Version May 2024
INTRODUCTION — Published reports of the ketogenic diet as an effective treatment for epilepsy date to the early 1920s [1]. This topic will review the ketogenic dietary therapies (KDTs) for the treatment of epilepsy. These include the classic ketogenic diet, the modified Atkins diet, the low glycemic index treatment, and the medium-chain triglyceride diet.
Other treatments for refractory seizures and epilepsy are reviewed separately. (See "Overview of the management of epilepsy in adults" and "Seizures and epilepsy in children: Initial treatment and monitoring" and "Vagus nerve stimulation therapy for the treatment of epilepsy" and "Initial treatment of epilepsy in adults" and "Surgical treatment of epilepsy in adults".)
MECHANISMS OF ACTION — The classic ketogenic diet is a high-fat, adequate-protein (1 gram/kg), low-carbohydrate diet that produces metabolic changes associated with the starvation state. Changes in plasma ketones, insulin, glucose, glucagon, and free fatty acids can occur within hours of starting the diet and can be quite profound [2]. Which of these metabolic changes are primarily responsible for the improvement in seizure frequency is not known, but the mechanism is likely multifactorial [2].
The ketogenic diet was so-called based upon the theory that ketone bodies (acetoacetate, acetone, and beta-hydroxybutyrate), created in the liver from long- and medium-chain fatty acids, are directly anticonvulsant when crossing the blood-brain barrier. However, the importance of ketosis in the mechanism of action of KDT is increasingly questioned. KDT is associated with increased mitochondrial biogenesis, oxidative phosphorylation, enhanced gamma-aminobutyric acid (GABA) levels, reduced neuronal excitability and firing, and stabilized synaptic function [2,3]. While these may be induced by ketosis, alternative mechanisms proposed include elevated plasma free fatty acids (including polyunsaturated fatty acids), reduced glucose fluctuations, increased activation of adenosine triphosphate (ATP)-sensitive potassium channels, caloric restriction, and elevated brain amino acids. Inhibition of the mammalian target of rapamycin pathway and decreased glutamatergic synaptic transmission have also been implicated [4].
In mice, the ketogenic diet induced significant changes in the gut microbiome that led to altered metabolite levels; there were reductions in systemic gamma-glutamylated amino acids and an increase in the ratio of hippocampal GABA to glutamate that correlated with seizure protection in mice [5].
INDICATIONS
Drug-resistant epilepsy — For patients with drug-resistant epilepsy, options include KDT as an adjunct or alternative to drug therapy. Antiseizure medications are the primary treatment for epilepsy and are effective in controlling seizures in most patients. However, approximately one-third of patients will not become seizure-free with medications and require nonpharmacologic treatments [6]. Although epilepsy surgery can be a treatment option for these patients, there are many families, guardians, and patients who are unwilling to consider surgery. Others may not be candidates for epilepsy surgery because the region that is causing the seizures overlaps with eloquent cortex and therefore cannot be resected without causing neurologic deficits. In these patients, vagus nerve stimulation (VNS), responsive neurostimulation (RNS), deep brain stimulation (DBS), and KDTs are rational options (algorithm 1). (See "Seizures and epilepsy in children: Refractory seizures" and "Evaluation and management of drug-resistant epilepsy".)
Based upon the available evidence, a 2018 expert consensus panel recommended that KDT should be offered to children with drug-resistant epilepsy after unsuccessful treatment trials of two antiseizure medications [7]. Some families or caregivers may wish to try it sooner, and in many of those situations it may be appropriate.
Note that KDT is usually not a substitute for antiseizure medications. In one study, 86 percent of children on KDT remained on antiseizure medications as well, even when the number or doses were reduced in many cases [8].
Efficacy — KDT is an effective treatment for patients with epilepsy, regardless of age or seizure type [7]. The major KDTs for the treatment of epilepsy are the classic (long-chain) ketogenic diet, the medium-chain triglyceride diet, the modified Atkins diet, the low glycemic index treatment, and the modified ketogenic diet. The composition of these diets is described below (see 'Composition' below); the evidence of their efficacy follows:
●Classic ketogenic diet – In 2008, the first randomized controlled trial of the classic ketogenic diet, 145 children (ages 2 to 16 years) with drug-resistant epilepsy (failure of at least two antiseizure medications) were assigned to immediate implementation of the classic ketogenic diet or to a control group in whom implementation was delayed for three months [9]. At three months, 103 remained in the trial, 54 in the active treatment group. In these patients, treatment with the diet was associated with a lower mean percentage of baseline seizures (62 versus 137 percent) and a higher percentage with greater than 50 percent seizure reduction (38 versus 6 percent).
A meta-analysis of 19 observational studies (1084 patients) found that after six months, approximately 60 percent of children started on the classic ketogenic diet had a greater than 50 percent seizure reduction, with 30 percent having greater than 90 percent seizure reduction [10]. Other meta-analyses and reviews have come to similar conclusions [10-12]. While more often considered a treatment in children rather than adults with epilepsy, the growing available (nonrandomized) evidence suggests that adults may also achieve benefit with the classic ketogenic diet [13,14]. Because the "classic" ketogenic diet must be strictly followed, adults are more commonly offered the modified Atkins diet, which is easier to use or follow. (See 'Modified Atkins diet' below.)
Data are limited for the efficacy of the ketogenic diet in infancy, although use is growing in this age group due to availability of ketogenic formulas on the market. In an open-label trial, infants 1 to 24 months of age with drug-resistant epilepsy (having four or more seizures per week and prior treatment with two or more antiseizure medications) were randomly assigned to the classic ketogenic diet (n = 78) or to further antiseizure medication treatment (n = 58); data were available for 61 and 47 infants, respectively, in the modified intention-to-treat analysis [15]. During weeks six to eight, the median number of seizures per day was similar between the ketogenic diet and antiseizure medication groups, as was the responder rate (defined by a ≥50 percent improvement from baseline in seizure frequency). The number of serious adverse events was greater in the ketogenic diet group, although the proportion of infants affected was similar between groups. Trial limitations include small patient numbers, early stopping due to slow recruitment, and relatively high loss to follow-up.
Some anecdotal and published evidence suggested that the classic ketogenic diet might be less effective in focal compared with generalized epilepsy [16]. However, there are many reports in which patients with focal epilepsy appear to respond to the classic ketogenic diet [17-21], and in the randomized clinical trial discussed above, the subgroups of focal versus generalized epileptic syndromes appeared to experience similar benefit [9]. On the other hand, patients with intractable focal epilepsy who are candidates for epilepsy surgery are more likely to achieve complete seizure remission with epilepsy surgery than with the classic ketogenic diet [19,20]. It is reasonable to attempt the classic ketogenic diet for children who are candidates for surgery, however, if the family or guardians are not ready or the child is very young. (See "Seizures and epilepsy in children: Refractory seizures", section on 'Epilepsy surgery' and "Surgical treatment of epilepsy in adults".)
●Medium-chain triglyceride diet – A randomized controlled study found no difference in efficacy or tolerability between the classic ketogenic diet and the medium-chain triglyceride diet [22]; however, there may be individual differences. Most centers believe the MCT diet to be equivalent to the classic ketogenic diet, but it is still primarily used in England.
●Modified Atkins diet – The efficacy of the modified Atkins diet has been reported in more than 25 studies since 2003, including at least two randomized trials [17,23-39]. Overall, 175 of approximately 390 total reported patients (45 percent) have had at least a >50 percent reduction in seizures after six months; of these, 96 (25 percent) had >90 percent seizure reduction. When this cohort is specifically analyzed for adults, the results are less robust, with 26 of 90 (29 percent) having >50 percent seizure reduction and 10 (11 percent) with >90 percent seizure reduction.
In one randomized trial in children, 102 patients ages 2 to 14 (mean age 5 years) with refractory epilepsy (daily seizures despite three or more antiseizure medications) were assigned to receive a modified Atkins diet or no dietary intervention for three months [36]. Four out of 50 patients discontinued the diet before the study endpoint (two due to frequent chest infections, one due to hyperammonemic encephalopathy, one due to patient and family preference). In an intention-to-treat analysis, treatment with the diet was associated with decreased mean seizure frequency as a percentage of baseline (59 versus 100 percent) and increased proportion of patients with >50 percent reduction in seizure frequency (52 versus 12 percent). Five children (10 percent) became seizure-free on the diet, compared with none in the control arm. In a second trial of 160 patients (80 adults and 80 adolescents) with drug-resistant epilepsy, more patients had a >50 percent reduction in seizure frequency at six months in the modified Atkins diet group compared with the control group (26.2 versus 2.5 percent, absolute risk reduction 23.7 percent, 95% CI 13.5–33.9) [38].
These results are similar to those reported for the classic ketogenic diet. When patients are able to remain on diet therapy for more than six months, rates of seizure reduction are also similar to those reported for the classic ketogenic diet [35]. A retrospective comparison of the two diets found that the classic ketogenic diet was more successful at three months (65 versus 20 percent achieved >50 percent seizure reduction); at six months the difference was no longer statistically significant (41 versus 20 percent) [30].
A case series reviewed the experience of 27 patients treated with the modified Atkins diet who subsequently were switched to the ketogenic diet. Ten patients achieved additional benefit with the ketogenic diet; five of these became seizure-free [40]. Additional benefit occurred in a greater proportion of children with myoclonic-astatic epilepsy than other epilepsy syndromes. The five children who had no response to the Atkins diet did not respond to the ketogenic diet. This information suggests that the modified Atkins diet may represent a "low dose" of dietary treatment and the classic ketogenic diet a "high dose" (rather than two unique diets).
Most experts perceive the modified Atkins diet as being equivalent and interchangeable with the classic ketogenic diet, and it is the primary diet for adolescents and adults [39].
●Low glycemic index treatment – This diet does not calculate fats but rather allows carbohydrates with glycemic indices <55. However, it is still a high-fat diet, although high ketosis is not achieved. Results from a single center were the first to demonstrate early benefit with this diet in 74 patients [41]. Using an intent-to-treat analysis, 32 percent had a >50 percent seizure reduction after three months of implementation.
The low glycemic index treatment is typically used in adults, adolescents, and children with Angelman syndrome [42].
●Modified ketogenic diet (UK) – This diet has been used in United Kingdom and is essentially a low ratio classic ketogenic diet. In a one-year prospective cohort study, 42 adults with epilepsy (mean age 37 years) were started on the modified ketogenic diet [43]. Retention rates at 3, 6 and 12 months were 60, 43, and 29 percent, respectively. Improvement in seizure frequency was reported in 60 percent, while a >50 percent reduction in seizure frequency in 38 percent, and worsening in seizure frequency in 30 percent. Discontinuation was most often attributed to adverse effects and dietary restrictiveness. In a clinical trial of 45 children who were randomly assigned to the classic or the modified ketogenic diet, there was no difference between groups for achieving a >50 percent reduction in seizure frequency (approximately 45 percent in each group) [44]. However, the modified ketogenic diet was reported to be better tolerated and to have fewer adverse effects.
Particularly responsive conditions — A number of epilepsy syndromes, listed below, appear to be particularly responsive to the KDTs, identified by the expert panel as conditions with at least three publications from at least two ketogenic dietary centers that consistently reported at least a 20 percent or greater improvement in efficacy above the "norm" for KDT (ie, a 40 to 50 percent chance of ≥50 percent seizure reduction); that is, conditions with 60 to 70 percent responder rates [7].
Doose syndrome — Doose syndrome (epilepsy with myoclonic-atonic seizures [EMAtS]), a generalized epilepsy of early childhood with a high frequency of intractability, has been reported to respond rapidly and often completely to KDT in a number of case series [45-49]. Up to 5 percent of patients with Doose syndrome may have glucose transporter 1 (GLUT-1) deficiency, another condition that responds to KDT as discussed below [50]. A three-center study of children with Doose syndrome found that KDT was associated with a 50 percent or greater reduction in seizures in 79 percent of cases, compared with only 26 percent with standard antiseizure medications [49].
Dravet syndrome — Intractable epilepsy is one of the characteristics of Dravet syndrome along with intellectual delay and increased seizures with fever and heat. Observational studies suggest that KDTs may reduce seizure frequency in many of these patients, especially the atypical absence seizures noted in this condition. Since there are many newer antiseizure medications for Dravet syndrome, KDTs are less widely used than in the past but are valuable therapies to still consider. (See "Dravet syndrome: Management and prognosis", section on 'Ketogenic dietary therapy'.)
GLUT-1 deficiency — GLUT-1 deficiency syndrome, caused by SLC2A1 mutations, is a genetic disorder characterized by impaired glucose transport across the blood brain barrier resulting in generalized epilepsy, developmental delay, and an associated movement disorder. It may also present with early-onset absence epilepsy. KDT is the most effective first-line treatment for this disorder and provides ketones as an alternative energy source for the brain [50-53]. In one study of 92 children, 95 percent had >50 percent seizure reduction after initiation of KDT and 80 percent had >90 percent seizure reduction [54]. Two-thirds of the patients were not on antiseizure medications. Two other studies have reported similar efficacy rates [55,56].
While seizures typically remit with this treatment, the effects on neurodevelopment are less apparent [23,24]. However, the diagnosis of GLUT-1 deficiency is often delayed, and developmental outcomes may be better with early initiation of dietary therapy [57,58]. Supplementation with triheptanoin, a medium-chain triglyceride, is also under investigation as a potential novel therapy for this disorder [59], but most GLUT-1 experts do not advocate it as a substitute for the ketogenic diet at this time. (See "Childhood absence epilepsy", section on 'Differential diagnosis'.)
Infantile epileptic spasms syndrome — A number of large observational studies have shown that the KDTs may control infantile epileptic spasms syndrome in patients refractory to first-line treatments. They are widely used therapies after first-line treatment with hormonal therapy or vigabatrin for Dravet syndrome. This is discussed in detail separately. (See "Infantile epileptic spasms syndrome: Management and prognosis", section on 'Ketogenic diet'.)
Pyruvate dehydrogenase deficiency — KDT may also serve to provide an alternative energy source for the brain in pyruvate dehydrogenase deficiency, a mitochondrial disease characterized by lactic acidosis, severe neurologic impairments, and occasionally, intractable epilepsy [60,61]. In uncontrolled reports, patients on the ketogenic diet appear to have an improved neurologic outcome (eg, cognition, vision) [62].
Tuberous sclerosis complex — Many patients with tuberous sclerosis complex are refractory to medical treatment but have multiple tubers that can be epileptogenic making surgery impossible. In these patients, KDT may help control seizures and should be considered along with other nonpharmacologic interventions. (See "Tuberous sclerosis complex: Management and prognosis", section on 'Epilepsy'.)
Gastrostomy- or formula-fed children — Poor adherence to the strict diet is a major limiting factor in the effective use of the KDTs. This is much less of a concern in patients who receive all of their nutrition via formula or gastrostomy feedings. In one case series, 12 patients with intractable epilepsy and a static encephalopathy were maintained on the ketogenic diet administered via a gastrostomy tube; only two patients did not improve; six patients achieved a more than 90 percent reduction in seizures [63]. A retrospective study found that 61 children who were administered the KDT via formula (31 formula fed infants; 30 gastrostomy tube fed children) had better rates of >90 percent seizure remission with the diet than children administered the diet through solid food: 59 versus 27 percent [64].
Super-refractory status epilepticus — There have been multiple case reports of KDT successfully treating prolonged, medically refractory nonconvulsive and focal status epilepticus in adults and children [65-67]. One case series describes efficacious use of KDT in seven of nine patients with febrile infection-related refractory epileptic encephalopathy (FIRES) [68]. An additional case report describes successful use of the modified Atkins diet in two children with nonconvulsive status epilepticus [69]. These cases all used ketogenic formulas provided through enteral (eg, gastric) tubes to administer KDT and achieve ketosis. In a prospective multicenter study of 15 patients with super refractory status epilepticus (ie, persisting for 24 hours or more despite appropriate anesthetic treatment, or returning after withdrawal of anesthetic agents), 14 patients completed ketogenic dietary treatment, and status resolved in 11 (73 percent) at a median of five days after starting ketogenic diet [70]. (See "Management of convulsive status epilepticus in children" and "Convulsive status epilepticus in adults: Management".)
Others — Other conditions identified by the expert panel as particularly responsive to KDT are the following [7]:
●Angelman syndrome (see "Microdeletion syndromes (chromosomes 12 to 22)", section on '15q11-13 maternal deletion syndrome (Angelman syndrome)')
●Complex I mitochondrial disorders
●Ohtahara syndrome (see "Overview of infantile epilepsy syndromes", section on 'Early infantile developmental and epileptic encephalopathy')
Moderately responsive conditions — In other conditions, KDT may be at least moderately beneficial, based upon lower-quality data or reports suggesting that the response is similar to but no better than the "norm" (ie, a 40 to 50 percent chance of ≥50 percent seizure reduction) [7]:
●Rett syndrome – KDT may improve intractable seizures in patients with Rett syndrome. However, in view of the frequent occurrence of growth failure in this syndrome, a ketogenic diet should be used with some caution. (See "Rett syndrome: Treatment and prognosis", section on 'Seizures'.)
●Landau-Kleffner syndrome – This epilepsy syndrome, which is now encompassed by the term DEE-SWAS, is sometimes associated with intractable epilepsy that may benefit from KDT [71]. The value for improving the EEG in this condition is somewhat controversial, however. (See "Epilepsy syndromes in children", section on 'Developmental and epileptic encephalopathy with spike-wave activation in sleep (DEE-SWAS)'.)
●Lennox-Gastaut syndrome (LGS) – This condition has many causes and is characterized by intractable epilepsy with multiple seizure types, especially tonic seizures; KDT appears to be helpful in many cases. (See "Lennox-Gastaut syndrome".)
●Absence epilepsy – Children with absence epilepsy may respond to KDT when antiseizure medications are incompletely effective. An ongoing clinical trial is comparing the modified Atkins diet with ethosuximide for the treatment of new-onset absence seizures. (See "Childhood absence epilepsy", section on 'Patients who fail or do not tolerate first-line therapy'.)
●Juvenile myoclonic epilepsy – In one case series of eight adolescents and adults with refractory juvenile myoclonic epilepsy, five patients had >50 percent seizure reduction after three months of a modified Atkins diet [37]. This diet may be considered as an alternative to valproate for adolescent or adult women with refractory juvenile myoclonic epilepsy who wish to avoid the teratogenicity of valproate. (See "Juvenile myoclonic epilepsy".)
●Others – Additional epilepsy conditions that may benefit from KDT are [7]:
•Adenylosuccinate lyase deficiency
•Cyclin-dependent kinase-like 5 (CDKL5) encephalopathy
•Cortical malformations (see "Seizures and epilepsy in children: Classification, etiology, and clinical features", section on 'Neurodevelopmental lesions')
•Epilepsy of infancy with migrating focal seizures
•Epileptic encephalopathy with continuous spike-and-wave during sleep (see "Epilepsy syndromes in children", section on 'Developmental and epileptic encephalopathy with spike-wave activation in sleep (DEE-SWAS)')
•Lafora body disease
CONTRAINDICATIONS — Fasting or implementation of KDT in individuals with inborn errors of metabolism that affect the transport or oxidation of long-chain fatty acids can lead to a devastating catabolic crisis. KDT can also exacerbate acute intermittent porphyria.
●Absolute contraindications to all forms of KDT include [7]:
•Primary carnitine deficiency
•Carnitine palmitoyltransferase I or II deficiency
•Carnitine translocase deficiency
•Fatty acid beta-oxidation defects
•Medium-chain acyl dehydrogenase deficiency
•Long-chain acyl dehydrogenase deficiency
•Short-chain acyl dehydrogenase deficiency
•Long-chain 3-hydroxyacyl-CoA deficiency
•Medium-chain 3-hydroxyacyl-CoA deficiency
•Pyruvate carboxylase deficiency
•Porphyria
Carnitine cycle disorders and long-, medium-, and short-chain fatty acid oxidation disorders are reviewed separately (see "Specific fatty acid oxidation disorders"), as are pyruvate carboxylase deficiency (see "Causes of hypoglycemia in infants and children", section on 'Pyruvate carboxylase deficiency') and porphyria. (See "Porphyrias: An overview".)
Patients with clinical features suggesting one of these diagnoses (eg, cardiomyopathy, hypotonia, exercise intolerance, myoglobinuria, easy fatigability) should be tested in order to exclude these conditions [7]. Recommended tests are included in the table (table 1). (See "Metabolic emergencies in suspected inborn errors of metabolism: Presentation, evaluation, and management".)
By contrast, patients with mitochondrial respiratory chain complex defects and epilepsy have been safely and effectively treated with KDT [72]. (See 'Particularly responsive conditions' above.)
●Relative contraindications to KDT include the following [7]:
•Inability to maintain adequate nutrition
•Inability of parent or caregiver to comply with dietary therapy
•Propofol concurrent use (risk of propofol infusion syndrome may be higher)
These conditions may make the diet difficult to implement, interfere with compliance, or increase the risk of adverse events.
In addition, patients with intractable focal epilepsy who are candidates for epilepsy surgery are likely to have a better response to epilepsy surgery than to KDT [19]. An electroencephalogram (EEG) and magnetic resonance imaging (MRI) are useful in the evaluation of potential epilepsy surgery candidates, but are likely to have been performed prior to consideration of KDT. However, a trial of KDT prior to surgery is an individual family/caregiver decision and should not be necessarily discouraged. (See "Seizures and epilepsy in children: Refractory seizures", section on 'Epilepsy surgery' and "Surgical treatment of epilepsy in adults".)
IMPLEMENTATION
Pretreatment evaluation — Prior to starting KDT, it is important to counsel patients, families, guardians, and caregivers about the implications and potential medical concerns associated with strict KDT, including anticipated seizure reduction, concomitant antiseizure medication use, and the psychosocial issues that may impact implementation [7]. When it comes to the diet, a discussion must include the cost and time involved in preparing special meals for the child or adult patient, avoidance of carbohydrates, the complexity of diet management, the importance of strict adherence to the diet, the need for additional supplements (see 'Supplementation' below), and potential adverse effects (see 'Adverse effects' below). If there is disagreement between parents or caregivers about using the diet, it is probably not worth starting.
It is important to assess potential food allergies or intolerances as well as cultural and religious dietary needs. A review of a three-day food record can also be helpful in identifying potential problems. However, most religions and cultures can be successfully accommodated with KDT.
Components of the pretreatment evaluation and laboratory studies are provided in the table (table 2) [7]. Patients must be screened for contraindications and evaluated for comorbid conditions that may complicate therapy. (See 'Contraindications' above.)
There is one more pretreatment concern: certain medical conditions may be aggravated by the diet. These include a history of kidney stones, hypercholesterolemia, liver disease, gastroesophageal reflux, constipation, cardiomyopathy, and chronic metabolic acidosis. A renal ultrasound and nephrology consult are recommended if there is a personal or family history of kidney stones. These conditions do not preclude the diet’s use, but make it more difficult. Prior history of pancreatitis or elevated cholesterol should be discussed at length with the family before initiating the diet.
Diet selection — The major KDTs for the treatment of epilepsy are the classic (long-chain) ketogenic diet, the medium-chain triglyceride diet, the modified Atkins diet, the low glycemic index treatment, and the modified ketogenic diet. The choice of a particular diet is individualized and often at the discretion of the parent. Most centers agree, however, that infants and children under two years of age should use the classic ketogenic diet, whereas adolescents and adults may have improved compliance if the modified Atkins diet or low glycemic index treatment is chosen.
Composition
Classic ketogenic diet — In the classic ketogenic diet, fat comes primarily from long-chain triglycerides obtained through standard foods; protein intake is based on that required for growth; and carbohydrates are restricted. The classic ketogenic diet consists of four parts fat to one part protein and carbohydrate (ie, 4:1 lipid to nonlipid ratio) [7]. Fat provides 90 percent of the calories. At most centers, dietitians match calories to the pre-ketogenic diet baseline. In the past, total calories were restricted to 80 to 90 percent of recommended values for age, but this did not result in clear benefit [73,74]. Modification of the diet to a somewhat lower fat content (3:1 lipid to nonlipid ratio from the classic 4:1 ratio) improves tolerability of the diet, but may reduce its efficacy during the initial three months [75].
Alternatively, the diet can be given as a liquid-based formula to bottle-fed infants and enterally fed patients. There are a number of commercially available products. Because the diet is easy to administer in this form, compliance and efficacy are generally excellent in this population [63,64]. (See 'Particularly responsive conditions' above.)
Medium-chain triglyceride diet — In an alternative form of the classic ketogenic diet, medium-chain triglycerides provided in an oil supplement (eg, coconut oil) are utilized as a major fat source [9]. These medium-chain triglycerides yield more ketones, are more efficiently absorbed, and are carried directly to the liver. As a result, less total fat is needed in the diet and more protein and carbohydrates can be allowed. The medium-chain triglyceride diet is widely used in England because the highly ketogenic nature of medium-chain triglyceride oils allows for additional carbohydrates. Many centers will also add medium-chain triglyceride oils to the classic ketogenic diet as a supplement (eg, 5 to 10 mL several times a day).
The traditional medium-chain triglyceride diet derives 60 percent of the energy from medium-chain triglycerides [7]. In some children, this needs to be reduced to 30 to 50 percent in order to minimize gastrointestinal symptoms, with the resulting decrement in energy source to be made up from long-chain fats.
Medium-chain triglycerides should be given with each meal or evenly divided up between several smaller meals in order to improve tolerance.
Modified Atkins diet — The modified Atkins diet is an alternative KDT designed to mimic some aspects of the classic ketogenic diet but allow more ad lib protein, fluids, and calories [25,76].
The diet is initiated as an outpatient without a fast. Carbohydrates are initially restricted to 20 grams per day with patients counseled to increase their use of high-fat foods (at the expense of protein). When implemented, most parents will provide an approximately 1:1 to 2:1 lipid to nonlipid ratio (although there is no weighing and measuring of foods recommended in this regimen) [76]. A ketogenic liquid supplement (KetoCal shake) can be used during the first month to increase the ketogenic ratio [77]. All children are started on a multivitamin and calcium, and lab studies are obtained identically to the classic ketogenic diet. After one to three months, the carbohydrate restrictions can be loosened and/or antiseizure medications reduced as clinically indicated.
This less restrictive diet has been used successfully in adults who are not typically offered the classic ketogenic diet [26,27], in countries with limited financial and dietitian resources [28], in children on the classic ketogenic diet for many years who desire fewer restrictions [25], and as an early treatment for epilepsy before intractability is established [76]. One center reported that the modified Atkins diet could be successfully maintained in selected adults using electronic communication (email) rather than in-person clinic visits [78]. This may be particularly useful in resource-poor regions of the world.
Ketosis will occur with the modified Atkins diet and may correlate with seizure control, but perhaps only initially, during the first month [17].
Low glycemic index treatment — An even less restrictive KDT, the low glycemic index diet restricts carbohydrates to 40 to 60 grams per day, does not restrict fluids or protein, and loosely monitors fat and calories [79]. It is started as an outpatient without a fasting period. As with the other KDTs, most children eat significantly more fat on this regimen than they did before. Unlike the modified Atkins diet, the type of carbohydrates is important in the low glycemic index diet, with only carbohydrates with low glycemic indices (<50) allowed. These carbohydrates include strawberries and whole grain breads as opposed to potatoes, white bread, and most citrus fruits. In patients on this diet, there is nearly no serum ketosis noted.
Modified ketogenic diet — The modified ketogenic diet, created in United Kingdom [80], uses a 1:1 ketogenic ratio (lower than normal), yet still monitors grams of fat, protein, and carbohydrates using a gram scale. It targets approximately 75 percent of calories from fat, 20 percent from protein, and 5 percent from carbohydrate, and is started at home. Although gram scales are used, simple food lists and recipes are provided to make it easier.
Starting treatment — Ideally, KDT should be supervised by a trained dietitian and neurologist in an epilepsy center with expertise in this form of therapy. Antiseizure medications should be switched to carbohydrate-free preparations (preferably tablets rather than liquid or chewable formulations). Dosing regimens are typically left unchanged for at least the first month of KDT [81]. It is also important to reduce or eliminate carbohydrates in medications where possible [82]. Another strategy is to compensate for potential carbohydrates in medications by increasing the ketogenic ratio. In one small case-control study, children remained on liquid antiseizure medications, compensated by an increased ketogenic ratio; these children had similar outcomes in achieving large ketosis, weight gain, and seizure reduction compared with controls on tablet formulations [83]. (See 'Antiseizure medication management' below.)
●The classic ketogenic diet and the medium-chain triglyceride diet are often started in the hospital without a fast period, but alternative approaches are recognized:
•Inpatient versus outpatient – The classic ketogenic diet is most often started in the hospital, which provides for close observation and rapid medical interventions if needed [7]. One of the most important reasons for the admission is to educate the parents and caregivers. Most hospitals will schedule lectures and teaching sessions during the admission. After an average hospital duration of three to four days, the child is discharged home.
Outpatient initiation of the classic ketogenic diet is an option for select children (≥12 months of age) who live close to medical care [7]. For outpatient initiation, children must first undergo metabolic testing and receive appropriate education about the diet. Some centers will initiate the classic ketogenic diet as an outpatient with classes held in various settings [73]. Although only 20 percent of an expert consensus panel routinely starts the classic ketogenic diet as an outpatient, the majority (92 percent) felt that it was possible to do this in select situations [7].
•Fasting versus nonfasting – The classic ketogenic diet was traditionally started following a 24-hour fasting period, but fasting is now considered optional [7]. Fasting may be appropriate when more rapid seizure control is needed. However, the initial fasting period is not required for long-term seizure control and increases the risk for hypoglycemia and lethargy [84,85]. We suggest avoiding fasting in children younger than two years and in older children who are poor eaters, those with multiple medical comorbidities, and children who have limited ability to communicate complaints of fatigue or nausea. In one study, 80 percent of children admitted for the classic ketogenic diet had some (often mild) side effects, typically resolving with minor interventions, but they can be more severe in those younger [85]. A randomized study that compared initial fasting with a more gradual initiation period found that there was less hypoglycemia, acidosis, or weight loss with a gradual approach, but the incidence of vomiting was unchanged [84]. Fasting does appear to lead to a more rapid onset of ketosis and seizure control, however [86,87].
The favored approach at many centers is to begin with full calories. This can be accomplished by gradually increasing the lipid to nonlipid ratio daily from 1:1, 2:1, 3:1, and finally to 4:1 to help the patient adapt to the higher fat concentration [84]. The classic ketogenic diet can also be started at full calories and a ratio of 4:1 on the first day; retrospective evidence suggests this approach does not reduce efficacy or increase adverse effects [74]. Alternatively, the diet can be started at a set ratio (eg, 4:1) and calories (not ratio) increased daily.
During the first one to two days of hospitalization, serum glucose is typically monitored several times a day (eg, before meals); small quantities of orange juice are provided if serum glucose falls below 30 mg/dL [7].
●The modified Atkins diet and the low glycemic index treatment are usually started in the outpatient setting without a fasting period but with group teaching sessions to help caregivers learn how to manage these diets at home [7]. (See 'Modified Atkins diet' above and 'Low glycemic index treatment' above.)
Supplementation — All children on KDTs are started routinely on a carbohydrate-free multivitamin with minerals (including selenium), calcium, and vitamin D supplement [7]. At Johns Hopkins and other centers, based on evidence for reduced kidney stone risk, all children are also started on a daily potassium citrate supplement (2 mEq/kg per day) [88]. Recommended and optional supplements for children are listed in the table (table 3).
ADVERSE EFFECTS
Burden — Parents and caregivers should be counseled as to the common, occasional, and rare adverse effects that can happen in the short-term when using KDTs. Fortunately, most adverse effects are predictable, often preventable, and only uncommonly in themselves lead to diet discontinuation [7]. Adverse effects and intolerance of the food restrictions are important to address early in the course of dietary therapy to prevent these issues from leading to premature discontinuation [89,90].
Common adverse effects — Relatively common adverse effects of all forms of KDT include [7,85]:
●Gastrointestinal symptoms – Diarrhea, constipation, nausea, vomiting, and exacerbation of gastroesophageal reflux are among the most common side effects, occurring in more than half of children on the ketogenic diet [7,35,36,85,91]. These are generally managed symptomatically with proton pump inhibitors, laxatives, and other treatments as needed.
●Dyslipidemia – While abnormal lipid parameters are frequently seen, significant hypercholesterolemia and hypertriglyceridemia are less common [35,90,92-96]. Most levels improve spontaneously even without dietary or other intervention, and only a minority of patients require medication.
While some centers are hesitant to use the KDT in children with baseline hyperlipidemia, there is evidence that dietary modifications can control lipid levels [93,95,96]. These modifications include increasing the consumption of medium-chain triglycerides and olive oil; adding omega-3 fatty acid or carnitine supplementation while reducing trans fat, saturated fat, and cholesterol; decreasing the lipid to nonlipid ratio of the diet; avoiding fatty meats, egg yolk, cream, butter, animal fat, palm oil and coconut oil; and using an exclusively formula-based ketogenic diet.
Atherogenic consequences of KDT, if any, are not known, but do not appear to happen in children who are followed years after KDT has been discontinued [7]. Studies are underway in adults on KDT specifically to investigate lipid abnormalities [97].
●Hypoglycemia – One or more episodes of hypoglycemia occur in approximately 25 percent of patients during the initiation week [85]. Pre-diet fasting may increase risk for hypoglycemia and lethargy, particularly in younger children (<2 years of age). (See 'Starting treatment' above.)
●Other laboratory abnormalities – Hyperuricemia, hypoproteinemia, hypomagnesemia, hyponatremia, hepatitis, and metabolic acidosis are sometimes seen and should be monitored for (see 'Follow-up and monitoring' below), but are usually minor [90]. Metabolic acidosis requiring treatment may be more common in children who are also taking topiramate or zonisamide [98]. Rare cases of severe hypercalcemia have been reported [99]. A study of 66 children with drug-resistant epilepsy found that thyroid hormone levels were unchanged with KDT [100].
●Growth failure – Most children on KDT fall into lower height and weight percentiles [101-104]. Younger children on KDT for longer periods of time appear to be the most at risk. In one small series, height velocity was most affected in those with the most pronounced ketosis [105]. The higher protein content in the medium-chain triglyceride diet does not appear to ameliorate this [106].
●Carnitine deficiency – Carnitine levels typically decline in the first months on KDT; however, carnitine deficiency is not common (<20 percent) [90,107]. Levels typically stabilize or improve spontaneously without supplementation. The expert panel recommended supplementation for those with low carnitine levels who are symptomatic with fatigue and lethargy (82 percent agreement) [7]. Only 16 percent of centers automatically use carnitine supplementation in all children on the ketogenic diet [7].
●Bone disease – Osteopenia, osteoporosis, and bone fractures are a concern for children maintained on KDTs, as they are for children on chronic antiseizure medications [90,102,108,109]. Prophylactic supplementation with calcium and vitamin D is mandatory for all children on KDTs, but sometimes fails to prevent continued bone loss. Although there is no consensus, many centers screen children who are on KDTs for two years or more with dual-energy x-ray absorptiometry (DXA) scans to evaluate for osteopenia [7]. (See "Overview of dual-energy x-ray absorptiometry".)
●Nephrolithiasis – Kidney stones may occur with KDTs in as many as 7 percent of children [90,110]; if fluid intake is normal, the incidence is likely lower, and in one study was reduced to less than 1 percent with the addition of a potassium citrate supplement [88]. The risk for kidney stones is highest in those with baseline urologic abnormalities or a personal or family history of nephrolithiasis [110]. Slightly more than half (56 percent) of the expert panel use potassium citrate in all patients empirically on KDT to prevent stones [7].
●Selenium deficiency – Selenium deficiency has been found in up to approximately 50 percent of children on KDTs [111]. Selenium deficiency has rarely produced an irreversible cardiomyopathy and sudden death in children on KDT [112-114]. Most centers check selenium levels; repletion with multivitamin with minerals including selenium is the most common strategy to avoid selenium deficiency [7].
●Neutropenia – A retrospective report of 102 children with epilepsy found that KDT was associated with an increased prevalence of neutropenia, generally mild, beginning at approximately six months of treatment and affecting up to 25 percent of patients [115]. However, there was no association of neutropenia with clinically significant infection.
Less common or rare adverse effects — Less common adverse effects of KDTs include:
●Cardiac abnormalities – As noted above, cardiomyopathy and prolonged QT intervals have been described in patients on KDT with and without selenium deficiency [116]; in the absence of selenium deficiency, the mechanism of cardiac abnormalities is unknown [7].
●Pancreatitis – Pancreatitis has been reported in a few children on KDT [90,117,118]. Most cases, but not all, had another risk factor for pancreatitis, usually valproate administration.
●Vascular changes – Several studies have documented a transient decrease in carotid distensibility in children on KDTs [119,120]. The clinical relevance of this is unclear.
Resolution with diet discontinuation — Once KDT is discontinued, most of the adverse effects resolve. In a follow-up study of 101 children who had discontinued the diet at a median of six years prior, height, weight, cholesterol, and food preferences were all normal [121]. There was no increased risk of late reported cardiovascular disease, bone fractures, or kidney stones. Most patients were eating normal foods years later as well.
ANTISEIZURE MEDICATION MANAGEMENT — KDTs generally do not significantly alter serum levels of the most commonly used antiseizure medications, and there are no particular antiseizure medications to avoid with KDTs [7]. Implementation of KDT was not associated with altered plasma concentrations of antiseizure medications in one case series of 51 children, suggesting that no dose adjustments or additional monitoring of drug levels beyond what is otherwise required is necessary in children on KDT [122]. However, more recent evidence in adults may indicate levels may fluctuate with KDT [123].
If seizure control is obtained on KDT, antiseizure medication withdrawal can begin as early as the first month of KDT. A slow, cautious taper is advised for medications like phenobarbital and benzodiazepines since rapid changes in these drugs can result in medication-withdrawal seizures [7].
KDT should be considered an adjunctive therapy for seizures; most patients on KDT remain on at least one antiseizure medication. In a retrospective series of 81 children, antiseizure medications were tapered during the first six months on KDT in 65 percent and within the first month on the diet in 37 percent [124]. Antiseizure medications were successfully eliminated in approximately 20 percent of patients. Transient worsening was more likely when the tapered drug was phenobarbital or a benzodiazepine.
It is unknown if any particular antiseizure medications modify the efficacy of KDT, and data are scarce. In one report of 115 patients on KDT, a greater than 50 percent reduction in seizure frequency was more common in patients on zonisamide, while those on phenobarbital were less likely to achieve this benefit [125]. Another study of 71 children suggested lower efficacy of KDT with combined use of lamotrigine [126]. The results of these small retrospective studies are not conclusive.
FOLLOW-UP AND MONITORING — Once children are started on KDT, they should be seen in clinic for follow-up at one month and then at 3, 6, 9, and 12 months during the first year of treatment [7]. Thereafter, follow-up should be every six months. Infants and those at higher risk for nutritional deficiency should have more frequent follow-up (at times monthly). Each visit provides an opportunity to adjust KDT and/or antiseizure medications if needed in order to improve seizure control and reduce adverse effects [127]. Email and phone contact in between clinic appointments is also recommended.
A clinical response to KDT is quick, usually within a few weeks [18,87]. Its efficacy can generally be established in individual patients within two to three months of initiation.
Routine monitoring recommendations are outlined in the table (table 4). Calorie intake, height, and weight are monitored at each visit. Calorie intake should be adjusted for excessive weight loss or gain. Laboratory studies, including complete blood count with platelets, comprehensive metabolic profile, fasting lipid profile, calcium, vitamin D, and magnesium are recommended every three months during the first year of therapy and every 6 to 12 months thereafter [7]. Additional laboratory studies may be useful for select patients, including serum selenium, zinc, beta-hydroxybutyrate levels, as well as renal ultrasounds and dual-energy x-ray absorptiometry (DXA) scans (table 4). European guidelines recommend monitoring serum beta-hydroxybutyrate levels for infants [128].
Measurement of urine ketones by parents is recommended several times per week to ensure the diet is being managed correctly, in a manner analogous to following antiseizure medication levels [7]. The value of serum ketone monitoring is unclear.
Some children may require laxatives if constipation is a problem, and/or extra selenium, salt, or carnitine, if they develop relevant abnormal laboratory parameters. The ketogenic ratio can be adjusted if there is loss of seizure control or decreased ketosis.
Monitoring for metabolic acidosis is suggested for patients on KDT who are taking carbonic anhydrase inhibitors (eg, topiramate and especially zonisamide) particularly in the first weeks/months after diet initiation [98]. While the combination of carbonic anhydrase inhibitors and the KDTs might be expected to increase the risk of kidney stones, this does not appear to be the case. In one series of 221 children on the diet, kidney stones were no more common in children on these antiseizure medications than on any other regimen [110].
The use of valproate in the setting of KDT may increase the risk of secondary carnitine deficiency [117,129], and there are rare reports of hepatotoxicity [90,130,131]. Increased monitoring of liver function tests and carnitine levels may be prudent in children on this therapeutic combination. In at least one case report of hepatic dysfunction arising in a child taking valproate, liver enzymes normalized when the drug was stopped and the diet was able to be resumed [130]. Pancreatitis has been reported a few cases of children on the ketogenic diet who were also taking valproate [90,117,118].
DIET DURATION AND DISCONTINUATION — Observational studies suggest that if successful, KDT usually will result in seizure reduction within two to three months [87]. We therefore recommend discontinuing KDT as a treatment failure only after it has been used for at least three months [7].
For patients who do experience benefit, there are no clear minimum or maximum duration periods for KDT, and the timing should be individualized [7]. In children who benefit from the diet, we re-evaluate the relative value of KDT after one and two continuous years of use, especially in those who have become seizure-free [7]. Eighty percent of those who discontinue KDT after two years of seizure freedom will continue to remain seizure-free [132]. Many centers obtain electroencephalography (EEG) to help inform the decision about withdrawing KDT [7]. Adults doing well with KDT may choose to remain on treatment for much longer, as their epilepsy is less likely to spontaneously improve.
The impact of discontinuing the diet in those with seizure frequency reduction but not seizure remission is not known. In one study, outcomes at three to six years were evaluated in 54 children who discontinued the diet within one year of initiation [89]. Almost half of these children had >50 percent reduced seizure frequency compared with baseline; 22 percent were seizure-free without surgery. Children who had had some improvement while on the diet were more likely to have a favorable long-term outcome. At our center, approximately 0.5 percent of KDT patients are on the diet for over six years; some have been on the diet for more than 20 years [133].
Discontinuation of KDT is done in a manner similar to antiseizure medications, with a slow wean over two to three months, reducing the fat to protein plus carbohydrate ratio (ie, lipid to nonlipid) from 4:1 monthly to 3:1, 2:1, and 1:1, followed by reintroduction of regular foods until ketosis is lost [7]. The benefit of this approach is that the ketogenic ratio can be increased should seizures recur or worsen during the weaning period. In one case series, seizure recurrence could be effectively managed in most (58 percent) children with resuming the diet, or adding antiseizure medications [132].
In some situations, particularly if the diet is either unsuccessful or is associated with worsening seizures, it can be weaned much more rapidly. In a retrospective chart review of 183 children on the diet, speed of diet discontinuation (<1 week, 1 to 6 weeks, or >6 weeks) was not associated with incidence of seizure worsening [134]. We have reduced the ratio weekly for some children, and even abruptly returned to standard diets for children in intensive care unit settings.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Seizures and epilepsy in adults" and "Society guideline links: Seizures and epilepsy in children".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Epilepsy in children (The Basics)" and "Patient education: Ketogenic diet (The Basics)" and "Patient education: Ketogenic diet and your child (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Ketogenic dietary therapy (KDT) is an effective treatment for patients with epilepsy, regardless of age or seizure type. A greater than 50 percent seizure reduction occurs in 38 to 60 percent of patients, with as many as 30 percent achieving a greater than 90 percent seizure reduction. (See 'Efficacy' above.)
●KDT is a valid treatment option for patients who have failed at least two antiseizure medications. There are a number of conditions in which KDT may be particularly effective and can be offered earlier in the course of treatment, including Doose syndrome, Dravet syndrome, glucose transporter 1 (GLUT-1) deficiency, infantile spasms, pyruvate dehydrogenase deficiency, and tuberous sclerosis complex; KDT may also be particularly effective for patients with drug-resistant epilepsy who are fed by gastrostomy tube or formula. Patients with intractable focal epilepsy who are candidates for epilepsy surgery are likely to have a better response to epilepsy surgery than to KDT (algorithm 1). (See 'Indications' above.)
●Before starting KDT, it is important to counsel patients, families, guardians, and caregivers about the implications related to seizure control, antiseizure medication use, and psychosocial issues (table 2). Patients should be screened for potential contraindications and/or factors that may complicate it use. (See 'Contraindications' above and 'Pretreatment evaluation' above.)
●The major KDTs for the treatment of epilepsy are the classic (long-chain) ketogenic diet, the medium-chain triglyceride diet, the modified Atkins diet, the low glycemic index treatment, and the modified ketogenic diet. The modified Atkins diet and a low glycemic index diet may be useful for adults as well as children. The choice of a particular diet should be based upon individual circumstances. Multivitamins (including minerals such as selenium), calcium, and vitamin D supplements are given to all patients on KDTs. Some children will require additional supplements (table 3). (See 'Composition' above and 'Diet selection' above and 'Supplementation' above.)
●KDT should be supervised by a trained dietitian and neurologist in an epilepsy center with expertise in this form of therapy. The classic ketogenic diet and the medium-chain triglyceride diet are often started in the hospital without a fast period, but alternative approaches are recognized. The modified Atkins diet and the low glycemic index treatment are usually started in the outpatient setting without a fasting period and may be useful for adults as well as children. (See 'Composition' above and 'Supplementation' above.)
●Patients are followed closely with frequent monitoring of clinical status and laboratory tests (table 4). Each visit provides an opportunity to adjust KDT and/or antiseizure medications if needed in order to improve seizure control and reduce adverse effects. Important adverse effects of KDT include gastrointestinal symptoms, dyslipidemia, hypoglycemia, constipation, growth failure, bone disease, and kidney stones. (See 'Follow-up and monitoring' above and 'Adverse effects' above.)
●Efficacy of the diet is usually apparent in the first one to three months. If seizure control is achieved, antiseizure medications may be slowly tapered. (See 'Antiseizure medication management' above.)
●KDT is generally recommended for a minimum of three months and maximum of two years; however, longer periods of effective use have been described. In the absence of severe adverse sequelae, slow taper is advised but the diet can be weaned more quickly if needed. (See 'Diet duration and discontinuation' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
