Reading Time: 13 minutes

The following is an early draft of a section written for a review paper 

Inflammation has a well-characterized role in promoting cancer [1] and an inverse relationship with longevity [2]. One prospective cohort study of 2,438 subjects aged 70-79 with an average follow-up of 5.5 years reported that elevated inflammatory markers IL-6, CRP, and TNF-α were strongly associated with both increased cancer incidence and cancer mortality after excluding cancer diagnoses within two years of baseline [3]. Another study of 7178 patients with stable cardiovascular disease and CRP < 10 mg/L and a median follow-up of 8.3 years also reported strong association between cancer and CRP levels, with sensitivity analyses excluding patients within 1, 2, and 5 years all showing similar results [4]. Another study with 7,017 participants aged 55 years or older with a mean follow-up of 10.2 years using CRP as a biomarker and excluding cancer diagnosis before 5 years follow-up reported similar results [5]. A large literature, moreover, demonstrates the relationship between genetic variants that produce higher lifetime inflammatory exposure and risk from a variety of cancers, including breast [6,7], non-Hodgkin’s lymphoma [8], gastric cancer [9], prostate cancer [10], lung cancer [11], and others. The relationship, too, between chronic infection and tumorogenesis is well-known, with about 18% of cancer globally due to chronic infection and a large portion of that to the inflammatory component of such infections [12]. Historical cohort data moreover suggest a link between childhood mortality from infectious disease (a proxy for cohort-wide inflammatory exposure or nutritional deficiency) and risk of death later, the latter of which may be mediated in part by increased cancer risk [13,14].

Accordingly, the ketogenic diet has been proposed to lower cancer risk by reducing inflammation [15].

Several studies have recently been published on the role of beta-hydroxybutyrate, the principal metabolite elevated during the ketogenic diet, on NLRP3 inflammasome-mediated inflammatory markers and disease [16–19]. NLRP3 is an innate immune sensor that is activated by toxins, ATP, excess glucose, ceramides, amyloids, urate, and cholesterol crystals and may have an important etiological role in type 2 diabetes, atherosclerosis, multiple sclerosis, Alzheimer’s disease, age-related functional decline, bone loss, and gout [16]. In 2015, investigators reported the results of a series of elegant experiments showing that beta-hydroxybutyrate inhibits activation of the NLRP3 inflammasome in response to LPS treatment in both rodents in vitro and in vivo and in human monocytes in vitro, via a potassium channel mediated mechanism [16]. In contrast however, a recent paper has reported the opposite effect, namely an increase in LPS activation when human monocytes were exposed to betahydroxybutyrate [20]. Marked methodological differences between the papers might account for these conflicting findings. In Youm et al., 2015, isolated monocytes were cultured in 11.1 mM glucose, then exposed to BHB and LPS for 1 hour and 4 hours, respectively. On the other hand, Neudorf et al., 2019 exposed for two hours to LPS human monocytes which were still in the original whole blood that had been drawn from subjects who consumed an exogenous ketone ester or ketone salt 2.5 hours earlier. Thus, the different findings might be accounted for by the differences in glucose in the media/blood, differences in LPS and BHB exposure time, differences in effects on the monocytes from non-monocyte cells in the whole blood versus lack of such effects in the media, and differences in exposure to other metabolites (such as fatty acids) and hormones (such as insulin) which may also be modulated in response to betahydroxybutyrate and may change the time-course of the phenotype in vivo compared to in vitro. Context may be a key determinant of the effect of betahydroxybutyrate on immune cell phenotypes; if so, this context-dependence needs to be adequately described before making strong claims about the impact of BHB on NLRP3 inflammasome activation.

Much preclinical data exists demonstrating the efficacy of the ketogenic diet in rodent models of inflammatory disease [21]. In one rat model of pain, KD decreased swelling and plasma extravasation [22]. However, as with previous studies discussed, this study is also confounded by profound protein restriction, with the ketogenic diet group having 5% of calories from protein, and the control group having 28% of calories from protein. Although some recent human studies suggest that higher protein intakes may be associated with lower inflammatory markers [23,24], some recent studies in rodents have shown the opposite [25,26], with one recent study in particular demonstrating particularly profound immunomodulatory effects of a protein-restricted diet in rodents [27]. This calls into question whether the anti-inflammatory effects seen in this study are due to ketosis or to protein restriction.

Another study showed that the KD alleviates motor dysfunction induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), a neurotoxin used to model Parkinson’s in rodents [28]. For diet, this study cited another paper, which itself provided scant detail about the composition of the diet [29]. In another rat model of fever, KD reduced inflammatory markers in the blood and lower fever [21]. Yet this study too used a profoundly protein-restricted diet, with the KD group having approximately 1/3 of the protein as a proportion of total calories as the controls. Another study showed that motor, learning, and brain defects in a mouse model of multiple sclerosis were ameliorated by the KD [30]. This study provided no information about the control diet, confounding interpretation. In a recent study using a mouse model of gout, no information was given about the control diet used [17].

Suffice to say, the lack of appropriate control diets for ketogenic diet rodent studies makes many or most of the studies conducted to date difficult to interpret.

The body of human study literature is equivocal. Although there are a number of relevant randomized controlled trials comparing low- and high-carbohydrate diets and inflammatory markers during weight loss, we could find no systematic reviews or meta-analyses on the topic, which has also been reported by a recent National Lipid Association report [31]. We will therefore briefly review of a random sampling of eleven relevant randomized clinical trials comparing a low- to a high-carbohydrate diet.

Four studies showing reduced inflammatory markers

Among these, four showed that a low-carbohydrate diet improved blood inflammatory markers: two showed lower blood C-reactive protein [32,33], a risk factor highly associated with cardiovascular disease risk and activation of  systemic inflammatory pathways [34], one showed for the low-carbohydrate group compared to the high-carbohydrate group lower levels of IL-8, TNF-alpha, MCP-1, I-CAM, and PAI-1, among a selection of 20 markers [35], and one showed for the low-carbohydrate group lower serum amyloid A levels (with null findings for other markers including CRP) [36]. Three of these studies [32,33,35] were only of three months in duration and therefore predictably showed greater weight loss in the low-carbohydrate group, an effect that is lost in longer-term studies (see above section on weight loss). Because weight loss is one of the best-established means of reducing blood inflammatory markers in the general population [37], differential weight loss in the arms of the low- versus high-carb trials confounds interpretation of these studies. One of these three studies [32] found no difference after adjusting for weight loss, while the effect on inflammatory markers remained after adjusting for weight loss in the other two studies [33,35]. In one of these remaining two studies of these three, baseline characteristics were not appropriately matched in ways that might be relevant to CRP levels, particularly with respect to metabolic markers and possibly CRP itself [33]. Nonetheless, the findings of Forsythe et al., 2008, the third of these studies, are suggestive. The study finding elevated serum amyloid A levels showed borderline statistical significance (p = 0.04) at a p=0.05 cutoff and did not adjust for multiple comparisons of inflammatory markers, of which there were nine reported, suggesting that this finding was not in fact statistically significant [36].

Four studies show no difference in inflammatory markers

Nonetheless, in three other studies in this sample, there was no difference in inflammatory markers between groups, despite significantly greater weight loss in the low-carbohydrate group [38–40], suggesting the opposite of the above studies: that a low-carbohydrate diet might mitigate some of the positive effects on inflammatory markers of weight loss. And in yet another study that showed no difference in weight loss, there was also no difference in C-reactive protein between groups [41].

Three studies show a worsening of inflammatory markers

Finally, in the three remaining studies in this sample, higher CRP was observed in the low-carbohydrate group compared to the high-carbohydrate group [42–44]. Importantly, unlike the previous studies, two of these three studies were in subjects that lost similar amounts of weight [43,44]. However, while weight loss was tightly controlled, with equal weight loss during the baseline diet and ketogenic diet phases, Rosenbaum et al., 2019 was conducted in crossover fashion, with 4 weeks on the baseline diet followed by 4 weeks on the ketogenic diet, with roughly 2 kg lost per 4-week phase and no washout period. The lack of randomization of diet order (or alternatively, the lack of a second crossover phase back to baseline diet) confounds interpretation of causality of C-reactive protein changes during the ketogenic diet period in this trial [44]. The results however of Ebbeling et al., 2012 are suggestive, and the impact on C-reactive protein on inflammatory markers remains to be reported from a new study by the same group (NCT02068885; see Ebbeling et al., 2018).

In summary, the evidence for changes in inflammatory markers is mixed, with studies with equivalent weight loss between groups showing worsening of these inflammatory markers [43,44]; one study with higher weight loss in the low-carbohydrate group also showing worsening [42]; other studies with higher weight loss in the low-carbohydrate group showing either neutral effects [38–40] or benefits to inflammatory markers [32,33,35], some of which remain after controlling for weight loss [33,35], and others of which that do not [32].

Subtle differences in dietary composition, adherence, or other aspects of study design could account for the heterogeneity of these study findings. Likewise, although only blood inflammatory markers were reported in these studies, it is conceivable that inflammatory markers in the organs or tissues (e.g. liver, muscle biopsy) could be markedly differentially expressed compared to blood, thereby causing blood-only inflammatory marker studies to overlook real effects. However, we are not aware of human randomized controlled trials assessing low-carbohydrate versus high-carbohydrate diets that have looked at inflammatory markers other than in the blood. Moreover, given that low-carbohydrate diets have not shown more weight loss at 12 months than high-carbohydrate diets (Churuangsuk et al., 2018; see section on weight loss), weight equivalence is an important experimental variable to control for in such studies. But even this approach poses a critical problem: weight loss at twelve months or more is not different between diets probably because of non-adherence to a low energy intake and perhaps to the diet itself [47], not necessarily because the diets equilibrate in their theoretical biological capacity to induce weight loss despite strict adherence. Thus, studies that evaluate differences in inflammatory markers between low- and high-carbohydrate diets with weight equivalence and strict adherence in the period of less than 6-12 months of duration are not necessarily representative of real-world differences between low- and high-carbohydrate diets at twelve months or longer, the latter of which will likely be substantially deviated from the original diet prescribed and thus may have a different inflammatory marker profile than a more strictly adhered-to version of the low-carbohydrate diet.

Therefore, as with weight loss, the crucial problem with the carbohydrate-restricted or ketogenic diet in modulating inflammatory markers, assuming it does so in humans, is adherence. If adherence is not possible in the long-term, then short-term nutrition studies looking at inflammatory markers may be evaluating a phenomenon that does not occur in the real world over a long enough period of time to be practically applicable. Thus, interventions that are expected to exert their effects in the long-term (such as through inflammatory signaling), should be validated as an adherable intervention by the population being studied under the conditions of the RCT in question, in order to make the results of that RCT of practical utility to recommendations, guidelines, or clinical practice.

On the basis of currently existing evidence, it cannot be concluded that carbohydrate-restricted or ketogenic diets reduce inflammatory markers in humans and thus cannot exert their effect on cancer prevention or treatment through an anti-inflammatory mode of action. It is possible that the ketogenic diet has tissue-specific anti-inflammatory action, but this can only be shown for a minority of tissues via biopsy. The majority of tissues for which this effect might be demonstrated will require a randomized clinical trial for a specific inflammatory disease state.

An early draft of a table that summarizes the above randomized controlled trials

Screen Shot 2019-10-08 at 9.56.24 PM.png

  1. Mantovani, A.; Allavena, P.; Sica, A.; Balkwill, F. Cancer-related inflammation. Nature 2008, 454, 436–44.
  2. Pilling, L.C.; Kuo, C.L.; Sicinski, K.; Tamosauskaite, J.; Kuchel, G.A.; Harries, L.W.; Herd, P.; Wallace, R.; Ferrucci, L.; Melzer, D. Human longevity: 25 genetic loci associated in 389,166 UK biobank participants. Aging (Albany. NY). 2017, 9, 2504–2520.
  3. Il’yasova, D.; Colbert, L.H.; Harris, T.B.; Newman, A.B.; Bauer, D.C.; Satterfield, S.; Kritchevsky, S.B. Circulating levels of inflammatory markers and cancer risk in the health aging and body composition cohort. Cancer Epidemiol. Biomarkers Prev. 2005, 14, 2413–2418.
  4. van’t Klooster, C.C.; Ridker, P.M.; Hjortnaes, J.; van der Graaf, Y.; Asselbergs, F.W.; Westerink, J.; Aerts, J.G.J. V; Visseren, F.L.J. The relation between systemic inflammation and incident cancer in patients with stable cardiovascular disease: a cohort study. Eur. Heart J. 2019.
  5. Siemes, C.; Visser, L.E.; Coebergh, J.W.W.; Splinter, T.A.W.; Witteman, J.C.M.; Uitterlinden, A.G.; Hofman, A.; Pols, H.A.P.; Stricker, B.H.C. C-reactive protein levels, variation in the C-reactive protein gene, and cancer risk: The Rotterdam Study. J. Clin. Oncol. 2006, 24, 5216–5222.
  6. Schuetz, J.M.; Grundy, A.; Lee, D.G.; Lai, A.S.; Kobayashi, L.C.; Richardson, H.; Long, J.; Zheng, W.; Aronson, K.J.; Spinelli, J.J.; et al. Genetic variants in genes related to inflammation, apoptosis and autophagy in breast cancer risk. PLoS One 2019, 14.
  7. Milanese, J.-S.; Tibiche, C.; Zaman, N.; Zou, J.; Han, P.; Meng, Z.; Nantel, A.; Droit, A.; Wang, E. Germline genomes of breast and lung cancer patients significantly predict clinical outcomes. bioRxiv 2018, 312355.
  8. Cerhan, J.R.; Ansell, S.M.; Fredericksen, Z.S.; Kay, N.E.; Liebow, M.; Call, T.G.; Dogan, A.; Cunningham, J.M.; Wang, A.H.; Liu-Mares, W.; et al. Genetic variation in 1253 immune and inflammation genes and risk of non-Hodgkin lymphoma. Blood 2007, 110, 4455–4463.
  9. Furuya, T.K.; Jacob, C.E.; Tomitão, M.T.P.; Camacho, L.C.C.; Ramos, M.F.K.P.; Eluf-Neto, J.; Alves, V.A.F.; Zilberstein, B.; Cecconello, I.; Ribeiro, U.; et al. Association between polymorphisms in inflammatory response-related genes and the susceptibility, progression and prognosis of the diffuse histological subtype of gastric cancer. Genes (Basel). 2018, 9.
  10. Cui, X.; Yan, H.; Ou, T.W.; Jia, C.S.; Wang, Q.; Xu, J.J. Genetic variations in inflammatory response genes and their association with the risk of prostate cancer. Biomed Res. Int. 2015, 2015.
  11. Pintarelli, G.; Cotroneo, C.E.; Noci, S.; Dugo, M.; Galvan, A.; Delli Carpini, S.; Citterio, L.; Manunta, P.; Incarbone, M.; Tosi, D.; et al. Genetic susceptibility variants for lung cancer: Replication study and assessment as expression quantitative trait loci. Sci. Rep. 2017, 7.
  12. Parkin, D.M. The global health burden of infection-associated cancers in the year 2002. Int. J. cancer 2006, 118, 3030–44.
  13. Finch, C.E.; Crimmins, E.M. Inflammatory exposure and historical changes in human life-spans. Science 2004, 305, 1736–9.
  14. Beltrán-Sánchez, H.; Crimmins, E.M.; Finch, C.E. Early cohort mortality predicts the rate of aging in the cohort: A historical analysis. J. Dev. Orig. Health Dis. 2012, 3, 380–386.
  15. Weber, D.D.; Aminzadeh-Gohari, S.; Tulipan, J.; Catalano, L.; Feichtinger, R.G.; Kofler, B. Ketogenic diet in the treatment of cancer – Where do we stand? Mol. Metab. 2019.
  16. Youm, Y.-H.; Nguyen, K.Y.; Grant, R.W.; Goldberg, E.L.; Bodogai, M.; Kim, D.; D’Agostino, D.; Planavsky, N.; Lupfer, C.; Kanneganti, T.D.; et al. The ketone metabolite β-hydroxybutyrate blocks NLRP3 inflammasome-mediated inflammatory disease. Nat. Med. 2015, 21, 263–9.
  17. Goldberg, E.L.; Asher, J.L.; Molony, R.D.; Shaw, A.C.; Zeiss, C.J.; Wang, C.; Morozova-Roche, L.A.; Herzog, R.I.; Iwasaki, A.; Dixit, V.D. β-Hydroxybutyrate Deactivates Neutrophil NLRP3 Inflammasome to Relieve Gout Flares. Cell Rep. 2017, 18, 2077–2087.
  18. Shang, S.; Wang, L.; Zhang, Y.; Lu, H.; Lu, X. The Beta-Hydroxybutyrate Suppresses the Migration of Glioma Cells by Inhibition of NLRP3 Inflammasome. Cell. Mol. Neurobiol. 2018, 38, 1479–1489.
  19. Trotta, M.C.; Maisto, R.; Guida, F.; Boccella, S.; Luongo, L.; Balta, C.; D’Amico, G.; Herman, H.; Hermenean, A.; Bucolo, C.; et al. The activation of retinal HCA2 receptors by systemic beta-hydroxybutyrate inhibits diabetic retinal damage through reduction of endoplasmic reticulum stress and the NLRP3 inflammasome. PLoS One 2019, 14.
  20. Neudorf, H.; Durrer, C.; Myette-Cote, E.; Makins, C.; O’Malley, T.; Little, J.P. Oral Ketone Supplementation Acutely Increases Markers of NLRP3 Inflammasome Activation in Human Monocytes. Mol. Nutr. Food Res. 2019, 63.
  21. Dupuis, N.; Auvin, S. Anti-Inflammatory Effects of a Ketogenic Diet: Implications for New Indications. In Ketogenic Diet and Metabolic Therapies: Expanded Roles in Health and Disease; Rho, J.M., Ed.; Oxford University Press (OUP): Oxford; New York, 2017; pp. 147–155.
  22. Ruskin, D.N.; Kawamura, M.; Masino, S.A. Reduced pain and inflammation in juvenile and adult rats fed a ketogenic diet. PLoS One 2009, 4.
  23. Kitabchi, A.E.; McDaniel, K.A.; Wan, J.Y.; Tylavsky, F.A.; Jacovino, C.A.; Sands, C.W.; Nyenwe, E.A.; Stentz, F.B. Effects of high-protein versus high-carbohydrate diets on markers of β- Cell function, oxidative stress, lipid peroxidation, proinflammatory cytokines, and adipokines in obese, premenopausal women without diabetes. Diabetes Care 2013, 36, 1919–1925.
  24. Hruby, A.; Jacques, P.F. Dietary Protein and Changes in Biomarkers of Inflammation and Oxidative Stress in the Framingham Heart Study Offspring Cohort. Curr. Dev. Nutr. 2019, 3.
  25. de Carvalho, T.S.; Sanchez-Mendoza, E.H.; Nascentes, L.M.; Schultz Moreira, A.R.; Sardari, M.; Dzyubenko, E.; Kleinschnitz, C.; Hermann, D.M. Moderate Protein Restriction Protects Against Focal Cerebral Ischemia in Mice by Mechanisms Involving Anti-inflammatory and Anti-oxidant Responses. Mol. Neurobiol. 2019.
  26. Hanjani, N.A.; Vafa, M. Protein restriction, epigenetic diet, intermittent fasting as new approaches for preventing age-associated diseases. Int. J. Prev. Med. 2018, 9.
  27. Orillion, A.; Damayanti, N.P.; Shen, L.; Adelaiye-Ogala, R.; Affronti, H.; Elbanna, M.; Chintala, S.; Ciesielski, M.; Fontana, L.; Kao, C.; et al. Dietary Protein Restriction Reprograms Tumor-Associated Macrophages and Enhances Immunotherapy. Clin. Cancer Res. 2018, 24, 6383–6395.
  28. Yang, X.; Cheng, B. Neuroprotective and anti-inflammatory activities of ketogenic diet on MPTP-induced neurotoxicity. J. Mol. Neurosci. 2010, 42, 145–153.
  29. Xu, X.; Sun, R.; Jin, R. Effect of ketogenic diet on hippocampus mossy fiber sprouting and GluR5 expression in kainic acid induced rat model. Chin. Med. J. (Engl). 2006, 119, 1925–9.
  30. Kim, D.Y.; Hao, J.; Liu, R.; Turner, G.; Shi, F.D.; Rho, J.M. Inflammation-mediated memory dysfunction and effects of a ketogenic diet in a murine model of multiple sclerosis. PLoS One 2012, 7.
  31. Kirkpatrick, C.F.; Bolick, J.P.; Kris-Etherton, P.M.; Sikand, G.; Aspry, K.E.; Soffer, D.E.; Willard, K.E.; Maki, K.C. Review of current evidence and clinical recommendations on the effects of low-carbohydrate and very-low-carbohydrate (including ketogenic) diets for the management of body weight and other cardiometabolic risk factors: A scientific statement from the National Lipid Association Nutrition and Lifestyle Task Force. J. Clin. Lipidol. 2019.
  32. O’Brien, K.D.; Brehm, B.J.; Seeley, R.J.; Bean, J.; Wener, M.H.; Daniels, S.; D’Alessio, D.A. Diet-induced weight loss is associated with decreases in plasma serum amyloid a and C-reactive protein independent of dietary macronutrient composition in obese subjects. J. Clin. Endocrinol. Metab. 2005, 90, 2244–9.
  33. Ruth, M.R.; Port, A.M.; Shah, M.; Bourland, A.C.; Istfan, N.W.; Nelson, K.P.; Gokce, N.; Apovian, C.M. Consuming a hypocaloric high fat low carbohydrate diet for 12 weeks lowers C-reactive protein, and raises serum adiponectin and high density lipoprotein-cholesterol in obese subjects. Metabolism. 2013, 62, 1779–87.
  34. Ridker, P.M. From C-Reactive Protein to Interleukin-6 to Interleukin-1: Moving Upstream to Identify Novel Targets for Atheroprotection. Circ. Res. 2016, 118, 145–156.
  35. Forsythe, C.E.; Phinney, S.D.; Fernandez, M.L.; Quann, E.E.; Wood, R.J.; Bibus, D.M.; Kraemer, W.J.; Feinman, R.D.; Volek, J.S. Comparison of low fat and low carbohydrate diets on circulating fatty acid composition and markers of inflammation. Lipids 2008, 43, 65–77.
  36. Rajaie, S.; Azadbakht, L.; Saneei, P.; Khazaei, M.; Esmaillzadeh, A. Comparative effects of carbohydrate versus fat restriction on serum levels of adipocytokines, markers of inflammation, and endothelial function among women with the metabolic syndrome: a randomized cross-over clinical trial. Ann. Nutr. Metab. 2013, 63, 159–67.
  37. Meydani, S.N.; Das, S.K.; Pieper, C.F.; Lewis, M.R.; Klein, S.; Dixit, V.D.; Gupta, A.K.; Villareal, D.T.; Bhapkar, M.; Huang, M.; et al. Long-term moderate calorie restriction inhibits inflammation without impairing cell-mediated immunity: A randomized controlled trial in non-obese humans. Aging (Albany. NY). 2016, 8, 1416–1431.
  38. Hu, T.; Yao, L.; Reynolds, K.; Whelton, P.K.; Niu, T.; Li, S.; He, J.; Bazzano, L.A. The effects of a low-carbohydrate diet vs. a low-fat diet on novel cardiovascular risk factors: A randomized controlled trial. Nutrients 2015, 7, 7978–7994.
  39. Sharman, M.J.; Volek, J.S. Weight loss leads to reductions in inflammatory biomarkers after a very-low-carbohydrate diet and a low-fat diet in overweight men. Clin. Sci. (Lond). 2004, 107, 365–9.
  40. Seshadri, P.; Iqbal, N.; Stern, L.; Williams, M.; Chicano, K.L.; Daily, D.A.; McGrory, J.; Gracely, E.J.; Rader, D.J.; Samaha, F.F. A randomized study comparing the effects of a low-carbohydrate diet and a conventional diet on lipoprotein subfractions and C-reactive protein levels in patients with severe obesity. Am. J. Med. 2004, 117, 398–405.
  41. Harvey, C.J.D.C.; Schofield, G.M.; Zinn, C.; Thornley, S.J.; Crofts, C.; Merien, F.L.R. Low-carbohydrate diets differing in carbohydrate restriction improve cardiometabolic and anthropometric markers in healthy adults: A randomised clinical trial. PeerJ 2019, 2019.
  42. Rankin, J.W.; Turpyn, A.D. Low carbohydrate, high fat diet increases C-reactive protein during weight loss. J. Am. Coll. Nutr. 2007, 26, 163–9.
  43. Ebbeling, C.B.; Swain, J.F.; Feldman, H.A.; Wong, W.W.; Hachey, D.L.; Garcia-Lago, E.; Ludwig, D.S. Effects of dietary composition on energy expenditure during weight-loss maintenance. JAMA – J. Am. Med. Assoc. 2012, 307, 2627–2634.
  44. Rosenbaum, M.; Hall, K.D.; Guo, J.; Ravussin, E.; Mayer, L.S.; Reitman, M.L.; Smith, S.R.; Walsh, B.T.; Leibel, R.L. Glucose and Lipid Homeostasis and Inflammation in Humans Following an Isocaloric Ketogenic Diet. Obesity 2019, 27, 971–981.
  45. Ebbeling, C.B.; Feldman, H.A.; Klein, G.L.; Wong, J.M.W.; Bielak, L.; Steltz, S.K.; Luoto, P.K.; Wolfe, R.R.; Wong, W.W.; Ludwig, D.S. Effects of a low carbohydrate diet on energy expenditure during weight loss maintenance: randomized trial. BMJ 2018, 363, k4583.
  46. Churuangsuk, C.; Kherouf, M.; Combet, E.; Lean, M. Low-carbohydrate diets for overweight and obesity: a systematic review of the systematic reviews. Obes. Rev. 2018, 19, 1700–1718.
  47. Guo, J.; Robinson, J.L.; Gardner, C.D.; Hall, K.D. Objective versus Self-Reported Energy Intake Changes During Low-Carbohydrate and Low-Fat Diets. Obesity 2019, 27, 420–426.


Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.


Reading Time: 15 minutes

This is an early, lightly edited draft of a section from an upcoming scientific review.

Early days: carbohydrate restriction for diabetes and weight loss

Historically, culturally, and clinically, what is called the ketogenic diet is in fact a conglomeration of a number of dietary applications. What today might seem like an obvious, singular concept in retrospect in fact represents multiple independent interventions that have, by their family resemblance, been placed under essentially two major semantic umbrellas: low-carbohydrate and ketogenic diet. Diets called low-carbohydrate or ketogenic vary widely in their composition and have a range of different physiological effects. In this section, a brief history of these various ketogenic diets will be elucidated.

The first case of the successful use of a therapeutic carbohydrate-restricted diet, although probably not a ketogenic one, was reported in 1797 by English physician John Rollo for the treatment of what we now call type 2 diabetes mellitus. Rollo reported that after confinement to one room without exercise was ordered, food should consist of the following: “Breakfast, 1½ pints of milk and 1 pint of lime-water, mixed together; and bread and butter. For noon, plain blood puddings, made of blood and suet only. Dinner, game, or old meats, which have been long kept; and as far as the stomach may bear, fat and rancid old meats, as pork. To eat in moderation. Supper, the same as breakfast.” Rollo’s patient would steadily improve, lose substantial weight, and being a captain in the Navy, resume military duty. Echoes of John Rollo’s methods would be heard in the case report literature throughout the 19th century [1], including the perhaps most famous account by William Banting as retold by Gary Taubes [2]. An indication of the depth of the influence of Rollo’s method, William Osler would write in The Principles and Practice of Medicine in 1892 that in order to treat diabetes mellitus “the carbohydrates in the food should be reduced to a minimum” [3]. In the 1910s, this treatment would indeed be embraced even for type 1 diabetes and the outcomes reported, despite the frequently cited (if not fully understood) concern of ketoacidosis, are impressive [4]. Recently, carbohydrate-restriction has been once again rediscovered as an adjunct to insulin for type 1 diabetes [5] and as a means of remission for type 2 diabetes [6].

A few of decades after Rollo, in 1825, a similar diet would be recommended for the treatment of obesity by a Frenchman Jean Anthelme Brillat-Savarin in his classic The Physiology of Taste. He writes: “a more or less rigid abstinence from everything that is starchy or floury will lead to the lessening of weight” [7]. Later, in 1905, President Taft was treated using carbohydrate by an English physician who advocated such a diet in bestselling popular works [8,9]. From there, the low-carb diet for weight loss would cycle in and out of fashion for the next two centuries, with the names of popularizers William Banting, Robert Atkins, and Gary Taubes among the most familiar in the United States today.

Even earlier days: fasting for epilepsy

But perhaps the most ancient and fundamental strand of the ketogenic diet story is thought to relate to its use in the treatment of epilepsy, starting with Hippocrates. However, just as the oft-cited “Let food by thy medicine and medicine thy food” is a fabrication [10], it is remarkably difficult to find in the Hippocratic corpus advocacy for fasting for the treatment of epilepsy, despite claims of such having become a mythical, permanent fixture in the literature [11–16], with several recent articles providing false citations [14,16], others falsely claiming that fasting was the only treatment for epilepsy advocated by Hippocrates [11,15], and few properly referencing the actual source document [17]. What was actually written by Hippocrates in the relevant passage bears strikingly little resemblance to the claims typically made about it:

On the fifth day, tongue severely affected; the convulsion came on and he was beside himself. When these things ceased, his tongue with difficulty returned to its own condition. On the sixth day, as he abstained from everything, both gruel and drink, there were no further seizures [18] (pg. 353).

On Epidemics, the book in which this account is to be found, was a case report series, not a series of recommendations, and this is an account with scant detail indeed (fasting is not mentioned in On the Sacred Disease, as is often claimed). What is clear moreover is that the primary intervention for epilepsy among the Hippocratics was not fasting. As epilepsy and ketogenic diet expert William Lennox wrote, directly quoting Hippocrates in his 1960 classic textbook: “Epilepsy in young persons is most frequently removed by changes of air, of country and of modes of life” [19]. Later physicians would later lament the vagueness of Hippocrates’s therapeutic prescriptions for epilepsy [20].

The relationship of fasting to epilepsy in antiquity becomes just as unclear when interpreting the commonly cited line in the King James Bible (e.g. Wheless, 2008): “This kind can come forth by nothing, but by prayer and fasting.” (Mark 9:29) This apparent quote is in fact is subject to interpretation: many modern translations do not include “and fasting”, leaving things at “by prayer.” This calls into question the supposed New Testament pedigree of the fasting for epilepsy concept. Some articles take this mistake further, claiming that Jesus actively instructed a boy suffering from convulsions to fast [16]. No translation of the verse is compatible with this claim.

Still, while fasting may have a peripheral role in the Hippocratic therapeutic armamentarium, it quickly comes to play a central role in Greek medicine two centuries later, as Alexandrian physician Erasistratus makes clear:

One inclining to epilepsy should be made to fast without mercy and be put on short rations, but should beware of too much bathing and things causing a powerful change [17].

Thereafter, fasting would figure in the medical texts from the Hellenistic period, transmitted to the Romans, playing an important role in Galen’s writing on epilepsy, and through physicians in the Middle Ages (who cited the Byzantine Bible rather than Greco-Roman medical knowledge) [17,20].

Thus, while neither Jesus nor Hippocrates endorsed fasting for the treatment of epilepsy (sorry), making the ketogenic-diet-for-epilepsy story less mythical and clear-cut than almost always misreported, still, it is clear that the concept of fasting for epilepsy is quite old if it could find its way, even as a fabrication, into the Byzantine manuscript or the works of Alexandrian physicians at all. And its persistence suggests that perceptive clinical acumen was enough to keep it alive as a therapeutic modality. However, that same experience was noisy enough as a whole that fasting was only one therapeutic modality among others, with most medical writers preferring other approaches [17,20]. Proper recognition of fasting’s place as an epilepsy treatment, much less the ketogenic diet, would have to wait for modern science. To imply that it was somehow always present and central to the wisdom traditions of the past, as many writers have, is an anachronistic distortion of the historical record.

Modern science: the rise and fall of the ketogenic diet for epilepsy

The first modern case report of fasting for epilepsy was published by Parisian physicians Guelpa and Marie in 1911 [21]. Word soon spread of an osteopathic physician Hugh W. Conklin using fasting for epilepsy [11]. In 1921, Geyelin would present the case reports of Hugh W. Conklin at that year’s American Medical Association conference. Conklin would follow in 1922 by publishing data showing cure rates of 90% in his juvenile patients and 50% in adults [13]. A wealthy New York corporate lawyer enlisted Conklin’s help to treat his son (H.T.H.), who had intractable epilepsy. After following Conklin’s starvation regimen, the boy’s seizures left him, permanently [11], which we now know occurs in a substantial minority of pediatric epilepsy patients treated with the KD for several years [22].

H.T.H.’s parents asked Stanley Cobb, a Harvard professor, to explain the success of starvation and would fund some of the first research into fasting for epilepsy. In a series of experiments, it was demonstrated that the fast was broken with carbohydrate or protein but not with 40% cream [23]. Wilder at Mayo Clinic in a concurrent article wrote: “If this is the mechanism responsible for the beneficial effect of fasting, it may be possible to substitute for that rather brutal procedure a dietary therapy which the patient can follow with little inconvenience and continue at home as long as seems necessary.” It was in this paper that the term “ketogenic diet” was born [24].

The ketogenic diet was a breakthrough for the long-term management of epilepsy and the first truly efficacious and sustainable treatment for the disease. For the majority of patients, fasting does not cure epilepsy and seizures recur upon cessation of fasting, making the treatment impractical, especially for growing children. But once it was recognized that ketogenesis drove the therapeutic effect, and that ketogenesis could be maintained in the fasted state with the addition of fat but not carbohydrate or protein, it became possible to conceive of a diet that maintained the ketogenic benefits of fasting while avoiding many of the downsides, namely excessive calorie restriction, growth retardation, and wasting.

The publication by Peterman of the precise macronutrient formulation of the diet for children—1 gram of protein per kilogram of bodyweight, 10-15 grams of carbohydrate, and the remainder fat—followed Wilder’s publication in 1924 [25]. Talbot and colleagues at Harvard then published the classic 3:1 and 4:1 fat-to-carbohydrate-and-protein gram ratios in 1926 [26]. These formulations are used in refractory cases to the present day [27].

With the discovery of diphenylhydantoin in 1938 and with attention focused on the growing number of antiepileptic medications (which were easier to administer), the ketogenic diet soon fell by the wayside. Only modest developments in the use of the diet occurred during the subsequent decades. Perhaps most notable, in 1971, Huttenlocher et al published on the use of medium-chain triglycerides (MCTs), which enables ketogenesis with less carbohydrate restriction but still shows clinically significant antiepileptic effects [28].

Re-emergence of the ketogenic diet for epilepsy

Use of ketogenic diets for epilepsy almost vanished until 1994 when NBC-TV’s Dateline aired the story of Charlie Abrahams, son of director Jim Abrahams, whose intractable epilepsy was successfully treated with the ketogenic diet at the Johns Hopkins Epilepsy Center, which was at that time one of the few treatment centers still using the diet. Charlie was among the minority of children for whom the ketogenic diet could achieve complete and permanent remission, even after discontinuing the treatment [22], after only a few years using the diet [11]. When Charlie became seizure-free, Jim Abrahams started the Charlie Foundation, distributing educational videotapes about the ketogenic diet to tens of thousands of doctors and directing a television drama called …First Do No Harm about a mother (Meryl Streep) who defies doctors’ orders and uses the ketogenic diet to treat her son with intractable epilepsy. Research interest in the ketogenic diet for epilepsy has since exploded [11]. A recent systematic review on modern clinical trials published by the Cochrane Collaboration concluded that the ketogenic diet resulted in clinically significant reduction in seizure activity in drug-resistant epilepsy and maintained that the diet was a valid treatment in such cases [29]. A recent review by an international consensus group noted that four randomized controlled trials have shown that KD is efficacious compared to placebo and has a response rate of 70% or better consistently reported in the literature for twelve epilepsy syndromes. The mechanisms of KD for epilepsy are thought to be “parallel, multiple, and potentially synergistic,” due to proposed changes in neurotransmitter synthesis, ion channel modulation, glucose utilization, mitochondrial bioenergetics, and more, depending on the pathobiology of the syndrome [27]

Since the resurgence of the ketogenic diet for epilepsy, several new permutations of the diet have been formulated for this purpose, perhaps most notably the modified Atkins diet (MAD), so named because while the original Atkins diet prescribes an increase in carbohydrate intake after the initial “induction period”, the modified Atkins diet does not. In contrast to the classic 4:1 ketogenic diet (90% fat), MAD maintains a much more liberal ratio of 1:1 fat-to-carbohydrate-and-protein, or 65% of calories from fat, 25% from protein, and 10% from carbohydrate. This makes it more palatable and less restrictive than the stricter, classic ketogenic diet [30]. Despite this liberalization, in clinical trials MAD showed similar efficacy to the classical ketogenic diet [30–32]. Another alternative to the classic ketogenic diet, the low glycemic index treatment (LGIT), was developed at Harvard at around the same time as MAD [33]. Each of these diets shows similar efficacy to the classical ketogenic diet and are used at discretion of the clinician and families [27], though a minority of patients may experience better seizure control on more highly restricted forms of the ketogenic diet, such as the classical 3:1 or 4:1 ketogenic diet [34].

Parallel developments in carbohydrate restriction

While all this was happening with the ketogenic diet for epilepsy, carbohydrate restriction and ketogenic dietary therapies were developing elsewhere.

  1. Mackarness’s [35], Atkins’s [36], and Taubes’s [2] popular works and David Ludwig’s research [37] on the use of carbohydrate restriction for obesity;
  2. Reaven’s [38] and Krauss’s [39] scientific work establishing carbohydrate restriction as a viable treatment for metabolic syndrome and atherogenic dyslipidemia;
  3. Mary and Steve Newport’s experiences with medium-chain triglycerides and ketone esters for Alzheimer’s disease [40];
  4. Stephen Phinney and Jeff Volek’s work on the ketogenic diet for exercise performance [41];
  5. Clinical data and single-arm trials demonstrating the efficacy of the ketogenic diet for type 2 diabetes [6,42];
  6. Reports of a very-low-carbohydrate/ketogenic diet for maintaining glycemic control in the normal range with fewer complications and insulin use in type 1 diabetes [43];
  7. Development of a number of exogenous ketones for inducing ketosis without carbohydrate restriction.

Ketogenic diet timeline

~400 BC – Hippocrates mentions fasting as curative of epilepsy in a case study [18]
~260 BC – Erasistratus suggests that fasting should be an important part of the treatment of epilepsy [17]
1797 – John Rollo publishes case studies using a low-carbohydrate diet for treatment of patients with diabetes mellitus [1]
1825 – Brillat-Savarin publishes The Physiology of Taste, part of which advocates a carbohydrate-restricted diet for the treatment of obesity [7]
1863 – Banting publishes Letter on Corpulence, Addressed to the Public, which advocates a diet of meat and vegetables and dietary restriction of starches and beer for the treatment of obesity [44]
1909 – Moreschi shows that tumors implanted into rats did not grow as quickly when calories are restricted [45]
1900-1930 – Explorers document the high-fat, low-carbohydrate diet of the Inuit in Alaska and Canada
1911 – Guelpa & Marie report that pediatric epilepsy can be treated by fasting [21]
1921 – Geyelin presents Conklin’s work at the 1921 American Medical Association conference, which makes fasting for epilepsy widely known to the medical community [11]
1921 – Wilder coins the term “ketogenic diet” for epilepsy and explains its efficacy [24]
1924 – Peterman defines the macronutrient requirement ketogenic diet as 1 grams protein daily/kilogram bodyweight, 10-15 grams carbohydrate daily, and the remainder of the diet as fat [25]
1927 – Talbott publishes the classic 4:1 and 3:1 ketogenic diet formulations [26]
1930 – Steffanson and Andersen consume an all-meat diet for a full year under the supervision of physicians at Bellevue Hospital in New York [46,47]
1938 – Merritt and Putnam discover diphenylhydantoin, ushering in the era of epilepsy pharmacotherapy, causing KD for epilepsy to wane [11]
1972 – Atkins publishes Dr. Atkins Diet Revolution, starting the modern low-carb/ketogenic diet for weight loss era [36]
1994 – Jim Abrahams tells Charlie’s story on NBC-TV’s Dateline, causing a resurgence in interest in KD for epilepsy [11]
2003 – Seyfried publishes mouse xenograft study showing that ketogenic diet impairs growth of glioblastoma multiforme
2005 – 73 academic centers in 41 countries have a ketogenic diet center for the treatment of epilepsy [48]
2007 – Taubes publishes Good Calories, Bad Calories, a sweeping historical and scientific popular work that provides a critical reference point for his followers, who have been highly successful in popularizing the ketogenic diet [2]
2011 – Phinney and Volek publish The Art and Science of Low Carbohydrate Living, a book summarizing the research on the ketogenic diet for practical application by laypeople to treat obesity and metabolic syndrome disease [49]
2012 – Seyfried publishes Cancer as a Metabolic Disease, a controversial book that defends Warburg’s hypothesis that the etiology of cancer was metabolic and mitochondrial; advocates for the ketogenic diet [50]
2012-2019 – Several investigators publish studies showing that a ketogenic diet produces superior outcomes for type 2 diabetes compared to standard of care [6,42,51–54]

Ketogenic and carbohydrate-restricted diets used today

Diet name F:C&P* Fat kcal % Carb + protein kcal % References
Classic 4:1 4:1 90% 10% [55–58]
Classic 3:1 3:1 87% 13%
MCT oil diet 1.9:1 50%/21%** 29%
Low glycemic index treatment 1:1 60% 30%
Modified Atkins ciet 0.8:1 65% 35%
Low-carbohydrate diet N/A N/A 50-150 g/d carbs
Very-low-carbohydrate ketogenic diet N/A N/A <20-50 g/d carbs
* F:C&P is the ratio of fat to protein and carbohydrate. This is calculated in grams, not kilocalories.
** Percent of calories from medium chain triglycerides / long chain triglycerides (MCT/LCT)
Ketogenic diets used in clinical practice may deviate substantially from the above.
  1. Allen, F.M. Total dietary regulation in the treatment of diabetes; The Rockefeller Institute for Medical Research,: New York :, 1919;
  2. Taubes, G. Good calories, bad calories : fats, carbs, and the controversial science of diet and health; Anchor Books, 2008; ISBN 1400033462.
  3. William Osler The principles and practice of medicine : designed for the use of practitioners and students of medicine; 1st ed.; 1892;
  4. Newburgh, L.H.; Marsh, P.L. The use of a high fat diet in the treatment of diabetes mellitus. Arch. Intern. Med. 1920, 26, 647.
  5. Lennerz, B.S.; Barton, A.; Bernstein, R.K.; Dikeman, R.D.; Diulus, C.; Hallberg, S.; Rhodes, E.T.; Ebbeling, C.B.; Westman, E.C.; Yancy, W.S.; et al. Management of Type 1 Diabetes With a Very Low-Carbohydrate Diet. Pediatrics 2018, 141, e20173349.
  6. Hallberg, S.J.; McKenzie, A.L.; Williams, P.T.; Bhanpuri, N.H.; Peters, A.L.; Campbell, W.W.; Hazbun, T.L.; Volk, B.M.; McCarter, J.P.; Phinney, S.D.; et al. Effectiveness and Safety of a Novel Care Model for the Management of Type 2 Diabetes at 1 Year: An Open-Label, Non-Randomized, Controlled Study. Diabetes Ther. 2018, 9, 583–612.
  7. Brillat-Savarin, 1755-1826.; Fisher, M.F.K. (Mary F.K. The physiology of taste, or, Meditations on transcendental gastronomy; Vintage Books, 2011; ISBN 9780307390370.
  8. Yorke-Davies, N.E. Foods for the fat; a treatise on corpulency and a dietary for its cure 1889, 138 p.
  9. Levine, D.I. Corpulence and correspondence: President William H. Taft and the medical management of obesity. Ann. Intern. Med. 2013, 159, 565–570.
  10. Cardenas, D. Let not thy food be confused with thy medicine: The Hippocratic misquotation. ESPEN. J. 2013, 8.
  11. Wheless, J.W. History of the ketogenic diet. Epilepsia 2008, 49, 3–5.
  12. Hartman, A.L.; Rubenstein, J.E.; Kossoff, E.H. Intermittent fasting: A “new” historical strategy for controlling seizures? Epilepsy Res. 2013, 104, 275–279.
  13. Bailey, E.E.; Pfeifer, H.H.; Thiele, E.A. The use of diet in the treatment of epilepsy. Epilepsy Behav. 2005, 6, 4–8.
  14. Magiorkinis, E.; Diamantis, A.; Sidiropoulou, K.; Panteliadis, C. Highights in the History of Epilepsy: The Last 200 Years. Epilepsy Res. Treat. 2014, 2014, 1–13.
  15. D’Andrea Meira, I.; Romão, T.T.; Do Prado, H.J.P.; Krüger, L.T.; Pires, M.E.P.; Da Conceição, P.O. Ketogenic diet and epilepsy: What we know so far. Front. Neurosci. 2019, 13.
  16. Taylor, M.K.; Swerdlow, R.H.; Sullivan, D.K. Dietary Neuroketotherapeutics for Alzheimer’s Disease: An Evidence Update and the Potential Role for Diet Quality. Nutrients 2019, 11, 1910.
  17. Temkin, O. The falling sickness : a history of epilepsy from the Greeks to the beginnings of modern neurology; ISBN 9780801848490.
  18. Hippocrates No Title. In Hippocrates, Vol 7; London: Heinemann; New York: Putnam: Paris; Toronto, 1923.
  19. Lennox, W. Epilepsy and related disorders by William Gordon Lennox, with the collaboration of Margaret A. Lennox.; Little  Brown ; Boston, 1960;
  20. Eadie, M.J.; Bladin, P.F. A disease once sacred : a history of the medical understanding of epilepsy; John Libbey, 2001; ISBN 9780861966073.
  21. Marie, A.; Guelpa, G. La lutte contre l’epilepsie par la desintoxication et par la reeducation alimentaire. Rev. Ther. Medico-Chirurgicale 1911, 78, 8–13.
  22. Martinez, C.C.; Pyzik, P.L.; Kossoff, E.H. Discontinuing the ketogenic diet in seizure-free children: recurrence and risk factors. Epilepsia 2007, 48, 187–90.
  23. Wheless, J.W. History and Origin of the Ketogenic Diet. In Epilepsy and the Ketogenic Diet; Humana Press, 2004; pp. 31–50.
  24. Wilder, R.M. High fat diets in epilepsy. Mayo Clin Bull 1921, 2, 308.
  26. TALBOT, F.B.; METCALF, K.M.; MORIARTY, M.E. A Clinical Study of Epileptic Children Treated by Ketogenic Diet. Bost. Med. Surg. J. 1927, 196, 89–96.
  27. Kossoff, E.H.; Zupec-Kania, B.A.; Auvin, S.; Ballaban-Gil, K.R.; Christina Bergqvist, A.G.; Blackford, R.; Buchhalter, J.R.; Caraballo, R.H.; Cross, J.H.; Dahlin, M.G.; et al. Optimal clinical management of children receiving dietary therapies for epilepsy: Updated recommendations of the International Ketogenic Diet Study Group. Epilepsia Open 2018, 3, 175–192.
  28. Huttenlocher, P.R.; Wilbourn, A.J.; Signore, J.M. Medium-chain triglycerides as a therapy for intractable childhood epilepsy. Neurology 1971, 21, 1097–103.
  29. Levy, R.G.; Cooper, P.N.; Jackson, C.F.; Martin-McGill, K.J.; Bresnahan, R. Ketogenic diets for drug-resistant epilepsy. Cochrane Database Syst. Rev. 2018.
  30. Sharma, S.; Jain, P. The modified atkins diet in refractory epilepsy. Epilepsy Res. Treat. 2014, 2014, 404202.
  31. Kossoff, E.H.; McGrogan, J.R.; Bluml, R.M.; Pillas, D.J.; Rubenstein, J.E.; Vining, E.P. A Modified Atkins Diet Is Effective for the Treatment of Intractable Pediatric Epilepsy. Epilepsia 2006, 47, 421–424.
  32. Kossoff, E.H.; Krauss, G.L.; McGrogan, J.R.; Freeman, J.M. Efficacy of the Atkins diet as therapy for intractable epilepsy. Neurology 2003, 61, 1789–1791.
  33. Pfeifer, H.H.; Thiele, E.A. Low-glycemic-index treatment: A liberalized ketogenic diet for treatment of intractable epilepsy. Neurology 2005, 65, 1810–1812.
  34. Wirrell, E.C. Ketogenic ratio, calories, and fluids: Do they matter? In Proceedings of the Epilepsia; 2008; Vol. 49, pp. 17–19.
  35. Mackarness, R. Eat Fat and Grow Slim: Or, Banting Up to Date; The Harvill Press, 1958;
  36. Atkins, R.C. Dr. Atkins’ diet revolution : the high calorie way to stay thin forever; Bantam Books, 1973; ISBN 0553271571.
  37. Ebbeling, C.B.; Leidig, M.M.; Feldman, H.A.; Lovesky, M.M.; Ludwig, D.S. Effects of a low-glycemic load vs low-fat diet in obese young adults: A randomized trial. J. Am. Med. Assoc. 2007, 297, 2092–2102.
  38. Jeppesen, J.; Schaaf, P.; Jones, C.; Zhou, M.Y.; Ida Chen, Y.D.; Reaven, G.M. Effects of low-fat, high-carbohydrate diets on risk factors for ischemic heart disease in postmenopausal women. Am. J. Clin. Nutr. 1997, 65, 1027–1033.
  39. Krauss, R.M.; Blanche, P.J.; Rawlings, R.S.; Fernstrom, H.S.; Williams, P.T. Separate effects of reduced carbohydrate intake and weight loss on. Am. J. Clin. Nutr. 2006, 83, 1025–1031.
  40. Newport, M.T.; Vanitallie, T.B.; Kashiwaya, Y.; King, M.T.; Veech, R.L. A new way to produce hyperketonemia: Use of ketone ester in a case of Alzheimer’s disease. Alzheimer’s Dement. 2015, 11, 99–103.
  41. Volek, J.S.; Noakes, T.; Phinney, S.D. Rethinking fat as a fuel for endurance exercise. Eur. J. Sport Sci. 2015, 15, 13–20.
  42. Athinarayanan, S.J.; Adams, R.N.; Hallberg, S.J.; McKenzie, A.L.; Bhanpuri, N.H.; Campbell, W.W.; Volek, J.S.; Phinney, S.D.; McCarter, J.P. Long-Term Effects of a Novel Continuous Remote Care Intervention Including Nutritional Ketosis for the Management of Type 2 Diabetes: A 2-Year Non-randomized Clinical Trial. Front. Endocrinol. (Lausanne). 2019, 10, 348.
  43. Turton, J.L.; Raab, R.; Rooney, K.B. Low-carbohydrate diets for type 1 diabetes mellitus: A systematic review. PLoS One 2018, 13, e0194987.
  44. Banting, W. Letter on corpulence, addressed to the public. 1869. Obes. Res. 1993, 1, 153–63.
  45. Kritchevsky, D. Caloric restriction and cancer. J. Nutr. Sci. Vitaminol. (Tokyo). 2001, 47, 13–9.
  46. McClellan, W.; Du BOIS, E.F. Clinical calorimetry XLV: Prolonged meat diets with a study of kidney function and ketosis; 1930;
  47. McClellan, W.S.; Du Bois, E.F. Clinical Calorimetry: XLVI. Prolonged Meat Diets with a Study of the Metabolism of Nitrogen, Calcium and Phosphorus. J. Biol. Chem. 1930, 87, 651–778.
  48. Kossoff, E.H.; McGrogan, J.R. Worldwide use of the ketogenic diet. Epilepsia 2005, 46, 280–9.
  49. Volek, J.; Phinney, S.D.; Kossoff, E.; Eberstein, J.A.; Moore, J. The art and science of low carbohydrate living : an expert guide to making the life-saving benefits of carbohydrate restriction sustainable and enjoyable; Beyond Obesity, 2011; ISBN 9780983490708.
  50. Seyfried, T.N. Cancer as a metabolic disease : on the origin, management, and prevention of cancer; John Wiley & Sons, 2012; ISBN 9780470584927.
  51. Saslow, L.R.; Kim, S.; Daubenmier, J.J.; Moskowitz, J.T.; Phinney, S.D.; Goldman, V.; Murphy, E.J.; Cox, R.M.; Moran, P.; Hecht, F.M. A randomized pilot trial of a moderate carbohydrate diet compared to a very low carbohydrate diet in overweight or obese individuals with type 2 diabetes mellitus or prediabetes. PLoS One 2014, 9, e91027.
  52. Saslow, L.R.; Mason, A.E.; Kim, S.; Goldman, V.; Ploutz-Snyder, R.; Bayandorian, H.; Daubenmier, J.; Hecht, F.M.; Moskowitz, J.T. An Online Intervention Comparing a Very Low-Carbohydrate Ketogenic Diet and Lifestyle Recommendations Versus a Plate Method Diet in Overweight Individuals With Type 2 Diabetes: A Randomized Controlled Trial. J. Med. Internet Res. 2017, 19, e36.
  53. Saslow, L.R.; Daubenmier, J.J.; Moskowitz, J.T.; Kim, S.; Murphy, E.J.; Phinney, S.D.; Ploutz-Snyder, R.; Goldman, V.; Cox, R.M.; Mason, A.E.; et al. Twelve-month outcomes of a randomized trial of a moderate-carbohydrate versus very low-carbohydrate diet in overweight adults with type 2 diabetes mellitus or prediabetes. Nutr. Diabetes 2017, 7, 304.
  54. Saslow, L.R.; Summers, C.; Aikens, J.E.; Unwin, D.J. Outcomes of a digitally delivered low-carbohydrate type 2 diabetes self-management program: 1-year results of a single-arm longitudinal study. J. Med. Internet Res. 2018, 20.
  55. What is a Ketogenic Diet? Keto Diet Facts, Research, and Variations Available online: (accessed on Oct 2, 2019).
  56. Schwartz, K.A.; Noel, M.; Nikolai, M.; Chang, H.T. Investigating the Ketogenic Diet As Treatment for Primary Aggressive Brain Cancer: Challenges and Lessons Learned. Front. Nutr. 2018, 5.
  57. Pfeifer, H.H.; Lyczkowski, D.A.; Thiele, E.A. Low glycemic index treatment: Implementation and new insights into efficacy. In Proceedings of the Epilepsia; 2008; Vol. 49, pp. 42–45.
  58. Westman, E.C.; Feinman, R.D.; Mavropoulos, J.C.; Vernon, M.C.; Volek, J.S.; Wortman, J.A.; Yancy, W.S.; Phinney, S.D. Low-carbohydrate nutrition and metabolism. Am. J. Clin. Nutr. 2007, 86, 276–284.


Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 10 minutes

The following is an early, edited draft of a subsection of an upcoming scientific review.

Definition of ketosis and composition of the ketogenic diet

Ketones are always being produced in the liver, released into the blood, and utilized by the peripheral tissues, even in the fed state [1,2]. Transgenic mice lacking SCOT, a gene critical for utilizing ketones, die within two days from ketoacidosis [3]. A calorie-restricted diet, even one that does not specifically restrict carbohydrate but reduces all macronutrient calories equally, will produce higher blood ketones and higher ketone utilization than a non-restricted diet [4,5]. Enhanced ketogenesis and ketosis also occurs after an overnight fast and after vigorous exercise [6]. Thus, in an important sense, ketosis and ketonemia is the constitutive state of most living organisms. Ketosis as a metabolic state is thus not dichotomous but a sliding scale.

There are three ketone bodies. These ketone bodies define ketosis. They are acetoacetate (AcAc), beta-hydroxybutyrate (BHB), and acetone. In the normal individual, levels of BHB and AcAc are below 0.1 mM, and acetone is undetectable. In human starvation, levels of BHB can reach nearly 6 mM, AcAc above 1 mM, with acetone concentrations similar to AcAc [7].

Picture1 2

Acetone has conventionally been thought of as a “minor” ketone body of relatively little biological importance, the product of spontaneous, non-enzymatic decarboxylation of AcAc in the blood [8]. Recent studies have suggested a minor role for acetone in the biosynthesis of acetate and therefore of cholesterol and fatty acids [9], and perhaps even a role in the anti-epileptic effects of the ketogenic diet [10]. Most research however has focused on the biological effects of BHB and AcAc.

Because ketone production and metabolism are often the means to a particular therapeutic endgoal, ketosis is often considered for practical purposes as having a “cutoff”. This cutoff has been defined by important figures in the field as around 0.5 mM [11,12], and this definition is generally accepted. According to this framework, nutritional ketosis is defined as between 0.5 mM to 3 mM and fasting ketosis as BHB between 5 and 10 mM, each of which are far below the 15-25 mM seen in diabetic ketoacidosis [12], which requires insufficient insulin signaling such as in type 1 diabetes to develop. Except in rare cases [13,14], in healthy people, nutritional ketosis does not lead to and has none of the harmful effects of ketoacidosis [12].

It is important to note that the above definition of nutritional ketosis is largely based on a carbohydrate-restricted diet similar to that advocated by Atkins. Substantially higher values than 3 mM are reported among those consuming MCT oil diets and classic ketogenic diets [15]; anecdote and some evidence suggests that ketogenic diets higher in unsaturated fatty acids might also produce higher blood ketones than on the typical animal-rich ketogenic dietary pattern [16–18].

Ketosis is increased and nutritional ketosis induced by carbohydrate, protein, and fat restriction (in that order of importance). The more one restricts carbohydrate and protein, the higher the expected blood ketone levels, mimicking the fasted state. The 3:1 and 4:1 classical ketogenic diets, so-called because they consist of a three-to-one and four-to-one ratio of fat grams to carbohydrate and protein grams, respectively, produce the highest degree of ketosis, while the modified Atkins diet and low-glycemic index treatment produces among the lowest.

In the diabetes and obesity fields and in the popular and scientific press more generally, ketogenic diets are defined somewhat differently than they are in the epilepsy field specifically (though terminology frequently crosses over). This is because specific blood ketone targets are often less the goal than carbohydrate restriction is. A review article written by many of the most influential contemporary scientific proponents of a carbohydrate-restricted diet for the treatment of obesity, diabetes, and other chronic diseases defines a low-carbohydrate diet as being in the range of 50-150 grams per day, and a very-low-carbohydrate ketogenic diet as in the range of <20-50 grams per day [19].

Major ketogenic diet types and their compositions are listed below.

Diet name F:C&P* Fat kcal % Carb + protein kcal % References
Classic 4:1 4:1 90% 10% [19–22]
Classic 3:1 3:1 87% 13%
MCT oil diet 1.9:1 50%/21%** 29%
Low glycemic index treatment 1:1 60% 30%
Modified Atkins diet 0.8:1 65% 35%
Low-carbohydrate diet N/A N/A 50-150 g/d carbs
Very-low-carbohydrate ketogenic diet N/A N/A <20-50 g/d carbs
* F:C&P is the ratio of fat to protein and carbohydrate. This is calculated in grams, not kilocalories.
** Percent of calories from medium chain triglycerides / long chain triglycerides (MCT/LCT)
Ketogenic diets used in clinical practice may deviate substantially from the above.

Why ketosis?

In most prokaryotes, ketone polymers in the form of complexed polyhydroxybutyrate (cPHB) are an important energy storage molecule and ketone bodies an important source of energy. This energy system may be as old as 2-3 billion years and may have evolved as an important energy substrate under conditions of oxygen deprivation, due to the higher ratio of ATP produced to oxygen consumed characteristic of ketone bodies [23]. While low levels of circulating cPHB are found in human serum, in cells (reportedly in granules), as essential non-protein components of ion channels, and may have a multitude of other important biological effects that have yet to be fully characterized [23,24], this energy storage form is not thought to be important source of energy in normal mammalian physiology. Conceivably, this is because monomeric BHB can be readily generated from fatty acids, which in mammals are stored in triglycerides; in contrast, prokaryotes do not have generate or use triglyceride molecules [23].

Although polymeric BHB is not an important energy storage form of BHB, BHB itself is readily produced and oxidized by mammals in periods of starvation as an alternative to glucose. Ketone bodies are an alternative to glucose as an energy source in mammals. Classic experiments by George Cahill and others in the 1960s and 70s demonstrated the role of ketone bodies and especially beta-hydroxybutyrate in the maintenance of cognitive function during starvation. In one study of three obese subjects fasted for 5-6 weeks each, arteriovenous metabolite concentration differences showed that a mean of 65% of energy substrate used by the brain directly derived from beta-hydroxybutyrate and acetoacetate, 55% and 10% respectively [25]. Indeed, evidence in rodents suggests that as ketosis proceeds, the proportion of energy derivable from ketone bodies also increases due to increases of the blood-brain barrier permeability to ketones [26,27]. After Cahill’s initial studies, another pair of studies conducted by Cahill and a group at UCLA almost concurrently showed that subjects subjected to long-term starvation and then injected with insulin to produce profound hypoglycemia showed no symptoms and intact cognition [28,29]. The subjects reached blood glucose readings as 0.5 mM, i.e. 6-fold lower than the 3.0 mM that in normal circumstances produces symptoms and is regarded as dangerous. These levels are typically regarded as fatal, and these studies were landmark demonstrations of BHB as an alternative fuel source in the brain.

Why is the brain in particular so effective at using ketone bodies?

Due to their ancient evolutionary pedigree, before the split between archaea and prokaryotes [23], production and metabolism of ketones are ubiquitous among animals, including fish [30] and insects [31]. Yet humans produce more ketone bodies in response to starvation than any other known species, and this is likely because the human brain accounts for a larger proportion of the body’s energy expenditure compared to the brains of other species [23]. The brain does not metabolize such fatty acids to any appreciable degree due to an oxidative defect in neurons and astrocytes, probably due to the high energetic demands of the brain and vulnerability to oxidative stress [32]. The brain is therefore heavily reliant on glucose. If this reliance on glucose persisted during the course of starvation, i.e. in the absence of exogenous glucose in the form of dietary carbohydrate, the body would need to provide for the brain’s glucose needs by breaking down the body’s protein (mainly skeletal muscle) and converting it to glucose.

This would quickly lead to muscle loss and rapid death. Since the brain requires 100-145g of glucose per day [33], and the theoretical maximum rate conversion of protein to glucose is 60% [34], and the body’s protein stores are approximately 6,000g but only about 3,000 can be mobilized before death, and the body would need to catabolize 170-240g of protein per day to sustain the energy needs of the brain alone, it follows a starving human would survive for a maximum of 18 days and a minimum of 12, depending on how much energy their brain required [1]. Furthermore, a human surviving just a short period of starvation, e.g. 1 week, would be rendered frail due to skeletal muscle loss, impairing physical function and probably odds of future survival.

Ketones, produced when exogenous glucose and thus insulin is sufficiently low, fill the gap, allowing mammals to survive for extended periods of time while fueling the brain and minimizing the impact to the body’s own protein and thus to skeletal muscle. Early experiments in starved subjects showed clearly that glucose infusions were sufficient to drastically reduce both nitrogen excretion and blood ketone levels, and that each of these increased in tandem on the day that the glucose infusions were stopped, after which nitrogen excretion then declined, a classic sign of compensatory response [23]. Correspondingly, upon BHB infusion, endogenous glucose production by the liver is decreased [35], tissue uptake of glucose is impaired in many tissue types [35,36], and blood levels of important gluconeogenic amino acids decline [36,37]. Extended fasting periods far exceed anything possible if the brain consumed strictly glucose; this is shown by the famous case report of A.B., a 27-year-old male who survived a 382-day therapeutic fast [38]. The opposite is also true: individuals homozygous for the CPT1A P479L variant show impairment in hepatic fatty acid oxidation are hypoketotic and hypoglycemic in response to carbohydrate restriction and cannot tolerate fasting [39]. Ketonemia, or high levels of blood ketones, is therefore critical to muscle maintenance during fasting or during periods of low-carbohydrate intake (such as occurs on the ketogenic diet).

  1. Owen, O.E. Ketone bodies as a fuel for the brain during starvation. Biochem. Mol. Biol. Educ. 2005, 33, 246–251.
  2. WERK, E.E.; McPHERSON, H.T.; HAMRICK, L.W.; MYERS, J.D.; ENGEL, F.L. Studies on ketone metabolism in man. I. A method for the quantitative estimation of splanchnic ketone production. J. Clin. Invest. 1955, 34, 1256–1267.
  3. Cotter, D.G.; d’Avignon, D.A.; Wentz, A.E.; Weber, M.L.; Crawford, P.A. Obligate role for ketone body oxidation in neonatal metabolic homeostasis. J. Biol. Chem. 2011, 286, 6902–10.
  4. Meidenbauer, J.J.; Ta, N.; Seyfried, T.N. Influence of a ketogenic diet, fish-oil, and calorie restriction on plasma metabolites and lipids in C57BL/6J mice. Nutr. Metab. 2014, 11.
  5. Knott, C.D. Changes in orangutan caloric intake, energy balance, and ketones in response to fluctuating fruit availability. Int. J. Primatol. 1998, 19, 1061–1079.
  6. Féry, F.; Balasse, E.O. Ketone body turnover during and after exercise in overnight-fasted and starved humans. Am. J. Physiol. 1983, 245, E318-25.
  7. Cahill, G.F.; Veech, R.L. Ketoacids? Good medicine? Trans. Am. Clin. Climatol. Assoc. 2003, 114, 149–61; discussion 162-3.
  8. Puchalska, P.; Crawford, P.A. Multi-dimensional Roles of Ketone Bodies in Fuel Metabolism, Signaling, and Therapeutics. Cell Metab. 2017, 25, 262–284.
  9. Gavino, V.C.; Somma, J.; Philbert, L.; David, F.; Garneau, M.; Bélair, J.; Brunengraber, H. Production of acetone and conversion of acetone to acetate in the perfused rat liver. J. Biol. Chem. 1987, 262, 6735–40.
  10. McNally, M.A.; Hartman, A.L. Ketone bodies in epilepsy. J. Neurochem. 2012, 121, 28–35.
  11. Miller, V.J.; Villamena, F.A.; Volek, J.S. Nutritional Ketosis and Mitohormesis: Potential Implications for Mitochondrial Function and Human Health. J. Nutr. Metab. 2018, 2018, 1–27.
  12. Volek, J.; Phinney, S.D.; Kossoff, E.; Eberstein, J.A.; Moore, J. The art and science of low carbohydrate living : an expert guide to making the life-saving benefits of carbohydrate restriction sustainable and enjoyable; Beyond Obesity, 2011; ISBN 9780983490708.
  13. Sloan, G.; Ali, A.; Webster, J. A rare cause of metabolic acidosis: ketoacidosis in a non-diabetic lactating woman. Endocrinol. diabetes Metab. case reports 2017, 2017.
  14. Wei, K.Y.; Chang, S.Y.; Wang, S.H.; Su, H.Y.; Tsai, C.L. Short-term starvation with a near-fatal asthma attack induced ketoacidosis in a nondiabetic pregnant woman. Med. (United States) 2016, 95.
  15. van Delft, R.; Lambrechts, D.; Verschuure, P.; Hulsman, J.; Majoie, M. Blood beta-hydroxybutyrate correlates better with seizure reduction due to ketogenic diet than do ketones in the urine. Seizure 2010, 19, 36–39.
  16. Beynen, A.C.; Katan, M.B. Why do polyunsaturated fatty acids lower serum cholesterol? Am. J. Clin. Nutr. 1985, 42, 560–563.
  17. Lundsgaard, A.M.; Holm, J.B.; Sjøberg, K.A.; Bojsen-Møller, K.N.; Myrmel, L.S.; Fjære, E.; Jensen, B.A.H.; Nicolaisen, T.S.; Hingst, J.R.; Hansen, S.L.; et al. Mechanisms Preserving Insulin Action during High Dietary Fat Intake. Cell Metab. 2019, 29, 50-63.e4.
  18. Fuehrlein, B.S.; Rutenberg, M.S.; Silver, J.N.; Warren, M.W.; Theriaque, D.W.; Duncan, G.E.; Stacpoole, P.W.; Brantly, M.L. Differential metabolic effects of saturated versus polyunsaturated fats in ketogenic diets. J. Clin. Endocrinol. Metab. 2004, 89, 1641–5.
  19. Westman, E.C.; Feinman, R.D.; Mavropoulos, J.C.; Vernon, M.C.; Volek, J.S.; Wortman, J.A.; Yancy, W.S.; Phinney, S.D. Low-carbohydrate nutrition and metabolism. Am. J. Clin. Nutr. 2007, 86, 276–284.
  20. What is a Ketogenic Diet? Keto Diet Facts, Research, and Variations Available online: (accessed on Oct 2, 2019).
  21. Schwartz, K.A.; Noel, M.; Nikolai, M.; Chang, H.T. Investigating the Ketogenic Diet As Treatment for Primary Aggressive Brain Cancer: Challenges and Lessons Learned. Front. Nutr. 2018, 5.
  22. Pfeifer, H.H.; Lyczkowski, D.A.; Thiele, E.A. Low glycemic index treatment: Implementation and new insights into efficacy. In Proceedings of the Epilepsia; 2008; Vol. 49, pp. 42–45.
  23. Cahill, G.F. Fuel Metabolism in Starvation. Annu. Rev. Nutr. 2006, 26, 1–22.
  24. Dedkova, E.N.; Blatter, L.A. Role of β-hydroxybutyrate, its polymer poly-β-hydroxybutyrate and inorganic polyphosphate in mammalian health and disease. Front. Physiol. 2014, 5 JUL, 1–22.
  25. Owen, O.E.; Morgan, A.P.; Kemp, H.G.; Sullivan, J.M.; Herrera, M.G.; Cahill, G.F.; Jr. Brain metabolism during fasting. J. Clin. Invest. 1967, 46, 1589–95.
  26. Puchowicz, M.A.; Xu, K.; Sun, X.; Ivy, A.; Emancipator, D.; LaManna, J.C. Diet-induced ketosis increases capillary density without altered blood flow in rat brain. Am. J. Physiol. Metab. 2007, 292, E1607–E1615.
  27. Gjedde, A.; Crone, C. Induction processes in blood-brain transfer of ketone bodies during starvation. Am. J. Physiol. Content 1975, 229, 1165–1169.
  28. Passonneau, J. V. Cerebral metabolism and neural function; Williams & Wilkins, 1980; ISBN 9780683067880.
  29. Drenick, E.J.; Alvarez, L.C.; Tamasi, G.C.; Brickman, A.S. Resistance to Symptomatic Insulin Reactions after Fasting. J. Clin. Invest. 1972, 51, 2757–2762.
  30. Zammit, V.A.; Newsholme, E.A. Activities of enzymes of fat and ketone-body metabolism and effects of starvation on blood concentrations of glucose and fat fuels in teleost and elasmobranch fish. Biochem. J. 1979, 184, 313–322.
  31. Kunieda, T.; Fujiyuki, T.; Kucharski, R.; Foret, S.; Ament, S.A.; Toth, A.L.; Ohashi, K.; Takeuchi, H.; Kamikouchi, A.; Kage, E.; et al. Carbohydrate metabolism genes and pathways in insects: Insights from the honey bee genome. Insect Mol. Biol. 2006, 15, 563–576.
  32. Schönfeld, P.; Reiser, G. Why does brain metabolism not favor burning of fatty acids to provide energy? Reflections on disadvantages of the use of free fatty acids as fuel for brain. J. Cereb. Blood Flow Metab. 2013, 33, 1493–9.
  33. Cahill, G.; Owen, O. Starvation and survival. Trans. Am. Clin. Climatol. Assoc. 1968, 79, 13–20.
  34. Gannon, M.C.; Nuttall, F.Q. Amino acid ingestion and glucose metabolism-A review. IUBMB Life 2010, 62, 660–668.
  35. Mikkelsen, K.H.; Seifert, T.; Secher, N.H.; Grøndal, T.; Van Hall, G. Systemic, cerebral and skeletal muscle ketone body and energy metabolism during acute hyper-D-β-hydroxybutyratemia in post-absorptive healthy males. J. Clin. Endocrinol. Metab. 2015, 100, 636–643.
  36. Robinson, A.M.; Williamson, D.H. Physiological roles of ketone bodies as substrates and signals in mammalian tissues. Physiol. Rev. 1980, 60, 143–187.
  37. Felig, P.; Owen, O.E.; Wahren, J.; Cahill, G.F. Amino acid metabolism during prolonged starvation. J. Clin. Invest. 1969, 48, 584–94.
  38. Stewart, W.K.; Fleming, L.W. Features of a successful therapeutic fast of 382 days’ duration. Postgrad. Med. J. 1973, 49, 203–9.
  39. Gillingham, M.B.; Hirschfeld, M.; Lowe, S.; Matern, D.; Shoemaker, J.; Lambert, W.E.; Koeller, D.M. Impaired fasting tolerance among Alaska native children with a common carnitine palmitoyltransferase 1A sequence variant. Mol. Genet. Metab. 2011, 104, 261–4.


Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 4 minutes

Prior to the discovery of insulin, a very low-carbohydrate and probably ketogenic diet was used by some respected physicians to treat type 1 diabetes [1]. Many considerations may argue in favor of treating type 1 diabetes with a very low-carbohydrate diet (with the advantage of modest insulin dosing) today, compared with current standard of care. Among people with type 1 diabetes, current standard of care results in: a higher rate of obesity than the general population [2], an average of two symptomatic hypoglycemia episodes per week [3], nearly universal retinopathy within two decades of diagnosis [4], a much higher rate of cardiovascular mortality [5], and a higher incidence of cancer [6]. A higher rate of obesity is probably due to higher-than-required insulin to “cover” glucose [2]; hypoglycemia is caused by insulin “overshoot” [3]; and retinopathy, excess cardiovascular mortality, and probably cancer are due to chronic hyperglycemia [4–6]. Correspondingly, HbA1c, a measure of average blood glucose, was 8.3% on average in a recent study in 25,833 children and adults with type 1 diabetes, while the goal according to guidelines is 7.0% for adults and 7.5% for children, and the cutoff for diabetes is 6.5%. Thus, the large majority of people with type 1 diabetes have a blood sugar that is excessively high, and they do not meet official targets [7].

In response to the risks and difficulties in managing type 1 diabetes, very-low-carbohydrate diets (defined as <50 grams of carbohydrate per day) have seen a resurgence in recent years, spearheaded in part by the advocacy and sophisticated glycemic management strategies of Richard Bernstein, a physician and type 1 diabetes patient himself [8]; others within academia have also played an important role [9]. Very-low-carbohydrate diets have the potential to mitigate many or all of the risks discussed above. A recent systematic review, which included single-arm trials with up to 18 months of follow-up, have found several advantages to using a very-low-carbohydrate diet for type 1 diabetes, including improvements of HbA1c into the normal range (no other intervention has shown this effectiveness apart from islet cell transplants) and reduced insulin use [10]. Although no studies with multi-year follow-ups have been done, such improvements in glycemia in persons with type 1 diabetes suggest the potential for large benefits to cardiovascular risk and mortality, in line with the findings of DCCT/EDIC, which showed such benefits via glucose lowering by conventional strategies [11].

Moreover, if the hyperglycemia is the major risk driving increased cancer risk, a cancer risk reduction benefit is plausible. Additionally, the ketogenic diet may conceivably confer some protection from the symptoms of hypoglycemia, since ketones have been shown to provide the brain with an alternative source of energy substrate during periods of extreme hypoglycemia [12,13]. Likewise, limiting or eliminating hyperglycemia would likely prevent the development of retinopathy [4,11]. Finally, much animal work suggests that the beta-hydroxybutyrate produced by ketogenesis might be able to independently enhance the benefits of improved glycemia on retinopathy and other diabetic sequelae, via anti-inflammatory effects from the activation of GPR109A [14] and inhibition of NLRP3 [15]. However, although promising, many of these mechanisms remain speculative in humans and need to be tested in formal trials.

A recent, highly publicized study conducted using an online survey found exceptional management of hemoglobin A1c when subjects were placed on a very low carbohydrate diet, with an average HbA1c of 5.67%, 69% of respondents reporting five or fewer hypoglycemic events per month, an incidence of just 1% per year of ketoacidosis, and in general an extremely low incidence of severe diabetes-related adverse events at just 2% per year [16]. However, the patients surveyed represent a highly selected population that is probably more motivated and educated than average, and these findings need to be replicated in the general population in well-designed clinical trials. Importantly, an upcoming randomized clinical trial including children and adults and with a carefully controlled meal delivery design may help to illuminate the potential role of very low-carbohydrate diets in the treatment of type 1 diabetes in conventional medical practices (NCT03710928). Thus, the use of very-low-carbohydrate or ketogenic diets for type 1 diabetes shows promise and further investigation is underway.


1.        Newburgh, L.H.; Marsh, P.L. The use of a high fat diet in the treatment of diabetes mellitus. Arch. Intern. Med. 1920, 26, 647.

2.        Mottalib, A.; Kasetty, M.; Mar, J.Y.; Elseaidy, T.; Ashrafzadeh, S.; Hamdy, O. Weight Management in Patients with Type 1 Diabetes and Obesity. Curr. Diab. Rep. 2017, 17.

3.        McCrimmon, R.J.; Sherwin, R.S. Hypoglycemia in Type 1 Diabetes. Diabetes 2010, 59, 2333–2339.

4.        Fong, D.S.; Aiello, L.; Gardner, T.W.; King, G.L.; Blankenship, G.; Cavallerano, J.D.; Ferris, F.L.; Klein, R. Retinopathy in Diabetes. Diabetes Care 2004, 27, S84–S87.

5.        Lind, M.; Svensson, A.-M.; Kosiborod, M.; Gudbjörnsdottir, S.; Pivodic, A.; Wedel, H.; Dahlqvist, S.; Clements, M.; Rosengren, A. Glycemic Control and Excess Mortality in Type 1 Diabetes. N. Engl. J. Med. 2014, 371, 1972–1982.

6.        Carstensen, B.; Read, S.H.; Friis, S.; Sund, R.; Keskimäki, I.; Svensson, A.M.; Ljung, R.; Wild, S.H.; Kerssens, J.J.; Harding, J.L.; et al. Cancer incidence in persons with type 1 diabetes: a five-country study of 9,000 cancers in type 1 diabetic individuals. Diabetologia 2016, 59, 980–988.

7.        Beck, R.W.; Tamborlane, W. V.; Bergenstal, R.M.; Miller, K.M.; DuBose, S.N.; Hall, C.A. The T1D exchange clinic registry. J. Clin. Endocrinol. Metab. 2012, 97, 4383–4389.

8.        Bernstein, R.K. Dr. Bernstein’s diabetes solution : the complete guide to achieving normal blood sugars; Little, Brown and Co, 2011; ISBN 0316182699.

9.        Feinman, R.D.; Pogozelski, W.K.; Astrup, A.; Bernstein, R.K.; Fine, E.J.; Westman, E.C.; Accurso, A.; Frassetto, L.; Gower, B.A.; McFarlane, S.I.; et al. Dietary carbohydrate restriction as the first approach in diabetes management: Critical review and evidence base. Nutrition 2015, 31, 1–13.

10.      Turton, J.L.; Raab, R.; Rooney, K.B. Low-carbohydrate diets for type 1 diabetes mellitus: A systematic review. PLoS One 2018, 13, e0194987.

11.      Nathan, D.M. The diabetes control and complications trial/epidemiology of diabetes interventions and complications study at 30 years: Overview. Diabetes Care 2014, 37, 9–16.

12.      Passonneau, J. V. Cerebral metabolism and neural function; Williams & Wilkins, 1980; ISBN 9780683067880.

13.      Drenick, E.J.; Alvarez, L.C.; Tamasi, G.C.; Brickman, A.S. Resistance to Symptomatic Insulin Reactions after Fasting. J. Clin. Invest. 1972, 51, 2757–2762.

14.      Gambhir, D.; Ananth, S.; Veeranan-Karmegam, R.; Elangovan, S.; Hester, S.; Jennings, E.; Offermanns, S.; Nussbaum, J.J.; Smith, S.B.; Thangaraju, M.; et al. GPR109A as an anti-inflammatory receptor in retinal pigment epithelial cells and its relevance to diabetic retinopathy. Invest. Ophthalmol. Vis. Sci. 2012, 53, 2208–2217.

15.      Youm, Y.-H.; Nguyen, K.Y.; Grant, R.W.; Goldberg, E.L.; Bodogai, M.; Kim, D.; D’Agostino, D.; Planavsky, N.; Lupfer, C.; Kanneganti, T.D.; et al. The ketone metabolite β-hydroxybutyrate blocks NLRP3 inflammasome-mediated inflammatory disease. Nat. Med. 2015, 21, 263–9.

16.      Lennerz, B.S.; Barton, A.; Bernstein, R.K.; Dikeman, R.D.; Diulus, C.; Hallberg, S.; Rhodes, E.T.; Ebbeling, C.B.; Westman, E.C.; Yancy, W.S.; et al. Management of Type 1 Diabetes With a Very Low-Carbohydrate Diet. Pediatrics 2018, 141, e20173349.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 12 minutes

The following is an early draft of a section of a scientific review article in preparation.

Observational studies are largely consistent in showing an association between low-carbohydrate dietary patterns and increased risk of cancer, cardiovascular, and all-cause mortality [1], which would seem to recommend against low-carbohydrate diets for the purpose of health or cancer prevention. One well-known study from Japan seemingly contradicts this trend [2]. However, the cutoff of the lowest decile of carbohydrate intake percent in this study as a fraction of total macronutrients was 53.5%, while the mean intake in this decile was 51.5%, suggesting a very narrow range of carbohydrate intake well outside the low-carbohydrate or carbohydrate-restricted range [1]; in this cohort from 1980, moreover, the Atkin’s diet was reportedly virtually unknown [2]. These results are not unlike those found in the Prospective Urban and Rural Epidemiological (PURE) study, which too has been recently cited as providing evidence in favor of low-carbohydrate diets: while there was a dose-response relationship between carbohydrate intake and mortality risk in PURE, the carbohydrate intake of those with the lowest all-cause mortality was slightly lower than 55% [3], which is exactly the median carbohydrate intake of the Acceptable Macronutrient Distribution Range recommended by the 2015-2020 Dietary Guidelines for Americans [4]. This is also almost the exact same intake value found in the recently published Atherosclerosis Risk in Communities (ARIC) study and its associated meta-analysis of previous studies, including PURE [5]. A recent editorial written by PURE principal investigators looked at all eight major cohort studies examining this question and concluded: “Taking all the studies into account, the message of moderation is perhaps the most convincing one of all— diets that focus too heavily on a single macronutrient, whether extreme protein, carbohydrate, or fat intake, may adversely impact health” [6].


Lending confidence to the conclusions of these studies are the sophisticated study designs in these studies. In ARIC, for example, sensitivity analyses ruled out an impact of major chronic disease on diet (reverse causation); a dose-response relationship between low-carbohydrate diet quintile and mean soft drink consumption for that quintile suggested that this dietary pattern was intentional for a large proportion of individuals rather than incidental; and control for a variety of disease confounders further reduced the potential for reverse causality [5]. However, these findings are qualified by their observational nature and residual confounding cannot be ruled out.

Moreover, individually, diets high in fiber have been shown in meta-analyses to be consistently associated with lower all-cause and specific-cause mortality [7], as well as lower rates of colorectal cancer [8–11], breast [12], and ovarian cancer [13,14]. Diets higher in fiber, fruit, and vegetables during adolescence have each been associated with lower rates of breast cancer in later life [15–17]. Diets high in whole grains have been associated with lower risk of colorectal cancer [10], breast cancer [18], inflammatory markers [19], and insulin sensitivity, possibly via novel betainized compounds [20], while diets higher in nuts show an inverse association with total mortality [21], with a recent meta-analysis pointing toward an association between nut intake and lower cancer risk [22]. A recent meta-analysis of observational trials suggested that diets high in whole grains and cereal fibers was inversely associated with type 2 diabetes [23], established as a robust risk factor for cancer in 121 cohorts and 20 million people [24], while red meat and sugar-sweetened beverages were positively associated with risk [23]. Likewise, a recent and largest meta-analysis to date published on the topic shows a robust inverse relationship between fiber intake and cancer and all-cause mortality [25]. Meanwhile, a recent meta-analyses has suggested a relationship between saturated fat intake and breast cancer [26]; a recent cohort study in prostate cancer patients a link between saturated fat intake and cancer aggressiveness; and many studies, a link red and processed meat and cancer-specific and all-cause mortality, which is widely regarded as causal for processed meat [21,27,28]. Epidemiological associations between consumption of minimally processed plant foods and low disease risk, on the one hand, and between consumption of animal foods and high disease risk are consistent with short-term biomarker RCTs and animal studies and current widely accepted theories of disease (refs).

On the other hand, recent observational studies have established a link between simple carbohydrate intake and survival in head-and-neck cancer patients [29], between random blood glucose, HbA1c, and fasting blood glucose readings and survival in patients with solid tumors [30], between glycemic index and load and risk of mortality [31], between fiber intake and survival after cancer diagnosis [32], and between dietary insulin load and survival after chemotherapy for stage 3 colon cancer [32]. In a cohort study of 1011 stage III colon cancer patients, dietary carbohydrate and glycemic load was associated with decreased disease-free survival and cancer recurrence in obese and overweight but not lean patients [33]. A review of the current evidence concluded that the evidence for an effect of glycemic index and load on cancer is inconsistent, with findings from the largest meta-analyses that are either positive or null [34]. Likewise, in a recent systematic review of cohort studies, most studies were reported to have shown a null association between sugar intake and cancer, but some associations were suggested for added sugars and sugary beverages [35]. This suggests a modest or contextually dependent effect of glycemic load, glycemic index, and/or carbohydrate on cancer.

Despite the consistent relationship of low-carbohydrate diets on cancer mortality in prospective cohort studies, and consistent with the above findings, when investigators have defined two populations of low-carbohydrate eaters—those with high animal-based and those with high plant-based scores [2,36]—the animal-based low-carbohydrate diet was associated with higher total and cancer mortality, while plant-based low-carbohydrate diets decrease it. For instance, in a Nurses’ Health Study/Health Professionals Follow-up Study cohort of 1575 patients diagnosed with colorectal cancer, a low-carbohydrate, plant-rich diet was associated with a 30% reduced risk of all-cause mortality and a 63% reduced risk of cancer mortality, while a low-carbohydrate, animal-rich diet was associated with a higher or neutral risk of all-cause and cancer mortality [37], consistent with a study from the same group showing that persons consuming diets high in fiber also show a reduced all-cause and cancer mortality after colorectal cancer diagnosis [32]. This finding has been consistently observed in the general population for total and cancer mortality in multiple, diverse cohorts [5,38], with the exception of NIPPON DATA80, which found no difference in mortality risk between plant- and animal-based diets with a low-carbohydrate score [2]. Similar findings have been observed for cohorts analyzed along the lines of animal vs. plant protein intake, with plant protein consistently associated with neutral [39] or lower risk [40–43] and animal protein with higher risk of total and cancer mortality [41–44].

Despite these findings, it should be pointed out that none of the quantiles analyzed in the above studies were in the ketogenic range, with ARIC reporting the lowest carbohydrate intake of all, in one analysis in the lowest quantile a mean of 26.3%, far above that normally required for ketogenesis. This lack of epidemiological analysis of very low-carbohydrate diets may be due to long-term adherence to ketogenic diets being uncommon. It may be possible, therefore, that while animal-based low-carbohydrate diets are associated with higher mortality out of the ketogenic range due to interactions with other components of the diet, animal-based ketogenic diets might not show this disadvantage. This possibility is suggested by two recent rodent longevity studies comparing the ketogenic diet to high-fat and control diets. The high-fat and control diets had comparable health effects, while when carbohydrate was completely excluded, a substantial increase in both healthspan and median lifespan was obtained [45,46], with one of these studies reporting a statistically significant reduction in cancer in the ketogenic diet group [46]. If a similar effect occurs in humans, then the reported findings of a higher mortality in those with lower-carbohydrate diets in the epidemiological literature could still be consistent with a longevity advantage of an animal-based ketogenic diet. Indeed, a recent exploratory report showed that a small glucose bolus given to subjects on a ketogenic diet produced markers of acute cardiovascular damage [47], suggesting one potential mechanism (in addition to elevation in apolipoprotein B from high saturated fat intake) for elevated cardiovascular disease risk for subjects in the cohort studies.

Two further considerations warrant caution. First, if as DIETFITS suggests (discussed above) [48] and the paucity of very low-carbohydrate subjects in the above-discussed cohorts points to, then long-term adherence to a very low-carbohydrate ketogenic diet is possibly very low. With substantial carbohydrate intake thwarting ketogenesis and placing subjects in the carbohydrate intake range indicated by these studies, then recommendations to consume a ketogenic diet might pose inherent health risk due to variation in adherence in a substantial proportion of the population to which such recommendations might be directed. In other words, if non-adherence with the ketogenic diet is the rule rather than the exception, population-level recommendations for a ketogenic diet are inappropriate if animal-based lower-carb but not ketogenic dietary intakes are the norm among ketogenic dieters. Second, if the above findings point to a health advantage of plants and a disadvantage of meat and other animal products, then these health effects would be conceivably maintained even at ketogenic macronutrient compositions, even if ketosis itself offers independent advantages. In other words, the above studies, while not in those on ketogenic dieters, nonetheless point toward a ketogenic diet higher in plants as being preferable to one higher in animal products. Therefore, if a ketogenic diet may be substantially cancer preventive or curative, one that is higher in fiber and plants and lower in saturated fats and animal products is, according to current evidence, most likely to be the most healthful version, absent RCT data to the contrary.

1.        Mazidi, M.; Katsiki, N.; Mikhailidis, D.P.; Sattar, N.; Banach, M. Lower carbohydrate diets and all-cause and cause-specific mortality: a population-based cohort study and pooling of prospective studies. Eur. Heart J. 2019.

2.        Nakamura, Y.; Okuda, N.; Okamura, T.; Kadota, A.; Miyagawa, N.; Hayakawa, T.; Kita, Y.; Fujiyoshi, A.; Nagai, M.; Takashima, N.; et al. Low-carbohydrate diets and cardiovascular and total mortality in Japanese: a 29-year follow-up of NIPPON DATA80. Br. J. Nutr. 2014, 112, 916–924.

3.        Dehghan, M.; Mente, A.; Zhang, X.; Swaminathan, S.; Li, W.; Mohan, V.; Iqbal, R.; Kumar, R.; Wentzel-Viljoen, E.; Rosengren, A.; et al. Associations of fats and carbohydrate intake with cardiovascular disease and mortality in 18 countries from five continents (PURE): a prospective cohort study. Lancet (London, England) 2017, 390, 2050–2062.

4.        Odphp 2015-2020 Dietary Guidelines for Americans; 2015;

5.        Seidelmann, S.B.; Claggett, B.; Cheng, S.; Henglin, M.; Shah, A.; Steffen, L.M.; Folsom, A.R.; Rimm, E.B.; Willett, W.C.; Solomon, S.D. Articles Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis. Lancet Public Heal. 2018, 3, e419–e428.

6.        de Souza, R.J.; Dehghan, M.; Anand, S.S. Low carb or high carb? Everything in moderation … until further notice. Eur. Heart J. 2019.

7.        Veronese, N.; Solmi, M.; Caruso, M.G.; Giannelli, G.; Osella, A.R.; Evangelou, E.; Maggi, S.; Fontana, L.; Stubbs, B.; Tzoulaki, I. Dietary fiber and health outcomes: an umbrella review of systematic reviews and meta-analyses. Am. J. Clin. Nutr. 2018, 107, 436–444.

8.        Ma, Y.; Hu, M.; Zhou, L.; Ling, S.; Li, Y.; Kong, B.; Huang, P. Dietary fiber intake and risks of proximal and distal colon cancers: A meta-analysis. Medicine (Baltimore). 2018, 97, e11678.

9.        Gianfredi, V.; Salvatori, T.; Villarini, M.; Moretti, M.; Nucci, D.; Realdon, S. Is dietary fibre truly protective against colon cancer? A systematic review and meta-analysis. Int. J. Food Sci. Nutr. 2018, 69, 904–915.

10.      Schwingshackl, L.; Schwedhelm, C.; Hoffmann, G.; Knüppel, S.; Laure Preterre, A.; Iqbal, K.; Bechthold, A.; De Henauw, S.; Michels, N.; Devleesschauwer, B.; et al. Food groups and risk of colorectal cancer. Int. J. Cancer 2018, 142, 1748–1758.

11.      Aicr; WCRF Wholegrains, vegetables and fruit and the risk of cancer;

12.      Chen, S.; Chen, Y.; Ma, S.; Zheng, R.; Zhao, P.; Zhang, L.; Liu, Y.; Yu, Q.; Deng, Q.; Zhang, K. Dietary fibre intake and risk of breast cancer: A systematic review and meta-analysis of epidemiological studies. Oncotarget 2016, 7, 80980–80989.

13.      Zheng, B.; Shen, H.; Han, H.; Han, T.; Qin, Y. Dietary fiber intake and reduced risk of ovarian cancer: a meta-analysis. Nutr. J. 2018, 17, 99.

14.      Huang, X.; Wang, X.; Shang, J.; Lin, Y.; Yang, Y.; Song, Y.; Yu, S. Association between dietary fiber intake and risk of ovarian cancer: a meta-analysis of observational studies. J. Int. Med. Res. 2018, 46, 3995–4005.

15.      Farvid, M.S.; Eliassen, A.H.; Cho, E.; Liao, X.; Chen, W.Y.; Willett, W.C. Dietary Fiber Intake in Young Adults and Breast Cancer Risk. Pediatrics 2016, 137, e20151226.

16.      Farvid, M.S.; Chen, W.Y.; Michels, K.B.; Cho, E.; Willett, W.C.; Eliassen, A.H. Fruit and vegetable consumption in adolescence and early adulthood and risk of breast cancer: population based cohort study. BMJ 2016, 353, i2343.

17.      Farvid, M.S.; Chen, W.Y.; Rosner, B.A.; Tamimi, R.M.; Willett, W.C.; Eliassen, A.H. Fruit and vegetable consumption and breast cancer incidence: Repeated measures over 30 years of follow-up. Int. J. Cancer 2019, 144, 1496–1510.

18.      Xiao, Y.; Ke, Y.; Wu, S.; Huang, S.; Li, S.; Lv, Z.; Yeoh, E.; Lao, X.; Wong, S.; Kim, J.H.; et al. Association between whole grain intake and breast cancer risk: a systematic review and meta-analysis of observational studies. Nutr. J. 2018, 17, 87.

19.      Xu, Y.; Wan, Q.; Feng, J.; Du, L.; Li, K.; Zhou, Y. Whole grain diet reduces systemic inflammation: A meta-analysis of 9 randomized trials. Med. (United States) 2018, 97.

20.      Kärkkäinen, O.; Lankinen, M.A.; Vitale, M.; Jokkala, J.; Leppänen, J.; Koistinen, V.; Lehtonen, M.; Giacco, R.; Rosa-Sibakov, N.; Micard, V.; et al. Diets rich in whole grains increase betainized compounds associated with glucose metabolism. Am. J. Clin. Nutr. 2018, 108, 971–979.

21.      Schwingshackl, L.; Schwedhelm, C.; Hoffmann, G.; Lampousi, A.-M.; Knüppel, S.; Iqbal, K.; Bechthold, A.; Schlesinger, S.; Boeing, H. Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies. Am. J. Clin. Nutr. 2017, 105, 1462–1473.

22.      Wu, L.; Wang, Z.; Zhu, J.; Murad, A.L.; Prokop, L.J.; Murad, M.H. Nut consumption and risk of cancer and type 2 diabetes: A systematic review and meta-analysis. Nutr. Rev. 2015, 73, 409–425.

23.      Neuenschwander, M.; Ballon, A.; Weber, K.S.; Norat, T.; Aune, D.; Schwingshackl, L.; Schlesinger, S. Role of diet in type 2 diabetes incidence: Umbrella review of meta-analyses of prospective observational studies. BMJ 2019, 366.

24.      Ohkuma, T.; Peters, S.A.E.; Woodward, M. Sex differences in the association between diabetes and cancer: a systematic review and meta-analysis of 121 cohorts including 20 million individuals and one million events. Diabetologia 2018, 61, 2140–2154.

25.      Reynolds, A.; Mann, J.; Cummings, J.; Winter, N.; Mete, E.; Te Morenga, L. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. Lancet (London, England) 2019, 393, 434–445.

26.      Brennan, S.F.; Woodside, J. V; Lunny, P.M.; Cardwell, C.R.; Cantwell, M.M. Dietary fat and breast cancer mortality: A systematic review and meta-analysis. Crit. Rev. Food Sci. Nutr. 2017, 57, 1999–2008.

27.      Wolk, A. Potential health hazards of eating red meat. J. Intern. Med. 2017, 281, 106–122.

28.      Etemadi, A.; Sinha, R.; Ward, M.H.; Graubard, B.I.; Inoue-Choi, M.; Dawsey, S.M.; Abnet, C.C. Mortality from different causes associated with meat, heme iron, nitrates, and nitrites in the NIH-AARP Diet and Health Study: population based cohort study. BMJ 2017, 357, j1957.

29.      Arthur, A.E.; Goss, A.M.; Demark-Wahnefried, W.; Mondul, A.M.; Fontaine, K.R.; Chen, Y.T.; Carroll, W.R.; Spencer, S.A.; Rogers, L.Q.; Rozek, L.S.; et al. Higher carbohydrate intake is associated with increased risk of all-cause and disease-specific mortality in head and neck cancer patients: results from a prospective cohort study. Int. J. Cancer 2018, 143, 1105–1113.

30.      Barua, R.; Templeton, A.J.; Seruga, B.; Ocana, A.; Amir, E.; Ethier, J.-L. Hyperglycaemia and Survival in Solid Tumours: A Systematic Review and Meta-analysis. Clin. Oncol. 2018, 30, 215–224.

31.      Shahdadian, F.; Saneei, P.; Milajerdi, A.; Esmaillzadeh, A. Dietary glycemic index, glycemic load, and risk of mortality from all causes and cardiovascular diseases: a systematic review and dose-response meta-analysis of prospective cohort studies. Am. J. Clin. Nutr. 2019.

32.      Song, M.; Wu, K.; Meyerhardt, J.A.; Ogino, S.; Wang, M.; Fuchs, C.S.; Giovannucci, E.L.; Chan, A.T. Fiber Intake and Survival After Colorectal Cancer Diagnosis. JAMA Oncol. 2018, 4, 71.

33.      Meyerhardt, J.A.; Sato, K.; Niedzwiecki, D.; Ye, C.; Saltz, L.B.; Mayer, R.J.; Mowat, R.B.; Whittom, R.; Hantel, A.; Benson, A.; et al. Dietary glycemic load and cancer recurrence and survival in patients with stage III colon cancer: findings from CALGB 89803. J. Natl. Cancer Inst. 2012, 104, 1702–11.

34.      Maino Vieytes, C.A.; Taha, H.M.; Burton-Obanla, A.A.; Douglas, K.G.; Arthur, A.E. Carbohydrate Nutrition and the Risk of Cancer. Curr. Nutr. Rep. 2019.

35.      Makarem, N.; Bandera, E. V.; Nicholson, J.M.; Parekh, N. Consumption of Sugars, Sugary Foods, and Sugary Beverages in Relation to Cancer Risk: A Systematic Review of Longitudinal Studies. Annu. Rev. Nutr. 2018, 38, 17–39.

36.      Fung, T.T.; Willett, W.C.; Stampfer, M.J.; Manson, J.A.E.; Hu, F.B. Dietary patterns and the risk of coronary heart disease in women. Arch. Intern. Med. 2001, 161, 1857–1862.

37.      Song, M.; Wu, K.; Meyerhardt, J.A.; Yilmaz, O.; Wang, M.; Ogino, S.; Fuchs, C.S.; Giovannucci, E.L.; Chan, A.T. Low-Carbohydrate Diet Score and Macronutrient Intake in Relation to Survival After Colorectal Cancer Diagnosis. JNCI Cancer Spectr. 2018, 2.

38.      Fung, T.T.; Van Dam, R.M.; Hankinson, S.E.; Stampfer, M.; Willett, W.C.; Hu, F.B. Low-carbohydrate diets and all-cause and cause-specific mortality: Two cohort studies. Ann. Intern. Med. 2010, 153, 289–298.

39.      Nilsson, L.M.; Winkvist, A.; Johansson, I.; Lindahl, B.; Hallmans, G.; Lenner, P.; Van Guelpen, B. Low-carbohydrate, high-protein diet score and risk of incident cancer; A prospective cohort study. Nutr. J. 2013, 12.

40.      Kelemen, L.E.; Kushi, L.H.; Jacobs, D.R.; Cerhan, J.R. Associations of dietary protein with disease and mortality in a prospective study of postmenopausal women. Am. J. Epidemiol. 2005, 161, 239–249.

41.      Levine, M.E.; Suarez, J.A.; Brandhorst, S.; Balasubramanian, P.; Cheng, C.W.; Madia, F.; Fontana, L.; Mirisola, M.G.; Guevara-Aguirre, J.; Wan, J.; et al. Low protein intake is associated with a major reduction in IGF-1, cancer, and overall mortality in the 65 and younger but not older population. Cell Metab. 2014, 19, 407–417.

42.      Song, M.; Fung, T.T.; Hu, F.B.; Willett, W.C.; Longo, V.D.; Chan, A.T.; Giovannucci, E.L. Association of animal and plant protein intake with all-cause and cause-specific mortality. JAMA Intern. Med. 2016, 176, 1453–1463.

43.      Virtanen, H.E.K.; Voutilainen, S.; Koskinen, T.T.; Mursu, J.; Kokko, P.; Ylilauri, M.P.T.; Tuomainen, T.P.; Salonen, J.T.; Virtanen, J.K. Dietary proteins and protein sources and risk of death: The Kuopio ischaemic heart disease risk factor study. Am. J. Clin. Nutr. 2019, 109, 1462–1471.

44.      Hernández-Alonso, P.; Salas-Salvadó, J.; Ruiz-Canela, M.; Corella, D.; Estruch, R.; Fitó, M.; Arós, F.; Gómez-Gracia, E.; Fiol, M.; Lapetra, J.; et al. High dietary protein intake is associated with an increased body weight and total death risk. Clin. Nutr. 2016, 35, 496–506.

45.      Newman, J.C.; Covarrubias, A.J.; Zhao, M.; Yu, X.; Gut, P.; Ng, C.P.; Huang, Y.; Haldar, S.; Verdin, E. Ketogenic Diet Reduces Midlife Mortality and Improves Memory in Aging Mice. Cell Metab. 2017, 26, 547-557.e8.

46.      Roberts, M.N.; Wallace, M.A.; Tomilov, A.A.; Zhou, Z.; Marcotte, G.R.; Tran, D.; Perez, G.; Gutierrez-Casado, E.; Koike, S.; Knotts, T.A.; et al. A Ketogenic Diet Extends Longevity and Healthspan in Adult Mice. Cell Metab. 2017, 26, 539-546.e5.

47.      Durrer, C.; Lewis, N.; Wan, Z.; Ainslie, P.N.; Jenkins, N.T.; Little, J.P. Short-term low-carbohydrate high-fat diet in healthy young males renders the endothelium susceptible to hyperglycemia-induced damage, an exploratory analysis. Nutrients 2019, 11.

48.      Gardner, C.D.; Trepanowski, J.F.; Del Gobbo, L.C.; Hauser, M.E.; Rigdon, J.; Ioannidis, J.P.A.; Desai, M.; King, A.C. Effect of Low-Fat vs Low-Carbohydrate Diet on 12-Month Weight Loss in Overweight Adults and the Association With Genotype Pattern or Insulin Secretion. JAMA 2018, 319, 667.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 5 minutes

So apparently Paul Saladinos and Mikhaila Peterson have recently been talking about me on a podcast.🤨

I haven’t had a chance to listen to the podcast, and I probably won’t. But apparently it had something to do with my statements that the benefits of the carnivore diet are caused by calorie restriction.

So I will make my thoughts clear about this. There are probably five main mechanisms of the carnivore diet.

Calorie restriction
Antigen elimination
FODMAPs, etc.

First. Calorie restriction.

People might remember that @joerogan brought up that point with Mikhaila. @chriskresser agreed on a later podcast. And so did @foundmyfitness on yet another later podcast.

So I’m not alone.

Why? Because calorie restriction has long been robustly associated with anti-inflammatory effects. For instance, here from CALERIE 2:

And this is only CHRONIC calorie restriction, producing a lean body type. Chronic calorie restriction causes chronic reductions in inflammation.

But what’s more, there is a long literature on the ACUTE effects of fasting on rheumatoid arthritis. You can find that literature here:…

In a word, fasting has long been known to acutely cause remission of the symptoms of rheumatoid arthritis and other autoimmune conditions, almost immediately. Start fasting, and watch the symptoms disappear. It works for many people.

Of course it is not sustainable in the long-term without periodic refeeds, which will cause symptoms to return.

What are the mechanisms for this? It could be ketones. There is a conflicting literature on the supposed anti-inflammatory effects of ketones, which I am writing about for a review right now.

It could also be inhibition of many anabolic/inflammatory proteins associated with lower cellular energy levels.

It could also be the withdrawal of an antigen that the body responds to too strongly.

We do not know, but calorie restriction has profound anti-inflammatory effects and would be expected to help substantially for people with inflammatory conditions.

Incidentally, a ketogenic diet might also mimic some of the effects of the fasted state, and in doing so, might have an important therapeutic benefit for autoimmunity–similar to how it does for epilepsy.

But this still remains speculative. A definitive test for this might be done in a clinical trial:

A ketogenic diet
A diet that is carbohydrate-restricted with a very low glycemic load but no ketosis
A baseline diet with supplemental ketone esters
A baseline diet

Run it in random crossover format and watch the symptoms. Maintain stable bodyweight. Make sure the diets are as identical as possible except for the characteristics above.

An RCT like this would pick apart the antigenic effects, versus the ketotic effects, versus the glycemic effects, versus full ketosis. Somebody write a grant~

A pilot would just test ketogenic versus baseline diet. Anyway…

Mechanism two. Protein.

Protein has profound metabolic effects. It increases lean muscle mass. It causes isocaloric remission of fatty liver. It promotes improved overall body composition.

Through these mechanisms, protein is also anti-inflammatory. More muscle and less liver and muscle fat will increase the “energy sink” of the body, and this will increase the homeostatic regulatory capacity of the body in response to energy intake, i.e. smaller post-prandial fatty acid and glucose spikes. This will spare the cardiovascular system and the immune system will secrete fewer inflammatory molecules. The brain will “feel” this. So will the joints, etc.

How BIG the impact of this is is anyone’s guess. We don’t know.

However, a caveat: it probably depends on the source of protein.

Three. Antigenic restriction.

This one is obvious. The carnivore diet is hypo-antigenic. It is an elimination diet.

We have long known that elimination diets, e.g. the Elemental Diet, can be used to successfully treat inflammatory conditions. See:…

The Elemental Diet is kind of like Medical Soylent. It is all the nutrients a person needs, but with no proteins–just dissociated amino acids. No complex carbs either–just glucose, etc. And vitamins, fatty acids, etc.

It is as hypo-antigenic as a diet can be.

I remember my Master’s advisor telling me about the elemental diet when I was a Paleoish Internet diet warrior. I was appalled. OMG NOT REAL FOOD.

It doesn’t matter.😆The point is that restricting antigens works. We know that.

Four. FODMAPs.

Fermentable oligosaccharides, disaccharides, monosaccharides and polyols, or FODMAPs, are carbohydrates in many plant foods that irritate the gut and cause inflammation. When restricted, they can treat irritable bowel syndrome, see here. FODMAPs and other carbohydrates like fiber might also modulate gut immunity in ways that science still has a long way to understanding.

Five. Placebo.

We know that the mind has a profound impact on immune function, predominantly via the vagus nerve. Example:

A really great book discussing this is Jo Marchant’s Cure. Highly recommended.

Incidentally, the same might be true for weight loss. Witness:

I do think that placebo might have a profound influence on how a diet is experienced. If a diet is expected to be anti-inflammatory, I do wonder if it might in fact be anti-inflammatory.

There is a psychological defense mechanism called conversion. I used to experience it. When I was a child, I used to have seizures. When the doctors gave me an EEG, we found that I was not having seizures.

I was having pseudoseizures. Thus the moniker “conversion”: conversion of psychological stress into bodily symptoms. According to psychoanalysts, conversion is an immature, narcissistic psychological defense. It focuses on the self.

So from my own personal experience, I really believe in the mind-body connection. It can be profound. It can create profound distortions in experience.

Jo Marchant’s book shows that these effects can extend beyond the psychological and manifest in physical illness (or cures) via modulation of immunity.

So these are my 4 mechanisms that I would hypothesize personally.

The reason this is important to talk about is that, once we know the mechanisms, we can then change and optimize the diet and have people helped even more by it.

There are stories floating about of people on a carnivore diet who are refusing statins but continuing to have CVD progression.

What if the carnivore diet does not work for everyone and could be harmful for some people? Does Mikhaila think that is possible?

If so, what if we could get all of the benefits of the carnivore diet by isolating mechanisms without some of the potential downsides?

I still strongly suspect that the high red meat content is not optimal. Red meat might be nutritious in some ways–sure–but I strongly suspect that large quantities are probably not optimal for whole body oxidative stress and cardiovascular disease risk.

So if we can pull apart the mechanisms and determine which are helping people on the carnivore diet, then we can use these mechanisms with more flexible dieting strategies, producing the same effect with fewer downsides.

Let me add one last thing.

There is a big difference in worldview between most doctors and many carnivore dieters (the same could probably be said of many plant-based dieters, etc.).

For the medical way of thinking, every intervention has benefits and risks. There is no perfect intervention. Knowing the mechanism of an intervention allows us to get rid of everything else that is not relevant to the mechanism, and thus reduce potential harms.

Thus, mechanisms are necessary to explore to try to maximize the benefits while minimizing the risks.

On the other hand, among many people in various dieting communities, lifestyle interventions are “wholly good”.

I don’t believe anything that humans do is wholly good. There are good and bad parts.

It is true that there are some things that humans do that are better than other things.

But I don’t think there are many “final goods” that solve all problems with no downsides, and I think we should be skeptical of anyone who claims to have found such a good.

And I think we are right to be skeptical. And I think if anyone says we shouldn’t be skeptical, that we should just believe–well, that is very problematic to say the least, and it is a lot like religion.

Science says that we should explore and understand. That is what I think we should do.

That is all.

This post was adapted and edited with some additional content from a popular Twitter thread, here.


Support me here.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 6 minutes

This represents an early draft of a section of an upcoming scientific review.

One of the most exciting series of findings involves reduction of xenograft growth in mouse cancer models among animals given the ketogenic diet. A much-celebrated meta-analysis of mouse cancer models by Klement et al., 2016 (using at least a 2:1 ketogenic ratio and included controls with 50% of energy without additional treatment and reported tumor growth and survival endpoints) showed a robust reduction in tumor growth was found in mice fed the ketogenic diet compared to mice fed standard rodent chow [1]. However, severe methodological problems may have caused this meta-analysis and the studies it includes to poorly represent the preclinical efficacy of the ketogenic diet. Among the twelve studies included in this meta-analysis, nine used ketogenic diets with protein content (as a percent of total kilocalories) ranging from 61% to as low as 22% of that in the control diets, or in other words, involving between a half and a four-fold protein restriction (with a mean protein intake of 42% that of the control diet) [2–11]. This is important because it has long been known that protein restriction causes a robust delay of chemically induced and xenograft tumor growth [12–16]. Methionine restriction, which also occurs during protein restriction, achieves the same effect [17–19], including among several types of glioma xenografts [20], the most prominent and much-celebrated cancer type to ostensibly respond to the ketogenic diet in the ketogenic diet literature [1]. Therefore, the positive results of these nine ketogenic diet studies are confounded by protein restriction, and it is unclear what role the ketogenic component of the diet played in the results reported.

Furthermore, while Klement et al., 2016 reported positive findings for the remaining three studies, two of these findings occurred in comparison to a high-fat diet [3,5], which has been shown in several studies to increase the rate of xenograft tumor growth [21,22]. In fact, in these same two papers, a comparison with a third group of mice fed standard rodent chow was reported, and this showed no significant difference with the ketogenic diet-fed mice (except for a small difference at day 39 in one paper, which favored the standard chow mice) [3,5]. Surprisingly, the comparison with this control group of mice, which would suggest no efficacy of the ketogenic diet beyond simply being better than a high-fat diet, was for some reason not reported by Klement et al., 2016, further biasing the meta-analysis.

The third of the three diets that matched for protein did see a beneficial effect for the ketogenic diet, but this version of the ketogenic diet was highly unusual, with non-ketogenic carbohydrate kilocalorie % and ketogenesis driven by supplemental medium-chain triglycerides (Martuscello et al., 2016). Interestingly, Martuscello’s MCT group had a 5% higher protein as % of macronutrients than control (21% vs 26%) and may have consumed more food: they gained more weight than the other groups.

This finding is promising, but future studies will need to confirm these results with less confounded feeding designs. Of note, one high-profile mouse study published last year in Nature also used a highly restricted protein intake in the ketogenic group—a mere 25% of the intake of the control group (Hopkins et al., 2018), perhaps exemplifying the prevalence of this design flaw in the ketogenic diet literature. Other recent papers, such as one that showed normalization of the metabolome in a breast cancer xenograft model with the administration of a ketogenic diet, are also confounded by profound protein restriction (Licha et al., 2019).

A list of studies discussed in this article can be found below:

This image has an empty alt attribute; its file name is Screen-Shot-2019-09-06-at-4.52.13-PM.png

Note: the 2013 Poff article listed in the spreadsheet above refers to a different study (but published in the same year) than that examined by Klement et al., 2016, with a slightly different macronutrient composition than that listed. This difference does not substantially change the analysis or conclusion.

1.        Klement, R.J.; Champ, C.E.; Otto, C.; Kämmerer, U. Anti-tumor effects of ketogenic diets in mice: A meta-analysis. PLoS One 2016, 11, 1–16.

2.        Zhou, W.; Mukherjee, P.; Kiebish, M.A.; Markis, W.T.; Mantis, J.G.; Seyfried, T.N. The calorically restricted ketogenic diet, an effective alternative therapy for malignant brain cancer. Nutr. Metab. (Lond). 2007, 4, 5.

3.        Freedland, S.J.; Mavropoulos, J.; Wang, A.; Darshan, M.; Demark-Wahnefried, W.; Aronson, W.J.; Cohen, P.; Hwang, D.; Peterson, B.; Fields, T.; et al. Carbohydrate restriction, prostate cancer growth, and the insulin-like growth factor axis. Prostate 2008, 68, 11–9.

4.        Otto, C.; Kaemmerer, U.; Illert, B.; Muehling, B.; Pfetzer, N.; Wittig, R.; Voelker, H.U.; Thiede, A.; Coy, J.F. Growth of human gastric cancer cells in nude mice is delayed by a ketogenic diet supplemented with omega-3 fatty acids and medium-chain triglycerides. BMC Cancer 2008, 8.

5.        Mavropoulos, J.C.; Buschemeyer, W.C.; Tewari, A.K.; Rokhfeld, D.; Pollak, M.; Zhao, Y.; Febbo, P.G.; Cohen, P.; Hwang, D.; Devi, G.; et al. The effects of varying dietary carbohydrate and fat content on survival in a murine LNCaP prostate cancer xenograft model. Cancer Prev. Res. (Phila). 2009, 2, 557–65.

6.        Rieger, J.; Bähr, O.; Maurer, G.D.; Hattingen, E.; Franz, K.; Brucker, D.; Walenta, S.; Kämmerer, U.; Coy, J.F.; Weller, M.; et al. ERGO: a pilot study of ketogenic diet in recurrent glioblastoma. Int. J. Oncol. 2014, 44, 1843–52.

7.        Maurer, G.D.; Brucker, D.P.; Bähr, O.; Harter, P.N.; Hattingen, E.; Walenta, S.; Mueller-Klieser, W.; Steinbach, J.P.; Rieger, J. Differential utilization of ketone bodies by neurons and glioma cell lines: a rationale for ketogenic diet as experimental glioma therapy. BMC Cancer 2011, 11, 315.

8.        Abdelwahab, M.G.; Fenton, K.E.; Preul, M.C.; Rho, J.M.; Lynch, A.; Stafford, P.; Scheck, A.C. The ketogenic diet is an effective adjuvant to radiation therapy for the treatment of malignant glioma. PLoS One 2012, 7, e36197.

9.        Hao, G.-W.; Chen, Y.-S.; He, D.-M.; Wang, H.-Y.; Wu, G.-H.; Zhang, B. Growth of human colon cancer cells in nude mice is delayed by ketogenic diet with or without omega-3 fatty acids and medium-chain triglycerides. Asian Pac. J. Cancer Prev. 2015, 16, 2061–8.

10.      Martuscello, R.T.; Vedam-Mai, V.; McCarthy, D.J.; Schmoll, M.E.; Jundi, M.A.; Louviere, C.D.; Griffith, B.G.; Skinner, C.L.; Suslov, O.; Deleyrolle, L.P.; et al. A Supplemented High-Fat Low-Carbohydrate Diet for the Treatment of Glioblastoma. Clin. Cancer Res. 2016, 22, 2482–95.

11.      Dang, M.T.; Wehrli, S.; Dang, C. V.; Curran, T. The ketogenic diet does not affect growth of Hedgehog pathway medulloblastoma in mice. PLoS One 2015, 10.

12.      Fontana, L.; Adelaiye, R.M.; Rastelli, A.L.; Miles, K.M.; Ciamporcero, E.; Longo, V.D.; Nguyen, H.; Vessella, R.; Pili, R. Dietary protein restriction inhibits tumor growth in human xenograft models of prostate and breast cancer. Oncotarget 2013, 4, 2451–2461.

13.      Hawrylewicz, E.J.; Huang, H.H.; Liu, J.M. Dietary protein, enhancement of N-nitrosomethylurea-induced mammary carcinogenesis, and their effect on hormone regulation in rats. Cancer Res. 1986, 46, 4395–9.

14.      Appleton, B.S.; Campbell, T.C. Inhibition of aflatoxin-initiated preneoplastic liver lesions by low dietary protein. Nutr. Cancer 1982, 3, 200–6.

15.      Appleton, B.S.; Campbell, T.C. Dietary protein intervention during the postdosing phase of aflatoxin B1-induced hepatic preneoplastic lesion development. J. Natl. Cancer Inst. 1983, 70, 547–9.

16.      Appleton, B.S.; Campbell, T.C. Effect of high and low dietary protein on the dosing and postdosing periods of aflatoxin B1-induced hepatic preneoplastic lesion development in the rat. Cancer Res. 1983, 43, 2150–4.

17.      Gao, X.; Sanderson, S.M.; Dai, Z.; Reid, M.A.; Cooper, D.E.; Lu, M.; Richie, J.P.; Ciccarella, A.; Calcagnotto, A.; Mikhael, P.G.; et al. Dietary methionine restriction targets one carbon metabolism in humans and produces broad therapeutic responses in cancer. bioRxiv 2019, 627364.

18.      Kokkinakis, D.M.; Schold, S.C.; Hori, H.; Nobori, T. Effect of long-term depletion of plasma methionine on the growth and survival of human brain tumor xenografts in athymic mice. Nutr. Cancer 1997, 29, 195–204.

19.      Latimer, M.N.; Freij, K.W.; Cleveland, B.M.; Biga, P.R. Physiological and molecular mechanisms of methionine restriction. Front. Endocrinol. (Lausanne). 2018, 9.

20.      Hoffman, R.M.; Kokkinakis, D.M.; Frenkel, E.P. Total Methionine Restriction Treatment of Cancer. Methods Mol. Biol. 2019, 1866, 163–171.

21.      O’Neill, A.M.; Burrington, C.M.; Gillaspie, E.A.; Lynch, D.T.; Horsman, M.J.; Greene, M.W. High-fat Western diet-induced obesity contributes to increased tumor growth in mouse models of human colon cancer. Nutr. Res. 2016, 36, 1325–1334.

22.      Lloyd, J.C.; Antonelli, J.A.; Phillips, T.E.; Masko, E.M.; Thomas, J.A.; Poulton, S.H.M.; Pollack, M.; Freedland, S.J. Effect of Isocaloric Low Fat Diet on Prostate Cancer Xenograft Progression in a Hormone Deprivation Model. J. Urol. 2010, 183, 1619–1624.

23.      Hopkins, B.D.; Pauli, C.; Xing, D.; Wang, D.G.; Li, X.; Wu, D.; Amadiume, S.C.; Goncalves, M.D.; Hodakoski, C.; Lundquist, M.R.; et al. Suppression of insulin feedback enhances the efficacy of PI3K inhibitors. Nature 2018, 560, 499–503.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 3 minutes

Nutrition science is challenging. In its own way, it is the most challenging of the sciences. But also the greatest.

(This quote is mainly lifted from Nicola Guess, who says that nutrition science drives people mad. She is right.)

Could it possibly be argued otherwise?

This difficulty comes from the fact that we are deeply biased about the act of eating, in a few ways:

Psychological. We are emotional about the act of eating, and we crave certainty about how we do it. We all eat, and eating is one of our most primary life functions. We form strong emotional attachments to the habits that fulfill primary life functions: food, sex, work, family, etc. When the emotional attachments to such habits of life–in this case, our eating habits–are challenged, we respond emotionally. The tendency to respond emotionally to biases our interpretation of the objective facts about nutrition. We also develop polarized thinking. Good-bad, right-wrong, natural-unnatural. Emotional, polarized thinking leads to a desire for certainty: if something so important is either good or bad, then we need it to be clearly one or the other, since making a mistake about something so important has serious implications. And yet…

Scientific. The objective facts of nutrition science, i.e. how we should eat, are unclear. Our strong emotions and craving for certainty about food are frustrated by the fact that the objective facts in nutrition science are incomplete and cannot fulfill these desires. We do not and will not have randomized controlled trials testing many of the most important hypotheses in nutrition science for the foreseeable future. Instead, we rely on surrogate or indirect tests of most of the hypotheses of nutrition science. Because these surrogate or indirect tests are many but also flawed, with different methods giving contradictory results, the results of many or most of these tests are debatable on the grounds that they do not and cannot provide the certainty that we psychologically crave. In fact, nutrition science does the very opposite and assures us that we can know little. What nutrition science can offer us (very little of certainty) is in direct conflict with deep-felt psychological needs (that is, a desire for certainty). This causes us to impute into nutrition science what is not there, or alternatively to become frustrated when definitive, practicable answers are not forthcoming–and to seek alternative sources of certainty, in the form of gurus, anecdotes, knowledge-through-self-knowledge, etc.

Politico-economic. The government and diet industry make authoritative but seriously conflicting recommendations. This leads to group identity and financial conflicts of interest. Because eating is one of our primary life functions, how we eat is important for health. To promote health, the government makes recommendations. Because we desire certainty and simplicity, they provide these recommendations in a simplified and authoritative form. Because these broad recommendations are not detailed, individualized, or explained well, an industry of practical advice has sprung up to provide more detailed, individualized, and better communicated recommendations. Because the advice of these authorities is conflicting, an emotionally charged conflict over food choices emerges. Shared feelings about food lead to the formation of groups of like-feeling people, in turn creating a sense of group identity. This sense of group identity further biases the way we receive the already fragmentary and difficult to interpret objective facts. Furthermore, because there are livelihoods to be made in this industry, financial conflicts of interest further bias the communication of the objective facts.

How should we evaluate and understand nutrition science so as to minimize these challenges and the impact of our biases on the interpretation of the facts? Is there any method or approach that is most reasonable that we can systematically follow to achieve this?

I think there might be. In evaluating this chapter, I hope to articulate the outlines of what that might look like. This systematic approach to the evidence will not enable us to fulfill our psychological desire for certainty about what we eat, but it might help people who want to understand how nutrition scientists think and work and where nutrition recommendations come from. Not much if anything of what I will write here will be new. It will only communicate what I have learned from my reflections on what nutrition scientists have told me and what I have read as a nutrition scientist-in-training.

But what I want to point out here is that nobody seems yet to be popularly endorsing or communicating a systematic approach for looking at nutrition science. That is part of the problem, and part of what this series is going to try to do.

For those who have followed this, I want this to be something bigger and more serious than just Diet Wars, or polyunsaturated or saturated fats, etc. That kind of argument is simply not interesting to me. I agreed to look at the chapter itself. So let us look at the chapter. And take it very seriously.


You can find me on Twitter and Instagram. You can help support me here.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 4 minutes

I have had a few exchanges with Paul Mason over the past couple of months. Most of the time he has explained concepts well, and I agree with his explanations.

What he gets wrong is the level of confidence that he has that these concepts are true. This is like much in the low-carbohydrate dieting space: plausible hypotheses are frequently passed as scientific fact, while everyone else is held to a much higher standard.

One would think that, given the consternation of the low-carbohydrate diet community at the weakness of evidence for, e.g. the American dietary guidelines, that this very same community would itself be very careful with the evidence. They are not. They have chosen to fight what they think to be fire with their own fire.

My own consternation came to a head recently, when, after seeing Paul tweet about his newly published chapter in Karim Khan and Peter Brukner’s Sports Medicine textbook, I pointed out that even the very first quote of Hippocrates in the book chapter was not something that Hippocrates actually said:

Snarky sure. So I beg forgiveness: snark is how I survive on Twitter, which is a veritable madhouse. Besides, in the Harvard Grant Study, humor was among the healthiest of ways to deal with psychological stress. So that’s my humor.

But, it’s also true: what does a misquote say about the factual soundness of the rest of the chapter? I mean, if your opening quote is contrived, what about the rest?

So Paul went on the offensive, and rightly…

This image has an empty alt attribute; its file name is Screen-Shot-2019-07-28-at-9.19.25-AM.png

I didn’t think I could give it a take that would result in a productive discussion. So I didn’t want to try. But Paul persisted, and I accepted.

I read the first few pages, and my eyes nearly rolled out of my head: it was exactly what I had expected. One weakly defensible hypothesis after another, presented as scientific fact, which, according to Paul, fools, conspirators, and other actors in bad faith from The Establishment had committed to suppressing through a deadening avalanche of bad science. And then, one criticism of dissenting views after another, presented as definitive. Next, references that provided weak support or used poor methods presented as conclusive. It goes on.

I thought: it will take 200 hours of digging, picking apart the distortions, and correcting them. And who will read it anyway? So I sat on it, planning to write something but procrastinating, mostly despairing of being able to do anything meaningful about this.

Paul decided to go on the offensive, again, here:

And here:

And the rest of that thread consists of Paul haranguing me about saturated fat and polyunsaturated fat and trying to pigeonhole me about my “views”. I mostly tried not to engage.

But what I realized is this: if I could do that work and break down Paul’s chapter, and demonstrate why it is a misuse of evidence, and people would actually commit to reading it, it might be worth my time. I wouldn’t be writing for myself. And if I feel confident that I can persuade intelligent people–and I do feel confident–then this could actually be quite a fulfilling exercise.

It would also help me to work out my own thoughts about what constitutes an impartial evaluation of nutrition science.

So I created a reading challenge. Here’s the gist of it. If you are a hardcore carnivore/LCHF type but open-minded, you can sign up for the reading challenge and commit to reading one hour of what I write.

Tit-for-tat with each such commitment, I myself will commit to an hour of research and writing.

So if I get 100 sign-ups, I will commit to researching and writing for 100 hours. And I work at a minimum rate of 10 hours per month, which is sustainable for me. So no concerns about spending tons of time writing for no reason:

Lo-and-behold, I received 90 commitments. So now I am committed to 90 hours. This will be the first post fulfilling that commitment. For those signed up, I will be sending out an email with an update when the first 10 hours are over and each month when the work is complete thereafter. When I have written an hour’s worth of reading content (as measured by an algorithm), people who have already signed up will also be able to sign up a second time.

If you are stumbling across this post from outside of Twitter, take a moment to read the Twitter thread. If you meet the criteria to sign-up and want to commit, please do. The sign-up form is at the end of the thread.

For those who do not meet the criteria but still want to follow this series and receive updates about it, please sign up using the form here.

Now, why the title? This is about Paul Mason and Daniel Freedman’s book chapter, isn’t it? It is. But after looking through the chapter, it is obvious to me that Paul’s book chapter actually represents “current low-carbohydrate thinking”, in general. All of the contentious claims about fiber, saturated fat, red meat, macronutrient intakes, epidemiology, etc.: everything is there.

Thus, I want to use Paul’s chapter to scrutinize “low-carb science” as a whole. But I don’t want to just say the chapter is wrong. I want to go deeper. I want to get to the heart of current methodological controversies within nutrition science itself. And I want to explain why “low-carb science” does it wrong. And I want to do it in the most minute, meticulous detail, point-by-point, error-by-error, explaining painstakingly why they are errors and what a better way might (or might not) be–to a degree nobody has done before.

It’s a big thing to take on. That’s why I asked for a commitment before starting this. Thank you for starting on this journey with me. If I continue to get commitments, I will be continuing this series throughout the entire remainder of my Ph.D. I would be honored and excited to do so.


You can find me on Twitter and Instagram. You can help support me here.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 4 minutes

A narrative has sprung up on the Internet, largely fueled by mainstream media organizations like Vice, CNN, etc., that meat consumption in the developed world is causing deforestation of the Amazon.

This narrative is false.

In this blog, I will cover two issues:

  1. Brazilian beef and who consumes it
  2. Brazilian soy and who consumes it

It is true that the Amazon is being deforested for wood, soy, and livestock. The soy in turn is being fed to livestock, so the Amazon is being deforested, more substantially, for wood and livestock.

But who is buying that soy? Who is buying that livestock? Who, in other words, is economically (and morally) responsible for environmental devastation that these commodities are a product of?

According to CNN, the breakdown of Brazilian beef exports is as follows:


So far, this is true.

Now, total production of beef in Brazil is ~22 billion pounds. 1.64 million metric tonnes or ~3.61 billion pounds is exported.

This means that 3.61/22 or 16.4% of Brazilian beef is exported. In turn, this means that Brazilians consume 84% of Brazilian beef domestically.

64 million pounds are exported to America, according to USDA:

This roughly confirms the CNN figures: 64 million is a little less than 2% of 3.61 billion. This amounts to ~0.3% of total Brazilian beef production.

On the other hand, the amount exported to China and Hong Kong combined amount to 43% of exports, which is 43% of 16%:

Or 7% of total Brazilian beef production.

This means that together, Brazilians and Chinese consume about 91% (84% + 7%) of Brazilian beef. So the narrative that the “world” is driving the fires in the Amazon is just wrong. The Brazilians and Chinese are driving the fires. And largely the Brazilians.

Europeans by the way consume about 7% of Brazilian exports, mainly Italians. This amounts to 1.1% of total Brazilian beef production.

So together, the EU and USA consume about 1.4% of Brazilian beef.

Brazilians and Chinese consume about 91%.

So we should definitely stop Brazilian beef importation completely, but we aren’t the main problem.

There is also a narrative about Brazilian soybeans, i.e. because Americans don’t send their soybeans to Brazil, we are somehow responsible. How do Americans and Europeans fit in there?

Almost 80% of Brazilian soybean exports are to China, and exports to China from the United States actually increased in 2019, while Brazilian exports to China dropped by almost 15%. (Source.)

But it gets worse, because while 80% of Brazilian soybean exports are to China, about 80% of Brazilian soybeans are NOT exported, probably mainly to feed Brazilian livestock. (Source.)

This means that about 96% of Brazilian soybeans are either consumed domestically (i.e. by livestock) or exported to China (also to feed livestock). I’m not sure that America and Brazil trade soybeans at all.

So if the Amazon rainforest fires are driven by beef and soybeans, and we want to blame anyone this, again, it’s overwhelmingly China, Hong Kong, and Brazil.

Together, China, Hong Kong, and Brazil consume about 91% of Brazilian beef and 96% of Brazilian soybeans.

Indeed, emissions from animal agriculture in the advanced world have declined for the past 30 or 40 years. However emissions from developing countries have massively increased, fueled by their own domestic production and importation of beef.


Industrialized countries should set an example and stop consuming so much beef. They should also help developing countries use more efficient means of production, and penalize countries that are high emitters.

But while about 15% of total global emissions are from livestock, in industrialized countries like America, only 4% of American emissions are from agriculture, and we have little role in the emissions from developing countries.

This is because about 92% of beef consumed in the United States is from American-produced beef. We simply don’t import very much beef, and the beef we do import is mainly from Canada and Mexico.

What this means is that if Americans want to combat global emissions from themselves, they should use less transport energy and less electricity. That is the bulk of emissions from Americans.

If they want to combat emissions and deforestation from China and Brazil, they should support economic and foreign policies that penalize China and Brazil for their bad and worsening environmental records and incentivize better behavior.

There is a notion that if Americans consumed less beef, we could send our soybeans to other countries so they could avoid deforestation. This is really questionable, though, because it implies that Americans should change our behavior to stop bad behavior from others–as if by being held hostage.

The fact is, Americans should consume less beef and reforest. And Brazilians and Chinese should also consume less beef and also reforest and stop deforestation.

Everyone should do something, but Americans are responsible for their own bad behavior, not the bad behavior of others. The bad behavior of Americans is mainly in using too much electricity and too many cars. The bad behavior of Brazilians is in burning down their own forests, largely for production of beef that they themselves consume.

Not every bad thing is linked to Americans eating meat. When you want to make everything about One Thing, that is called an ideology. Ideologies are bad because they distort and prevent behavior that can actually achieved the desired outcomes.

— Kevin

Like what I write? Support me at

Find me on Twitter and Instagram.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.