Metabolic & Cardiometabolic · Gut & Microbiome
sourdough long fermentation degrades wheat fructans lowering FODMAP load
In plain terms: Does slow-fermented sourdough bread lower the FODMAPs that upset sensitive guts, and is it better tolerated?
Strong support Metabolic & Cardiometabolic 🐭 Non-human evidence🔬 Includes disconfirming
RefutedContestedStrong support
consensus score 0.82
Yes — long sourdough (and yeast) fermentation measurably degrades wheat fructans/FODMAPs, and small human trials show the resulting bread is better tolerated in wheat-sensitive/IBS eaters, though effect size and clinical magnitude are modest.
📅 Last reviewed: 2026-07-15 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: Cells in a dish (In-vitro)
MechanismIn-vitroAnimalObservationalRCTMeta-analysis
How the studies fall
13 support 1 contradict 0 tested null 1 mixed · 15 sources, 11 independent groups
What the evidence shows
Long sourdough fermentation measurably degrades wheat fructans, which are essentially the whole FODMAP load of wheat bread, and each additional day of fermentation lowers them further. The evidence is mostly food-chemistry measurement rather than clinical: the human tolerance trials are small, so the laboratory effect is better established than the symptom benefit.
The evidence (17)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Pejcz E, Lachowicz-Wiśniewska S, Nowicka P, Wojciechowicz-Budzisz A, Harasym J 2023 · Foods | in-vitro | supports | moderate | Wheat breads were baked with sourdough fermented 24, 48 or 72 hours - spontaneous, or inoculated with the lactic-acid bacteria L. casei or L. plantarum - and fructans measured in the loaf. Bread made without sourdough held the most: 0.22 g fructans and 0.23 g FODMAPs per 100 g. By 72 hours fructans fell to 0.09-0.10 g/100 g however the sourdough was fermented, a 58-62% cut against that control. At 24 hours the inoculated breads (0.16-0.17 g/100 g) held more fructans than the spontaneous one, so fermentation time rather than the added strain drove the fall. One flour lot, duplicate assays. |
| Habuš M et al 2022 · Foods | in-vitro | supports | moderate | Wheat bran contained 2.64% fructans by dry weight, amaranth bran 0.96%. Fermenting the bran for 24 h reduced fructans in every treatment: a commercial sourdough starter of mixed lactic acid bacteria and yeast cut wheat-bran fructans by 93% to 0.18% dw, bakery yeast plus Limosilactobacillus fermentum by 83% to 0.33% dw, and bakery yeast alone by 63%. The fermentation itself released fructose and formed mannitol in every arm, so those have to be tracked; the estimated total FODMAP content of a 30 g baked snack made with the bran still stayed below the IBS cut-off. |
| Shewry PR, America AHP, Lovegrove A, Wood AJ, Plummer A, Evans J, van den Broeck HC, Gilissen L, Mumm R, Ward JL, Proos Z, Kuiper P, Longin CFH, Andersson AAM, Philip van Straaten J, Jonkers D, Brouns F 2022 · Food Chemistry | in-vitro | supports | low | Wholemeal flours blended from bread wheat, emmer and spelt were baked into bread by both yeast and sourdough fermentation, with composition measured at the flour, dough and bread stages. Fructans and raffinose - described by the authors as the major FODMAPs the grain itself contributes - were reduced by both yeast and sourdough fermentation, with yeast having the greater effect. Sourdough also raised organic acids and mannitol, itself a FODMAP. This is bench food chemistry on the bread; no people were tested. |
| Pejcz E, Spychaj R, Gil Z 2019 · Foods 2019;8(12):663 | in-vitro | supports | moderate | Factorial bake: light (450) vs wholegrain (1650) flour x yeast-only vs 24 h pure-culture L. plantarum sourdough (50% of flour) + yeast x 60-150 min final fermentation; fructan by Megazyme kit in two replicates. Sourdough breads 33% less fructan than yeast-only; 150 vs 60 min fermentation −37% (mean 0.44 → 0.28 g/100 g d.m.). Lowest 0.23/0.25 g/100 g (sourdough, 150 min), 60%/56% below yeast-only 60 min. Only fructans assayed — mannitol, which the authors note Lactobacillus can form from released fructose, was not measured. Sourdough breads scored worse sensorially. |
| Pejcz E, Lachowicz-Wiśniewska S, Nowicka P, Wojciechowicz-Budzisz A, Spychaj R, Gil Z 2021 · Molecules | in-vitro | supports | moderate | In wheat sourdough from a single flour lot, fructans were essentially the whole FODMAP load, and every extra day of fermentation cut them further. Fructan fell from 1.15 g per 100 g dry matter unfermented to 0.18 with spontaneous fermentation, 0.10 with Lactobacillus plantarum and 0.07 with L. casei after 72 hours - a 91 to 93 percent drop in total FODMAPs. Inoculation was no better than spontaneous fermentation; fermentation time drove the effect. Measured in the dough, not in baked bread, with two analytical replicates per value. |
| Laatikainen R, Koskenpato J, Hongisto SM, Loponen J, Poussa T, Huang X, Sontag-Strohm T, Salmenkari H, Korpela R 2017 · Nutrients | in-vitro | supports | moderate | Randomised, double-blind, parallel-group 7-day trial: 26 adults with IBS (13 per arm) ate 150 g/day of either a wheat sourdough bread fermented >12 h or a yeast-fermented wheat bread made from the same flour. By enzymatic assay the sourdough bread held 0.06 g fructans/100 g vs 0.23 g/100 g in the yeast bread - 'the FODMAP content of our sourdough wheat bread was reduced by 74%'. Fructan intake from bread was 0.08 (IQR 0.07-0.09) vs 0.31 (0.24-0.3) g/day, p < 0.001. The breads also differed in recipe (added gluten, improvers) and baking scale, so fermentation is not isolated. |
| Calasso M, Francavilla R, Cristofori F, De Angelis M, Gobbetti M 2018 · Nutrients 2018;10(12):1873 | in-vitro | contradicts | moderate | Wheat flour was fermented 8 h at 30 °C with ten selected lactobacilli plus fungal proteases, then made into bread and pasta and assayed with a Megazyme fructan kit. Fructan content was not significantly lower after that fermentation: bread 0.96 vs 0.90 g/100 g (p = 0.182), against a control that was itself a sourdough bread, and pasta 0.74 vs 0.70 g/100 g (p = 0.099), against unfermented flour. The fermentation was tuned to halve gluten, not fructans, and the authors write that no conclusion can be drawn for FODMAPs. |
| Koj K, Pejcz E 2023 · Molecules 2023;28(4):1910 | in-vitro | mixed | moderate | Rye wholemeal sourdoughs were fermented 24 and 48 h - spontaneously or inoculated with S. boulardii, L. plantarum, L. rhamnosus or B. coagulans - then baked, against a control bread made without sourdough (1% lactic acid instead). Fructans fell during every sourdough fermentation, and from 3.82% in the flour to 0.62-1.46% in the breads. But the finished sourdough breads matched the no-sourdough control (1.32, 1.31, 1.15, 1.33 vs 1.46% dry matter); only S. boulardii went significantly lower, to 0.62%. Two replicates per value. |
| Menezes LAA, De Marco I, Neves Oliveira Dos Santos N, Costa Nunes C, Leite Cartabiano CE, Molognoni L, Pereira GVM, Daguer H, De Dea Lindner J 2021 · Int J Food Sci Nutr | in-vitro | supports | moderate | Six screened lactic-acid bacteria (L. brevis, W. minor, L. plantarum, Leuc. citreum, L. fermentum, C. farciminis) used as type II sourdough starters for wheat bread. Sourdough breads 'showed a significant reduction of fructans' vs baker's-yeast-only bread; 'Including specific cultures as starters in sourdough reduced fructans content by >92%' - the >92% is attributed to specific cultures, and which strains, the per-strain values and the replicate basis cannot-tell from the abstract. Total fructans are the only FODMAP class the abstract reports quantifying (mannitol, excess fructose not reported). No people studied; the 'suitable for IBS patients' conclusion is the authors' inference, not a tested outcome. |
| Fraberger V, Call LM, Domig KJ, D'Amico S 2018 · Nutrients | in-vitro | supports | moderate | Thirteen yeast strains from traditional sourdoughs each fermented a wheat-flour suspension for 8 h 15 min. Total fructans fell in every strain but by very different amounts: 10-30% for Candida species, Hanseniaspora uvarum, Torulaspora pretoriensis and Wickerhamomyces anomalus; about 50% for Kazachstania exigua; 60-80% for baker's yeast, Torulaspora delbrueckii and Kluyveromyces marxianus; and 83% for a Saccharomyces cerevisiae sourdough isolate. Fructose rose markedly as the fructans broke down. Yeasts were tested alone, with no lactic acid bacteria and no baked bread. |
| Pitsch J, Sandner G, Huemer J, Huemer M, Huemer S, Weghuber J 2021 · Foods | in-vitro | supports | high | Rye and wheat sourdoughs were back-slopped weekly for 105 days and their fructo-oligosaccharide (FOS) content measured by HPLC against the flours they were made from. Wheat sourdough lost 60% of short-chain (DP 2-7) and 99% of long-chain (DP >7) FOS at 4 C, and 66%/79% at 10 C; rye lost 92%/80% at 4 C and 89%/98% at 10 C (all p<0.0001). Parallel batches at 20 C and 30 C moulded and were abandoned. The figures are for dough, not baked bread, and the rye sourdough also built up mannitol, itself a FODMAP, to about 21 g/kg. |
| Sokmen O, Dundar AN, Bagci U, Togay SÖ, Sahin OI, Saricaoglu FT 2025 · Food Science & Nutrition | in-vitro | supports | moderate | Wheat doughs fermented 24 h at 25 C with single Lactobacillus isolates from traditional Turkish sourdoughs, compared with a baker's-yeast control dough (not unfermented flour) holding 612 ppm fructan and 3710 ppm FODMAPs (glucose + fructose + mannitol). Nine of twelve isolates left significantly less fructan than the yeast control: Balikesir-1Y-1 (L. plantarum) 318 ppm (48% lower) with FODMAPs 40% lower; Tekirdag-1K-2 and Edirne-1K-2 (L. parabuchneri) 382-387 ppm fructan (37%) with FODMAPs about 64% lower. Three isolates (Balikesir-1Y-2, Bursa-1Y, Sakarya-3Y) did not differ from control on fructan; Bursa-1Y cut FODMAPs 74% yet left fructans at 624 ppm. One 24 h fermentation, doughs only, no bread baked; Table 3 and Table 4 FODMAP figures for the same isolates disagree. |
| Acín Albiac M, Di Cagno R, Filannino P, Cantatore V, Gobbetti M 2020 · Microbial Cell Factories | in-vitro | supports | moderate | Durum wheat flour model doughs (DY 160) fermented 24 h at 30 °C, triplicate: from 1003 ± 40 mg fructans/100 g flour, conventional sourdough LAB cut fructans ~56% (L. plantarum DC400) and ~40% (F. sanfranciscensis SD8); unstarted dough unchanged. Fructophilic bee strains 75-81%; with invertase 3 h, conventional pair 83-87%, FLAB complete. F. sanfranciscensis and FLAB released mannitol (a FODMAP); L. plantarum did not. Doughs only; no bread baked or eaten. |
| Cuffe MS, Staudacher HM, Aziz I, Adame EC, Krieger-Grubel C, Madrid AM, Ohlsson B, Black CJ, Ford AC 2025 · Lancet Gastroenterol Hepatol | meta-analysis | supports | high | Network meta-analysis of 28 randomised trials (2338 patients) comparing whole-diet interventions against habitual diet in irritable bowel syndrome. A low FODMAP diet reduced global IBS symptoms (RR of symptoms not improving 0·51, 95% CI 0·37-0·70; from 24 trials) and bloating or distension (0·55, 0·37-0·80; from 23 trials); for bowel habit no diet was superior to any control. The interventions were restrictions of FODMAP-containing foods across the whole diet; the paper does not report bread type, sourdough or any fermentation process, and measures no fructan content. |
| Reidzane S, Gramatina I, Galoburda R, Komasilovs V, Zacepins A, Bljahhina A, Kince T, Traksmaa A, Klava D 2022 · Foods | in-vitro | supports | moderate | Hull-less barley and wholegrain wheat breads were fermented 4 h with spontaneous barley sourdough, germinated barley sourdough or baker's yeast, and fructans measured enzymatically. Wholegrain wheat flour held 1.47% fructans by dry weight; the yeast-fermented wheat bread fell to 0.27% dw, a decrease of 81.6%, while both sourdough wheat breads kept significantly more fructan than the yeast loaf. In barley bread the spontaneous barley sourdough gave the largest fall, 89.6%. The authors conclude fructans decreased faster with yeast than with barley sourdough. |
| Di Stasio L, De Caro S, Marulo S, Di Renzo T, Ferranti P, Reale A, Mamone G 2025 · Foods. 2025 Apr 20;14(8):1418 | in-vitro | supports | moderate | Neapolitan pizza dough raised with a sourdough starter (L. brevis, F. sanfranciscensis, Leuconostoc pseudomesenteroides plus yeast) or with baker's yeast alone, fermented up to 48 h; fructans measured in the baked pizza before and after simulated digestion. Fructans fell with fermentation time in both, faster with sourdough: by 24 h, 0.93 to 0.29 g/100 g with sourdough (p < 0.0001) vs 0.91 to 0.36 with yeast alone (p = 0.0003); after simulated digestion, 0.22 vs 0.42 g/100 g at 16 h. By 48 h both reached about 0.3 g/100 g. Protein digestibility was unchanged by starter or time. |
| Loponen, Gänzle 2018 · Foods | mechanism | supports | low | Narrative review of FODMAP metabolism in dough fermentation; it reports no measurements of its own. Straight-dough processing leaves '1-1.5% fructans in wheat bread and about 3% in rye bread'; yeast invertase reaches only fructans of 'a DP of four or less', while heterofermentative lactobacilli turn released fructose into the polyol mannitol at '10-20 mmol/kg in wheat and 50 mmol/kg in rye'. The authors conclude conventional fermentations 'result in decreased levels of FODMAPs but may generate FODMAPs', so near-zero-FODMAP bread needs fructanase-positive, mannitol-fermenting strains. |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.