Dietary carbohydrates: role of quality and quantity in chronic diseaseBMJ 2018; 361 doi: https://doi.org/10.1136/bmj.k2340 (Published 13 June 2018) Cite this as: BMJ 2018;361:k2340
All rapid responses
Currently, there is considerable interest in the duplication of the AMY1 gene and its evolutionary history, but there are surprisingly few good studies of its physiological role in humans. In 4 linked studies (American Journal of Clinical Nutrition, 2018, in press), Atkinson et al. showed that starchy foods are digested faster and produce higher postprandial glycaemia in individuals with high AMY1 copy number (CN).
Briefly, in Study 1, we genotyped 201 healthy subjects and determined glucose tolerance, insulin sensitivity, salivary -amylase activity, BMI and macronutrient intake. As expected, AMY1 CN correlated positively with salivary -amylase activity (r = 0.62, p < 0.0001, n = 201) but not with BMI, glucose tolerance or insulin sensitivity. Interestingly, mean CN was higher in Asian individuals vs people of European descent (~8 vs 6 copies, respectively). In Study 2, a pool of 114 subjects tested 6 starchy foods, 3 sugary foods, 1 mixed meal and 2 reference glucose solutions, containing either 50 g or 25 g of available carbohydrate. CN was strongly correlated with normalized glycaemic responses to all starchy foods (explaining 26-61% of inter-individual variation), but not to sucrose or fruit.
In Study 3, we compared glycaemic and insulin responses to 2 starchy foods vs glucose in 40 individuals at extremes of high and low AMY1 CN. Individuals in the highest vs lowest decile of CN produced modestly higher glycaemia (+15%, p = 0.018), but not insulinaemia, after consuming 2 starchy foods. In Study 4, we compared breath hydrogen and methane responses over 8 h in 30 individuals at extremes of CN. Low CN individuals displayed >6-fold higher breath methane levels in the fasting state, and after starch ingestion, than high CN individuals (p = 0.001), suggesting differences in large bowel flora and metabolism.
These studies represent the first large-scale, systematic analysis of the physiological and phenotypic impact of AMY1 CN variation in healthy individuals. They raise interesting questions for future research, specifically in relation to the carbohydrate intake for health and ‘precision nutrition’.
Competing interests: No competing interests
Ludwig et al in their comprehensive and cogent review importantly discuss patient factors which impact upon the metabolic effect of carbohydrate (1). However the authors state that those with a high copy number of salivary amylase gene (AMY1), and hence greater salivary amylase quantity and activity, experience higher insulin and glucose spikes following a starch load. However the evidence does not unequivocally support this.
Mandel et al found almost the exact opposite. Those with a higher copy number variant enjoyed a lower post-prandial serum glucose level. This was mediated by an earlier insulin spike in this cohort (2). The reference proffered by Ludwig et al to support their assertion originates not from peered review literature, like the work of Mandel, but rather is a presentation at a scientific meeting. Further a review of studies to date suggest that a those with a high copy number of salivary amylase gene AMY1 tend to have a lower risk of obesity, insulin resistance and diabetes (4)(5). Patient factors may significantly affect the physiological ramifications a high carbohydrate diet in conjunction with glycaemic index and glycaemic load.
(1) Ludwig DS, Hu FB, Tippy L, Brand-Miller J. Dietary carbohydrates: role of quality and quantity in chronic disease. BMJ. 2018 Jun 13;361:k2340
(2) Mandel, A., Breslin, P. (2012). High endogenous salivary amylase activity is associated with improved glycemic homeostasis following starch ingestion in adults. Journal of Nutrition, 142, 853–858
(3) Atkinson FS, Hancock D, Petocz Physiological significance of higher amy1 gene copy number on postprandial responses to starchy foods in caucasian adults. Journal of Nutrition & Intermediary Metabolism2014;1:15doi:10.1016/j.jnim.2014.10.044
(4) Fernández CI, Wiley AS. Rethinking the starch digestion hypothesis for AMY1 copy number variation in humans. Am J Phys Anthropol. 2017; 163:645-657.
(5) Falchi M, El-Sayed Moustafa JS, Takousis P, Pesce F, Bonnefond A, Andersson-Assarsson JC, Sudmant PH, Dorajoo R, Al-Shafai MN, Bottolo L, Ozdemir E, So HC, Davies RW, Patrice A, Dent R, Mangino M, Hysi PG, Dechaume A, Huyvaert M, Skinner J, Pigeyre M, Caiazzo R, Raverdy V, Vaillant E, Field S, Balkau B, Marre M, Visvikis-Siest S, Weill J, Poulain-Godefroy O, Jacobson P, Sjostrom L, Hammond CJ, Deloukas P, Sham PC, McPherson R, Lee J, Tai ES, Sladek R, Carlsson LM, Walley A, Eichler EE, Pattou F, Spector TD, Froguel P. Low copy number of the salivary amylase gene predisposes to obesity. Nat Genet. 2014 May;46(5):492-7.
Competing interests: No competing interests
Dear Sir / Madam,
Thank you very much for your thought provoking series on nutrition. I read Ludwig et al.’s paper: Dietary Carbohydrates: role of quality and quantity in chronic disease with interest and felt attention should be drawn to two important potential inaccuracies.
1. The conclusion that low carbohydrate diets produce greater weight loss than lower fat diets is based on studies that compared very low to low carbohydrate intake to moderately low fat intake and thus lower significantly overall calorie consumption in the case of the former (e.g. Bueno et al., 2013 compared diets providing <10% of energy from carbohydrates with diets providing <30% of energy from fats and Mansoor et al., 2016 compared diets starting with <4% and increasing to <10% of energy intake in the forms of carbohydrates with diets mostly providing < 30% of energy intake from fats). In contrast, the low fat whole food plant based diets used by Esselstyn (1999) and by Ornish (1990) to reverse coronary artery disease, provided no more than 10% of energy intake from fat.
2. Like all publications on low carbohydrate diets the authors concluded that long term data on the safety of low carbohydrate diets are lacking. This is no longer true and in the absence of very long RCTs, one has to rely on morbidity as well as mortality data from prospective cohort studies. In the first meta-analysis on low carbohydrate diets and all cause mortality, Noto et al. (2013) confirmed that those with high scores for low carbohydrate diet had higher all cause mortality (RR = 1.31, 95% CI 1.07 – 1.59), with a non significant trend towards higher risk of cardiovascular mortality (RR = 1.10, 95% CI 0.98 – 1.24). Separate analyses examining low carbohydrate or high protein score resulted in similar results.
Fung and colleagues (2010) followed 85,168 women from 1980 and 44,548 men from 1986 until 2006. They documented 12,555 deaths (including 2,458 cardiovascular deaths & 5,780 cancer deaths) and established increased mortality risk associated with low carbohydrate diets, especially animal food based low carbohydrate diets, after adjusting for confounders:
For low carbohdrate diets overall HR All Cause Mortality 1.12 (95% CI 1.01 – 1.24). For animal product based low carbohydrate diets HR All Cause Mortality 1.23 (95% CI 1.11 – 1.37) p = 0.051; HR Cardiovascular Mortality 1.14 (95% CI 1.01 – 1.29) p = 0.029; and HR Cancer Mortality 1.28 (95% CI 1.02 – 1.60) p = 0.089 . For vegetable based low carbohydrate diets HR All Cause Mortality 0.80 (95% CI 0.75 – 0.85) p < 0.001; and HR Cardiovascular Mortality 0.77 (95% CI 0.68 – 0.87) P < 0.001
An earlier cohort of Swedish women (42,237 adults, followed up for 12 years) by Lagiou et al.(2012) found that low carbohydrate – high protein diets were associated with higher total as well as cardiovascular mortality. Decreasing carbohydrate by one decile increased total mortality by 6% (RR = 1.06, 95% CI 1.00 – 1.12) and increased cardiovascular mortality by 13% (RR = 1.13, 95% CI 0.96 – 1.32). Increasing protein by one decile increased total mortality non significantly (RR = 1.02, 95% CI 0.99 – 1.05) and increased cardiovascular mortality by 16% (RR = 1.16, 95% CI 1.05- 1.29).
An updated report by Lagiou et al. (2012) after 15.7 years follow up of 43,396 Swedish women, confirmed higher risk of cardiovascular disease for every tenth decrease in carbohydrate intake (IRR = 1.04, 95% CI 1.00 – 1.08), or every tenth increase in protein intake (IRR = 1.04, 95% CI 1.02 – 1.06) or every 2 units increase in low-carbohydrate-high protein score (IRR = 1.05, 95% CI 1.02 – 1.08).
It is important to start highlighting the potentially very significant health risks associated with low carbohydrate (and thus high protein and / or high fat) diets that depend on increased intake of animal products.
Dr Werner Pretorius
Consultant Liaison Psychiatrist
NHS Lothian Diabetes Mental Health Service
Bueno NB, de Melo IS, de Oliveira SL, da Rocha, Ataide T. (2013). Very-low-carbohydrate ketogenic diet v. low-fat diet for long-term weight loss: a meta-analysis of randomised controlled trials. British Journal of Nutrition . 110:1178.
Esselstyn, C. (1999). Updating a 12-Year Experience with Arrest and Reversal Therapy for Coronary Heart Disease (an Overdue Requiem for Palliative Cardiology). American Journal of Cardiology. 84, 339-41.
Fung, T., van Dam, R., Hankinson, S., Stampfer, M, Willett, W., & Hu, F. (2010). Low Carbohydrate Diets and All cause and Specific Mortality: Two cohort studies. Annals of Internal Medicine. 153(5), 288-298.
Lagiou, P., Sandin, S., Lof, M., Trichopoulos, D., Adami, H., & Weiderpass, E. (2012). Low carbohydrate-high protein diet and mortality in a cohort of Swedish wome. Journal of Internal medicine. 261(4), 366-374.
Lagiou, P., Sandin, S., Lof, M., Trichopoulos, D., Adami, H. O., & Weiderpass, E. (2012). Low carbohydrate-high protein diet and incidence of cardiovascular diseases in Swedish women: prospective cohort study. Bmj, 344, e4026.
Mansoor N, Vinknes KJ, Veierød MB, Retterstøl K. (2016). Effects of low-carbohydrate diets v. low-fat diets on body weight and cardiovascular risk factors: a meta-analysis of randomised controlled trials. British Journal of Nutrition. 115, 466–479.
Noto, H., Goto, A., Tsujimoto, T., & Noda, M. (2013). Low-carbohydrate diets and all-cause mortality: A systematic review and meta-analysis of observational studies. PLOS ONE. 8(1), e55030.
Ornish et al. (1990). Can Lifestyle Changes Reverse Coronary Heart Disease? Lancet. 336, 129-33.
Competing interests: No competing interests