Xanthan gum (E415)

References

1. Regulator

   EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS), Mortensen A, Aguilar F, Crebelli R, Di Domenico A, Frutos MJ, et al. Re-evaluation of xanthan gum (E 415) as a food additive EFSA Journal, 2017.

   doi:10.2903/j.efsa.2017.4909PubMed 32625570

   Funding and finding

   Funding

   European Food Safety Authority; requestor: European Commission; question number EFSA-Q-2011-00514.

   Finding

   Xanthan gum is a high molecular weight polysaccharide produced by pure-culture fermentation of a carbohydrate with strains of Xanthomonas campestris and recovered with ethanol or propan-2-ol; a backbone of beta-(1->4) linked D-glucose carries on every second glucose a C3 trisaccharide side chain of beta-D-mannose-(1->4)-beta-D-glucuronic acid-(1->2)-alpha-D-mannose, with pyruvate on 30 to 50% of the terminal mannose residues as a 4,6-acetal and acetyl on 60 to 70% of the internal mannose residues; reported molecular weight about 1,000,000 to 2,000,000 g/mol; CAS 11138-66-2. Authorised in 79 EU food categories, mostly at quantum satis as a Group I additive, with 10,000 mg/kg in jams, jellies and marmalades individually or combined with other listed gums, 1,200 mg/L in dietary foods for infants and young children for special medical purposes (categories 13.1.5.1 and 13.1.5.2) and no use in jelly mini-cups. Xanthan gum is unlikely to be absorbed intact and is not metabolised by gastrointestinal enzymes, but is partially fermented by intestinal microbiota to short-chain fatty acids; in rats given 14C-labelled xanthan gum, unchanged or only slightly modified polysaccharide accounted for about 98% of faecal radioactivity and a maximum of 15% of the label was metabolised to carbon dioxide within 100 hours. Refined non-brand-loyal exposure reached 64 mg/kg body weight per day in children aged 3 to 9 years, 38 mg/kg bw per day for child consumers of food supplements and, for brand-loyal consumers of foods for special medical purposes, 115 mg/kg bw per day in infants aged 12 weeks to 11 months. Repeated intake by adults of up to 15,000 mg per person per day (214 mg/kg bw per day) for at least ten days was well tolerated, with abdominal discomfort in some individuals, judged undesirable but not adverse. The Panel concluded that there is no need for a numerical ADI and no safety concern at the refined exposure; the re-evaluation does not apply to infants under 12 weeks of age. Human studies cited: five men took 10.4 to 12.9 g per day (149 to 184 mg/kg bw per day) for 23 consecutive days, with raised faecal wet and dry weight, shorter transit time, serum cholesterol 10% lower (p < 0.05) and higher faecal bile acid concentrations (p < 0.05) (Eastwood et al., 1987); 18 healthy men took 15 g per day for 10 days, with increases in stool output (p < 0.01), frequency of defaecation (p < 0.05) and flatulence (p < 0.01) and variable effects on transit time (Daly et al., 1993). Studies submitted by Abbott Nutrition in neonatal piglets dosed 0, 375, 750 or 3,750 mg/kg bw per day for 20 days from postnatal day 2 gave a NOEL of 375 mg/kg bw per day from faecal changes and a NOAEL of 750 mg/kg bw per day from minimal to moderate intestinal histopathology at the high dose; post-marketing surveillance by Abbott Nutrition of hydrolysate formulae containing about 750 mg/L over an estimated 131 million patient treatment days reported fewer than one adverse event report per 10,000 patient treatment days. JECFA (1986) and the SCF (1990) allocated an ADI 'not specified'. Use as an ingredient of food for patients with dysphagia is noted. Reports of 22 infants who developed late-onset necrotising enterocolitis after a xanthan gum-containing thickener (Beal et al., 2012; Woods et al., 2012) were judged not related to the food additive use of xanthan gum in the manufacture of infant formula, the ingested doses being unknown but expected in gram amounts. Recommendations: revise the lead limit and add limits for mercury, cadmium and arsenic in the EU specifications, and collect usage data, since Mintel lists xanthan gum in more products than industry reported.

2. Independent research

   Ostrowski MP, La Rosa SL, Kunath BJ, Robertson A, Pereira G, Hagen LH, et al. Mechanistic insights into consumption of the food additive xanthan gum by the human gut microbiota Nature Microbiology, 2022.

   doi:10.1038/s41564-022-01093-0PubMed 35365790

   Funding and finding

   Funding

   US National Institutes of Health (DK118024 and DK125445 to ECM, UL1TR002240 to MPO, AI124255 to VBY); Research Council of Norway, FRIPRO programme (250479 to PBP and SLLR, 302639 to LHH); University of Michigan Center for Gastrointestinal Research (NIDDK 5P30DK034933) for proteomics; work at the US Department of Energy Joint Genome Institute, a DOE Office of Science User Facility, under contract DE-AC02-05CH11231.

   Finding

   In an initial survey of faecal cultures from 80 adults under gentamicin selection, one culture degraded xanthan gum; in a second survey of 60 adults without antibiotic, about half of the cultures grew on xanthan gum and lost its viscosity. Across 20 degrading communities passaged ten times, the only operational taxonomic unit present above 0.5% in all of them was an uncultured Ruminococcaceae genus, R.UCG13, whose locus was conserved at above 95% amino acid identity. A secreted R.UCG13 glycoside hydrolase family 5 enzyme, RuGH5a, cleaves the xanthan backbone at the reducing end of the unbranched glucose, yielding pentasaccharide repeating units, and requires its three additional domains for activity; cell-free culture supernatant released pentameric and decameric oligosaccharides, while the intracellular fraction was needed for further saccharification; it hydrolyses native and lyase-treated xanthan with comparable specificity, unlike earlier xanthanases that prefer lyase-treated substrate by 600-fold or more. A polysaccharide lyase family 8 enzyme acted on the oligosaccharides but not on the polymer, removing the terminal pyruvylated mannose; a GH88 removed the resulting unsaturated glucuronic acid, two GH38 enzymes hydrolysed the trisaccharide, two carbohydrate esterases removed acetyl groups and a GH94 phosphorylase cleaved cellobiose, completing saccharification. Bacteroides intestinalis isolated from the same culture grew on the oligosaccharides but not on polymeric xanthan gum, and every gene of its polysaccharide utilisation locus rose more than 100-fold on the oligosaccharides compared with glucose. The R.UCG13 locus was detected in faecal metagenomes of all industrialised populations sampled and in none of the Yanomami, Hadza or Burkina Faso datasets, whose small size leaves false negatives possible; an expanded search of 21,762 metagenomes, 102,860 genome bins and 72,491 isolate genomes returned 35 hits for the locus, all in human microbiome datasets, including senior adults, children and a 12-month-old infant; the B. intestinalis locus was absent from the Japanese and Danish or Spanish datasets. In germ-free mice colonised with a human microbiota containing R.UCG13, its expansion depended on dietary xanthan gum, and a gavaged oligosaccharide-consuming B. intestinalis strain displaced the endogenous strain in the xanthan gum group. A xanthan utilisation locus with similar architecture was found in mouse faecal metagenomes.

3. Regulator

   Joint FAO/WHO Expert Committee on Food Additives. Xanthan gum: specification monograph, in Compendium of food additive specifications, Addendum 7 FAO Food and Nutrition Paper 52, Addendum 7, 53rd meeting of JECFA, Rome, 1 to 10 June 1999, 1999.

   https://www.fao.org/4/X3860E/X3860E41.htm

   Funding and finding

   Funding

   Food and Agriculture Organization of the United Nations and World Health Organization.

   Finding

   Xanthan gum carries INS No. 415 and CAS number 11138-66-2. Functional uses: thickener, stabiliser, emulsifier, foaming agent. Definition: a high molecular weight polysaccharide gum produced by pure-culture fermentation of a carbohydrate with Xanthomonas campestris, purified by recovery with ethanol or isopropanol, dried and milled; it contains D-glucose and D-mannose as the dominant hexose units along with D-glucuronic acid and pyruvic acid, is prepared as the sodium, potassium or calcium salt, and its solutions are neutral. ADI 'not specified', established at the 30th JECFA in 1986.

4. Independent research

   Chazelas E, Deschasaux M, Srour B, Kesse-Guyot E, Julia C, Alles B, et al. Food additives: distribution and co-occurrence in 126,000 food products of the French market Scientific Reports, 2020.

   doi:10.1038/s41598-020-60948-wPubMed 32132606

   Funding and finding

   Funding

   PhD grant from Paris 13 University (E. Chazelas); post-doctoral fellowship from the Fondation pour la Recherche Médicale (ARF201809007046, M. Deschasaux); French Ministry of Health (DGS Arrêté 29 Novembre 2019); French National Cancer Institute (INCa_14059); European Research Council (ERC CoG ADDITIVES_864219). Researchers were independent from funders. Funders had no role in the study design, the collection, analysis, and interpretation of data, the writing of the report, and the decision to submit the article for publication. Open Food Facts is a non-profit association.

   Finding

   53.8% of products contained at least one additive and 11.3% five or more. Appendix 7, products containing each additive: citric acid E 330 18,302; lecithins E 322 13,102; E 14XX modified starches 10,100; sodium carbonates E 500 8587; ascorbic acid E 300 7919; xanthan gum E 415 7327; sodium nitrite E 250 7144; diphosphates E 450 6529; pectins E 440 6504; mono- and diglycerides of fatty acids E 471 6060. Specific starches: E 1442 209, E 1422 173, E 1414 55, E 1450 30, E 1412 27, E 1420 19, E 1404 18, E 1440 5, E 1410 1. The footnote and the Discussion state that the number of products containing each specific modified starch may be underestimated because packaging often states only 'modified starch'. E 500 had the highest squared loading (0.64) in cluster 3, 'stabilizers and emulsifiers mostly used in biscuits and cakes'; sodium carbonate (E 500), diphosphates (E 450) and lecithins (E 322) co-occurred in 1667 products, mostly biscuits. The authors describe E 500, E 450, E 503 and E 422 as widely used in biscuits and cakes as acidity regulators, stabilizers and emulsifiers. Additive doses were not available.

5. Independent research

   Chazelas E, Druesne-Pecollo N, Esseddik Y, de Edelenyi FS, Agaesse C, De Sa A, et al. Exposure to food additive mixtures in 106,000 French adults from the NutriNet-Santé cohort Scientific Reports, 2021.

   doi:10.1038/s41598-021-98496-6PubMed 34608173

   Funding and finding

   Registration

   NCT03335644

   Participants

   106,489

   Funding

   NutriNet-Santé: Ministère de la Santé, Santé Publique France, INSERM, INRAE, CNAM and University Sorbonne Paris Nord. E. Chazelas: doctoral funding from University Sorbonne Paris Nord; C. Debras: French National Cancer Institute. Project: European Research Council (Horizon 2020, grant agreement No 864219), French National Cancer Institute (INCa_14059), French Ministry of Health (arrêté 29.11.19) and IdEx Université de Paris (ANR-18-IDEX-0001). Researchers were independent from funders; funders had no role in the study design, the collection, analysis and interpretation of data, the writing of the report, and the decision to submit the article for publication.

   In-kind support

   Some of the 2677 ad hoc laboratory assays of additive doses were ordered by the consumers' association UFC Que Choisir.

   Finding

   Intakes of 90 additives were estimated from repeated 24-h records matched to brand-level composition data (OQALI, Open Food Facts, Mintel GNPD), with doses from 2677 laboratory assays, EFSA dose data by food category or Codex GSFA levels; descriptive analyses were weighted to the 2016 French census. Table 1, % consumers; median, 95th percentile and mean (SD) in mg/day; mean (SD) in mg/kg bw/day: modified starches E 14xx 91.48%; 1304.66, 4270.38, 1596.45 (1394.44); 24.33 (21.84). Citric acid E 330 91.15%; 1380.84, 6193.84, 1956.51 (2231.41); 29.84 (34.31). Lecithins E 322 86.57%; 35.36, 168.33, 53.94 (71.99); 0.83 (1.1). Xanthan gum E 415 79.74%; 303.60, 1401.31, 439.26 (508.65); 6.62 (7.62). Mono- and diglycerides of fatty acids E 471 78.07%; 101.90, 547.24, 160.8 (197.42); 2.44 (2.98). Sodium carbonates E 500 67.10%; 560.00, 4950.00, 1267.46 (1913.51); 19.62 (29.73), with median 8.51 and 95th percentile 76.73 mg/kg bw/day. The ADI column reads NA for all six. Diphosphates E 450 64.36%; 62.86, 715.67, 178.33 (308.05); 2.74 (4.67); ADI 40 mg/kg bw/day, exceeded by 0.02% of all participants and 0.03% of consumers. Additives without a specified ADI can be used quantum satis, that is with no limitations other than current good manufacturing practice. The authors state that information on the health impact of regular and cumulative intake of food additives in humans, and on effects of their mixtures, is still missing.

6. Regulator

   EFSA Panel on Food Additives and Flavourings (FAF), Younes M, Aquilina G, Castle L, Degen G, Engel KH, et al. Re-evaluation of xanthan gum (E 415) as a food additive in foods for infants below 16 weeks of age and follow-up of its re-evaluation as a food additive for uses in foods for all population groups EFSA Journal, 2023.

   doi:10.2903/j.efsa.2023.7951PubMed 37151988

   Funding and finding

   Funding

   European Food Safety Authority; requestor: European Commission; question number EFSA-Q-2018-00097.

   Finding

   For infants below 16 weeks of age fed special formulae for special medical purposes (food category 13.1.5.1), exposure at the maximum permitted level of 1,200 mg/L was estimated at 240 mg/kg body weight per day for mean and 312 mg/kg bw per day for high-level consumption, and at 183 and 238 mg/kg bw per day using the mean use levels reported by industry. The clinical studies submitted were of low validity and gave no reference point; the neonatal piglet studies gave a NOAEL of 750 mg/kg bw per day, and the resulting margin of exposure of 2.4 was considered sufficient because xanthan gum is not absorbed and an infant-specific animal model was used. Use in formulae for infants below 16 weeks up to 1,200 mg/L was concluded not to raise concerns. The Panel quoted the US Food and Drug Administration report of 22 infants who developed necrotising enterocolitis after use of a xanthan gum-containing thickener for gastro-oesophageal reflux or dysphagia: 21 were preterm with a median gestational age of 30.5 weeks, the thickener was consumed for a median of 13 days (range 1 to 31) before onset, 19 developed necrotising enterocolitis after 37 weeks post-menstrual age, 14 required surgery and 7 died; a further three extremely preterm infants were reported separately. In May 2011 the FDA warned parents and healthcare providers to avoid feeding the product to infants born before 37 weeks of gestation. The Panel judged that the described cases are not related to the food additive use of xanthan gum in the manufacture of infant formula and that the associated doses are unknown but expected in gram amounts, and noted that the opinion does not cover ad libitum addition of xanthan gum to mother's milk or infant formula. Xanthan gum forms a colloidal dispersion and will not be present in the gastrointestinal tract in the pristine particulate form, so a conventional risk assessment applies; the presence of small particles including nanoparticles could be neither confirmed nor excluded. Recommended amendments to the EU specifications include lowering the lead limit, adding limits for cadmium, mercury and arsenic, adding Cronobacter sakazakii to the microbiological criteria, requiring absence of residual enzyme activity and replacing 'soluble' with 'dispersible'. Highest quantified toxic element levels in commercial samples were lead 0.11, cadmium 0.01, mercury 0.007 and arsenic 0.1 mg/kg; with lead at the current specification limit of 2 mg/kg the margin of exposure for infants below 16 weeks at the 95th percentile was 0.8.

7. Independent research

   Silva Rischiteli AB, Silva-Neto AF, Maza PK, Boldarine VT, Ribeiro DA, Moura CFG, et al. Xanthan gum intake modifies the colon microbiota profile and causes mild colon inflammation in rats PLoS One, 2026.

   doi:10.1371/journal.pone.0347232PubMed 41984887

   Funding and finding

   Funding

   CMO received funding from FAPESP (The São Paulo Research Foundation), number 2018/16470-0. The funder acts as a funding agency that supports research projects financially. It had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

   Finding

   Body weight gain, food intake, food efficiency, adiposity, glycaemia and serum lipids did not differ between groups. The histological inflammation score of the distal colon was higher than control at 925 and 1,850 mg/kg diet (2.0 ± 1.2 and 1.2 ± 1.3 against 0.2 ± 0.4, p ≤ 0.05), with mononuclear, chiefly lymphocytic, infiltration of the intestinal wall on haematoxylin and eosin staining. Colonic IL-1beta was 18.29 ± 2.68 pg/mg protein at 925 mg/kg against 11.41 ± 0.94 in controls (p ≤ 0.05) and 8.87 ± 1.25 at 1,850 mg/kg (p ≤ 0.05 against 925 mg/kg); IL-6 and TNF-alpha concentrations did not differ between groups (TNF-alpha median 0.04 pg/mg protein in every group). TNF-alpha and claudin-2 immunostaining scores were higher than control at all three doses (TNF-alpha 2.0 ± 0.7, 1.4 ± 0.8 and 1.4 ± 0.5 against 0.4 ± 0.5). Claudin-2 content (1.04 ± 0.43 against 0.31 ± 0.03 ng/mg protein) and ZO-1 content (83.69 ± 10.69 against 46.04 ± 5.29 pg/mg protein) were higher than control only at 462.5 mg/kg; occludin did not differ. Relative abundance of Firmicutes and Bacteroidetes, their ratio and alpha diversity were unchanged; Elusimicrobiota abundance was higher at 1,850 mg/kg than in the control and 462.5 mg/kg groups (p < 0.04). Doses were set by body surface area conversion to mimic a gluten-free diet and mild or moderate oropharyngeal dysphagia managed with a xanthan gum thickener.

8. Independent research

   Salame C, Javaux G, Sellem L, Viennois E, de Edelenyi FS, Agaësse C, et al. Food additive emulsifiers and the risk of type 2 diabetes: analysis of data from the NutriNet-Santé prospective cohort study The Lancet Diabetes & Endocrinology, 2024.

   doi:10.1016/S2213-8587(24)00086-XPubMed 38663950

   Funding and finding

   Registration

   NCT03335644

   Participants

   104,139

   Follow-up

   mean 6.8 years

   Funding

   NutriNet-Santé public institutions (Ministère de la Santé, Santé Publique France, INSERM, INRAE, CNAM, University Sorbonne Paris Nord); European Research Council (grant agreement 864219, ADDITIVES); French National Cancer Institute (INCa_14059); French Ministry of Health; IdEx Université de Paris (ANR-18-IDEX-0001); Bettencourt-Schueller Foundation Research Prize 2021. The funders had no role in study design; the collection, analysis, and interpretation of data; the writing of the report, or the decision to submit the manuscript for publication.

   Finding

   1056 incident cases of type 2 diabetes. Hazard ratios per 100 mg/day: total carrageenans 1.03 (95% CI 1.01 to 1.05), E407 1.03 (1.01 to 1.05); mean total carrageenan intake 59.0 mg/day. Associations were also reported for E340 (1.15, 1.02 to 1.31, per 500 mg/day), E472e, E331, E412, E414 and xanthan gum E415 (hazard ratio 1.08, 95% CI 1.02 to 1.14, per increment of 500 mg/day, p = 0.013). Xanthan gum was found in 5% of foods according to EFSA data cited by the authors. Appendix eTable3 (https://ars.els-cdn.com/content/image/1-s2.0-S221385872400086X-mmc1.pdf), per 500 mg/day, fully adjusted and mutually adjusted for other emulsifiers: E471 0.99 (0.91 to 1.08), p = 0.84, minimally adjusted 1.07 (0.98 to 1.16); total mono- and diglycerides of fatty acids 1.00 (0.95 to 1.05). E471 was consumed by 80.3%, mean 165.4 mg/day. eTable3, per 500 mg/day, fully adjusted: diphosphates E450 1.01 (0.98 to 1.04), p = 0.49; tripotassium phosphate E340 1.15 (1.02 to 1.31), p = 0.023, with a false-discovery-rate-adjusted p of 0.10 (Figure 4 legend); trisodium phosphate E339 1.03 (0.92 to 1.14), p = 0.64. E450 was consumed by 70.4%, mean 244.5 (SD 351.7) mg/day; E340 by 5.2%, mean 6.9 (SD 90.5) mg/day. Absolute risks by intake level are not reported.

9. Funding unknown

   Eastwood MA, Brydon WG, Anderson DM. The dietary effects of xanthan gum in man Food Additives and Contaminants, 1987.

   doi:10.1080/02652038709373610PubMed 3549377

   Funding and finding

   Participants

   5

   Follow-up

   23 consecutive days after a 7-day control period

   Funding

   not readable. The full text is behind a paywall and no open copy was located; the PubMed record lists no grant support and indexes the paper as Research Support, Non-U.S. Gov't.

   Finding

   Each man consumed 15 times the then acceptable daily intake of 10 mg/kg body weight: the lightest and heaviest took 10.4 and 12.9 g of xanthan gum per day, as a pre-hydrated gel, on each of 23 consecutive days. Faecal wet and dry weight rose and intestinal transit time fell. Serum cholesterol was 10% lower and faecal bile acid concentrations were higher (both p < 0.05 as reported by EFSA). Plasma biochemistry, haematological indices, urinalysis, glucose tolerance and insulin tests, serum immunoglobulins, triglycerides, phospholipids, HDL cholesterol and breath hydrogen and methane did not change significantly. Numbers were read in the PubMed abstract and in the EFSA 2017 re-evaluation, which converts the dose to 149 to 184 mg/kg body weight per day for a 70 kg adult.

10. Funding unknown

    Daly J, Tomlin J, Read NW. The effect of feeding xanthan gum on colonic function in man: correlation with in vitro determinants of bacterial breakdown British Journal of Nutrition, 1993.

    doi:10.1079/bjn19930089PubMed 8329363

    Funding and finding

    Participants

    18

    Follow-up

    two 10-day periods separated by 3 days

    Funding

    not stated. The paper carries no funding or acknowledgement statement; the PubMed record lists no grant support. Author affiliation: Centre for Human Nutrition, University of Sheffield, Northern General Hospital.

    In-kind support

    Xanthan gum (Keltrol T) supplied by Kelco International Ltd, San Diego, a xanthan gum manufacturer, in 5 g portions.

    Finding

    Xanthan gum 15 g per day, as three 5 g portions in flavoured jelly, was taken for 10 days after a 10-day control period. Stool weight rose from 190.3 (SE 13.6) to 242.8 (15.8) g/day (P < 0.001), stool dry weight from 53.2 to 63.4 g/day (P < 0.008), defaecation frequency from 1.14 to 1.31 per day (P < 0.035) and flatulent episodes from 9.6 to 13.9 per day (P < 0.001); transit time fell from 46.7 to 40.7 hours, not significantly. Before feeding, faecal homogenates from 12 of the 18 men reduced the viscosity of xanthan gum in vitro and 7 abolished it; after 10 days of feeding, 16 of 18 reduced and 11 abolished it, with greater viscosity loss (50.8 against 24.1 mPa.s, P = 0.045) and more hydrogen and short-chain fatty acids (P = 0.049), indicating bacterial adaptation. Reduction in viscosity in vitro did not correlate with the change in stool weight or transit time.

11. Funding unknown

    Osilesi O, Trout DL, Glover EE, Harper SM, Koh ET, Behall KM, et al. Use of xanthan gum in dietary management of diabetes mellitus The American Journal of Clinical Nutrition, 1985.

    doi:10.1093/ajcn/42.4.597PubMed 4050722

    Funding and finding

    Participants

    13

    Follow-up

    12 weeks of muffin feeding, xanthan gum during one half

    Funding

    not readable. The full text is behind a paywall and no open copy was located; the PubMed record lists no grant support and indexes the paper as Research Support, U.S. Gov't, Non-P.H.S.

    Finding

    Xanthan gum 12 g per day was fed in muffins during either the first or the second half of a 12-week muffin-feeding period. Blood was taken before the study and at the end of the xanthan and xanthan-free periods, before and 2 hours after an oral glucose load. In the participants with diabetes, feeding xanthan gum lowered fasting and post-load serum glucose and fasting total plasma cholesterol, and tended to lower fasting and post-load gastrin and gastric inhibitory polypeptide and fasting total and VLDL triglyceride and cholesterol in the VLDL and LDL fractions. Participants reported fullness after the xanthan muffins and no severe digestive symptoms. The abstract, the only part readable, reports directions of effect without figures; EFSA's re-evaluations of xanthan gum do not cite this study.

12. Funding unknown

    Beal J, Silverman B, Bellant J, Young TE, Klontz K. Late onset necrotizing enterocolitis in infants following use of a xanthan gum-containing thickening agent The Journal of Pediatrics, 2012.

    doi:10.1016/j.jpeds.2012.03.054PubMed 22575248

    Funding and finding

    Participants

    22

    Funding

    not readable. The full text is behind a paywall and no open copy was located; the PubMed record lists no grant support. The corresponding author is affiliated with the Center for Food Safety and Applied Nutrition, US Food and Drug Administration.

    Finding

    Adverse event reports submitted to the US Food and Drug Administration suggested a possible association between necrotising enterocolitis and ingestion of a commercial feed thickener by premature infants; review in 2011 of 22 cases with exposure revealed a distinct illness pattern. As quoted by EFSA in 2023: the thickener was used for gastro-oesophageal reflux or dysphagia in every case; 21 of 22 infants were preterm with a median gestational age of 30.5 weeks; it was consumed for a median of 13 days (range 1 to 31) before onset; 19 developed necrotising enterocolitis after 37 weeks post-menstrual age; 14 required surgery and 7 died; the daily intake was not recorded. The abstract was read in PubMed and the case details in the EFSA 2023 opinion, which quotes this report.

13. Independent research

    Payen de la Garanderie M, Hasenbohler A, Dechamp N, Javaux G, Szabo de Edelenyi F, Agaësse C, et al. Food additive mixtures and type 2 diabetes incidence: Results from the NutriNet-Santé prospective cohort PLoS Medicine, 2025.

    doi:10.1371/journal.pmed.1004570PubMed 40198579

    Funding and finding

    Registration

    NCT03335644

    Participants

    108,643

    Follow-up

    mean 7.7 years (SD 4.6)

    Funding

    NutriNet-Santé public institutions: Ministère de la Santé, Santé Publique France, INSERM, INRAE, Conservatoire National des Arts et Métiers and University Sorbonne Paris Nord; European Research Council (grant agreement 864219, ADDITIVES); French National Cancer Institute (INCa_14059); French Ministry of Health (arrêté 29.11.19); IdEx Université de Paris (ANR-18-IDEX-0001); Bettencourt-Schueller Foundation Research Prize 2021; the laboratory of one author is supported by a European Research Council Starting Grant (ERC-2018-StG-804135 INVADERS) and the INSERM programme Microbiote. Researchers were independent from funders. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

    Finding

    1,131 participants were diagnosed with type 2 diabetes. Nonnegative matrix factorisation identified five additive mixtures. Mixture 2, characterised by modified starches, pectin, guar gum, carrageenan, polyphosphates, potassium sorbates, curcumin and xanthan gum (E415 loading 0.16), was associated with type 2 diabetes incidence: hazard ratio 1.08 (95% CI 1.02 to 1.15) per increment of one standard deviation of the mixture score (SD 8.0), p = 0.006. Mixture 5, dominated by additives of sweetened drinks, gave a hazard ratio of 1.13 (1.08 to 1.18), p < 0.001. Mixtures 1, 3 and 4 showed no association. Associations persisted after adjustment for each emblematic additive in turn, so no single additive drove them, and did not differ by overall diet quality. Of 28 two-by-two interactions among the eight additives of mixture 2, three suggested synergism and four antagonism at p < 0.1. Absolute risks are not reported, but S1 Appendix Table H gives cases by tertile of the mixture 2 score (column labelled participants / incident cases; the values, summing to 1,131, are cases / participants): 260 / 36,215, 406 / 36,214 and 465 / 36,214, hazard ratios 1.28 (1.09 to 1.49) and 1.26 (1.08 to 1.48) against the lowest tertile, p-trend 0.007. With the mixture 2 score residualised on modified-starch intake (Table J), the hazard ratio was 1.06 (1.00 to 1.12). The authors state that residual confounding and exposure or outcome misclassification cannot be excluded and that causality cannot be established from this single observational study.

14. Regulator

    US Food and Drug Administration. Xanthan gum, 21 CFR 172.695: food additives permitted for direct addition to food for human consumption Code of Federal Regulations, title 21, section 172.695, read in the electronic Code of Federal Regulations, edition of 1 October 2026, 2026.

    https://www.ecfr.gov/current/title-21/section-172.695

    Funding and finding

    Funding

    US Food and Drug Administration; the text is federal regulation codified by the Office of the Federal Register.

    Finding

    The food additive xanthan gum may be safely used in food under prescribed conditions: it is a polysaccharide gum derived from Xanthomonas campestris by a pure-culture fermentation process and purified by recovery with isopropyl alcohol, containing D-glucose, D-mannose and D-glucuronic acid as the dominant hexose units and manufactured as the sodium, potassium or calcium salt; the strain must be non-pathogenic and non-toxic in man or other animals and the process must render the additive free of viable cells. Specifications: residual isopropyl alcohol not above 750 parts per million, a minimum viscosity of 600 centipoises for a 1% solution with 1% potassium chloride at 75 °F, a positive locust bean gum gel test and more than 1.5% pyruvic acid by weight. The additive is used in accordance with good manufacturing practice as a stabiliser, emulsifier, thickener, suspending agent, bodying agent or foam enhancer in foods for which standards of identity do not preclude such use; no numerical limit on use is set. The container label must bear the name of the additive, the designation 'food grade' and adequate directions for use.

15. Textbook or database

    Japan Food Chemical Research Foundation. List of existing food additives Japan Food Chemical Research Foundation, English translation of the Japanese list, revision of 25 August 2025, 2025.

    https://www.ffcr.or.jp/en/tenka/list-of-existing-food-additives/list-of-existing-food-additives.html

    Funding and finding

    Funding

    not stated. The page is published by the Japan Food Chemical Research Foundation and states that it translates the Japanese list first compiled by public notice of the Minister for Health and Welfare on 16 April 1996 and since revised, the latest revision announced by the Consumer Affairs Agency.

    Finding

    Existing food additives are substances of natural origin permitted for use and distribution in Japan as an exception to the designation system of the Food Sanitation Act, because of long use in Japan; the status was set by the 1995 amendment of the Act, which made all additives subject to designation. Entry 76 of the list is xanthan gum, described as a substance composed mainly of polysaccharides obtained from the cultured solution of bacteria belonging to Xanthomonas. Five entries are lecithins: 153 vegetable lecithin, mainly lecithin from rape seeds or soybeans; 312 yolk lecithin, from egg yolk; 251 fractionated lecithin, mainly sphingomyelin, phosphatidylinositol, phosphatidylethanolamine and phosphatidylcholine from vegetable or yolk lecithin; 119 enzymatically modified lecithin, mainly phosphatidylglycerol from either; and 122 enzymatically decomposed lecithin, mainly phosphatidic acid and lysolecithin from either. The revision announced by the Consumer Affairs Agency on 25 August 2025 removed 30 additives and left 327 on the list.