Effect of inulin on breath hydrogen, postprandial glycemia, gut hormone release, and appetite perception in RYGB patients: a prospective, randomized, cross-over pilot study

Roux-en-Y gastric bypass (RYGB) effectively treats obesity and type 2 diabetes, implicating gastrointestinal mechanisms and gut peptides (GLP-1, PYY) in regulating satiation and meal-related glycemia. The role of the large intestinal microbiome and its metabolites, particularly short-chain fatty acids (SCFAs: acetate, propionate, butyrate), remains unclear after RYGB. Postoperative studies report altered gut microbiota and reduced fecal SCFAs, despite SCFAs being mostly absorbed in the colon and only a small fraction excreted. In obesity, fecal SCFAs are often elevated, and animal data suggest increased energy harvest from SCFAs. Under typical physiology, meals reach the colon after 3–6+ hours, challenging the involvement of SCFAs in acute, meal-related responses; instead, fermentable fibers may exert a second-meal effect on glycemia and appetite via colonic SCFA production. The authors hypothesized that RYGB anatomical changes accelerate transit of fermentable fiber to the large intestine, leading to more rapid fermentation and SCFA production, enhancing GLP-1 and PYY secretion and improving meal-related glycemia and appetite. They tested this by comparing inulin versus equicaloric maltodextrin (MDX) before and ~8 months after RYGB, assessing breath hydrogen, plasma SCFAs, gut hormones, glycemia, and appetite.

Prior research indicates RYGB alters gut microbial composition and metabolic activity with reductions in fecal SCFAs, though most SCFAs are absorbed, complicating interpretation of fecal measures. Obesity is associated with altered SCFAs and potential increased energy harvest from SCFAs in animal models. In healthy individuals, the timing of colonic fermentation suggests SCFAs may influence second-meal rather than immediate postprandial responses. Studies of fermentable fibers report second-meal effects on glycemia and appetite, and longer-term inulin supplementation can increase breath hydrogen and GLP-1/PYY while reducing hunger. Acute studies found inulin can raise circulating SCFAs at 4–6 h without increasing GLP-1/PYY, and resistant starch may lower second-meal glycemia without raising SCFAs similarly. Mechanistically, SCFAs can activate GPR41/GPR43 on enteroendocrine cells to modulate appetite and glycemia. After RYGB, increased GLP-1 and PYY and improved glycemia are well documented. These data frame the current study’s focus on whether accelerated fermentation post-RYGB influences acute and second-meal responses.

Design: Explorative prospective, randomized, cross-over pilot study in 8 patients with obesity (1 male, 7 female) undergoing RYGB. Each participant was studied twice preoperatively and again 6–8 months postoperatively, receiving in random order on separate days one of two test drinks after an overnight fast (>10 h). At t=180 min, a fixed snack was given to assess second-meal effects. Interventions: Test drink: 300 ml orange juice (27 g carbohydrate sugars) plus either 25 g oligofructose-enriched inulin (Orafti Synergy1, ~92% inulin) or an equicaloric 15.5 g maltodextrin (MDX). Consumption within 5 min. At 180 min, snack: cereal bar plus 125 ml orange juice (~150 kcal; 12 g sugars) to stimulate gastrocolic reflex. Surgery details: Laparoscopic RYGB with small gastric pouch, 30 mm linearly stapled gastrojejunostomy, 150 cm antecolic alimentary limb, 70 cm biliopancreatic limb; same surgical team. Measurements: Blood samples at baseline and regular intervals over 300 min for plasma glucose (glucose oxidase), insulin (chemiluminescent microparticle immunoassay), total GLP-1 and total PYY (ELISA). Breath hydrogen via ambulatory hydrogen breath test. Plasma SCFAs quantified by UHPLC-MS/MS; LOQ set at 3.3× LLOD (s/n ≥ 33). Appetite perceptions (satiety, hunger, fullness, desire to eat, prospective consumption, nausea) via validated 100-mm visual analog scales. Rheology: viscosity of test solutions measured at 4, 25, 37 °C; inulin drink showed significantly higher viscosity at 37 °C vs MDX. Statistics: No power calculation (unknown effect size). Preliminary checks for outliers (box plot), normality (Shapiro–Wilk), and homoscedasticity (Levene). Linear mixed models (LMMs) with maximum likelihood estimation were fitted with within-subject predictors (test condition and time), random intercepts, and compared nested models with/without interaction using chi-squared tests. Estimated marginal means and pairwise contrasts with Bonferroni adjustment assessed: (i) pre- vs post-op within inulin or MDX conditions and (ii) within-group differences from baseline at each time. Wilcoxon signed-rank tests assessed pre- vs post-op differences in BMI, fasting glucose, and fasting insulin. Correlations: AUCs (trapezoidal) relating second-meal desire-to-eat (210–300 min) to early-phase (60–180 min) or second-meal (210–300 min) breath hydrogen using Pearson’s r. Analyses in R (v4.0.3) with emmeans; figures organized in Adobe Illustrator. Ethics: Approved by Cantonal Ethical Committee of Zurich (BASEC-Nr. 2017-02287). Trial registered at ClinicalTrials.gov (NCT03573258). Written informed consent obtained.

• Anthropometrics and fasting measures: BMI decreased from 41.5 ± 3.3 to 29.4 ± 4.2 kg/m² (P < 0.001); fasting glucose from 5.8 ± 0.7 to 5.1 ± 0.4 mmol/l (P = 0.046); fasting insulin from 25.4 ± 9.5 to 8.0 ± 7.1 mU/ml (P = 0.005) at ~8 months post-RYGB.
• Blood glucose: Pre-op, inulin increased glucose at 30 min; MDX at 30 and 60 min (P ≤ 0.05). Post-op, glucose increments were reduced: with inulin at 30 and 60 min and with MDX at 60 and 90 min vs pre-op (P ≤ 0.05). Notably, post-op inulin elicited no significant increase from baseline at 30 min, whereas MDX still did at 30 min, indicating a stronger early glucose-lowering effect with inulin.
• Plasma insulin: Pre-op increases with inulin at 30 min and after snack (210 min); with MDX between 30–90 min (P ≤ 0.05). Post-op, insulin increments were reduced: with inulin between 30–90 min and 210–240 min; with MDX between 30–120 min and at 240 min (P ≤ 0.05).
• Gut hormones: Pre-op, neither inulin nor MDX affected GLP-1 or PYY. Post-op, both inulin and MDX significantly increased total GLP-1 and PYY at 30 min vs baseline (P ≤ 0.05). PYY at 30 and 60 min was higher post-op vs pre-op (P ≤ 0.05). No increase in GLP-1 or PYY after the 180-min snack either pre- or post-op.
• Breath hydrogen: MDX had no effect pre- or post-op. Inulin increased breath hydrogen pre-op from 180–300 min vs baseline (P ≤ 0.05). Post-op, the inulin-induced increase began earlier (150–300 min vs baseline; P ≤ 0.05) but was less pronounced in magnitude, indicating accelerated fermentation after RYGB.
• Plasma SCFAs: Lactate, acetate, 2-hydroxybutyrate, and isovalerate detected in most subjects, but no effects of inulin, MDX, or surgery. Propionate, butyrate, isobutyrate, 2-methylbutyrate, valerate, and hexanoate were inconsistently detected.
• Appetite: Pre-op, desire-to-eat increased from 240–300 min vs baseline with MDX (P ≤ 0.05) but not with inulin. Post-op, desire-to-eat ratings were reduced for both treatments; the reduction appeared more pronounced with inulin after the 180-min snack, with ratings falling below baseline (not observed with MDX). Other appetite ratings showed similar patterns without significant differences.
• Correlation: Post-op, second-meal desire-to-eat AUC (210–300 min) inversely correlated with early-phase breath hydrogen AUC (60–180 min) with inulin (R = -0.85, p = 0.007). No other significant correlations were found.
• Model performance: Interaction terms improved LMM fits for most outcomes (e.g., glucose, insulin, GLP-1, PYY, appetite) with conditional R² ranging from ~66% to 85%; for breath hydrogen, improved only for inulin. Overall, RYGB enhanced early GLP-1/PYY responses and reduced glycemic and insulinemic excursions and appetite. Inulin fermentation occurred earlier post-op, and inulin appeared to potentiate early glucose-lowering and second-meal appetite suppression despite no detectable plasma SCFA changes.

The study tested whether RYGB accelerates the delivery and fermentation of inulin in the large intestine, thereby increasing SCFA production, stimulating GLP-1/PYY, and improving meal-related glycemia and appetite. Findings show that RYGB augmented early GLP-1 and PYY responses and reduced postprandial glucose, insulin, and appetite with both inulin and MDX. Inulin-specific effects included earlier onset of breath hydrogen increases post-op, indicating accelerated colonic fermentation, and an apparent potentiation of early-phase glucose lowering and second-meal appetite suppression compared with MDX. The strong inverse correlation between early-phase breath hydrogen and second-meal desire-to-eat post-op supports a physiological link between accelerated fermentation and appetite suppression after RYGB. Absence of measurable increases in plasma SCFAs or snack-induced GLP-1/PYY suggests either that acute, single-dose interventions are insufficient to elevate circulating SCFAs post-RYGB or that SCFAs are extensively metabolized before reaching systemic circulation. Technical factors (e.g., EDTA tubes and analytical sensitivity) may also limit SCFA detection. The more pronounced early glycemic effect with inulin likely reflects increased beverage viscosity at body temperature slowing intestinal glucose absorption and attenuating insulin secretion, rather than a hormonal mechanism, consistent with literature on viscous fibers. In post-RYGB anatomy, where gastric emptying control is altered, chyme viscosity may be a key determinant of early postprandial glycemia. Collectively, results indicate RYGB accelerates inulin fermentation without detectable systemic SCFA changes, and prebiotic inulin may enhance the surgery’s benefits on early glycemia and later appetite. Longer-term fiber supplementation and extended sampling may be required to reveal SCFA-mediated endocrine effects in this population.

RYGB accelerates large intestinal fermentation of inulin, as evidenced by earlier breath hydrogen rises, and augments early GLP-1/PYY secretion while reducing postprandial glucose, insulin, and appetite. Although no increases in circulating SCFAs or snack-induced GLP-1/PYY were detected, inulin appeared to further potentiate early-phase glucose lowering and second-meal appetite suppression post-RYGB, with appetite improvements correlating inversely with early breath hydrogen. These findings suggest that accelerated fermentation after RYGB may confer additional appetite-suppressive benefits and that prebiotic fibers could be a useful adjunct in post-bariatric dietary management. Future research should include larger, sex-balanced cohorts; longer-term prebiotic supplementation; extended postprandial sampling to capture second-meal effects; assessment of additional hormones (e.g., ghrelin, CCK) and splanchnic GLP-1 activity; optimization of SCFA measurement protocols; and exploration of viscous fibers (e.g., beta-glucans, pectins) to leverage viscosity-driven glycemic benefits post-RYGB.

• Small sample size (n=8) with sex imbalance (7 female, 1 male) limits statistical power and generalizability; potential sex differences in glycemic/incretin responses not addressed.
• Acute study with sampling limited to 5 hours; may miss later second-meal effects and longer-term adaptations to prebiotic fermentation.
• Plasma SCFA detection challenges: extensive first-pass metabolism, low systemic levels, use of EDTA tubes (potential acetate contamination), and analytical sensitivity may have obscured true changes.
• No measurements of other relevant hormones (e.g., ghrelin, CCK, leptin) or regional (splanchnic) GLP-1 activity that could mediate appetite/glycemic effects.
• Viscosity differences between inulin and MDX beverages at 37 °C introduce a confounding mechanism (slowed absorption) for early glycemic effects.
• Pilot nature with multiple comparisons despite corrections; findings need confirmation in larger trials.