May 28, 2018
Ketogenic Diet shifts gut biome to improve seizures

Jacob Schor, ND, FABNO

A new study confirms what clinicians have known for nearly a century. Wu, Wang, Fan et al reported in April 2018 that a ketogenic diet was helpful in about 60% of the 52 kids had followed the diet to treat seizures. These weren’t just any epileptic kids; they were notable because their seizures were not responding to drug treatment. The trial lasted 12 weeks but those who responded did so in the first two weeks. The kids ranged in age from 3 months old to 7 years.

The treatment was considered effective in 29 of the 52 kids or 56% of the cases at the end of 12 weeks of treatment. Though actually in those who benefited so effects quickly. Fifteen of the kids were doing better in the first week. At the end of the study 14 participants were seizure free, 9 had a marked reduction in the number of seizures, and 6 kids had were having half the number of seizures or less than before treatment started. The treatment was ‘ineffective’ in 23 kids but keep in mind the bar for being effective was set high, a 50% reduction in the number of seizures from baseline. [1]

Modern interest in diet to treat epilepsy began with fasting; two French physicians, Guelpa and Marie, wrote a paper in 1911 and reported seeing improvements in 20 children they had treated but provided little detail.[2] Hugh Conklin, a doctor in Battle Creek, Michigan popularized therapeutic fasting in the United States in the early 1900s. Conklin worked as an assistant to Bernarr Macfadden, the ‘fitness guru’ behind Physical Culture magazine. Macfadden claimed fasting could cure just about anything but especially epilepsy. Macfadden was a character; he apparently shortened his first name from Bernard so he could pronounce it like a lion’s roar.

Conklin fasted epileptic patients in his practice and Rawle Geyelin an endocrinologist in New York noticed and adopted the practice as well. He reported his patient experiences in 1921. [3] This spread further to Harvard where Cobb and Lennox began studying fasting. They were the first to observe how rapidly improvement happened; the effects of starvation caused seizure improvement in just 2-3 days. [4]

Two important observations on fasting were published in 1921. Woodyatt noted that in normal people a diet that had too low a proportion of carbohydrate and too high a proportion of fat caused similar changes as a starvation diet, that is, the appearance of acetone and beta-hydroxybutyric acid in the blood.[5] Wilder at the Mayo Clinic suggested that similar benefits to fasting were achieved if ketonemia was reached through dietary manipulation and that fasting was not necessary.[6] Wilder went on to coin the term ‘ketogenic diet’ and treated patients at the Mayo Clinic using it.

In 1925 Peterman, also at the Mayo Clinic published the formula that is still used today. One gram of protein per kilogram body weight in children and only 10-15 grams of carbohydrate is consumed per day with the remainder of calories coming from fat. [7]

The ketogenic diet was described in almost every medical textbook on epilepsy published between 1940 and 1980. Livingston at Johns Hopkins reported that in over 1,000 children he had treated, 52% had complete symptom control and 27% had symptom improvement following the diet. [8] These percentages are remarkably similar to those reported in the new Wu study.

Awareness of the ketogenic diet and research interest was revived by a Dateline TV show broadcast on NBC in October 1994. The TV program followed the story of a 2-year old boy named Charlie, treated at Johns Hopkins by Millicent Kelly (a dietitian who had worked with Dr. Livingston). Charlie became seizure-free and his father formed The Charlie Foundation, which disseminated informational videos for parents about the ketogenic diet, published a book about the diet [9] and even produced a TV movie (starring Merryl Streep) and funded research that showed the diet produced significant benefit. [10,11]

Although the ketogenic diet has been effective in treating childhood seizures for nearly a hundred years ago, no one has been certain why it works.

In late May 2018, a study published in the journal Cell explained that the ketogenic diet’s impact on epilepsy is related to its effect on the human intestinal biome. That sentence needs to be written a second time; the ketogenic diet may reduce seizures because it changes the gut biome. This is exciting news, a paradigm shift in our understanding on how diet affects mental function.

Wu et al in discussing their findings do not mention this biome idea in their discussion; they instead expressed uncertainty why the diet works suggesting that shifting the brain to using ketones as an energy source or perhaps the caloric restriction might have something to do with the benefits. Neither explanation was convincing.

Earlier mice experiments, have demonstrated that ketogenic diets prevent development of epilepsy[12] , improve symptoms of autism [13] , improve motor symptoms in Alzheimer’s disease [14] , and reduce epileptic activity in the brain. [15]

This biome theory appeared in the May 24, 2018 issue of Cell. Christine Olson and colleagues at Elaine Hsiao’s lab at UCLA claim that the ketogenic diet quickly alters the gut biome in a specific way so that it provides protection against both electrically induced seizures and spontaneous tonic-clonic seizures in mouse models of epilepsy.

The ketogenic diet does not provide seizure protection to germ free mice, raised in a germ free environment or who have been heavily treated with antibiotics. The effect requires gut bacteria to be present. However transplanting these germ free mice with populations of Akkermansia and Parabacteroides bacteria confers protection against seizures.

Olson et al propose a simple mechanism of action. The high fat, low carb ketogenic diet shifts the gut biome decreasing diversity and increasing populations of Akkermansia muciniphilia and Parabacteroides spp. bacteria. These bacteria decrease gamma-glutamyltranspeptidase activity, decreasing gamma-glutamyl amino acids in the blood, which in turn increases GABA levels in the brain. Increased GABA in the brain offers the protection against seizures.

[we don’t have permission for this graphic. I put it in just so I could look at it.]

Hsiao’s lab has been producing a steady stream of interesting research related to the gut biome and it’s impact on the brains of mice and humans for years.

In 2013 Hsiao reported that in a mouse model of autism, alterations in microbiota and GI barrier could be corrected using Bacteroides fragilis, which “…corrects gut permeability, alters microbial composition, and ameliorates defects in communicative, stereotypic, anxiety-like and sensorimotor behaviors.”[16] In simpler words, modifying the gut biome reduced the autism-like symptoms. Hsiao’s work on autism continues. Because of her work it is now widely accepted that “… immune dysregulation and gastrointestinal issues are common comorbidities” in the autism spectrum. [17, 18]

UCLA has already granted licensing rights to a start up company that plans to develop a probiotic treatment for epilepsy. The idea is that the right formulation of bacteria will modulate GABA, “… thereby conferring the neuroprotective effects of the ketogenic diet in a pill. The pill regimen would be easier to comply with than the diet and potentially have fewer side effects.” [19]

There may be other strategies to increase gut populations of these bacteria and so increase GABA. Metformin, a drug used to treat type-2 diabetes, has been reported to increase populations of both these bacterial species in mice.[20] Yang et al reported in 2017 that chronic use of metformin does have some anti-seizure effect on mice.[21] Consumption of certain ‘resistant starches’ designed to reach the large intestine and ‘feed’ certain bacteria may also increase populations of these bacteria. [22]

We really don’t understand the relationships between various bacteria species and disease as yet. We are just getting hints as to the profound impact they may have.

Both Akkermansia muciniphila and Acinetobacter calcoaceticus, were found to be four times more abundant in multiple sclerosis patients than in healthy people, while Parabacteroides distasonis is four times more abundant in healthy people than in MS patients. Akkermansia and Acinetobacter are associated with inflammatory responses in MS while Parabacteroides appears to have an anti-inflammatory action. [23] We probably do not want to rush into force-feeding all of our patients with every species of bacteria that improve life for a mouse.

Treatment of epilepsy may be on the verge of shifting to a focus on altering the gut biome by using a combination of probiotics, ketogenic diet and possibly eating foods containing resistant starches. If this strategy does indeed increase GABA levels in the brain, a long list of other possible conditions might be improved by such a treatment strategy.


1. Wu Q, Wang H, Fan YY, et al. Ketogenic diet effects on 52 children with pharmacoresistant epileptic encephalopathy: A clinical prospective study. Brain Behav. 2018 Apr 18;8(5):e00973.

2. Guelpa G, Marie A. (1911)La lutte contre l‘e’pilepsie par la de’ sintoxication et par la re’e’ducation alimentaire. Rev Ther Medico-Chirurgicale78:8–13.

3. Geyelin HR. (1921) Fasting as a method for treating epilepsy. Med Rec99:1037–1039.

4. Penfield W. Erickson TC. (1941) Epilepsy and cerebral localization: a study of the mechanism, treatment, and prevention of epileptic seizures. Charles C.Thomas, Baltimore . pp. 504–509.

5. Woodyatt RT. (1921) Objects and method of diet adjustment in diabetics. Arch Intern Med 28:125–141.

6. Wilder RM. (1921) The effect on ketonemia on the course of epilepsy. Mayo Clin Bull 2:307.

7. Peterman MG. (1925) The ketogenic diet in epilepsy. JAMA 84:1979–1983.

8. Livingston S. (1972)Comprehensive management of epilepsy in infancy, childhood and adolescence. Charles C. Thomas, Springfield , IL , pp. 378–405.

9. Freeman JM, Kelly MT,Freeman JB. (1994) The epilepsy diet treatment: an introduction to the ketogenic diet. Demos, New York .

10. efficacy of the ketogenic diet 1998: a prospective evaluation of intervention in 150 children. Pediatrics 102:1358–1363.

11. Wheless JW. History of the ketogenic diet. Epilepsia. 2008;49 Suppl 8:3-5.

12. Lusardi TA, Akula KK, Coffman SQ, et al. Ketogenic diet prevents epileptogenesis and disease progression in adult mice and rats. Neuropharmacology. 2015 Dec;99:500-9.

13. Ruskin DN, Svedova J, Cote JL, et al. Ketogenic diet improves core symptoms of autism in BTBR mice. PLoS One. 2013 Jun 5;8(6):e65021.

14. Brownlow ML, Benner L, D’Agostino D, Gordon MN, Morgan D. Ketogenic diet improves motor performance but not cognition in two mouse models of Alzheimer’s pathology. PLoS One. 2013 Sep 12;8(9):e75713

15. Forero-Quintero LS, Deitmer JW, Becker HM.
Reduction of epileptiform activity in ketogenic mice: The role of monocarboxylate transporters. Sci Rep. 2017 Jul 7;7(1):4900.

16. Hsiao EY, McBride SW, Hsien S, et al. Microbiota modulate behavioral and physiological abnormalities associated with neurodevelopmental disorders. 2013 Cell. 2013 Dec 19;155(7):1451-63.

17. Hsiao EY. Gastrointestinal issues in autism spectrum disorder. Harv Rev Psychiatry. 2014 Mar-Apr;22(2):104-11.

18. Vuong HE, Hsiao EY. Emerging Roles for the Gut Microbiome in Autism Spectrum Disorder. Biol Psychiatry. 2017 Mar 1;81(5):411-423.


20. Lee H, Lee Y, Kim J, et al. Modulation of the gut microbiota by metformin improves metabolic profiles in aged obese mice. Gut Microbes. 2017 Nov 20:1-11.

21. Yang Y, Zhu B, Zheng F, et al. Chronic metformin treatment facilitates seizure termination. Biochem Biophys Res Commun. 2017 Mar 4;484(2):450-455.

22. Graf D, Di Cagno R, Fåk F, et al. Contribution of diet to the composition of the human gut microbiota. Microb Ecol Health Dis. 2015; 26: 10.3402/mehd.v26.26164.

23. Cekanaviciute E, Yoo BB, Runia TF, et al. Gut bacteria from multiple sclerosis patients modulate human T cells and exacerbate symptoms in mouse models. PNAS October 3, 2017. 114 (40) 10713-10718;