While evaluation of nutraceutical strategies to improve outcomes for TBI is important, it will not be addressed in this review. Instead, the review will focus on the potential for a ketogenic diet to provide therapeutic benefit among individuals who have suffered a TBI.
The ketogenic diet is by no means a “fad”. Papers objectively evaluating its effects in epilepsy can be found as far back as 1911 and again in 1922.(1, 2) “Dr Atkins’ Diet Revolution was first published in 1972, and low carb ketogenic diets have been somewhat of a mainstay among the lay public seeking weight loss since the republication of Dr Atkins book in 2001, entitled “Dr Atkins’ New Diet Revolution”.
Extensive preclinical data highlights an impressive, positive impact of ketogenic diets following TBI, and will be reviewed below, as well as limited yet important human evidence for the same outcomes. Further supporting the role of ketogenic diets in TBI is extensive evidence of benefit from the strategy in epilepsy, on which topic over 50 human trials were identified, with an outdated meta analysis and a couple of interesting modern trials referenced.(3, 4, 5) Similarly, evidence of benefit from a ketogenic diet in human trials of neurodegenerative disease support the role of the strategy in TBI, with a recent review showcasing approximately 10 such human trials.(6)
For completeness and towards discussion of safety, a significant number of human trials have also successfully implemented ketogenic diets for management of both type I and type II diabetes, with a small sample of such trials referenced here.(7, 8, 9) As will be discussed below, impact of the ketogenic diet on glucose control is highlighted as a major basis for application of the diet in TBI.
The classic ketogenic diet of the early 1900’s delivered 90% of energy as fat, 7% of energy as protein, and 8% of energy as carbohydrate (CHO). The medium chain triglyceride (MCT)- based diet delivers 70% of energy as fat, 10% of energy as protein, and 20% of energy as CHO. The Atkins style ketogenic diet recommends 70% of energy as fat, 25% of energy as protein, and 5% of energy as CHO.(5) We generally prescribe the Atkins- style of the program, based on concerns that greater than 5% of energy as CHO will not successfully achieve ketosis.
Two ketone bodies are available for CNS metabolism; beta-hydroxybutyrate and acetoacetate. (10) Blood concentration of ketones in a state of typical CHO intake is very low (<0.3mmol/L), with glucose concentrations of approximately 4-6 mmol/L. (6) Internet sources (11) claim blood ketone concentration of 1.5-3.0 mmol/L is optimal for weight loss, while Wlodarek (2019) suggests the brain will begin to use ketones as a source of energy at a blood level of 4 mmol/L, and that up to 8mmol/L is normal and safe. Hans Krebs referred to this as “physiological ketosis”. In type I diabetes, blood ketones can achieve >25 mmol/L, with concomitant low/ near zero insulin and dramatically elevated blood glucose, which begins to impact (lower) blood PH. This true metabolic acidosis can be fatal. (6)
Neuronal Impact of Ketone Metabolism
Why would transition from reliance on glucose to ketones for energy within the CNS be neuroprotective? Oxidative stress and impaired ATP generation by mitochondria are considered central to the pathogenesis of neurodegeneration. Animal models have revealed that transition to metabolism of ketones lowers oxidative stress and improves ATP production within neurons. (6, 10) It has been proposed that ketones produce more ATP per unit of oxygen than glucose. Also, transition to ketone metabolism appears to increase NADH oxidation and inhibit mitochondrial permeability, further contributing to reduction in oxidative stress and improved ATP generation. In a rat model, transition to ketosis was accompanied by increased glutathione and glutathione peroxidase activity in the hippocampus. Transition to ketosis has also been shown to increase brain-derived neurotropic factor (BDNF). (6)
Interestingly, it is a common standard of care to deny patients with CNS injury glucose for a period of 24-48 hours. It has long been accepted that hyperglycemia exacerbates complications of ischemia, TBI, and experimental TBI, and thus insulin is administered to tightly control glucose in such settings. (10, 12, 13, 14)
Animal Data in TBI
Prins (2008) and Wlodarek (2019) provide excellent reviews of the extensive preclinical research on ketogenic diets in animal models of TBI.(6, 10) In brief, outcomes reproducibly demonstrated in animals include: improved survival, reduced neuronal cell death, decreased cerebral edema, decreased infarct size, decreased reactive oxygen species formation within the brain, increased NADH oxidation, improved ATP production, and improved outcomes on cognitive testing. Clearly, we would hope such outcomes can be reproduced in humans, yet the invasive nature of obtaining such outcomes does not permit such evaluation.
Human Data in TBI
Limited human data does exist, and it will be reviewed here. Ritter (1996) first conducted a series of animal experiments to determine macronutrient composition that delivered best outcomes in their animal model, then applied this macronutrient breakdown in a human trial. (15) Twenty severely head injured patients were randomized to ketogenic or control diets and followed for two weeks following the head injury. The ketogenic diet achieved significant decreases in blood glucose levels and blood lactate levels, with increases in blood ketone levels. Urinary nitrogen balance was also better in the group receiving the ketogenic diet.
In another study, ketogenic diets were administered to seven patients with spinal cord injury. (16) The intervention improved upper-extremity motor scores, as well as reducing several markers of inflammation, notably fibrinogen.
While human data on the use of ketogenic diets in TBI is limited, several facts lend strong support to its clinical application in TBI. First, metabolic consequences of transitioning neuronal metabolism to ketones creates an environment clearly lending itself to improved neuronal cell function and reductions in generation of harmful substrates. Second, a well- established body of human trials suggests very important impact of the strategy in epilepsy and neurodegenerative disease. Mechanistically, the manner by which the diet benefits these two states directly supports the application of the diet in TBI. Lastly, while not reviewed here, the diet has been reproducibly demonstrated to be safe across a broad range of populations.
In my practice prescription to follow a ketogenic diet is standard for patients whom have suffered a TBI. Unlike its use in epilepsy, type I diabetes, or neurodegeneration, the prescription for the diet in TBI is temporary. Once the patient demonstrates two to three symptom free weeks, the diet is transitioned back to a standard CHO containing diet.
- Guelpa, G., and Marie, A. Rev. Ther. Med. Chir.1911; 78, 8–13.
- Conklin, H.W. Jam. Osteopatic. Assoc. 1922; 26, 11–14.
- Ashrafi, M.R., et al. Acta Neurol Belg. 2017;117:175-182.
- IJff, D.M., et al. Epilepsy Behav. 2016 Jul; 60:153-157.
- Henderson, C.B., et al. J Child Neurol. 2006; 21:193-8.
- Włodarek, D., et al. Nutrients. 2019; 11. pii: E169.
- Hallberg, S.J., et al. Diabetes Ther. 2018; 9:583-612.
- Lennerz, B.S., et al. Pediatrics.2018, e20173349; DOI: 10.1542/peds.2017-3349
- Westman, E.C., et al. Expert Rev Endocrinol Metab. 2018; 13:263-272.
- Prins, M.L. J Cereb Blood Flow Metab. 2008; 28(1):1-16.
- Diet Doctor. Accessed April 2019. https://www.dietdoctor.com/lose-weight-by-achieving-optimal-ketosis
- Béjot Y., et al. Stroke. 2012; 43:243-5.
- Lam, A.M., et al. J Neurosurg. 1991 Oct;75(4):545-51.
- Melo, J.R., et al. Acta Neurochir (Wien). 2010 Sep;152(9):1559-65.
- Ritter, A.M., J Neurotrauma. 1996; 13:473-85.
- Yarar-Fisher C., et al. Spinal Cord Ser Cases. 2018 21; 4:88.