delayed spike for the HG. Some of the RH migraineurs ended up with such low blood glucose that they were close to fainting (two of them so far reached low 50s). Yet, because they don’t meet the “minimum three” criteria from the five listed above, many doctors flatly refuse to test them for metabolic syndrome. Some even disregard the at-home 5-hour blood glucose test as silly—not sure why, after all, there is no better method currently available. This just shows why migraineurs are never considered as having metabolic disease until it is late in the game, even though there are hundreds of academic articles showing the connection.

Given that migraineurs end up with metabolic syndrome (with most of the five points eventually) later in their lifetime, and since they are undiagnosed prediabetics for a very long time, could it be that migraine is actually a metabolic disease? Since IR, HG, and RH are part of metabolic syndrome and nearly all migraineurs have them from a young age, shouldn’t we investigate what the connection may be?

Observing that migraineurs end up with metabolic syndrome is one thing; finding if migraineurs have predisposition to metabolic diseases as a result of their genetics that includes metabolic gene variants within their migraine gene variants, would be an entirely different thing. Shall we go and find these genes?

The Genetics of Migraines

As a migraineur who is also a scientist, finding out as much as possible about migraines was easier for me than for those researchers who never had migraines. I knew what to ask and what to look for and could also tell what made sense and what didn’t. While most everyone’s first instinct is to say, “what works for one person may not work for another,” hold that thought! In the case of migraineurs this isn’t true. Over the many years of research, I found that migraineurs are like siblings. Nearly all of their symptoms and reactions to treatments are identical. If you want to fully understand why, you need to look at the genetic variances of migraineurs—all of these variances are identical or at least very similar to each other among migraineurs. How do I know this? Many migraineurs had genetic testing and shared their genetic data with me so I could compare their genetic variances with those “average” variances listed for migraine-genes.

True migraines start with a hormonal disruption as a result of overstimulation of the sensory neurons and end in electrolyte dysfunction that prevents action potential in some part(s) of the brain, which results in what is referred to as cortical spreading depression. This is well understood by studying the genetic template of migraine, where the first 30 or so genetic variances are all associated with ionic channels operating electrolyte management and maintenance. Other key genetic variants of importance among migraine genes are glucose transport, renin-angiotensin-aldosterone system (RAAS)10; see variants associated with RAAS as AGT, AGTR1, AGTR2 and BDKRB2 in the migraine genes and here the same associated with type 2 diabetes and here with hypertension, ATP binding, glucagon receptor, and several mitochondrial variants all with very high score in migraine genetic variances. These variants collectively lead to one key problem all migraineurs eventually face: metabolic syndrome5,11. Is it possible then that migraine, while it is clearly genetic, with its metabolism and energy associated variances, is actually a sign (symptom) of metabolic syndrome? In other words, does the genetic makeup of migraineurs lead directly to metabolic syndrome? The answer to this is likely yes. The reason? When migraineurs change their nutritional regime by excluding or greatly reducing all carbohydrates, they reverse their metabolic health condition, achieve full remission, and with that their migraines also disappear.

Clearly, migraineurs’ genetic makeup is such that factors responsible for type 2 diabetes are major components of their migraines. What are those factors? Metabolic syndrome in migraine is not diagnosed until the migraineur had the disease for many years because as we described, migraineurs don’t carry three out of five typical metabolic syndrome traits. The one common variable is chronic insulin resistance (CIR). While academic literature is quite a war zone when it comes to the causes of CIR, in migraineurs the cause is very simple to determine and the condition is easy to reverse.

Reversing Chronic Insulin Resistance in Migraineurs

All people with chronic insulin resistance have common symptoms when they are late with a meal or skip a meal: feeling grumpy, nervous, ravenous, sometimes shaky, sweaty while cold, anxious, and may even faint. Feeling hungry when blood glucose levels are above the body’s preferred 99 mg/dl (about 1 teaspoon of glucose in the entire body) is not normal but those with hyperglycemia feel hungry even when their blood glucose levels are way over 100. To understand the importance of this, we should understand what hunger actually is. I am not covering the biochemical and hormonal elements of hunger. Hunger is a state in which the body is running short of glucose for whatever reason. If the blood glucose level is way above normal, feeling hunger is not warranted. This sort of hunger actually signals insulin resistance because while the blood has plenty of glucose to feed the organs that need glucose, glucose doesn’t get there. It cannot get there because insulin is resistant—meaning it is not sensitive enough to pick glucose up to carry it to the cells. Therefore, eating when hungry while the level of blood glucose is still high, causes more trouble and the hunger will not be sated.

Reversing this type of insulin resistance (hyperglycemia) is very simple: stop eating and keep on checking your blood glucose levels. My migraineurs are asked to not eat until their blood glucose levels dips well below 99 mg/dl—preferably between 70 and 95 mg/dl. Getting down to this low blood glucose may take several days of not eating! If a migraineur is not on any psychotropic or heart medicines, the ketogenic diet I specifically formulated for migraineurs is the best option.

If the migraineur has reactive hypoglycemia, fasting is not an option, since their blood glucose level can drop below healthy minimum levels, endangering their lives. In this case, the migraineur moves to the zero carbs diet (all meat and fat, ratio dependent upon certain factors). Without any exogenous easy glucose, the liver must convert glucose from protein for all of the body’s glucose needs and this reduces insulin, since insulin is not spiked by glucose in this case.

Normal hunger is *just hungry* and no urgency. Normal hunger is healthy, and you can continue being hungry for several days without eating without any problems. That’s normal. A normal hunger doesn’t affect your blood sugar in any shape or form, because the liver manufactures the proper amount of glycogen via gluconeogenesis and ketones for the ketogenic metabolic process. Migraine is 100% preventable by a metabolic process that doesn’t use carbohydrates at all or only in minimal amounts. Since a large body of science suggests that metabolic syndrome is driven by excess carbohydrate intake12 (see here and here and here), and it is reversible by carbohydrate restriction (not caloric restriction), it follows that migraineurs recover from their migraines because they are able to reverse their genetic predispositions to carbohydrate intolerance and glucose sensitivity.

Since migraineurs are carbohydrate intolerant and glucose sensitive, they end up with metabolic syndrome if they continue to consume the – for them – offensive substances. This also suggests that each migraineur should be scrutinized by medical professionals to eliminate all opportunities for the development of metabolic syndrome by educating them to remove carbohydrates from their diet.

Support Hormones Matter

More people than ever are reading Hormones Matter, a testament to the need for independent voices in health and medicine. We are not funded and accept limited advertising. Unlike many health sites, we don’t force you to purchase a subscription. We believe health information should be open to all. If you read Hormones Matter, like it, please help support it. Contribute now.

Yes, I would like to support Hormones Matter. 

Additional Sources

1          Bhoi, S., Kalita, J. & Misra, U. Metabolic syndrome and insulin resistance in migraine. The Journal of Headache and Pain 13, 321-326, doi:10.1007/s10194-012-0416-y (2012).

2          Casucci, G., Villani, V., Cologno, D. & D’Onofrio, F. Migraine and metabolism. Neurological Sciences 33, 81-85, doi:10.1007/s10072-012-1047-4 (2012).

3          Fava, A. et al. Chronic migraine in women is associated with insulin resistance: a cross-sectional study. European Journal of Neurology 21, 267-272, doi:10.1111/ene.12289 (2014).

4          Guldiken, B. et al. Migraine in metabolic syndrome. Neurologist 15, doi:10.1097/NRL.0b013e31817781b6 (2009).

5          Sachdev, A. & Marmura, M. J. Metabolic Syndrome and Migraine. Frontiers in Neurology 3, 161, doi:10.3389/fneur.2012.00161 (2012).

6          Salmasi, M., Amini, L., Javanmard, S. H. & Saadatnia, M. Metabolic syndrome in migraine headache: A case-control study. Journal of Research in Medical Sciences : The Official Journal of Isfahan University of Medical Sciences 19, 13-17 (2014).

7          Shaw, S. W., Johnson, R. H. & Keogh, H. J. Metabolic changes during glucose tolerance tests in migraine attacks. J Neurol Sci 33, 51-59, doi:10.1016/0022-510X(77)90181-2 (1977).

8          Sinclair, A. J. & Matharu, M. Migraine, cerebrovascular disease and the metabolic syndrome. Annals of Indian Academy of Neurology 15, S72-S77, doi:10.4103/0972-2327.100015 (2012).

9          Winsvold, B. S. et al. Migraine, headache and development of metabolic syndrome: An 11-year follow-up in the Nord-Trøndelag Health Study (HUNT). PAIN® 154, 1305-1311, doi:https://doi.org/10.1016/j.pain.2013.04.007 (2013).

10        Atlas, S. A. The Renin-Angiotensin Aldosterone System: Pathophysiological Role and Pharmacologic Inhibition. Journal of Managed Care Pharmacy 13, 9-20, doi:10.18553/jmcp.2007.13.s8-b.9 (2007).

11        Bhoi, S. K., Kalita, J. & Misra, U. K. Metabolic syndrome and insulin resistance in migraine. J Headache Pain 13, doi:10.1007/s10194-012-0416-y (2012).

12        Volek, J. S., Fernandez, M. L., Feinman, R. D. & Phinney, S. D. Dietary carbohydrate restriction induces a unique metabolic state positively affecting atherogenic dyslipidemia, fatty acid partitioning, and metabolic syndrome. Prog Lipid Res 47, doi:10.1016/j.plipres.2008.02.003 (2008).