FROM THE ACADEMY Practice Paper

Practice Paper of the Academy of and Dietetics: Classic and Modified Ketogenic Diets for Treatment of

ABSTRACT Ketogenic (KD) is an established form of treatment for both pediatric and adult with intractable epilepsy. is a term that refers to any diet therapy in which dietary composition would be expected to result in a ketogenic state of human metabolism. While historically considered a last-resort therapy, classic KDs and their modified counterparts, including the modified and low treatment, are gaining ground for use across the spectrum of disorders. Registered are often the first line and the most influential team members when it comes to treating those on KD therapy. This paper offers registered dietitian nutritionists insight into the history of KD therapy, an overview of the various diets, and a brief review of the literature with regard to efficacy; provides basic guidelines for practical implementation and coordination of care across multiple and community settings; and describes the role of registered dietitian nutritionists in achieving successful KD therapy. J Acad Nutr Diet. 2017;117:1279-1292.

INCE ANCIENT TIMES, PRO- specifically , forcing the similar to those first proposed by the longed periods of have body to utilize for energy.4 Mayo Clinic group9—approximately 1 g been used to treat epilepsy.1 Two parties first conceptualized the /kg of body weight, 10 to 15 g The first modern reports using modern-day ketogenic diet (KD) inde- carbohydrate/day, and the remaining S 4 5 fasting in epilepsy were by the French pendently in 1921. Woodyatt of Rush calories from fat. Guelpa and Marie in 19112 Medical College in Chicago noted that Use of KDs was common practice in and by Dr. H. Rawle Geylin, an Amer- the and b-hydrox- the treatment of epilepsy through the ican endocrinologist at New York Pres- ybuteric acid were formed through 1920s and 1930s until the discovery of byterian .3 Researchers at the starvation on a diet low in carbohy- in 1938.4 As pharmaceuti- Harvard Medical School were the first drate and high in fat. Dr R.M. Wilder of cals grew in number, the KD fell out of to report improvements in seizure con- the Mayo Clinic, in Rochester, MN,6,7 favor due to the perceived complexity trol after 2 to 3 days of fasting, propos- proposed that a special diet be uti- of adherence. Although there was brief ing that a change in metabolism lized for the treatment of in interest in a version of the KD rich in occurred in the absence of food, efforts to achieve without medium-chain (MCT) in inducing that occurs with the 1980s, this was short-lived due to prolonged starvation. This was the the gastrointestinal side effects of a 4 2212-2672/Copyright ª 2017 by the origin of the classic KD. diet composed of 60% total calories Academy of Nutrition and Dietetics. Ketogenic diet is a term that refers to from MCT oil.10 In 1994, the KD therapy http://dx.doi.org/10.1016/j.jand.2017.06.006 any diet therapy in which dietary grew in popularity after a highly pub- composition would be expected to licized story, and eventual movie titled result in a ketogenic state of human First Do No Harm, about a boy who metabolism. A KD is generally defined quickly became seizure-free on the The Continuing Professional Education (CPE) 4 quiz for this article is available for free to as a high-fat, low-carbohydrate, mod- KD. The resurgence in use of the KD Academy members through the MyCDRGo app erate protein diet that aims to force the led to the development of less- (available for iOS and Android devices) and via body to breakdown fat instead of restrictive versions of the classic KD www.eatrightPRO.org. Simply log in with your Academy of Nutrition and Dietetics or , both which provide adenosine intended to enhance compliance; these Commission on Dietetic Registration username triphosphate synthesis, essentially diets are known as the Modified Atkins and password, go to the My Account section of 11,12 “ ” mimicking the metabolic state of star- Diet (MAD) and the low glycemic My Academy Toolbar, click the Access Quiz 13 link, click “Journal Article Quiz” on the next vation or fasting. KDs do not actually index treatment (LGIT). page, then click the “Additional Journal CPE induce starvation; instead, they are quizzes” button to view a list of available precisely calculated to maintain quizzes. Non-members may take CPE quizzes by OVERVIEW OF EPILEPSY sending a request to [email protected]. adequate nutrient intake to prevent the There is a fee of $45 per quiz (includes quiz and malnutrition associated with starva- Epilepsy is a chronic neurologic disor- copy of article) for non-member Journal CPE. tion, therefore, ensuring healthy der that causes seizures, or a disruption CPE quizzes are valid for 1 year after the issue 8 date in which the articles are published. growth and development. Calcula- in the electrical communication of the tions for classic KDs to this day remain .14 While seizures are considered a

ª 2017 by the Academy of Nutrition and Dietetics. JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1279 FROM THE ACADEMY

symptom, epilepsy has been defined as simple, where consciousness is main- decreasing and due to having two or more unprovoked sei- tained, or complex, where conscious- adequate energy intake by way of fat zures at least 24 hours apart.15 ness is lost, followed by a period of consumption, prevents , According to the Epilepsy Founda- confusion. The type of seizure(s) and resulting in increased b-oxidation, and a tion, 65 million people have epilepsy location of origin in the brain are often rise in bodies, which become the worldwide,14 of which one-third are the main determinant of treatment.14 main energy source for .16 considered to have uncontrolled sei- While some seizures can be controlled Decreased glycolysis alone has been zures that are refractory (uncontrolla- by anti-epileptic drugs (AEDs), others found to play a role in seizure reduction, ble) to standard medical treatment.14 are considered refractory, and may with increased seizure activity noted Although the etiology is often un- require treatment via alternative treat- with reintroduction of known, possible causes outside of ment modalities, including diet therapy, and subsequent rise in glycolysis.19 mitochondrial and genetic disorders surgical resection, and vagal nerve As carbohydrate intake decreases include an insult to the brain, such as stimulation. and fat intake increases on a KD, blood traumatic brain injury, central nervous glucose stabilizes, and ketone produc- system tumors, infections, and sub- OVERVIEW OF KD THERAPY tion from both endogenous and dietary stance abuse.14 sources rise, offering a steady fuel Determining the type, duration, and Potential Mechanisms of Action source for the neurons, decreasing the intensity of the seizures is important in Although mechanism(s) by which the likelihood of disruptions in energy 8,17,19 the diagnosis and treatment of epilepsy. KD impacts seizure control are not availability. The produces fi Classi cation can be complicated, completely understood, results from three types of , including — although two broad categories exist rodent and human studies offer multi- b-hydroxybutyrate (BOHB), which primary generalized seizures and par- ple hypotheses, which can be classified is measured in the serum; acetoacetate, tial seizures. Primary generalized sei- into two categories: 1) alterations in measured in the urine; and acetone, zures involve the entire brain, energy metabolism, including a measured on the breath. beginning with widespread abnormal decrease in glucose concentration with Alterations in . The electrical activity, and can be charac- an increase in oxidation and second mechanism by which KDs may terized as either tonic clonic or ketone production; and 2) alterations absence seizures. During tonicclonic reduce seizure activity is through alter- in production, release ations in neurotransmitters, in manners (convulsive) seizures, consciousness is and uptake.16-18 lost and involuntary movement oc- similar to AEDs in many cases. Ketone 14 bodies, specifically acetoacetate and curs, whereas during absence sei- Alterations in Energy Metabo- 20 zures, the individual may lose lism. As dietary carbohydrates are BOHB, have been found to inhibit g- 14 aminobutyric acid receptor-induced sei- awareness and appear to be dazed. reduced, blood glucose decreases and 21 Partial seizures involve only one area ketone levels rise. Figure 1 offers a visual zures. KDs have been found to increase of the brain, with symptoms varying depiction of differences in fuel sources production and synaptic release of g- depending on the area affected. Partial between a typical Western diet and a KD. aminobutyric acid, thereby reducing seizures can be classified as either The KD reduces the supply of glucose, neuronal excitation and seizure activity by decreasing the conversion of gluta- mate to aspartate,22,23 as well as poten- Western Diet Ketogenic Diet tially blocking neuronal uptake of glutamatethroughthepresenceof insulin release serum acetoacetate.20 In addition, BOHB and acetoacetate may result in mem- branehyperpolarizationduetoincreases in adenosine triphosphate potassium channel activity, potentially reducing the release of neurotransmitters, and inhibi- tion of action potentials.22 Furthermore, ketones have been found to reduce Fuel Source Fuel Source reactive oxygen species and inflamma- tion that results from seizure activity.24 Although no one mechanism has Breakfast Lunch Dinner conclusively been deemed responsible Breakfast Lunch Dinner for improved seizure control experi- enced with KD therapy, these mecha- = blood glucose nisms likely work in conjunction to = blood ketones control seizures in those who are affected by epilepsy. Figure 1. Variation in primary fuel source between a typical Western diet and a ketogenic diet. On a traditional Western diet, blood glucose rises after carbohydrate- rich (left), while on a ketogenic diet, carbohydrate intake is limited to only small Types of KDs quantities of those with low glycemic response, resulting in rises in serum ketone All KDs aim to reduce net carbohy- concentrations (right). drate intake and increase fat intake to

1280 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS August 2017 Volume 117 Number 8 FROM THE ACADEMY

alter energy metabolism. Several var- the MCT diet is less common and 60% to 70% of total calories.8,27 Multi- iations of the KD have been found to sometimes limited by the unpleasant and supplementation be successful in the treatment of epi- gastrointestinal side effects with con- is also recommended for patients on lepsy; including the classic KD, MAD, sumption of high concentrations of MAD and LGIT and will be discussed in MCT diet, and LGIT. Macronutrient MCT oil. Instead, smaller amounts of greater detail later in the paper composition of each diet in compari- MCTs are incorporated into other ver- (Figure 2). son to the 2015-2020 Dietary Guide- sions of the KD to enhance ketosis. Due lines for Americans can be found in to limited use of the MCT diet, this Efficacy 25,26 Table 1, along with a sample paper will focus primarily on the Impact on Seizure Frequency. While 1,700-kcal menu for the classic KD, classic KD, MAD, and LGIT. a comprehensive review is beyond the MAD, and LGIT in Tables 2, 3,and4. scope of this paper, results from two MAD and LGIT. In the early 2000s, recent reports29,30 offer insight into Classic KD and MCT Diet. The classic the MAD was first utilized at Johns the benefits of KDs. Generally, efficacy KD is the most restrictive, requiring all Hopkins Hospital, and later the LGIT at is reported as a 50% improvement in foods and beverages be carefully Massachusetts General Hospital in ef- seizure frequency, which is consistent calculated and precisely weighed on a forts to ease implementation and with measures of efficacy among gram scale.8,27 The classic KD offers adherence to KDs. These diets do not pharmaceutical outcome research for higher ketogenic potential and are require gram scales; instead, portions epilepsy. The KD and its variations may prescribed as a ratio of grams of fat to are measured through standard be effective for approximately half of combined grams of carbohydrate and household measurements. On the those who trial it for drug-resistant protein, generally as 4:1 or 3:1, but also MAD, net daily carbohydrate intake is epilepsy. A randomized clinical trial as low as 2:1, 1:1 ratios; while MAD, limited to 10 to 15 g for pediatric pa- published in 200829 revealed that 44% LGIT, and MCT, are typically ratios of tients and 20 g among adolescents and of children had 50% improvement in 2:1 or 1:1. Ratios refer to grams of fat to adults,11,12 while on the LGIT, daily seizure control. A systematic review of combined grams of carbohydrate and carbohydrates are limited to 40 to 60 g/ randomized controlled trials conduct- protein (Table 1). day from foods with a glycemic index ed among 427 children and adoles- The MCT diet is more liberal in car- <50 to prevent rapid changes in blood cents indicate that when following a bohydrates than the classic KD due to glucose and insulin levels.13 Carbohy- 4:1 classic KD, seizure freedom was high intake of ketone-boosting MCT- drates are encouraged to come from observed in up to 55% of patients after rich , comprising up to 60% of total foods with high fiber contents, such as 3 months of KD therapy and 85% re- calories with a slightly more liberal nonstarchy vegetables, nuts, and seeds. ported seizure reduction.30 Seizure carbohydrate content. Consumption of Although protein is not restricted on freedom was achieved in 10% of chil- MCTs results in higher ketogenic po- either version, intake above the needs dren following an MAD, with 60% tential due to ease of digestion and of the average adult (0.8 to 1.2 g/kg reporting a reduction in seizure activ- absorption, as they do not require bile actual or adjusted weight for an ity. Outcomes for adults are more salts for digestion; instead, MCTs are adult)28 or above the dietary reference challenging to generalize due to absorbed directly through the enter- intake for age in pediatrics and ado- limited publications; however, findings ocyte, rapidly transported into portal lescents may impact ability to maintain from a recent meta-analysis indicate circulation, and subsequently con- ketosis. Fat is encouraged on the MAD that among 270 adults with intractable verted to ketones by the liver.8 Use of diet and the LGIT, ideally composing epilepsy, 52% of those following a

Table 1. Comparison of macronutrient composition and initiation requirements between various ketogenic diets and the 2015-2020 Dietary Guidelines for Americansa

Diet Fat Carbohydrate Protein Hospital admission

ƒƒƒƒƒƒƒƒƒƒƒƒƒrange (%)ƒƒƒƒƒƒƒƒƒƒƒƒƒ! 2015-2020 Dietary Guidelines for Americans 20-35 45-65 10-35 No Ketogenic diet ratiob 4:1 90 2-4 6-8 Yes 3:1 85-90 2-5 8-12 Variesc 2:1 80-85 5-10 10-15 Variesc Modified Atkins diet (1:1 ratiob) 60-65 5-10 25-35 No Low glycemic index treatment (1:1 ratiob) 60-70 20-30 10-20 No Medium-chain diet (1:1 ratiob) 60-70 20-30 10 Yes

aBased on data from The Charlie Foundation for Ketogenic Therapies25 and US Department of Health and Human Services.26 bRatio refers to grams of calories from fat: carbohydrateþprotein. cAdmission requirement may vary based on institution.

August 2017 Volume 117 Number 8 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1281 FROM THE ACADEMY

a classic KD and 34% on the MAD expe- Table 2. Sample menu for the classic 3:1 ketogenic diet rienced 50% reduction in seizure frequency.31 The authors concluded Grams net Fat that diet compliance (defined as either carbohydrate (in grams) self-reported positive urinary ketosis or by diet recall) was higher among Breakfast those following MAD (56%) compared to Egg Scramble (To prepare: Melt butter in frying pan; classic KD (38%).31 While efficacy rates scramble all items together on medium heat.) vary, possibly due to fluctuations in 71 g raw egg mixed well 0.51 6.75 compliance, populations that may experience higher success rates include 17 g heavy cream 0.51 6.12 patients with West syndrome and Len- 28 g butter 0.02 22.71 nox Gastaut syndrome.28 Generally, fi 29 g feta cheese 1.2 6.17 classic KDs offer slightly higher ef cacy, however, compliance is greater among 21 g spinach 0.3 0.08 modified KDs, such as MAD and LGIT, 10 g mushrooms, chopped 0.23 0.02 and therefore may be a better long-term 10 g olive oil 0 22.71 therapy, particularly among those older than 2 years of age. Breakfast Subtotal: 2.76 64.56 Lunch Limitations. Although current research regarding efficacy of the KD Cobb Salad (To prepare: Toss all salad ingredients on seizure control among the pediatric together in a bowl, top with olive oil and population has been well established, red wine vinegar.) results among use for seizure control in 72 g mixed greens 0.9 0.22 adults are limited by small sample sizes, lack of randomization, heteroge- 18 g avocado, sliced 0.33 2.77 neity, high attrition rate, short study 68 g hard-boiled egg, chopped 0.76 7.21 duration, and lack of description of di- 14 g finely chopped bacon 0.42 6.3 etary intake. Larger randomized controlled trials using various types of 15 g hard cheese shredded 0.27 4.55 KDs are needed to better describe the 31 g olive oil 0 31 benefits of KD therapy, particularly 15 g red wine vinegar 0 0 among adults with epilepsy. More research is needed to describe other Lunch Subtotal: 2.68 52.05 potential benefits and side effects of Dinner KDs, including impact on seizure Chicken and Zucchini “Pasta” (To prepare: severity, quality of life, and long-term effects on health, as these diets Slice zucchini thinly into “noodles” and become more common among the sauté in olive oil. Mix half the pesto into the adult population. In addition, a detailed zucchini and spread the other half on top description of actual foods consumed of chicken. Basil Pesto recipe available at by those on a KD (vs simple macronu- Ketodietcalculator.org) trient breakdown) is needed to better 39 g baked chicken breast 0 1.4 understand how to achieve success and provide the most healthful therapy 80 g sliced or spiraled zucchini 1.69 0.26 with the fewest complications. 28 g olive oil 0 28 32 g basil pesto 0.62 16.7 KD Team Members Dinner Subtotal: 2.76 46.36 A well-trained, interdisciplinary team of health care practitioners is needed to Snacks initiate, manage, and best meet the Celery & Cream Cheese complex and varying needs of patients 10 g stalk of celery, sliced 0.14 0 on KD . This team should include a neurologist and a nurse and 30 g full-fat cream cheese 1.1 10.3 registered dietitian (RDN) Snacks Subtotal: 1.24 0 both specializing in KDs, and should Daily Total: 9.44 173.27 also include other clinical and community-based RDNs, epileptolo- aApproximate daily total: 1,700 kcal; 173.27 g fat : 9.44 g net carbohydrate + 45 g protein ¼ 3:1 diet ratio. Nutrition gists, nurses, nurse practitioners, information obtained from: www.ketodietcalulator.org. , social workers, case

1282 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS August 2017 Volume 117 Number 8 FROM THE ACADEMY

managers/discharge planners, and fi a Table 3. Sample menu for the modi ed Atkins diet families. Case managers can be an important part of the diet therapy team Grams net Fat when it comes to planning for b carbohydrate (in servings ) discharge and establishing connections with durable medical equipment pro- Breakfast viders for patients receiving enteral or Egg Scramble (To prepare: Melt butter to ensure neces- in frying pan; scramble all items sary product and equipment is pro- together on medium heat.) vided. In some settings, the nutrition and dietetics technician, registered, 2 large eggs 1 1 under the supervision of the RDN, can 1 2 Tbsp heavy cream /2 1 function in support of the RDN by 1 Tbsp butter 0 1 working as a liaison between foodser- vice and the RDN concerning the de- 1/ cup feta cheese 2 1/ 4 2 livery of food and nutrition services to 1 1 /2 cup spinach /2 0 patients on KD therapy. 1 /2 cup mushrooms, chopped 1 0 Successful KD implementation re- quires good interaction among the KD Breakfast Subtotal: 5 31/ 2 team, the patient, and his or her Lunch support systems. The importance of Cobb Salad (To prepare: Toss all salad and family support is crit- ingredients together in a bowl, top ical for success. They provide not only social support and encouragement, with olive oil and red wine vinegar.) but often implement the KD, and 1 1 1 /2 cups mixed greens /2 0 therefore must have a firm grasp on 1 /2 cup avocado, sliced 2 1 not only the diet, but also how to identify and act quickly to minimize 1 hard-boiled egg, sliced 1 1/ 2 symptoms of intolerance and prevent fi 1 1 Tbsp nely chopped bacon 0 /2 a potential complication. 1 /4 cup blue cheese or cheddar 11 cheese, shredded PRACTICAL IMPLEMENTATION 2 Tbsp olive oil 0 2 In 2009, the International Ketogenic Diet Study Group published guidelines 1 Tbsp red wine vinegar 0 0 for the clinical of children 1 32 Lunch Subtotal: 4 /2 5 receiving the KD ; however, these have Dinner not been updated and do not include fi “ ” speci c recommendations for adults Chicken and Zucchini Pasta (To prepare: undergoing KD therapy. Offering more “ ” Slice zucchini thinly into noodles guidance is the newly revised resource and sauté in olive oil. Mix half the by Kossoff and colleagues28: The Keto- pesto into the zucchini and spread genic and Modified Atkins Diets: Treat- the other half on top of chicken.) ments for Epilepsy and Other Disorders, 6th edition. Much of the information in 1 medium baked chicken breast 0 0 the following sections is based on rec- 1 1 cup sliced or spiraled zucchini 2 /2 0 ommendations from this resource from 1 Tbsp olive oil 0 1 the Johns Hopkins group, as well as a manual published by The Charlie Foun- 2 Tbsp pesto 1 1 dation for Ketogenic Therapies, and was 1 Dinner Subtotal: 3 /2 2 provided during ketogenic RDN Snacks training,33 as well as the clinical expe- rience of the authors. Celery & Cream Cheese 1 stalk of celery, sliced 1 1 Contraindications 2 Tbsp full-fat cream cheese 2 0 It is crucial to assess patients for 1 1 Sugar-Free Gelatin, /2 cup /2 0 potential contraindications for KD 1 Snacks Subtotal: 3 /2 1 therapy before initiation (Figure 3). Daily Total: 161/ 111/ Those with a history of certain meta- 2 2 bolic disorders that limit fat meta- a 1 1 Approximate daily total: 1,700 kcal, 16 /2 g net carbohydrate, 75 g protein, 150 g fat (11 /2 servings). bolism or carnitine production should b1 serving¼14 g of fat. not be initiated on KD therapy.32

August 2017 Volume 117 Number 8 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1283 FROM THE ACADEMY

a Although not contraindications, addi- Table 4. Sample menu for the low glycemic index treatment tional factors to consider with sug- gested workup plans can be found in Grams net Fat Figure 4. Choosing the right candidate b carbohydrate (in servings ) and involving the patient and family in selecting the most appropriate diet is Breakfast crucial when implementing KD ther- Egg Scramble (To prepare: Melt butter in frying pan. apy in order to optimize compliance Scramble all items together on medium heat.) and prevent unintended complica- 2 large eggs 1 1 tions (Figure 5). 1 1 1 Tbsp heavy cream /2 /2 1 Tbsp butter 0 1 Initial Consultation Initiation procedures based on diet 1/ cup feta cheese 2 1/ 4 2 type can be found in Figure 6. Before 1 1 /2 cup spinach /2 0 KD implementation, a consultation 1 between the patient and KD team is /2 cup mushrooms, chopped 1 0 needed.34 During this consultation the 1 medium grapefruit 18 0 team will conduct a full medical and Breakfast Subtotal: 23 3 nutritional assessment for appropri- Lunch ateness of KD therapy, as well as to determine the most appropriate diet. Cobb Salad (To prepare: Toss all salad ingredients Social factors impacting diet are also together in a bowl. Top with olive oil and considered at this time. While some red wine vinegar.) patients arrive with a baseline under- 1 1 1 /2 cups mixed greens /2 0 standing of KD therapy, the initial 1 1 consult offers the KD team time to offer /4 cup avocado, sliced 1 /2 1 1 detailed KD education and guidelines, 1 hard-boiled egg, sliced /2 /2 and establish both patient and team 1 1 Tbsp finely chopped bacon 0 /2 expectations. It is crucial to assess and 1 confirm patient and/or /4 cup blue cheese or cheddar cheese, shredded 1 1 comprehension of initiation protocols, 1 Tbsp olive oil 0 1 required testing, and anticipated 1 Tbsp red wine vinegar 0 0 follow-up schedule to prevent confu- 1 sion and complications. Lunch Subtotal: 3 3 /2 Dinner Baseline Data: Anthropometric and Chicken and Zucchini “Pasta” (To prepare: Slice Biochemical Values. It is imperative zucchini thinly into “noodles” and sauté in to obtain accurate baseline weight and olive oil. Mix half the pesto into the zucchini height/length measurements to deter- and spread the other half on top of chicken.) mine appropriate protein and energy requirements. These values are 1 medium baked chicken breast 0 0 the basis on which individual diet 1 1 cup sliced or spiraled zucchini 2 /2 0 regimens are calculated. Baseline 1 Tbsp olive oil 0 1 biochemical values (Figure 6) are ob- tained to address abnormalities and 2 Tbsp pesto 1 1 screen for contraindications or areas of 1 Dinner Subtotal: 3 /2 2 concern before KD therapy, as well as Snacks to be used as a comparison after diet implementation (Figure 3).32 Celery & Cream Cheese 1 3 small stalks of celery, sliced /2 0 Supplementation 2 Tbsp full-fat cream cheese 1 1 and Carbohydrate Composition of Yogurt & Strawberries Medications. Supplementing with a daily with minerals is 8 oz plain/unsweetened Greek yogurt (4% milkfat) 8 1 recommended to ensure micronutrient 1 28,32-35 /2 cup strawberry halves (mix into yogurt) 5 0 needs are met. Those on higher 1 Snacks Subtotal: 14 /2 2 ratio (3:1 and 4:1) KDs require supple- mentation with additional micro- Daily Total: 44 101/ 2 nutrients based on age-appropriate a 1 28 Approximate daily total: 1,700 kcal, 44 g net carbohydrate, 75 g protein, 140 g fat (10 /2 servings). dietary reference intakes (Figure 2). In b1 serving¼14 g of fat. addition, supplementation is

1284 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS August 2017 Volume 117 Number 8 FROM THE ACADEMY

Recommended supplements: Primary carnitine deficiency Multivitamin with minerals Carnitine palmitoyltransferase I or II deficiency and trace minerals Carnitine translocase deficiency with vitamin D b-oxidation defects B Medium-chain acyl dehydrogenase deficiency Optional supplements to consider B Long-chain acyl dehydrogenase deficiency based on specific patient needs: B Short-chain acyl dehydrogenase deficiency Selenium B Long-chain 3-hydroxyacyl-CoA deficiency Magnesium B Medium-chain 3-hydroxyacyl-CoA deficiency Phosphorus Pyruvate carboxylase deficiency Vitamin D Iron Probiotic Figure 3. Absolute contraindications to using ketogenic diet therapies. Based on data 32 Ecosapentanoic acid/ from Kossoff and colleagues. docosahexaenoic acid Medium chain triglyceride oil prepared and well informed and there available, although at this time none Laxatives is a system in place for the patient and are hypoallergenic. Blenderized for- Carnitine caregiver to access the KD team in the mulas are an option for the caregiver event of an adverse effect. All forms of with the time and ability to prepare Citrates KD therapy necessitate an intensive recipes designed by a ketogenic RDN Table salt/light salt educational session. Educational for- on a daily basis. Many are not Digestive enzymes mats vary by institutions, with some equipped to offer blenderized formulas Figure 2. Dietary supplementation for providing one-on-one sessions, and in the hospital setting, therefore, a patients on ketogenic diets. Based on others employing a classroom-based backup recipe using commercially data from Kossoff and colleagues,32 The environment with multiple patients. produced modular components may be Charlie Foundation for Ketogenic Ther- necessary during a hospital admission. apies,33 and Neal and colleagues.35 Planned Inpatient Admissions. The During admission, it is important to rationale behind admission for diet monitor for and treat symptoms of initiation is to manage potential acute acidosis, , and excessive recommended for those found to be side effects and provide ample time for or persistent ketosis (Figure 2). If deficient. Vitamin D deficiency is a po- education over multiple days. Histori- acidosis occurs, supplemental bicar- tential side effect of certain AEDs.36 cally, the classic KD is initiated in the bonate should be provided. Anti- Correction of the deficiency has been inpatient setting with a variable period epileptic medications that promote found to have an ef- of fasting; however, research indicates acidosis should be evaluated and fect.37 All supplements should be in that fasting does not improve efficacy adjusted, if able, and/or the diet ratio tablet or powder form when possible to and may increase the risk of side ef- lowered. Thresholds for treating meta- minimize carbohydrate consumption. fects, potentially increasing length of bolic acidosis vary by institution Ketogenic formulas are available for hospital stay.39,40 (serum bicarbonate <17 to 20 mEq/L), enterally fed individuals and are forti- Initiation of classic and MCT KD with acidosis generally treated with 2 fied with , but may need therapy occurs by gradually titrating to 3 mEq bicarbonate per kilogram of additional supplementation to meet the macronutrient composition during the bodyweight.32 Blood glucose should be dietary reference intakes for age. course of 3 to 4 days.28 The two initi- monitored every 4 to 8 hours with the All medications must be assessed for ation methods are as follows: 1) goal of >40 to 50 mg/dL (>2.2 to 2.8 carbohydrate content before and dur- replace one traditional with a mmol/L), depending on facility proto- ing KD therapy, as many can add sig- ketogenic meal on day 1, increasing to col. Glucose levels <40 to 50 mg/dL nificant carbohydrates, particularly full KD therapy by day 3; or 2) increase (<2.2 to 2.8 mmol/L) are treated with when taken multiple times daily.38 In the strength of the KD ratio daily as 15 to 30 mL juice and reassessed after general, medications in liquid, syrup, tolerated (1:1 ratio on day 1, then 2:1 30 to 60 minutes.33 If hypoglycemia and elixir formulations may contain ratio on day 2, and so forth, until goal persists, the team should consider carbohydrates in the form of sugars or ketosis is achieved). A slow KD intro- lowering the KD ratio or increasing sugar alcohols, and therefore, may duction allows the gastrointestinal calories.34 disrupt ketosis. Pharmacists should be tract to acclimate to changes in Blood and urine ketone monitoring consulted for recommendations for macronutrient composition and in- during hospital admission, as well in low-carbohydrate formulations of duces ketosis gradually, which can be thehomeenvironment,variesby medications when necessary. easier for the patient. institution. Assessment of urinary Diet initiation protocols for patients ketones (acetoacetate) may be less Initiation of KD Therapy requiring enteral nutrition are similar accurate than serum BOHB levels.41,42 KD therapy can be initiated in both the to those who eat by mouth, providing a Capillary BOHB has been found to inpatient and outpatient environments gradual increase in diet ratio. Multiple have high sensitivity, specificity, and a as long as the patient or caregiver is commercial ketogenic formulas are positive predictive value for diabetic

August 2017 Volume 117 Number 8 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1285 FROM THE ACADEMY

Concern Suggested Workup

Inability to maintain adequate nutrition or hydration Obtain gastrointestinal consult Failure to thrive Obtain swallow evaluation Dysphagia Consider need for gastrostomy tube placement Gastrointestinal issues (chronic diarrhea, vomiting, Increase fat/kcal before initiation reflux) Trial of 4:1 ketogenic formula Not able to meet fluid goals Provide recipes/foods to trial Extreme picky eating/limited food acceptance Behavioral feeding consult

Concerning medical history Obtain cardiology, nephrology, or hepatology consult Extreme for clearance Cardiomyopathy Adjust fluid minimums Renal /renal calculi Add citrate, consider bicitrate to alkalize urine, avoid/ Liver disease wean drugs like and Baseline Wean insulting medications if possible, increase fluid minimums, consider beginning with lower diet ratio

Social constraints Connect family with social worker to discuss access to Access to food and kitchen services, for example, but not limited to, durable Caregiver support and compliance medical equipment, Special Supplemental Program for Multiple caregivers/unstable home environment Women, Infants, and Children, respite care, in home supportive services and/or formula company’s assistance programs Registered dietitian nutritionist can discuss meal/food options feasible for family Figure 4. Considerations for determining appropriateness of initiation of ketogenic diet therapy and suggested further workup before diet initiation.

, and negative predictive however, there are occasions where achieve maximum ketogenic potential. value for identifying diabetic ketoa- emergent KD therapy is warranted. Given the critical nature of status epi- cidosis compared to urinary ketone Although limited, the available lepticus and febrile infection-related testing,42 although no studies have research for use of KD therapy for sta- epilepsy syndrome, minimum protein been published comparing BOHB to tus epilepticus is promising in both requirements may be temporarily urine ketones among patients pediatric44 and adult45 populations. sacrificed with the goal of achieving receiving ketogenic therapy. In addi- Status epilepticus is defined as contin- and maintaining ketosis. Levocarnitine tion, urine ketones levels may be uous or near-continuous seizure activ- may be initiated empirically for those influenced by hydration status,43 as ity without returning to baseline receiving or those found to has been found to have a neurologic functioning.28 KD therapy have free carnitine deficiency, with small, negative association with urine for status epilepticus appears to be dosing beginning at 50 mg/kg/day osmolality, although this study was most efficacious among those with divided into three doses based on rec- conducted in dogs. Generally, BOHB is underlying autoimmune and/or in- ommendations from The Charlie assessed daily for level of ketosis flammatory conditions, such as infan- Foundation for Ketogenic Therapies during hospital encounters. Reference tile spasms (West syndrome) and manual.33 Carnitine supplementation ranges for blood ketones (BOHB) can febrile infection-related epilepsy can improve ketosis and may need to vary by laboratory, although the goal syndrome.46 be continued for the duration of KD is positive ketosis.28 Not all patients The goal of emergent KD therapy is therapy. A complete metabolic panel experience symptoms of excessive to achieve ketosis as quickly as and serum BOHB are monitored daily ketosis, and therefore do not need to possible.33 As all fluids, medications, until ketosis is established and levels be treated. Persistent hypoglycemia and supplements are transitioned to stabilize. or symptomatic excessive ketosis carbohydrate-free products (if avail- despite multiple interventions may able), the patient is gradually transi- Initiation in the Outpatient Envi- be indicative of an underlying meta- tioned to full calories provided by the ronment. Less-restrictive versions of bolic condition and warrants further KD during the course of 1 to 3 days, the KD, such as 2:1 or 1:1 classic KD, investigation. generally via enteral nutrition sup- MAD, and LGIT can be initiated in port,28,45,46 achieving ketosis within the home environment, however, Emergent Admissions. Most admis- 3 to 5 days. The KD should be pro- this requires a well-informed patient sions for KD initiation are planned; vided at the highest ratio possible to with a good support system. To

1286 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS August 2017 Volume 117 Number 8 FROM THE ACADEMY

and LGIT therapies as tolerated over Are seizures well the course of a few days to weeks, controlled? depending on comfort level and tolerance. No Yes Overall, monitoring in the home environment is less rigid than the inpatient setting. Some institutions Are any of the absolute Consider KD therapy if require daily home blood glucose and contraindicaƟons to KD medicaƟon must be blood ketone monitoring, while others therapy present? simply monitor urine ketones daily, with the goal of moderate or large ke- No Yes tones. The gold standard for home ke- tone monitoring is blood ketone (BOHB) due to higher accuracy, which may be Is the paƟent >5 years KD therapy is not used to fine-tune the diet and achieve of age? recommended improved seizure control.47 When BOHB measurements are not possible fi No Yes due to nancial burden, urine ketones measured daily may be utilized. Corre- lation to seizure control has been Consider iniƟaƟng the Are there concerns about observed between urine and blood classic KD at a compliance or ability to (BOHB) ketones at low values, although 4:1 or 3:1 raƟo implement a KD? poor correlation has been noted at higher values.48 In addition, patients or caregivers are instructed to keep a log of No Yes daily dietary intake, weekly weight changes, and seizures, to help identify Consider iniƟaƟng the Would compliance be potential areas for improvement and classic KD at a enhanced using standard minimize side effects. 3:1, 2:1 or 1:1 raƟo measuring tools or less rigid counƟng? Monitoring and Management. Mo- nitoring and management strategies vary by institution, though they are No Yes generally more intense during the initial weeks and months of KD ther- KD therapy is not Consider iniƟaƟng apy. Those on a classic KD typically follow-up in an outpatient clinic with recommended the MADa or LGITb the KD team monthly for the first 3 Figure 5. Ketogenic diet (KD) therapy initiation decision tree. aMAD¼modified Atkins months. Children under 1 year of age diet. bLGIT¼low glycemic index treatment. generally follow-up within 2 weeks, based on the clinical judgment of the KD team and individual institutional optimize success of home initiation, seizure control, and laboratory values. protocols.32 Follow-up for the MAD and the patient, family, and/or caregivers The diet can be maintained at the ratio LGIT generally occurs 1 to 3 months must have a firm understanding of found to offer seizure control. after initiation. Patients are encouraged basic nutrition, an individualized diet Education, instruction, and initiation to contact the KD team by phone or prescription, as well as expectations for the MAD and LGIT are similar e-mail if questions arise. Follow-up and knowledge of how to manage po- and conducted in the outpatient envi- timing is similar across diet types af- tential complications, and have access ronment. Patients are educated on how ter 3 months. Assuming no complica- to the KD team for any urgent issues. to identify sources of protein, fat, and tions are experienced and the diet Initiation of lower-ratio classic KD carbohydrate, how to count grams of maintains efficacy, monitoring con- therapy in the outpatient home envi- net carbohydrate (total grams of car- tinues to occur every 3 to 6 months for ronment follows similar principals as bohydrate minus grams of fiber) for the duration of therapy, but is adjusted their higher-ratio counterparts, and those following MAD, and to identify based on patient need to enhance usually occurs during the course of foods with a low glycemic index (<50) compliance and tolerance.28 several weeks. Success can be for those following LGIT, in an effort to At each monitoring visit, a complete enhanced with close communication prevent fluctuations in blood glucose nutrition assessment is conducted to between the KD team and the patient and insulin levels. Each patient may be assess nutritional adequacy, and or caregiver. For the classic KD, the diet given an individualized diet prescrip- biochemical values are obtained until is started at a 1:1 diet ratio with in- tion that specifies net carbohydrate, stable or the diet is discontinued to creases in the ratio weekly based on protein, and fat recommendations. Pa- assess for potential complications patient symptoms, tolerance to diet, tients are encouraged to start the MAD (Figure 6). Biochemical values are

August 2017 Volume 117 Number 8 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1287 FROM THE ACADEMY

Laboratory values Pre-diet Daily during 1 and 3 mo post Every 3 mo Every 6 to baseline admission diet initiation until stable 12 mo Urine organic acids X Plasma amino acids X Complete metabolic panel X X X X X Complete blood count with XX X XX platelets Liver profile X X X X Ionized calcium X X X X Magnesium X X X X Phosphate X X X X Pre-albumin X X X X Lipid panel (fasting) X X X X Vitamin D-3 X X X Free and total carnitine X X X X b-hydroxybutyrate X X X X X Selenium X X X X X X X Urinalysis X X X X Urine calcium X X X X Urine creatinine X X X X A, E, and B-12 XX Copper XX Folate/ferritin XX Figure 6. Standard laboratory assessment recommendations throughout various states of ketogenic diet therapy. Protocols may vary by institution, individual patient, and diet type. Based on data from The Charlie Foundation for Ketogenic Therapies.33

monitored regularly for abnormal for development of atherosclerosis.56 benefits with the challenges of values. One value of particular concern Among the general adult population, following the KD when determining for many RDNs starting patients on KD researchers report improvements in continuation. While seizure freedom is therapy is the potential impact of diet cardiovascular risk factors and man- the ultimate goal, patients may report therapy on lipid profiles. While fluctu- agement of type 2 when other factors that impact choice to ations are likely to occur initially, these following low-carbohydrate diets57-59; continue KD therapy, even if seizure values generally remain similar to however, it is unknown whether this frequency is not dramatically baseline, or actually may improve on KD remains true among those with epi- improved. These factors may include therapy, specifically increases in high- lepsy. Along with laboratory values, experiencing shorter, milder seizures, density lipoprotein and reductions in tolerance to diet, compliance, side ef- improved postictal states (period of triglyceride levels.49-53 Serum low- fects, and weight trends are assessed. In altered level of consciousness after a density lipoprotein values occasionally the pediatric population, growth pa- seizure), increased mental clarity, or rise with KD therapy,54 though it is un- rameters are also monitored to assure improvements in cognition or level of clear whether it is the particle number that linear growth and weight gain in- alertness. If the decision is made to or size that increases. Among healthy crease proportionally over times with discontinue the diet based on overall adults following low-carbohydrate di- diet titrations as needed to ensure challenges or lack of desired benefits, ets for , low-density lipo- appropriate growth is maintained.28 the KD should be weaned gradually. protein values increase due to an Diet efficacy is assessed at each clinic increase in particle size.55 Larger, more visit, and is determined based on pa- Fine-Tuning. Before initiation, pa- buoyant low-density lipoprotein parti- tient or family expectations of KD tients and families are asked to give a 3- cles may be associated with a lower risk therapy, carefully weighing the month commitment to KD therapy.

1288 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS August 2017 Volume 117 Number 8 FROM THE ACADEMY

During this period (and often beyond), encouragement during times when resumed at the last point where it fine-tuning will likely be required to noncompliance is most likely, and may was effective. Once ketosis is lost and enhance the efficacy of the KD. Evidence enhance KD success. In addition, if seizures remain stable with correlating ketone level and diet efficacy because undesirable gastrointestinal discontinuation, patients are encour- is limited47; however, some patients side effects are another common aged to continue to adhere to an benefit from higher levels of ketosis, or reason for diet discontinuation, offer- overall low in processed increased blood ketone (BOHB) levels, ing recommendations to prevent these foods, specifically sugars. while others at milder or lower levels. effects, such as methods for ensuring fi fl Therefore, ratios may be adjusted to adequate ber and uid intake, can ROLE OF THE RDN optimize ketone levels and potentially enhance compliance and KD diet efficacy, if necessary. MCT oil may maintenance. also be gradually incorporated and For those receiving enteral nutrition Role of the Ketogenic RDN titrated to enhance ketosis. For those support, initial KD administration via Ketogenic RDNs require highly experiencing undesirable weight change continuous feedings may be better specialized training to ensure appro- or large fluctuations in blood glucose or tolerated with a transition to bolus priate implementation and monitoring ketones, a calorie adjustment may be feeds once tolerance has been estab- of KD therapy. RDNs with demon- beneficial.34 Monitoring of free carnitine lished. Standard ketogenic enteral for- strated and documented education and levels and initiating supplementation mulas and modular, such as MCT oil training with KD therapy are an inte- may increase KD efficacy (Figure 2).35 For and protein powders, may be necessary gral part of the multidisciplinary team, those on prolonged KD therapy, short to enhance ketosis and meet protein and are vital in designing and main- periods of may needs. Soy or peptide-based formulas taining a successful KD program. enhance ketosis and potentially increase are available if food allergies or Ketogenic RDNs are involved in every efficacy.60 are of concern and, as aspect of therapy, from assessing discussed previously, blenderized KD appropriateness of KD therapy, educa- Adjusting for Tolerability and enteral regimens may be utilized if tion, providing recommendations for Enhancing Success. Mild side effects desired by the family; however, close KD regimen, initiation, management of and tolerance concerns can occur dur- monitoring and calculation by a keto- symptoms, to diet discontinuation. ’ ing the first few days and weeks after genic RDN is needed to maintain Ketogenic RDNs primary role is to KD initiation. Intolerance often presents appropriate KD ratio and micronutrient safely and effectively design a KD to as fatigue, headaches, nausea, con- goals. optimize seizure control; this requires stipation, hypoglycemia, or acidosis. If careful diet manipulation and plan- these symptoms occur, an oral citrate or Weaning and Discontinuation. Length ning, and can be demanding, as there sodium bicarbonate can be added to of KD therapy often dictates length of are often many questions and addi- buffer acidosis, and/or the diet ratio can time over which the KD is weaned, and tional communication with the patient be decreased to improve tolerability guidelines for weaning may vary by pa- and caregivers. and palatability. It is important to note tient and/or institution. KD therapy is that many oral citrate products contain usually implemented for a minimum of 3 Appropriate staffing ratios. Staffing significant amounts of carbohydrate, to 6 months.32,35 Patients generally ratios vary widely. Unfortunately, no which must be calculated in the diet. follow KD therapy for several years. documented consensus as to the The ketogenic and medical teams ManychoosetocontinuetheKDdueto optimal ratio of patients to RDNs ex- should work together to resolve the continued efficacy and/or improvements ists; therefore, it is difficult to provide acidosis, potentially adjusting medica- in other areas of life, such as mental recommendations for the number of tion if necessary. Once tolerance has clarity and alertness.28 If compliance is full-time equivalents that would be been established, the diet may be not possible, the patient no longer necessary to maintain a successful KD adjusted to increase ketogenic potential wishes to continue KD therapy, or KD program. With the rigorous demands if needed for enhanced efficacy. therapy is deemed ineffective early on, of maintaining patients on KD therapy Close communication among the early discontinuation is possible. and potential for frequent, emergent epilepsy nurse, ketogenic RDN, and Diet discontinuation should occur hospital admissions, it is beneficial to patient or caregiver during the few gradually and under continued have a minimum of two trained and weeks after initiation may enhance supervision of the KD team to pre- competent ketogenic RDNs on staff, success, as compliance is the most vent the potential for rebound including a ketogenic RDN with pedi- important factor in successful KD seizures.32,35 For those on KD therapy atric experience if the program in- implementation. Poor understanding for fewer than 3 months, carbohy- cludes pediatric patients. and compliance will likely result in drate content can be increased grad- reduced efficacy and KD discontinua- ually by 1 to 5 g net carbohydrate per Models for RDN Reimbursement. tion. Offering encouragement via close week, or by a 0.5 to 1.0 decrease in Reimbursement and staffing models for monitoring and open lines of commu- diet ratio per week until ketosis is RDNs specializing in KD therapy have nication, as well as providing education lost.28,33 For those on long-term yet to be standardized; therefore, materials, including sample meal plans therapy, this process should occur models across other RDN specialties and recommendations for eating over the course of weeks to months. If may serve as a reference.61 Due to the outside of home and during social oc- seizures or other side effects occur need for highly specialized training, casions, offer the patient support and with weaning, the KD should be detailed diet education, close and

August 2017 Volume 117 Number 8 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1289 FROM THE ACADEMY

frequent monitoring, and risk for Academy of Nutrition and Dietetics Nutrition Care Manual: www.nutrition complications, ketogenic RDNs may spend more time with each patient, caremanual.org overall seeing fewer patients than Kossof EH, Turner Z, Doerrer S, Cervenka MC, Henry BJ. The Ketogenic and standard RDNs. Inpatient models do Modified Atkins Diets: Treatment for Epilepsy and Other Disorders. 6th ed. New not offer appropriate comparison, and York: demosHEALTH; 2016. most outpatient models may not The Charlie Foundation for Ketogenic Therapies: www.charliefoundation.org appropriately categorize time involved Matthew’s Friends: www.matthewsfriends.org to maintain a successful KD program. Carson Harris Foundation: www.carsonharrisfoundation.org One major barrier for RDN reimburse- ment is the cost for care. Many clinics Carley Eissman Foundation: www.carleyeissmanfoundation.com use a fee-for-service model in which Keto Hope Foundation: www.ketohope.org insurance companies reimburse the KetoDietCalculator: https://ketodietcalculator.org clinic or clinician; however, this is only Nutricia: www.myketocal.com possible for an overall small number of Cambrooke Therapeutics: www.ketovie.com diagnoses62 for which KD therapy is not included. Even for reimbursable Figure 7. Ketogenic references for registered dietitian nutritionists (RDNs). These diagnoses, frequency of RDN visits is resources were determined to provide quality ketogenic recommendations by limited, regardless of patient need and RDNs practicing ketogenic diet therapy. clinical judgment. Reimbursement for KD therapy poses additional consider- ketone ranges, avoidance of organizations, or referring the patient ation due to the time necessary for carbohydrate-containing medications to RDNs with KD training. intensive education, particularly with (if possible in nonemergent situations), patients who require a hospital meal and fluid schedules, and in- admission or have limited nutrition gredients and equipment necessary for SUMMARY STATEMENT knowledge.63 Ongoing and further dietary management. For school-aged RDNs play a unique and critical role in advocacy is critical to expand RDN children or adolescents or adults the assessment, initiation, manage- reimbursement for KD therapy. residing in managed-care environ- ment, and treatment of patients ments, coordination between keto- following KD therapy. Once an area of Cost-Savings Analysis. The RDN is genic RDNs and the school or home nutrition rarely utilized and consid- an integral part of the KD team in both environment is essential for KD ered unfeasible for most patients, the inpatient and outpatient settings success. particularly adults, use of KD therapy and may not only improve clinical RDNs in the community may is rapidly expanding. It is the re- outcomes, but also increase overall cost interact with patients following a KD sponsibility of all RDNs to be knowl- savings.64 While the cost benefitof and must be informed on the basics edgeable on the basics of KD therapy, RDN involvement on the KD team has outlined in this review. The knowledge their potential role in management, not been established, use of KD therapy of KD therapy needed for RDNs in the and how to locate experts in the among children and adolescents with community may include basic under- field, particularly as this effective intractable epilepsy that experienced standing of how to achieve and and novel treatment expands outside improved seizure control on KD ther- maintain ketosis, monitor for compli- the realm of epilepsy, including apy has resulted in an overall signifi- cations, and when and how to contact possible management of malignant cant reduction in health care costs, ketogenic experts for further guid- brain tumors and other various forms including reduced medication costs ance. This is particularly important in of , , Parkinson’sdisease, when compared to pre-KD costs.65-69 the hospital environment, as the pri- Alzheimer’s disease, traumatic brain Further research is warranted to mary KD treatment team can offer injuries, mitochondrial disorders, and determine the cost benefits of KD recommendations and adjustments, as for . therapy overall, specifically examining well as offer basic education for those the cost savings when ketogenic RDNs unfamiliar with KD therapy. If initia- References tion of KD therapy is desired, it is are part of the treatment team. 1. Wheless JW. History and origin of the crucial for the RDN and health care ketogenic diet. In: Stafstrom CE, Rho JM, teams unfamiliar with KD therapy to eds. Epilespy and the Ketogenic Diet.Totowa, Care Coordination: From Clinic to reach out to organizations, such as The NJ: Humana Press, Inc; 2004:31-50. Community Charlie Foundation for Ketogenic 2. Guelpa G, Marie A. La lute contre l’e’ ’ pilepse par la de’ sintoxication et par la Each patient should be provided with Therapies or Matthew s Friends, and re’e’ducation alimentaire. Rev Ther Med multiple customized letters to share local experts for guidance to find Chir. 1911;78:8-13. with other members of their medical appropriate recipes, baking mixes, 3. Geyelin HR. Fasting as a method for support team. These letters help equipment, menu ideas, and cooking treating epilepsy. Med Rec. 1921;999: ensure that KD guidelines are followed demonstration (Figure 7). If outpatient 1037-1039. in a variety of settings, and can vary KD therapy is desired, RDNs are 4. Wheless JW. History of the ketogenic diet. Epilepsia. 2008;49(suppl 8):3-5. based on the audience. Important in- responsible for receiving appropriate 5. Woodyatt RT. Objects and method of diet formation to include is not limited to training and demonstrated, docu- adjustment in diabetes. Arch Intern Med. fasting protocols, blood glucose and mented competency from these 1921;28(2):125-141.

1290 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS August 2017 Volume 117 Number 8 FROM THE ACADEMY

6. Wilder RM. The effect of ketonemia on ketogenic-diet/explore-1/introducing-the- beta-hydroybutyrate with dipstick urine the course of epilepsy. Mayo Clin Bull. diet. Accessed June 16, 2016. tests in the detection of ketone bodies. 1921;2:307. 26. US Department of Health and Human Turk J Emerg Med. 2016;14(2):47-52. 7. Kossoff E, Wang H. Dietary therapies for Services. Dietary Guidelines for Ameri- 42. Brooke J, Stiell M, Ojo O. Evaluation of epilepsy. Biomed J. 2013;36(1):2-8. cans. Office of Disease Prevention and the accuracy of capillary hydroxy- 8. Cervenka MC, Kossoff EH. Dietary treat- website. http://health. butyrate measurement compared with ment of intractable epilepsy. Continuum gov/dietaryguidelines/. Accessed June other measurements in the diagnosis of (Minneap, Minn). 2013;19(3 Epilepsy): 14, 2016. : A systematic re- 756-766. 27. Klein P, Janousek J, Barber A, view. Int J Environ Res . 2016;13(9):837. 9. Peterman MG. The ketogenic diet in epi- Weissberger R. Ketogenic diet treatment lepsy. JAMA. 1925;84(26):1979-1983. in adults with refractory epilepsy. Epilepsy 43. Ayoub JA, Beaufrere H, Acierno MJ. Asso- Behav. 2010;19(4):575-579. ciation between urine osmolality and 10. Sills MA, Forsythe WI, Haidukewych D, 28. Kossoff EH, Turner Z, Doerrer S, specific gravity in dogs and the effect of MacDonald A, Robinson M. The medium- commonly measured urine solutes on chain triglyceride diet and intractable Cervenka MC, Henry BJ. The Ketogenic and Modified Atkins Diets: Treatment for Epi- that association. Am J Vet Res. epilepsy. Arch Disease Child. 1986;61(12): th 2013;74(12):1542-1545. 1169-1172. lepsy and Other Disorders.6 ed. New York: demosHEALTH; 2016. 11. Kossoff EH, Dorward JL. The modified 44. Fung EL, Chang SK, Yam KK, Yau PY. Atkins diet. Epilepsia. 2008;49(suppl 8): 29. NealEG,ChaffeH,SchwartzRH,etal.The Ketogenic diet as a therapeutic option in 37-41. ketogenic diet for the treatment of super-refractory status epilepticus. childhood epilepsy: A randomized Pediatr Neonatol. 2015;56(6):429-431. 12. Kossoff EH, McGrogan JR, Bluml RM, controlled trial. Lancet Neurol. 2008;7(6): 45. Thakur KT, Probasco J, Hocker S, et al. Pillas DJ, Rubenstein JE, Vining EP. 500-506. A modified Atkins diet is effective for the Ketogenic diet for adults in super re- treatment of intractable pediatric epi- 30. Martin K, Jackson CF, Levy RG, Cooper PN. fractory status epilepticus. . lepsy. Epilepsia. 2006;47(2):421-424. Ketogenic diet and other dietary treat- 2014;82(8):655-670. ments for epilepsy. Cochrane Database 13. Pfeifer HH, Thiele EA. Low-glycemic- Systematic Reviews. 2016;2:CD001903. 46. Kossoff E, Nabbout R. The use of dietary index treatment: A liberalized ketogenic therapy for status epilepticus. J Child fi diet for treatment of intractable epilepsy. 31. Ye F, Li XJ, Jiang WL, Sun HB, Liu J. Ef cacy Neurol. 2013;28(8):1049-1051. Neurology. 2005;65(11):1810-1812. of and patient compliance with a keto- genic diet in adults with intractable epi- 47. Van Delft R, Lambrechts D, Verschuure P, 14. Shafer PO. About epilepsy: The basics. The lepsy: A meta-analysis. J Clin Neurol. Hulsman J, Majoie M. Blood beta- Epilepsy Foundation website. http:// 2015;11(1):26-31. hydroxybutyrate correlates better with www.epilepsy.com/learn/about-epilepsy- seizure reduction due to ketogenic diet basics. Published September 2014. 32. Kossoff EH, Zupec-Kania BA, Amark PE, than to ketones in the urine. Seizure. Accessed October 1, 2015. et al. Optimal clinical management of 2010;19(1):36-39. children receiving the ketogenic diet: 15. Fisher RS, Acevedo C, Arzimanoglou A, Recommendations of the International 48. Hartman AL, Gasior M, Vining EPG, fi et al. ILAE of cial report: A practical Ketogenic Diet Study Group. Epilepsia. Rogawski MA. The neuropharmacology of clinical definition of epilepsy. Epilepsia. 2009;50(2):304-317. the ketogenic diet. Pediatr Neurol. 2014;55(4):475-482. 2007;36(5):281-292. 33. The Charlie Foundation for Ketogenic 16. Klein P, Tyrlikova I, Mathews GC. Dietary Therapies. Professional’s Guide to the 49. Cervenka MC, Patton K, Eloyan A, Henry B, treatment in adults with refractory epi- Ketogenic Diet. 2016. Santa Monica, CA: Kossoff EH. The impact of the modified lepsy: A review. Neurology. 2014;83(21): The Charlie Foundation for Ketogenic Atkins diet on lipid profiles in adults 1978-1985. Therapies; 2008. with epilepsy. Nutr Neurosci. 2016;19(3): 131-137. 17. Rho JM. How does the ketogenic diet 34. Zupec-Kania BA, Spellman E. An overview induce anti-seizure effects? Neurosci Lett. of the ketogenic diet for pediatric epilepsy. 50. Kapetanakis M, Liuba P, Odermarsky M, 2017;637:4-10. Nutr Clin Pract. 2008;23(6):589-596. Lundgren J, Hallbook T. Effects of keto- genic diet on vascular function. Eur J 18. Greene AE, Todorova MT, McGowan R, 35. Neal EG, Zupec-Kania B, Pfeifer HH. Carni- Paediatr Neurol. 2014;18(4):489-494. Seyfriend TN. Caloric restriction inhibits tine, nutritional supplementation and seizure susceptibility in epileptic EL mice discontinuation of ketogenic diet thera- 51. Smith M, Politzre N, Macgarvle D, by reducing blood glucose. Epilepsia. pies. Epilepsy Res. 2012;100(3):267-271. McAndrews ME, Del Campo M. Efficacy 2001;42(11):1371-1378. fi 36. Berggvist AG, Schall JI, Stallings VA. and tolerability of the modi ed Atkins 19. Bough KJ, Rho JM. Anticonvulsant mech- Vitamin D status in children with intrac- diet in adults with pharmacoresistant anisms of the ketogenic diet. Epilepsia. table epilepsy, and impact of the keto- epilepsy: A prospective observational 2007;48(1):43-58. genic diet. Epilepsia. 2007;48(1):66-71. study. Epilepsia. 2001;52(4):775-780. 20. Juge N, Gray JA, Omote H, et al. Metabolic 37. Hollo A, Clemens Z, Armondi A, Lakatos P, 52. Samaha FF, Iqbal N, Seshadri P, et al. control of vesicular glutamate transport Szucs A. Correlation of vitamin D defi- A low-carbohydrate as compared with a and release. . 2010;68(1): ciency improves seizure control in epi- low-fat diet in severe . N Engl J 99-112. lepsy: A pilot study. Epilepsy Behav. Med. 2003;348(21):2074-2081. 21. Keith HM. Factors influencing experi- 2012;24(1):131-133. 53. Foster GD, Wyatt HR, Hill JO, et al. mentally produced convulsions. Arch 38. Lebel D, Morin C, Laberge M, Achim N, A randomized trial of a low-carbohydrate Neurol. 1933;29(1):148-154. Carmant L. The carbohydrate and caloric diet for obesity. N Engl J Med. 22. Ma W, Berg J, Yellen G. Ketogenic diet content of concomitant medications for 2003;348(21):2082-2090. metabolites reduce firing in central neu- children with epilepsy on the ketogenic 54. Chen W, Kossoff EH. Long-term follow-up rons by opening KATP channels. diet. Can J Neurol Sci. 2001;28(4):322-340. of children treated with the modified J Neurosci. 2007;27(14):3618-3625. 39. Bergqvist AG, Schall JI, Gallagher PR, Atkins diet. J Child Neurol. 2012;27(6): 23. Yudkoff M, Daikhin Y, Nissim I, et al. Cnaan A, Stallings VA. Fasting versus 754-758. Response of brain metabolism gradual initiation of the ketogenic diet: 55. Sharman MJ, Kraemer WJ, Love DM, et al. to ketosis. Neurochem Int. 2005;47(1-2): A prospective, randomized clinical A ketogenic diet favorably affects serum 119-128. trial of efficacy. Epilepsia. 2005;46(11): biomarkers for cardiovascular disease in 1810-1819. 24. Jarrett SG, Milder JB, Liang L, Patel M. The normal-weight men. J Nutr. 2002;132(7): ketogenic diet increases mitochondrial 40. Kim DW, Kang HC, Park JC, Kim HD. Benefits 1879-1885. of the nonfasting ketogenic diet compared glutathione levels. J Neurochem. 56. Song TJ, Cho HJ, Chang Y, et al. Low- with the initial fasting ketogenic diet. Pe- 2008;106(3):1044-1051. density lipoprotein particle size predicts a diatrics. 2004;114(6):1627-1630. 25. The Charlie Foundation for Ketogenic poor outcome in patients with athero- Therapies. What is a ketogenic diet? 41. Sever M, Aksay E, Gocan T, Yalcin N, thrombotic . J Clin Neurol. https://www.charliefoundation.org/explore- Eren ES, Ustuner F. Comparing finger-stick 2015;11(1):80-96.

August 2017 Volume 117 Number 8 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1291 FROM THE ACADEMY

57. Nordmann AJ, Nordmann A, Briel M, et al. 62. Jortberg BT, Fleming MO. Registered die- 66. de Kindren RJ, Lambrechts DA, Wijnen BF, Effects of low-carbohydrate vs low-fat titian nutritionists bring value to emerging et al. An economic evaluation of the diets on weight loss and cardiovascular health care delivery models. JAcadNutr ketogenic diet versus care as usual in risk factors: A meta-analysis of random- Diet. 2014;114(12):2017-2022. children and adolescents with intractable ized trials. Arch Intern Med. 2006;166(3): 63. MacCracken KA, Scalisi JC. Development epilepsy: An interim analysis. Epilepsia. 285-293. and evaluation of a ketogenic diet 2016;57(1):41-50. 58. Bazzano LA, Hu T, Reynolds K, et al. Effects program. J Am Diet Assoc. 1999;99(12): 67. de Kindren RJ, Postulart D, of low-carbohydrate and low-fat diets: A 1554-1558. Aldenkamp AP, et al. Cost-effectiveness randomized trial. Ann Intern Med. of the ketogenic diet and vagus nerve 64. Academy of Nutrition and Dietetics Evi- 2014;161(5):309-318. stimulation for the treatment of children dence Analysis Library. What is the evi- with intractable epilepsy. Epilepsy Res. 59. Tay J, Luscombe-Marsh ND, Thompson CH, dence to support the cost-effectiveness, 2015;110:119-131. et al. Comparison of low- and high- cost benefit or economic savings of carbohydrate diets for type 2 diabetes outpatient MNT services provided by an 68. de Kinderen RJ, Labmrechts DA, management: A randomized trial. Am J RDN? https://www.andeal.org/topic.cfm? Postulart D, et al. Research into the (cost-) Clin Nutr. 2015;102(4):780-790. cat¼4085&evidence_summary_id¼250816 effectiveness of the ketogenic diet among 60. Hartman AL, Rubenstein JE, Kossoff EH. &highlight¼cost%20&home¼1. Accessed children and adolescents with intractable Intermittent fasting: A “new” historical June 13, 2016. epilepsy: Design of a randomized controlled trial. BMC Neurol. 2011;11:10. strategy for controlling seizures? Epilepsy 65. Mandel A, Ballew M, Pina-Garza JE, Res. 2013;104(3):275-279. Stalmasek V, Clemens LH. Medical costs 69. Gilbert DL, Pyzik PL, Vining EP, 61. Patrick S. What went right? The story of are reduced when children with intrac- Freeman JM. Medication cost reduction in US and medical nutrition ther- table epilepsy are successfully treated children on the ketogenic diet: Data from apy. Nestle Nutr Workshop Ser Clin Perform with the ketogenic diet. J Am Diet Assoc. a prospective study. J Child Neurol. Programme. 2009;12:137-158. 2002;102(3):396-398. 1999;14(7):469-471.

AUTHOR INFORMATION Practice Papers should not be used to indicate endorsement of products or services. All requests to use portions of the paper or republish in its entirety must be directed to the Academy at [email protected]. This paper will be up for review in December 2021. Authors: Kelly Roehl, MS, RDN, LDN, CNSC (Rush University Medical Center, Chicago, IL); Sarika L. Sewak, MPH, RDN (UCLA Medical Center/Mattel Children’s Hospital, Los Angeles, CA). STATEMENT OF POTENTIAL CONFLICT OF INTEREST Kelly Roehl wrote an introduction to a ketogenic cookbook and received a one-time contract fee. Sarika L. Sewak reported no potential conflicts. FUNDING/SUPPORT There is no funding to disclose. Reviewers: Sharon Denny, MS, RD (Academy Knowledge Center, Chicago, IL); Jennifer Noll Folliard, MPH, RDN (Academy Policy Initiatives & Advocacy, Washington DC); Sarah Picklo Halabu, RDN, LDN, CDE (Academy Publications and Resources, Chicago, IL); Rosa Hand, MS, RDN, LD, FAND (Academy Research, International and Scientific Affairs, Chicago, IL); School Nutrition Services dietetic practice group (Carol Longley, PhD, RD, LD, Retired-Western Illinois University, Macomb, IL); Pediatric Nutrition dietetic practice group (Jessica M. Lowe, MPH, RDN, CSP, University of Southern California, Los Angeles, CA); Denise Potter, RDN, CSP, CDE (University of Michigan , Ann Arbor, MI); Mary Pat Raimondi, MS, RD (Academy Policy Initiatives & Advocacy, Washington, DC); Beth Zupec-Kania, RDN (Ketogenic Therapies LLC, Elm Grove, WI). Academy Positions Committee Workgroup: Valaree M. Williams, MS, CSO, RDN (chair) (University of Colorado Health System, Aurora, CO); Brenda E. Richardson, MS, RDN, LD, CD, FAND (Dietary Consultants, Inc, Salem, IN); Laura Dority, MS, RD, LD (content advisor) (Medical University of South Carolina, Charleston, SC). We thank the reviewers for their many constructive comments and suggestions. The reviewers were not asked to endorse this practice paper.

1292 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS August 2017 Volume 117 Number 8