Perspective | Published:

Classification of obesity targeted personalized dietary weight loss management based on carbohydrate tolerance

European Journal of Clinical Nutritionvolume 72pages13001304 (2018) | Download Citation

Introduction

There is a huge variability in weight loss responses to diets in overweight and obese individuals, and 3–6 months weight loss on a low-fat or a low-carb diet typically varies from a gain of few kg to a 15–20 kg weight loss. Efforts to develop pretreatment prognostic markers to identify the obese patients who will benefit from a particular weight loss have been unsuccessful. The lack of identifiable biological predictors has been interpreted as being due to differences in dietary adherence and psychosocial factors, and some yet unknown factors (genes, epigenetics, pollutants, etc.) [1]. Moreover, randomized controlled trials (RCT) comparing diets differing in macronutrient content, e.g., low-fat, low-carb, or high protein, have not identified one diet that is substantially superior to others. This has led to the view that all calories are equal, and that focus should be on using willpower and practical tools to reduce caloric intake.

However, recent discoveries have shown that the large variability in weight loss response observed in obese individuals prescribed the same diet is partly due to alterations in the mechanisms linking glucose metabolism to appetite control, and that prevalent physiological alterations in glucose metabolism are responsible for powerful differences in responsiveness to carbohydrate intake ranging from induction of marked satiety signals to almost absent satiety effect. The discovery of this interplay between glucose metabolism and diet composition has opened a novel clinical management strategy that can improve dietary weight loss in overweight, obese, and diabetic individuals.

As a clinical implication, we have discovered that individual weight loss responsiveness to diets depends on glucose metabolic traits that clinically can be characterized by fasting glucose and insulin levels for each patient before initiation of treatment. Based on these measurements an optimal diet composition can be tailored to enhance satiety, adherence, and weight loss.

The role of glucose and insulin in meal-induced satiety

As part of the established appetite regulatory system the central nervous system (CNS) monitors blood glucose, i.e., the glucostatic regulation. This states that glucose utilization by critical cells in the hypothalamus generates a signal to brain areas controlling appetite and food intake. When brain glucose uptake and utilization is decreasing or low, hunger is elicited and eating ensues if food is available. As the eating episode progresses glucose in the blood increases, leading to increased hypothalamic glucose utilization, ultimately causing the individual to feel satiated and to stop eating [2]. In vivo evidence of the physiological role in normal appetite regulation in humans has previously been difficult to obtain for methodological reasons, although strong associations between postprandial responses in plasma glucose and insulin responses, and satiety and food intake on the other hand, have been observed. In the study by Verdich et al. [3], we observed a strong inverse association between postprandial insulin response to a breakfast meal (50% of energy from carbohydrate) and subsequent spontaneous energy intake at an ad libitum lunch meal (Fig. 1). The gastrointestinal hormones GLP-1 and GIP were also associated with appetite and energy intake, but in a stepwise regression analysis the integrated postprandial insulin response was found to explain 67% of the variation in ad libitum energy intake, leaving no further contribution by other hormones. We also found that this relation was blunted among the obese individuals. In a subsequent individual data meta-analysis of 136 normal weight, overweight, and obese individuals, we found no association between postprandial glucose increases and appetite and subsequent energy intake, but strong associations between postprandial insulin responses and appetite and subsequent energy intake [4]. Again, these findings were present among the normal weight individuals and were blunted among the overweight and obese individuals. These early findings suggest that insulin or insulin-mediated glucose uptake is an important determinant of meal-induced satiety and for energy intake at a subsequent meal in the normal insulin-sensitive individuals, and that these effects are blunted in overweight and obese individuals. This suggests a role for insulin resistance in weight regulation.

Fig. 1
Fig. 1

Relationship between 3 h insulin response following a 2.5 MJ test meal (abscissa) and ad libitum energy intake at lunch (ordinate) in 12 lean male subjects. Reproduced from Verdich et al. [3]

Altered glucostatic appetite regulation in insulin resistance

There is increasing recognition of the brain as an important organ of insulin-mediated glucose uptake, and that the brain is also an important contributor to the whole-body insulin resistance that occurs with increasing levels of body fatness and inactivity. Moreover, a number of adverse CNS effects, caused by impaired brain glucose uptake due to insulin resistance, have been linked to cognitive decline and reduced memory function in type 2 diabetics [5]. Persons with diabetes have a higher risk of developing Alzheimer’s disease and other dementias, which strongly suggests that insulin resistance not only exerts adverse effects on peripheral tissues and organs, but also involves the CNS, with consequences for important brain functions [6]. There is strong evidence to suggest that the adverse effects of insulin resistance also affect appetite regulation, e.g., genetic deletion of the brain insulin receptor causes central insulin resistance and obesity in rodents [7].

A recent study used magnetic resonance (MR) spectroscopy scanning in the occipital lobe to measure the change in intracerebral glucose levels during a 2-h hyperglycemic clamp [8]. The intracerebral glucose uptake was significantly different across groups during hyperglycemia after controlling for age and sex. Compared with lean participants, brain glucose increments were lower in participants with obesity and type 2 diabetes (Fig. 2), and there was a positive association between perceived satiety/fullness and cerebral glucose uptake.

Fig. 2
Fig. 2

Intracerebral glucose uptake across groups during hyperglycemia after controlling for age and sex. Normal weight, obese, and type 2 diabetic participants underwent 1H MR spectroscopy scanning to measure change in intracerebral glucose levels during a 2-h hyperglycemic clamp (glucose ~220 mg/dl). Numbers presented are when plasma and intracerebral glucose levels had reached steady state (between 60–120 min). Individuals with obesity (n = 10) and those with T2DM (n = 6) had significantly reduced increments in brain glucose concentrations compared with lean controls (n = 9), and individuals with poorly controlled T2DM showed a further blunting of brain glucose levels compared with obese individuals. Illustration based on information by Hwang et al. [8]

These results indicate that glucose uptake in the brain is compromised in obese and further in T2D individuals. In conclusion, a blunted carbohydrate-induced satiety may operate in insulin-resistant individuals due to impaired postprandial cellular glucose uptake in the brain and perhaps also in other organs and tissues. Consequently, important pathways in normoglycemic individuals may cause carbohydrates to induce rapid satiety, and the impaired or delayed CNS glucose uptake diminished in prediabetic and diabetic individuals may be responsible for overconsumption of calories when challenged by carbohydrate-rich foods.

Linking glucose metabolism to clinical outcomes

Our hypothesis is that meals containing carbohydrate may be very satiating in insulin-sensitive overweight (type A), less so in more insulin resistant obese prediabetics (type B), and even less in obese individuals with type 2 diabetes (type C) (Table 1). In diabetic patients satiety signals may be more dependent on other satiety hormones, e.g., CCK, GLP-1, and PYY, released mainly in response to fat and protein reaching the small intestine [9]. Naturally, the capacity to increase insulin secretion to overcome insulin resistance in response to a meal should also play a role [3]. If this is the case, high-carbohydrate (low fat) meals would be more effective for producing weight loss in normoglycemic obese [10] and less effective in prediabetic obese individuals, while type 2 diabetics would benefit more from diets with reduced carbohydrate content (and more fat and protein) [11,12,13].

Table 1 A simple classification of overweight and obese individuals based on fasting glucose concentration

Clinical evidence for use of glucose metabolic biomarkers to classify obesity

Over the past several decades numerous trials have compared low-fat, high-carbohydrate diets with low-carbohydrate, high-fat diets for the management of overweight and obesity, assuming that a single dietary strategy is appropriate for all individuals. These comparisons have shown very similar weight loss on low-fat and low-carb diet, and many scientists now say that macronutrient composition does not really matter. However, we have reanalyzed a number of these trials and have shown that if one considers each patient’s glucose metabolism and based on fasting glucose classify them into normoglycemic (type A), prediabetic (type B), and type 2 diabetic (type C), the optimal diet and weight loss responsiveness are completely different [14, 15]. Our studies show that providing specific diets for weight management based on pretreatment glycemic (and insulinemic) status holds great promise for moving forward with personalized nutrition.

In the NUGENOB study, a RCT comparing hypocaloric low-fat (high carb) versus low-carb (high fat) diet in 770 obese individuals for 10 weeks, we initially reported that both diets produced the same mean weight loss of 7.5 kg [16]. However, the reanalysis showed that type A subjects lost 0.4 kg more on the low-fat than on the low-carb diet (P = 0.03); the diets were similarly effective for type B, but the low-carb diet produced a 2.0 kg more (P = 0.07) than the low-fat diet of the diabetic obese [14]. The diabetic subgroup, however, was not large; so we reanalyzed the large-scale Spanish PREDIMED trial that enrolled individuals with overweight and obesity to investigate primary prevention of cardiovascular disease with an ad libitum Mediterranean diet without any focus on weight management [17]. In a reanalysis of this study we confirmed that type 2 diabetics over 5 years lost significantly more weight on this high fat, low-carbohydrate diet compared to normoglycemic participants (1.6 kg versus 0.2 kg, P < 0.001) [18].

The results in type 2 diabetics are not surprising. We have previously argued that the diet recommendations for type 2 diabetics should not be the same as for the healthy normal weight population [11], and more and more systematic reviews and meta-analyses point in that direction [13]. Carbohydrate restriction, as part of a healthy, palatable diet seems to produce immediate improvements in glycemic control and better satiety [12], which is likely to improve weight control in the long run. However, there is clearly a need for better trials with more controlled and palatable diets to increase adherence and assess efficacy.

There is also good news for prediabetic obese subjects (type B). Their appetite control is clearly much more susceptible to the diet’s glycemic index, fiber, and whole grain content. To illustrate the effect size, we reanalyzed the previously published Diogenes trial [19]. The initially reported effects of diet group on body weight in the total population of overweight and obese were quite small: after a weight loss of 11 kg on an 800 kcal/day diet, the weight-reduced subjects on low glycemic load diet regained 1.9 kg less than those on the high glycemic load diet over the subsequent 6 months. However, when the subjects were divided into subgroups based on their baseline fasting glucose levels [15], consistent and more pronounced effects of the different diets were detected. Prediabetics regained mean 5.8 kg more on the high than low glycemic-load diet, whereas normoglycemics regained 1.4 kg more (Fig. 3). This 4.4 kg difference in weight change between normoglycemic and prediabetic obese over 6 months was highly statistically significant (P = 0.001), and extremely clinically important.

Fig. 3
Fig. 3

Changes in bodyweight among participants during an initial 8-week low calorie diet followed by a 26-week weight maintenance diet high or low in glycemic load. a Normoglycemic participants. b Participants with prediabetes. Based on data from Hjorth et al. Am J Clin Nutr 2017 [14]

The findings in the DioGenes study were subsequently confirmed in the SHOPUS study in which we examined the effect on cardiovascular health effect of either a new Nordic Diet based on Nordic foods, loaded with fruit, vegetables, fiber, and whole grains or an Average Danish Diet (Western control diet). All foods were provided free of charge from a supermarket at the University campus with scanning of barcodes insuring dietary composition according to assigned intervention diets [20]. When stratifying according to type A and B, we found that type B lost 6.0 kg more on the new Nordic Diet than on the control diet, whereas type A individuals lost only 2.2 kg more, the group difference of 3.8 kg being highly significant (P = 0.001) [14].

Notably, the study designs of DioGenes, PREDIMED, and SHOPUS were typically ad libitum, so no caloric restrictions were prescribed to the participants. As there was no difference in changes in physical activity, this quite substantial decrease in spontaneous energy intake must be due to differences in satiation and satiety, but the exact mediation by substrates or neuro-hormonal pathways to influence appetite control needs further studies.

Future perspectives

The observations that markers of insulin resistance and secretion capacity are so important for the role of carbohydrate in inducing weight loss (or weight gain in type 2 diabetics) suggest that cellular glucose uptake and intracellular levels may be the link between carbohydrate ingestion, attenuated satiety along with deterioration of glucose metabolism, caloric intake, and weight changes.

We believe that there is sufficient current evidence to systematically conduct prospective trials of personalized nutrition for weight loss based on these three classifications of overweight and obese subjects.

It is obvious that the classification into three groups (Table 1) is rather simplistic and that more subtypes are likely to be identified, especially when also taking fasting insulin into consideration. Recent subclassifications of type 2 diabetes [21, 22] will be relevant to study in relation to identifying the optimal diet for each category. It is also obvious that the definition and classification of the prediabetic and type 2 diabetic states is highly dependent on the use of different criteria (HgbA1c or plasma glucose, and cut-offs). In our analyses we compared different biomarkers of insulin metabolism, and found the combination of fasting glucose and insulin to provide the best predictor. However, future studies will definitely sophisticate the battery of biomarkers.

These observations call for more studies to demonstrate the pathways involved, and also how differences in the organ and tissue sites of insulin resistance contribute. In a number of trials we have found that fasting plasma insulin, a marker of insulin secretion capacity, adds further information to identify the optimal diet for the three types [14, 15], e.g., so for the prediabetic individuals with low insulin are more carbohydrate intolerant than those with higher insulin.

However, little is known about the contribution of different organs and tissues to link between impaired glucose uptake and satiety. Is it only brain insulin resistance and the ability to counterbalance with increased insulin resistance that is important for appetite control, or are other tissues also producing satiety hormones signaling their degree of glucose uptake and stores to the brain?

There is also a need for prospective randomized trials to validate the clinical findings, we have made based on reanalysis of existing trials, and to further develop the understanding of the importance of the microbiota and its interplay with glucose metabolism.

References

  1. 1.

    Elobeid MA, Allison DB. Putative environmental-endocrine disruptors and obesity: a review. Curr Opin Endocrinol Diabetes Obes. 2008;15:403–8.

  2. 2.

    Woods SC, Metabolic signals and food intake. Forty years of progress. Appetite. 2013;71:440–4.

  3. 3.

    Verdich C, Toubro S, Buemann B, Lysgard Madsen J, Juul Holst J, Astrup A. The role of postprandial releases of insulin and incretin hormones in meal-induced satiety-effect of obesity and weight reduction. Int J Obes. 2001;25:1206–14.

  4. 4.

    Flint A, Gregersen NT, Gluud C, Møller BK, Raben A, Tetens I, et al. Associations between postprandial insulin and blood glucose responses, appetite sensations and energy intake in normal weight and overweight individuals: a meta-analysis of test meal studies. Br J Nutr. 2007;98:17–25.

  5. 5.

    Riederer P, Korczyn AD, Ali SS, Bajenaru O, Choi MS, Chopp M, et al.The diabetic brain and cognition. J Neural Transm. 2017;124:1431–54. https://doi.org/10.1007/s00702-017-1763-2.Epub ahead of print.

  6. 6.

    Heni M, Wagner R, Kullmann S, Gancheva S, Roden M, Peter A, et al. Hypothalamic and striatal insulin action suppresses endogenous glucose production and may stimulate glucose uptake during hyperinsulinemia in lean but not in overweight men. Diabetes. 2017;66:1797–806. https://doi.org/10.2337/db16-1380

  7. 7.

    Brüning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, et al. Role of brain insulin receptor in control of body weight and reproduction. Science. 2000;289:2122–5.

  8. 8.

    Hwang JJ, Jiang L, Hamza M, Sanchez Rangel E, Dai F, Belfort-DeAguiar R, et al. Blunted rise in brain glucose levels during hyperglycemia in adults with obesity and T2DM. JCI Insight. 2017;2:95913 https://doi.org/10.1172/jci.insight.95913.

  9. 9.

    Belza A, Ritz C, Sørensen MQ, Holst JJ, Rehfeld JF, Astrup A. Contribution of gastroenteropancreatic appetite hormones to protein-induced satiety. Am J Clin Nutr. 2013;97:980–9.

  10. 10.

    Astrup A, Hjorth MF. Low-fat or low carb for weight loss? It depends on your glucose metabolism. EBioMedicine. 2017;22:20–21.

  11. 11.

    Feinman RD, Pogozelski WK, Astrup A, Bernstein RK, Fine EJ, Westman EC, et al. Dietary carbohydrate restriction as the first approach in diabetes management: critical review and evidence base. Nutrition. 2015;31:1–13.

  12. 12.

    Samkani A, Skytte MJ, Kandel D, Kjaer S, Astrup A, Deacon CF, et al. A carbohydrate-reduced high-protein diet acutely decreases postprandial and diurnal glucose excursions in type 2 diabetes patients. Br J Nutr. 2018;119:910–7. https://doi.org/10.1017/S0007114518000521

  13. 13.

    Snorgaard O, Poulsen GM, Andersen HK, Astrup A. Systematic review and meta-analysis of dietary carbohydrate restriction in patients with type 2 diabetes. BMJ Open Diabetes Res Care. 2017;5:e000354 https://doi.org/10.1136/bmjdrc-2016-000354

  14. 14.

    Hjorth MF, Ritz C, Blaak EE, Saris WH, Langin D, Poulsen SK, et al. Pretreatment fasting plasma glucose and insulin modify dietary weight loss success: results from 3 randomized clinical trials. Am J Clin Nutr. 2017;106:499–505. https://doi.org/10.3945/ajcn.117.155200

  15. 15.

    Hjorth MF, Due A, Larsen TM, Astrup A, Pretreatment fasting plasma glucose modifies dietary weight loss maintenance success: results from a stratified RCT. Obesity. 2017;25:2045–8.https://doi.org/10.1002/oby.22004.

  16. 16.

    Petersen M, Taylor MA, Saris WHM, Verdich C, Toubro S, Macdonald I, et al. Randomized, multi-center trial of two hypo-energetic diets in obese subjects: high- versus low-fat content. Int J Obes. 2006;30:552–60. NUGENOB Consortium

  17. 17.

    Estruch R, Ros E, Salas-Salvadó J, Covas MI, Corella D, Arós F, Gómez-Gracia E, Ruiz-Gutiérrez V, Fiol M, Lapetra J, Lamuela-Raventos RM, Serra-Majem L, Pintó X, Basora J, Muñoz MA, Sorlí JV, Martínez JA, Martínez-González MA, PREDIMED Study Investigators. Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J Med. 2013;368:1279–90.

  18. 18.

    Estruch R, Corella D, Salas-Salvado J, Hjorth MF, Astrup A, Zohar Y, et al. Pretreatment fasting plasma glucose determines weight loss on high-fat diets: the PREDIMED study. Abstract. ADA; 2017.

  19. 19.

    Larsen TM, Dalskov S, van Baak M, Jebb S, Papadaki A, Pfeiffer AFH, et al.Diet, Obesity, and Genes (Diogenes) Project. Diets with high or low protein content and glycemic index for weight-loss maintenance. N Engl J Med. 2010;362:2102–13.

  20. 20.

    Poulsen SK, Due A, Jordy AB, Kiens B, Stark KD, Stender S, et al. Health effect of the new nordic diet in adults with increased waist circumference: a 6-mo randomized controlled trial. Am J Clin Nutr. 2014;99:35–45.

  21. 21.

    Udler MS, Kim J, von Grotthuss M, Bonàs-Guarch S, Mercader JM, Cole JB, et al. Clustering of type 2 diabetes genetic loci by multi-trait associations identifies disease mechanisms and subtypes. Preprint at bioRxiv: https://doi.org/10.1101/319509. 2018.

  22. 22.

    Vistisen D, Witte DR, Brunner EJ, Kivimäki M, Tabák A, Jørgensen ME, et al. Risk of cardiovascular disease and death in individuals with prediabetes defined by different criteria: the Whitehall II Study. Diabetes Care. 2018;41:899–906. https://doi.org/10.2337/dc17-2530

  23. 23.

    American Diabetes Association. Standards of medical care in diabetes—2016. Diabetes Care. 2016;39(Suppl 1):S1–S106.

Download references

Author information

Affiliations

  1. Department of Nutrition, Exercise and Sports, University of Copenhagen, Copenhagen, Denmark

    • Arne Astrup
    •  & Mads F. Hjorth

Authors

  1. Search for Arne Astrup in:

  2. Search for Mads F. Hjorth in:

Conflict of interest

AA and MFH are co-inventers on a pending provisional patent application on the use of biomarkers for prediction of weight-loss responses, and co-founders/owners of the University of Copenhagen spin-out company Personalized Weight Management Research Consortium ApS (Gluco-diet.dk). AA is consultant or a member of advisory boards for Basic Research, USA; Beachbody, USA; BioCare Copenhagen, Denmark; Gelesis, USA; Groupe Éthique et Santé, France; McCain Foods Limited, USA; Nestlé Research Center, Switzerland; and Weight Watchers, USA. AA and MFH are co-authors of a number of diet/cookery books, including personalized nutrition for weight loss, published in several languages.

Corresponding author

Correspondence to Arne Astrup.

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/s41430-018-0227-6