Gut-derived appetite hormones do not explain energy intake differences in humans following low-carbohydrate versus low-fat diets

Abstract

Objective: The objective of this study was to explore how dietary macronutrient composition influences postprandial appetite hormone responses and subsequent energy intake.

Methods: A total of 20 adults (mean [SEM], age 30 [1] years, BMI 27.8 [1.3] kg/m², n = 8 with normal weight, n = 6 with overweight, n = 6 with obesity) consumed a low-fat (LF) diet (10% fat, 75% carbohydrate) and a low-carbohydrate (LC) diet (10% carbohydrate, 75% fat) for 2 weeks each in an inpatient randomized crossover design. At the end of each diet, participants consumed isocaloric macronutrient-representative breakfast test meals, and 6-h postprandial responses were measured. Ad libitum energy intake was measured for the rest of the day.

Results: The LC meal resulted in greater mean postprandial plasma active glucagon-like peptide-1 (GLP-1; LC: 6.44 [0.78] pg/mL, LF: 2.46 [0.26] pg/mL; p < 0.0001), total glucose-dependent insulinotropic polypeptide (GIP; LC: 578 [60] pg/mL, LF: 319 [37] pg/mL; p = 0.0004), and peptide YY (PYY; LC: 65.6 [5.6] pg/mL, LF: 50.7 [3.8] pg/mL; p = 0.02), whereas total ghrelin (LC: 184 [25] pg/mL, LF: 261 [47] pg/mL; p = 0.0009), active ghrelin (LC: 91 [9] pg/mL, LF: 232 [28] pg/mL; p < 0.0001), and leptin (LC: 26.9 [6.5] ng/mL, LF: 35.2 [7.5] ng/mL; p = 0.01) were lower compared with LF. Participants ate more during LC at lunch (244 [85] kcal; p = 0.01) and dinner (193 [86] kcal; p = 0.04), increasing total subsequent energy intake for the day compared with LF (551 [103] kcal; p < 0.0001).

Conclusions: In the short term, endogenous gut-derived appetite hormones do not necessarily determine ad libitum energy intake.

Figure 1.

Postprandial responses to isocaloric low carbohydrate (LC) or low fat (LF) meals following habituation to each diet in a randomized crossover design. Mean (range) of energy in the test meals was 777 (532 to 1043) kcal. Data are mean ± SEM. n=20. p-values from paired t-test of mean postprandial plasma concentrations. (A) active glucagon-like peptide-1 (GLP-1), (B) total glucose-dependent insulinotropic polypeptide (GIP), (C) peptide YY (PYY), (D) leptin, (E) total ghrelin, (F) active ghrelin.

Figure 2.

Total energy intake (EI), and intake from lunch, dinner, and snacks throughout the day after isocaloric low carbohydrate (LC) or low fat (LF) meals following habituation to each diet in a randomized crossover design. Mean (range) of energy in the test meals was 777 (532 to 1043) kcal. Data are mean ± SEM and individual responses. n=20. p-values from paired t-test or Wilcoxon test. (A) Energy intake (kcal), (B) Relative Total EI as a proportion of 24h energy expenditure (24hEE), sedentary energy expenditure (SedEE), and sleeping energy expenditure (SEE), (C) Mass intake (g).

Figure 3.

Comparisons between dietary macronutrient induced changes in gut hormone responses and pharmacological or bariatric surgery induced changes. (A) estimated mean active GLP-1 steady state average exposure concentrations, C_avg, achieved by low carbohydrate (LC) or low fat (LF) diet were orders of magnitude lower than the both oral and subcutaneous semaglutide using values from Overgaard et al. (B) peak active GLP-1 and PYY concentrations following a LC or LF test meal were orders of magnitude lower than peak concentrations observed during a mixed-meal test following Roux-en-Y Gastric Bypass surgery (RYGB) using data from Tan et al. For 3A, data are median ± 90% exposure ranges, for 3B, data are mean ± SEM.