Plant-rich mixed meals based on Paleolithic diet principles have a dramatic impact on incretin, peptide YY and satiety response, but show little effect on glucose and insulin homeostasis: an acute-effects randomized study

Abstract: There is evidence for health benefits from Paleolithic diets; however, there are a few data on the acute effects of rationally designed Paleolithic-type meals. In the present study, we used Paleolithic diet principles to construct meals comprising readily available ingredients: fish and a variety of plants, selected to be rich in fiber and phyto-nutrients. We investigated the acute effects of two Paleolithic-type meals (PAL 1 and PAL 2) and a reference meal based on WHO guidelines (REF), on blood glucose control, gut hormone responses and appetite regulation. Using a randomized cross-over trial design, healthy subjects were given three meals on separate occasions. PAL2 and REF were matched for energy, protein, fat and carbohydrates; PAL1 contained more protein and energy. Plasma glucose, insulin, glucagon-like peptide-1 (GLP-1), glucose-dependent insulinotropic peptide (GIP) and peptide YY (PYY) concentrations were measured over a period of 180 min. Satiation was assessed using electronic visual analogue scale (EVAS) scores. GLP-1 and PYY concentrations were significantly increased across 180 min for both PAL1 (P= 0·001 and P< 0·001) and PAL2 (P= 0·011 and P= 0·003) compared with the REF. Concomitant EVAS scores showed increased satiety. By contrast, GIP concentration was significantly suppressed. Positive incremental AUC over 120 min for glucose and insulin did not differ between the meals. Consumption of meals based on Paleolithic diet principles resulted in significant increases in incretin and anorectic gut hormones and increased perceived satiety. Surprisingly, this was independent of the energy or protein content of the meal and therefore suggests potential benefits for reduced risk of obesity.

Free Full-text

Alex’s Notes: The human diet has changed dramatically in a very short period of time. What started as a high-protein, energy-dilute diet based on fruits, vegetables, nuts, and meat changed to one based on cereal grains 7,000 to 10,000 years ago with the advent of agriculture. More recently, just in the last century, our diets became even more dissociated from our evolutionary roots with industrialization and more intensive food processing methods that bestowed to us various seed oils and packaged foods.

This isn’t to say that agriculture, industrialization, or intensive food processing are bad, but rather it should serve as a wake-up call that perhaps we have moved too far from what many propose to be a nutrition strategy better suited for our genome. Consider the differences between our diet and indigenous tribes in the Amazon for example; people who consume no less than 130 different species of food foraged for in the rainforest.

With the increasing rates of chronic diseases, there has been great discussion lately on returning to a “Paleolithic” diet as a possible solution, especially knowing that diet plays a central role in all aspects of health. In the last ten years, several published articles have suggested Paleolithic diets to improve blood lipids and blood pressure in healthy individuals and those with metabolic syndrome in as little as two weeks, improve glycemic control independently of changes in waist circumference and to a greater degree than a Mediterranean diet, promote greater satiety per caloric eaten than a Mediterranean diet, reduce weight and blood lipids levels more than a Nordic diet, and result in weight loss without intentional caloric restriction. Despite these promising results, no studies have yet looked into the “why” portion of the equation.

Thus, the current study investigated the acute effects of the consumption of Paleolithic meals on metabolic processes influencing glucose tolerance.

Ultimately, 21 healthy, normal-weight male subjects came to the researchers’ laboratory three times over six weeks (2-week washout between trials) after a 12-hour overnight fast to consume one of three different meals and have blood-borne measurements made for the following three hours.

Total Energy (kcal) 380 550 380
Protein, g (%) 13 (14%) 41 (29%) 16 (17%)
Fat, g (%) 11 (26%) 18 (28%) 11 (26%)

Carbohydrate, g (%)

Fiber, g

57 (60%)


66 (43%)


65 (57%)


Foods used in meal Long-grain white rice, mango, carrots, farmed Atlantic salmon, & olive oil Strawberries, apples, yellow sweet peppers, sweet onion, eggplant, mushrooms, farmed Atlantic salmon, raisins, courgettes, cinnamon, capers, flax seed oil, haddock (PAL1 only), & blanched almonds (PAL1 only)

The table above summarizes the three test meals. The REF meal was based on WHO macronutrient recommendations, and the PAL2 was designed to match it with Paleolithic-based ingredients. PAL1 was based on protein and fat ratios that were suggested to by typical of hunter-gatherers. Needless to say, none of the meals use low-quality processed foods. Even the REF is basically paleo with the exception of the rice, and the use of whole foods that were prepared the day of testing is a strong aspect of this study.

Three meals later

In contrast to other studies, there were no significant differences among the meals for the total blood glucose or insulin responses over the two hour post-meal period. However, this makes sense considering that the available carbohydrate in each meal was very similar. Nonetheless, there were differences in the pattern of glucose and insulin levels. All groups had similar peak glucose concentrations by 30 minutes, but the peak insulin response at this time was significantly greater in the REF (+28%) and PAL2 (+17%) meals compared to PAL1. This led to a rapid fall of blood glucose levels to below fasting values by the 60 minute mark, which returned to baseline by 90 minutes in the REF group and stayed below fasting levels in the PAL1 condition throughout the 120 minutes. Conversely, PAL2 declined to fasting levels by the 60 minute mark and stayed there. There were also no differences among conditions with regard to insulin sensitivity.

For gut hormones, over the entire three hour post-meal period, there were no differences in ghrelin or amylin, but GLP-1 and PYY were dramatically higher and GIP significantly lower for both PAL meals compared to the REF. For satiety, both PAL conditions were consistently rated as more filling than the REF condition, although the participants like the REF meal more than the PAL meals.

Together, the data suggests that some aspect of the PAL meals renders glucose less accessible.

This is only a hypothesis though, because glucose absorption and uptake were not measured. Nonetheless, while delayed gastric emptying alone could explain the reduced GIP response for the PAL meals, the elevated GLP-1 response suggests the delivery of unabsorbed glucose to the lower duodenum. The authors speculate that a possible explanation for this could be flavonoids in the PAL meals, inhibiting glucose absorption in the jejunum.

All meal conditions used whole-foods with the only real differences between the REF and PAL meals being a greater reliance on fruit and fibrous vegetables for carbohydrates instead of white rice. This led to significantly greater amounts of fiber and polyphenols, which together likely accounted for the differences in the blood glucose response. Surprisingly, PAL1 did not fare better than PAL2 despite the greater amount of protein and was in fact less satiating per kcal. Overall, this preliminary study suggests that meals containing more vegetables may aid in glucose management simply through increasing the amount of polyphenols and fiber consumed. The fact that PAL1 and PAL2 were “paleo” is not entirely relevant since the only non-paleo thing about the REF group was the white rice.

Network Affiliates

Quick Links I

Our Location

2908 Brownsboro Rd
Suite 103
Louisville, KY 40206
(502) 690-2200

SHR Newsletter

Subscribe to our FREE newsletter
to receive the latest updates in your inbox!
SHR Newsletter
Internet Radio