• 
• 

Three Part Question

In [children with treatment-resistant epilepsy] is a [ketogenic diet] effective in [reducing the number of seizures]?

Clinical Scenario

An eight year-old girl was seen in an epilepsy clinic with a diagnosis of left cerebral hemi-atrophy with right hemiparesis and complex partial seizures. Eight months previously, she was on 800mg sodium valproate and 12.5mg lamotrigine and still having 18 seizures a week. She then spent seven weeks as an inpatient under specialist observation. A ketogenic diet was recommended and commenced five months ago. Since then she has less seizures, but has a major episode requiring midazolam every two weeks. Her seizures now involve intermittent paraesthesia on the left side. Her medication has been decreased to 400mg sodium valproate and 5mg lamotrigine. The patient is struggling with the diet (particularly the Liquigen medium chain triglyceride) and urinary ketones have been ranging from trace to high. The current plan is to continue with the diet, increase the valproate to 500mg and gradually take her off the lamotrigine.

Search Outcome

Cochrane - 1 result.
PubMed - "Epilepsy/diet therapy"[MeSH]"
First set of limit: 2 results, 1 relevant but not an RCT.
Second set of limits: 21 results, 7 relevant
Third set of limits: 10 results. 1 systematic review

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Lefevre & Aronson (2000)	Child subjects with various forms of epilepsy.	Systematic review of 9 retrospective and 2 prospective cohort studies	Seizure elimination and seizure frequency reduction	Elimination: 16% (95% CI:11.0-21.7). >90% reduction: 32% (95% CI: 25.3-39.8). >50% reduction in 56% (95% CI: 41.2-69.7)
Vining et al (1998)	51 paediatric patients from 7 centres	Prospective uncontrolled trial. 2b	Seizure frequency reduction (SFR) at 3, 6 and 12 months	3 mo: 25% had >90% SFR, 29% had 50-90% SFR, 33% had <50% SFR. 6 mo: 29% had >90% SFR, 24% had 50-90% SFR, 16% had <50% SFR. 12 mo: 22% had >90% SFR, 18% had 50-90% SFR, 8% had <50% SFR.	By 12 months only 24 patients remained on the diet.
Freeman et al (1998)	150 epileptic patients aged 1-16	Before-after uncontrolled trial. 2b	Seizure frequency reduction at 3, 6 and 12 months	>90% seizure reduction, 3 mo: 34%, 6 mo: 32%, 12 mo: 27%	By 12 months, 83 patients remained on the diet.
Freeman & Vining (1999)	17 paediatric Lennox-Gastaut syndrome patients	Prospective uncontrolled trial. 2b	Seizure frequency reduction over first 5 days of diet	All children had >50% reduction in seizures by day 5 of diet	Small number of subjects. Individual results not reported. The paper also found that glucose drinks eliminated ketosis whilst on the diet. This information is being used in the first blinded RCT of the ketogenic diet.
Hemingway et al (2001)	All 150 patients from Freeman et al (1998) were followed up.	Prospective uncontrolled trial. 2b	Seizure frequency reduction and medication requirements 3-6 years after initiation of the diet	13% seizure free, 14% had 90-99% seizure reduction	No children still on diet. Aimed to analyse the 83 patients who had dieted for 1 year. Results may be compounded by natural progressive improvement in epilepsy over the years.
Kossoff et al (2002)	23 children with infantile spasms	Retrospective uncontrolled trial. 2b	Seizure frequency reduction (SFR) and medication requirements at 3, 6, 9 and 12 months	At 3, 6, 9, and 12 months, 38%, 39%, 53%, and 46% respectively had >90% SFR, and 67%, 72%, 93%, and 100% were >50% improved.	13 patients remained on the diet at 1 year. Present the proportion of benefiting patients as a percentage of those still on the diet rather than those originally enrolled in the trial.
Coppola et al (2002) Italy	56 children and young people aged 1-23 years, with refractory partial or generalised epilepsy	Prospective uncontrolled trial. 2b	Seizure frequency reduction at 3, 6, 9 and 12 months	At 3, 6 and 12 months respectively, >50% improvement was found in 37.5%, 26.8% and 8.9% of patients.	Only 5 (<10%) patients remained on the diet at 1 year. Adherence was especially poor in the older age groups.
Chul Kang et al (2005) Korea	199 epileptic children, mean age 57.9 months	Retrospective uncontrolled trial. 2b	Seizure frequency reduction and medication requirements at 3, 6, 9 and 12 months	At 6 mo and 12 mo respectively, 68% and 46% remained on the diet. Of the original 199 patients, 58% and 41% had >50% seizure reduction and 33% and 25% became seizure free.
Hosain et al (2005)	12 gastrostomy fed epileptic children	Prospective uncontrolled study. 2b	Seizure frequency reduction at 12 an 18 months	Mean reduction in seizure frequency was 61% and 66% respectively.	11 patients still on the diet at 12 months. The fact that the patients are gastrostomy-fed elminates one of the principle problems with this therapy: adherence.

Comment(s)

The studies identified were highly relevant to the clinical question and show that a significant proportion of patients have a reduction in seizure frequency whilst on the diet. A number of them suggest that there are very few patients who will show no improvement if they remain on the diet. However, a major drawback for most of these trials is the high drop-out rates (usually around 50% by 1 year). Patients who discontinue the diet are mainly those who noticed no improvement on it. The diet is indeed arduous. Note also that much of the published literature originates from one centre (John Hopkins Hospital, Baltimore, references 1,2,3,5 & 10), though the research now emerging from other centres seems to be concordant. The diet is well-tolerated, and most studies report only mild, transient side effects which can include gastroesophageal reflux and kidney stones. At present, the research on this topic consists of retrospective and prospective studies (ie level 2b). The single relevant record retrieved from the Cochrane library (ref 11) explains that as yet there are no randomised controlled trials on the efficacy of the ketogenic diet. Until recently it has not been possible to conduct a blinded RCT in this area. This is because the diet needs to be supervised by clinicians, so they would need to know who was on the diet. Freeman & Vining (1999) found that the ketotic state produced by the diet can be eliminated if the patient drinks a dextrose solution. This discovery has enabled the first double-blinded RCT to take place, with all patients on the diet but control patients will also drink the sugar solution (ref 10). The results of this trial are not yet published, but should demonstrate whether ketosis has an anti-epileptic effect or not.

Clinical Bottom Line

Evidence suggests that the ketogenic diet has a significant anti-epileptic effect. However, it is arduous and should only be considered in medication-resistant epilepsy. Access is currently very limited in the UK, as the diet requires expert supervision. The diet is physically safe, but contraindicated in patients with an inborn error of metabolism.

References

1. Lefevre F, Aronson N Ketogenic diet for the treatment of refractory epilepsy in children: A systematic review of efficacy Pediatrics 2000 Apr; 105(4):E46
2. Vining EP, Freeman JM, Ballaban-Gil K, et al A multicenter study of the efficacy of the ketogenic diet Arch Neurol 1998 Nov; 55(11): 1433-7.
3. Freeman JM, Vining EP, Pillas DJ, et al The efficacy of the ketogenic diet-1998: a prospective evaluation of intervention in 150 children Pediatrics 1998 Dec; 102(6): 1358-63
4. Freeman JM, Vining EP Seizures decrease rapidly after fasting: preliminary studies of the ketogenic diet Arch Pediatr Adolesc Med 1999 Sep; 153(9): 946-9.
5. Hemingway C, Freeman JM, Pillas DJ, et al. The ketogenic diet: a 3- to 6-year follow-up of 150 children enrolled prospectively. Pediatr Neurol 2005 Feb; 32(2): 81-3.
6. Kossoff EH, Pyzik PL, McGrogan JR, et al Efficacy of the ketogenic diet for infantile spasms Pediatrics 2002 May; 109(5):780-3
7. Coppola G, Veggiotti P, Cusmai R, et al The ketogenic diet in children, adolescents and young adults with refractory epilepsy: an Italian multicentric experience Epilepsy Res 2002 Feb; 48(3): 221-7
8. Chul Kang H, Joo Kim Y, Wook Kim D, et al Efficacy and safety of the ketogenic diet for intractable childhood epilepsy: Korean multicentric experience Epilepsia 2005 Feb; 46(2): 272-9
9. Hosain SA, La Vega-Talbott M, Solomon GE Ketogenic diet in pediatric epilepsy patients with gastrostomy feeding Pediatr Neurol 2005 Feb; 32(2): 81-3
10. Freeman Efficacy of the Ketogenic Diet – a Blinded Study
11. Levy R, Cooper P Ketogenic diet for epilepsy The Cochrane Database of Systematic Reviews 2003, Issue 3. Art. No.: CD001903. DOI: 10.1002/14651858.CD001903.