1) Vitamins, minerals and trace elements
Is there a risk of deficiency on ketogenic therapy?
Historically, ketogenic diets were not always fully supplemented with micronutrients, however early reports of problems in children date back to 1979. During this year there were two published papers about nutritional deficiency while on the ketogenic diet; the first reported two patients who presented with optic neuropathy caused by thiamine deficiency (Hoyt & Billson, 1979), and the second reported reduced bone mass due to vitamin D deficiency in five patients (Hahn et al, 1979).
More recently, selenium deficiency was found in nine children on the ketogenic diet, including one who developed cardiomyopathy (Bergqvist et al, 2003). This research group have also reported poor vitamin D status in children on the diet which can compromise bone health because of the resulting loss of bone mineral content (Bergqvist et al, 2007 and 2008). One case report describes a nine year old girl on the ketogenic diet who developed scurvy due to vitamin C deficiency (Willmott & Bryan, 2008). Plasma magnesium levels can be lowered (Kang et al, 2004) which may be a particular problem in children on the classical ketogenic diet despite micronutrient supplementation (Christodoulides et al, 2011).
Risk of nutritional deficiency may be increased by a limited food intake pre-ketogenic treatment in a child with severe disability or the effects of multiple anticonvulsant drugs, and the restrictive nature of any type of ketogenic diet makes it necessary to be fully nutritionally supplemented with vitamins, minerals and trace elements. Although the medium chain triglyceride (MCT) ketogenic diet, the modified Atkins diet and the low glycaemic index treatment are less restrictive than the classical ketogenic diet, it is still essential that all children using any of these diets are fully assessed by a dietitian who can advise on their nutritional adequacy and recommend the necessary supplementation based on a child’s nutritional requirements. A diet history that accurately records food intake over a few days should ideally be done at least once a year, and the prescribed supplementation checked regularly, so that the provision of micronutrients can be assessed by the dietitian to ensure that all requirements are met, and no nutrient is being given in unnecessary excess.
What are nutritional requirements?
The UK guidelines on nutritional requirements of children aged between 0-18 years are based on the report of the COMA Panel on Dietary Reference Values (DRVs), Department of Health Report on Health and Social Subjects No 41, published in 1991. These recommendations refer to groups, and any individual is likely to have requirements which fall within a range of recommended intakes for their age group. For this reason, an upper and lower value for requirements is given for each nutrient, these are termed reference nutrient intake (RNI) and lower reference nutrient intake (LRNI) respectively. The RNI of a nutrient would be considered the amount that would meet the requirements of nearly all of the people in a group. Many nutrients also have a value given for estimated average requirement (EAR), about half a group of people would be expected to have a requirement above this level, and about half below. If there is limited information on requirements for a particular nutrient, a value for safe intake may be used. So, when assessing vitamin and mineral requirements, the dietitian would want to be sure that a prescribed diet and supplementation met the RNI for age for as many nutrients as possible. In children who are very small for their age it may be more appropriate to use the requirements for their height age, rather than actual age. This would be assessed on an individual basis by the dietitian.
What supplements do we use?
In the UK, there are limited carbohydrate-free micronutrient supplements that are available to be prescribed for children on a ketogenic diet. The two currently most suitable and nutritionally complete are Phlexy-vits (Nutricia), available in powder sachet and tablet form and recommended for children over 11 years, and FruitiVits (Vitaflo), available in powder sachet form and recommended for children aged 3-10 years. Other types of supplement can be bought over the counter and may be more palatable, however must be carbohydrate-free and discussed with the dietitian as may not provide the full range of micronutrients needed e.g. inadequate calcium and phosphate. Calcium supplements may not be necessary on an MCT diet if adequate amounts of milk are consumed as part of the prescription, but this should be individually assessed by the dietitian.
The dose of any chosen supplement will be calculated by the dietitian to avoid risk of either under- or over supplementation of any nutrient. Whereas vitamins, minerals and trace elements do have a vital role in the body, and we must ensure children on the diet are not deficient, it must be noted that adding in further supplements additional to those recommended by the dietitian could in fact prove harmful. Any other supplements should therefore always be discussed with the dietitian/medical team. Children on a ketogenic diet will have regular blood monitoring to check nutritional status. This should include fat-soluble vitamins due to the risk of high levels of vitamins A and E (Christodoulides et al, 2011). Children who show deficiencies of particular nutrients on blood monitoring may need additional supplementation of that nutrient, a common example is vitamin D (although requirements of this vitamin will vary with the seasons, as it is synthesised in the body when exposed to sunlight).
What is carnitine?
Carnitine is a small water soluble compound. It is absorbed well from food, with the main dietary sources from protein such as milk, meat and eggs. It can also be synthesised in the body, formed from two amino acids, lysine and methionine. Over 90% of body stores are in muscle. L-carnitine is the biologically active form of carnitine. Carnitine has an essential role in fat metabolism; it combines with long chain fatty acids to form acylcarnitines (esters) to enable their transport into the cell mitochondria for oxidation. This process by which carnitine facilitates the transfer of long chain fat into the mitochondria is often referred to as the carnitine shuttle, as once carnitine has transported the fatty acid esters across the inner mitochondrial membrane, it is shuttled back across this membrane for the process to be repeated. Once inside the mitochondria, oxidation of fatty acids occurs in stages, with two carbons removed at each stage to form acetyl coA; this either enters the Krebs cycle, or is used to synthesise ketone bodies. The intermediates of this oxidation process can combine with carnitine in the mitochondria, forming acylcarnitines. Acetyl CoA also combines with carnitine within the mitochondria to form acetyl carnitine; this then leaves the mitochondria with ketone bodies.
Medium chain fatty acids have a direct passage into the cell mitochondria for oxidation, and so have no need for the carnitine shuttle.
Carnitine and the ketogenic diet
Carnitine is thought to be important on the ketogenic diet because the high fat intake means more fatty acids need to be transported into the mitochondria for oxidation, requiring more carnitine and therefore increasing risk of depletion of body carnitine stores. This risk may be magnified as food restrictions could reduce dietary carnitine intake and is of particular importance in diets with high long chain fat intake rather than MCT. An additional risk in some individuals is that long term use of the medication sodium valproate also can lead to carnitine deficiency. If carnitine is deficient, it will be difficult to achieve adequate ketosis on the ketogenic diet, due to impaired ketone body synthesis; energy levels may also be impaired.
There have been limited studies examining whether carnitine deficiency does occur on the ketogenic diet. Berry Kravis et al, in 2001, reported a study which looked at plasma total carnitine levels in 46 patients (age range 1-24 years) who were on the classical ketogenic diet; this included 38 who were followed from diet initiation, and an additional eight already on the diet at the time of the study. Of the 38 patients monitored from diet initiation, three were started on carnitine supplementation at baseline due to low levels, and five others needed supplementation later in diet treatment (3 after 1 month, 2 after 6 months). One of the additional eight patients already on the diet needed carnitine supplementation due to low levels after 1 year. So, out of all the ketogenic diet patients who were not started on carnitine when starting the diet, 6 (18%) went on to have low total carnitine levels and need supplementation later. None of them showed any clinical signs of carnitine deficiency, and did not show any worsening of seizure control with low carnitine levels. The average total carnitine in patients who were never carnitine supplemented was lower after one and six months on the diet than at baseline, but this then increased again by 12 and 24 months. The conclusions from this study were that although total carnitine does decrease over the first few months of ketogenic diet treatment, and in some patients, dip into the deficiency range, it then normalises after the first months, with no evidence of a continued decline in levels.
One other study, reported in 2005 by Coppola et al, measured plasma free carnitine levels in 164 epilepsy patients (age 1mo-26 years). None of the 11 patients who were on the classical ketogenic diet developed abnormal levels of free carnitine.
Assessing carnitine status and supplementation
The two studies discussed above used total and free carnitine. Total carnitine includes free carnitine and all the acylcarnitines, i.e, all carnitine-fatty acid esters including the intermediates of the fat oxidation process that have combined with carnitine. In a state of ketosis, even β-hydroxybutyrate will combine with carnitine to form β-hydroxybutryl carnitine – these will all be included in the acylcarnitine fraction. Another suggested measure of carnitine status is the ratio of plasma acylcarnitine to free carnitine. A consensus paper on carnitine supplementation in childhood epilepsy suggested a free carnitine level of less than 20µmol/litre or an acyl:free carnitine ratio greater than 0.4 (after 1 week post term) indicated a deficiency (DeVivo et al, 1998). These were arbitrary values, and different centres may use other age-dependent ranges, frequently a lower free carnitine cut-off for deficiency. We do not know what measures accurately determine status as plasma levels are not a true reflection of total body stores which are mostly in muscle. Although free carnitine does give some useful indication of status in patients on the ketogenic diet, the acyl:free ratio does not. Because of the increase in fat metabolism and ketosis that occur while on the diet, as discussed above, levels of acylcarnitines including acetyl carnitine will be greatly increased and this will result in an elevated ratio. This is a normal consequence of being on the ketogenic diet and is likely to reflect the level of ketosis, rather than an indication of carnitine status. Further supplementation with carnitine will have no effect on reducing the ratio and may even cause an increase due to formation of acetyl carnitine. It has been suggested that the ratio may normalise slightly with time on the ketogenic diet with adaptation to the ketotic state (Berry-Kravis et al, 2001).
Despite current consensus that children on the ketogenic diet should not be routinely supplemented with carnitine unless showing biochemical or symptomatic deficiency (Kossoff et al, 2009), a number of families of children using the diet do choose to use either medically prescribed or bought over-the-counter carnitine supplements, regardless of biochemical status, with reports of improved well-being, energy levels and seizure control. As true carnitine deficiency will impair oxidation of fatty acids in the mitochondria and ketone production, a drop in ketone levels would be expected. Anecdotal reports do suggest ketone levels may improve with additional carnitine supplementation especially if previously showing an unexplained drop, even if plasma carnitine levels are normal, indicating it may be a useful additional tool for dietary fine-tuning. Any supplements used should be of the L-carnitine form, be commenced at a low dose and increased gradually. DeVivo et al (1998) recommended supplementing patients with biochemical deficiency at 100mg per kg body weight per day, in three or four divided doses, up to maximum of 2g/day. There may be poor absorption, diarrhoea or an increase in seizures if high doses are started without gradual build-up. A starting dose of 10mg/kg is frequently used for patients on the ketogenic diet, which is increased as needed; many children do not require above 50mg/kg per day.
3) Essential fatty acids (‘healthy oils’)
What are essential fatty acids?
Essential fatty acids (EFA) are fatty acids that we need for our health but cannot be synthesised in the human body from any other fatty acids provided in our diet. They belong to the class of fatty acids called polyunsaturated fatty acids (PUFAs). There are two types of EFA, omega-3 and omega-6. These names refer to the chemical structure of the fatty acid; both types are unsaturated, that is, they contain carbon-carbon double bonds, the type is determined by the final double bond being either at the n-3 or n-6 position. The main essential omega-3 fatty acid is alpha-linolenic acid (ALA), and the main essential omega-6 fatty acid is linoleic acid. Although the human body cannot synthesise either of these fatty acids from scratch, it can use them to synthesise other essential fatty acids. ALA is a precursor of the longer chain omega-3 fatty acids eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA), and linoleic acid is a precursor of the longer chain omega-6 fatty acids gamma-linolenic acid (GLA), dihomogamma-linolenic acid (DGLA) and arachadonic acid (AA). These longer chain fatty acids can also be provided directly from dietary sources.
Biological roles of EFA
Both omega-3 and omega-6 EFA have a role in maintaining normal growth and development including that of the brain, and are important components of all cell membranes in the body. However the two classes of EFA are metabolically and functionally separate, and often have important opposing physiological functions. AA (omega-6), DGLA (omega-6), and EPA (omega-3) are used to synthesise eicosanoids in the body, these are signalling molecules that exert complex control over many bodily systems, mainly in inflammation or immunity. There are four families of eicosanoids—the prostaglandins, prostacyclins, the thromboxanes and the leukotrienes. The eicosanoids derived from AA tend to increase inflammation (an important component of the immune response), blood clotting, and cell proliferation, while those derived from EPA and DGLA decrease those functions. The amounts and balance of omega-3 and omega-6 fats in a diet will affect the body’s eicosanoid-controlled functions and are therefore important in maintaining optimum health. It is widely believed that Western diets tend to have too much omega-6, particularly in relation to omega-3 fatty acids, and that this imbalance can increase risk of cardiovascular disease, cancer, osteoporosis and other inflammatory disorders.
Food sources of EFA
The main food sources of the different EFA are shown in the table below:
Dark, green leafy vegetables, certain nuts and seeds and their oils (flaxseed/linseed oil, hempseed oil, walnut oil)
|EPA||Oily fish, fish oil supplements|
|DHA||Oily fish, fish oil supplements|
Commonly used polyunsaturated vegetable cooking oils, including sunflower, safflower, corn, cottonseed, and soybean. Processed foods containing these oils. Nuts, sesame and sunflower seeds
|GLA||Plant based oils including evening primrose oil, borage seed oil and blackcurrant seed oil. Some hempseed oils.|
|AA||Egg yolk, meats|
The oil highest in omega-3 fats is flaxseed (linseed), with over 50% of fatty acids as omega-3, and a ratio of 0.3:1 omega-6: omega-3. Hempseed oil also has a good balance, with about 20% omega-3, and a ratio of approximately 3:1 omega-6: omega-3. Walnut oil is often recommended as a good source of EFA: this is lower in omega-3, about 3-11% of fatty acids, with a ratio of approximately 5:1 omega-6: omega-3.
Eye Q liquid is widely used as a dietary supplement in children. This contains fish oil (EPA and DHA, omega-3), and evening primrose oil (GLA, omega-6), and also vitamin E. Some children on the ketogenic diet may also use the nutritional products Calogen and/or Ketocal (Nutricia). These are both supplemented with omega-3 and omega-6 EFA, the amount provided will depend on the amount of the product used.
Recommended intakes of EFA
It is recommended that a healthy diet should consist of approximately 2 – 4 times more omega-6 than omega-3 fatty acids however a typical Western diet tends to contain 15 – 20 times more omega-6 than omega-3 fatty acids due to the amount of vegetable oils and processed foods eaten. Healthy eating guidelines recommend lowering omega-6 intake and increasing omega-3 by reducing processed foods, including oily fish, and replacing some of the commonly used vegetable oils with oils higher in omega-3 fats or olive oil (monounsaturated oil, so contains relatively low amounts of both omega-3 and omega-6 fatty acids, but known to be very beneficial for cardiovascular health) (Simopoulos et al, 2000; Hibbeln et al, 2006). Although there are no recommendations for exact amounts of EFA in the diet of children, current UK department of Health dietary reference values suggest that ALA (omega-3) should provide approximately 0.2% of total dietary energy, and linoleic acid (omega-6) approximately 1% of total dietary energy. Although in most cases, the longer chain omega-6 fatty acids GLA and DGLA and omega-3 fatty acids EPA and DHA can be formed in the body from linoleic acid and ALA, there is increasing evidence that there may also be additional requirement for these longer chain fatty acids to be provided directly from the diet in some cases.
Is it safe to use evening primrose oil supplements in epilepsy?
There was concern for a number of years that evening primrose oil might cause seizures and advice was given not to use this in epilepsy. Evidence was based on two studies published in the early 1980s on use of evening primrose oil in schizophrenia treatment. A re-examination of this data (Puri, 2007) found the original reports to be spurious and concluded the oil might actually have benefits in epilepsy, suggesting that any contra-indications for use of evening primrose oil in epilepsy be removed from all medical formularies.
Recommendations for use of EFA supplements while on the ketogenic diet
To ensure a good balance between omega-3 and-6 fatty acids, the oil source should be varied wherever possible, and if using large amounts of a polyunsaturated vegetable cooking oil such as those listed in the above table (omega-6 sources), a small amount of an omega-3 source added to the diet as well, such as flaxseed/linseed oil, hempseed oil or walnut oil. The amounts of these omega-3 oils used can be very small, e.g. 2-3ml a day, but this should ensure a child is receiving the correct balance of EFA. This is also important for children following the MCT diet, especially if receiving a large percentage of their energy from the MCT source, to ensure that adequate EFA are provided. General healthy eating recommendations should also be included wherever possible on any type of ketogenic diet within the constraints of the prescription, and these include regular oily fish and dark green vegetables, both good omega-3 sources. Olive oil, although not containing any EFA, can also be included, as has known benefits on cardiovascular health.
There have been questions about whether any one particular type of EFA rich oil is better to use as a supplement on the ketogenic diet. Although flaxseed oil contains the highest proportion of omega-3 fatty acids, adequate supply can easily be provided from small amounts of walnut oil which is more widely available. There is no evidence of advantages of any one oil type. A study examining the effects of hempseed and flaxseed oil on healthy adult volunteers found no differences in fasting serum total or lipoprotein lipids, plasma glucose or insulin level or haemostatic factors between the two oils (Schwab et al, 2006).
Supplemental oils may need to be accounted for in the energy prescription, and should not be used in quantities greater than the small amounts already mentioned as they provide additional fat-soluble vitamins, notably A and E. There have also been concerns about the risk of increased bleeding time with long-term use of excessive amounts of a high omega-3 oil source in the diet. As with any type of dietary supplement, use should always be discussed with a dietitian and medical team before commencing.
Bergquist AGC, Chee CM, Lutcha L, Rychik J, Stallings VA (2003). Selenium deficiency associated with cardiomyopathy: a complication of the ketogenic diet. Epilepsia 44 (4):618-620.
Bergqvist AG, Schall JI, Stallings VA (2007). Vitamin D status in children with intractable epilepsy, and impact of the ketogenic diet. Epilepsia 48 (1):66-71
Bergqvist AG, Schall JI, Stallings VA, Zemel BS (2008). Progressive bone mineral content loss in children with intractable epilepsy treated with the ketogenic diet. Am J Clin Nutr. 88 (6):1678-1684
Berry-Kravis et al (2001). Carnitine levels and the ketogenic diet. Epilepsia 42 (11), 1445-1451.
Christodoulides SS, Neal EG, Fitzsimmons G, Chaffe HM, Jeanes YM, Aitkenhead H, Cross JH. The effect of the classical and medium chain triglyceride ketogenic diet on vitamin and mineral levels. J Hum Nutr Diet. 2012; 25(1):16-26.
Coppola et al (2006). Plasma free carnitine in epilepsy children, adolescents and young adults treated with old and new antiepileptic drugs with or without ketogenic diet. Brain Dev 28 (6), 358-365.
DeVivio et al (1998). L-carnitine supplementation in childhood epilepsy: current perspectives. Epilepsia 39, 1216-1225.
Hahn T, Halstead L, DeVivo D. (1979), Disordered mineral metabolism produced by ketogenic diet therapy, Calcified Tissue Int 28, 17-22.
Hibbeln JR, Nieminen LR, Blasbalg TL, Riggs JA, Lands WE (2006). Healthy intakes of n−3 and n−6 fatty acids: estimations considering worldwide diversity. Am J Clin Nutr 83(6):1483S–1493S.
Hoyt CS, Billson FA (1979). Optic neuropathy in ketogenic diet. British J Opthalmol 63 (3), 191-194.
Kang HC, Chung DE, Kim DW, Kim HD. Early- and late-onset complications of the ketogenic diet for intractable epilepsy. Epilepsia 2004; 45:1116-1123.
Kossoff et al (2009). Optimal clinical management of children receiving the ketogenic diet: Recommendations of the International Ketogenic Diet Study Group. Epilepsia 2009; 50(2):304-317.
Puri BK (2007). The safety of evening primrose oil in epilepsy. Prostaglandins Leukot Essent Fatty Acids 77(2):101-103.
Schwab U, Callaway J, Erkkilä A, Gynther J, Uusitupa M, Järvinen T (2006). Effects of hempseed and flaxseed oils on the profile of serum lipids, serum total and lipoprotein lipid concentrations and haemostatic. Eur J Nutr 45 (8):470-477.
Simopoulos AP, Leaf A, Salem Jr N (2000). Statement on the essentiality of and recommended dietary intakes for n−6 and n−3 fatty acids. Prostaglandins, Leukot Essent Fatty Acids 63:119–121
Willmott NS, Bryan RA (2008). Case report: scurvy in an epileptic child on a ketogenic diet with oral complications. Eur Arch Paediatr Dent 9 (3):148-52.