https://orcid.org/0000-0003-2489-6430
Abstract
Emulsifiers are implicated in the pathogenesis of inflammatory bowel disease (IBD). Few studies have examined emulsifier intake in people with existing IBD. We aimed to describe the frequency of exposure to 6 selected emulsifiers in a contemporary cohort of people with IBD and compare intake with healthy controls (HCs).
Baseline food records from participants in an Australian prospective cohort study examining the microbiome of IBD patients and HCs were analyzed. Exposure to inflammatory emulsifiers polysorbate-80 (P80); carboxymethylcellulose (CMC); carrageenan; xanthan gum (XG); lecithin (soy and sunflower) and mono- and diglycerides of fatty acids (MDGs) were determined by examining ingredient lists. Frequency of emulsifier exposure between groups (IBD vs HC, Crohn’s disease [CD] vs ulcerative colitis [UC], IBD children vs adults, active disease vs remission) was examined after controlling for confounders.
Records from 367 participants were analyzed (n = 176 IBD, of which there were 101 CD, 75 UC, and 191 HC patients). In total, 5022 unique food items were examined, with 18% containing 1 or more emulsifier of interest. Inflammatory bowel disease participants had significantly higher total daily emulsifier exposure compared with HCs (2.7 ± 1.8 vs 2.3 ± 1.6, P = .02). In IBD participants, emulsifiers with the highest daily exposure were MDGs (1.2 ± 0.93), lecithin (0.85 ± 0.93), and XG (0.38 ± 0.42). There were no recorded exposures to P80.
Inflammatory bowel disease participants were exposed to more emulsifiers than HCs. Intake of inflammatory emulsifiers were low or nonexistent, suggesting their presence in the food supply are not as common as frequently stated.
What is already known? Some emulsifiers are associated with gut inflammation in vitro.
What is new here? This study found exposure to inflammatory emulsifiers in Australians with inflammatory bowel disease was very low or nonexistent, and exposure did not differ between those with IBD and healthy controls.
How can this study help patient care? The results suggest that the presence of inflammatory emulsifiers in the Australian food supply are not as common as frequently stated.
Introduction
The modern food supply includes many non-nutritional ingredients and additives. These include emulsifiers,1 which are estimated to be added to 55% of all ultra-processed foods.2 Emulsifiers are particularly useful as they impart stability, improve the texture, and increase shelf life.3 They enable the homogenization of immiscible liquids such as oil droplets in water (eg, mayonnaise and dressings) or water droplets in oil (eg, margarine and butter).1 Emulsifiers also include naturally occurring substances (eg, lecithin) and synthetic compounds (eg, polysorbates).
Emerging evidence from experimental studies suggests several emulsifiers may induce microbiota-mediated gut inflammation. Preclinical studies with emulsifiers, polysorbate-80 (P80) and carboxymethylcellulose (CMC) have shown induced thinning of the mucosal layer and intestinal microbial changes promoting the development of colitis in mice.4,5 Furthermore, when the altered microbiota was transplanted to unexposed germ-free mice, it was sufficient to induce a similar inflammatory response. These same effects have also been observed in other recent animal models6–8 (summarized in Table 1). Other studies have shown that in addition to P80 and CMC, carrageenans and gum emulsifiers cause microbiota alterations (density and composition) as well as increase expression of pro-inflammatory markers, lipopolysaccharide, and flagellin.15 The inflammatory effects of carrageenan and its subsequent negative impact on intestinal microbiota has also been observed in mice models.13,14,16
| Author | Emulsifier/s | Model | Effect on Microbiota | Inflammatory Effects |
|---|---|---|---|---|
| Bhattacharyya, 201711 | Carrageenan | Human RCT (n = 12) | Not measured | Increase in faecal calprotectin and interleukin-6 in the carrageenan-containing diet group (n = 5) but not in the carrageenan-free group (n = 7) |
| Chassaing, 20154 | CMC P80 | Mice | Reduction in microbial diversity and reduced mucus thickness | Induced low-grade intestinal inflammation and colitis, increased levels of inflammatory markers (bioactive lipopolysaccharide and flagellin) |
| Chassaing, 20175 | CMC P80 | M-SHIME (mucosal simulator of the human microbiota) | Altered intestinal microbiota composition and reduced diversity | Increased proinflammatory potential, evidence by elevated bioactive flagellin |
| Chassaing, 202212 | CMC | Human RCT (n = 16) | Altered intestinal microbiota composition and reduced diversity, increased microbiota encroachment of the mucus layer | Not measured |
| Furuhashi, 20206 | P80 | Mice | Decreased the α-diversity in the small intestine | Increased expression of pro-inflammatory cytokine (interleukin-1β) |
| Mi, 202013 | Carrageenan | Mice | Altered microbial composition | Increased abundance of inflammatory bacteria, aggravated intestinal inflammation |
| Munyaka, 201614 | Carrageenan | Mice | Decreased bacterial richness and composition | Induced colitis in mice. |
| Naimi, 202115 | A total of 20 dietary emulsifiers | Human microbiota-maintained ex vivo | P80, CMC, carrageenans and gums: Induced dysbiosis and decreased microbial bacterial density | Xanthan gum: increased lipopolysaccharide and flagellin levels Carrageenans: increased flagellin levels |
| Sandall, 20202 | All dietary emulsifiers (n = 65) | Uncontrolled human feasibility study (n = 20) | Not measured | Not measured |
| Shang, 201716 | Carrageenan | Mice | Decreased abundance of anti-inflammatory bacterium (A. muciniphila) in the gut | Induced low-grade colitis |
| Viennois, 20207 | CMC P80 | Mice | Not measured | Promoted chronic inflammation and metabolism dysregulation |
| Zangara, 20218 | CMC | Mice | Elevated flagella expression by microbes | Accelerated onset of colitis |
| Author | Emulsifier/s | Model | Effect on Microbiota | Inflammatory Effects |
|---|---|---|---|---|
| Bhattacharyya, 201711 | Carrageenan | Human RCT (n = 12) | Not measured | Increase in faecal calprotectin and interleukin-6 in the carrageenan-containing diet group (n = 5) but not in the carrageenan-free group (n = 7) |
| Chassaing, 20154 | CMC P80 | Mice | Reduction in microbial diversity and reduced mucus thickness | Induced low-grade intestinal inflammation and colitis, increased levels of inflammatory markers (bioactive lipopolysaccharide and flagellin) |
| Chassaing, 20175 | CMC P80 | M-SHIME (mucosal simulator of the human microbiota) | Altered intestinal microbiota composition and reduced diversity | Increased proinflammatory potential, evidence by elevated bioactive flagellin |
| Chassaing, 202212 | CMC | Human RCT (n = 16) | Altered intestinal microbiota composition and reduced diversity, increased microbiota encroachment of the mucus layer | Not measured |
| Furuhashi, 20206 | P80 | Mice | Decreased the α-diversity in the small intestine | Increased expression of pro-inflammatory cytokine (interleukin-1β) |
| Mi, 202013 | Carrageenan | Mice | Altered microbial composition | Increased abundance of inflammatory bacteria, aggravated intestinal inflammation |
| Munyaka, 201614 | Carrageenan | Mice | Decreased bacterial richness and composition | Induced colitis in mice. |
| Naimi, 202115 | A total of 20 dietary emulsifiers | Human microbiota-maintained ex vivo | P80, CMC, carrageenans and gums: Induced dysbiosis and decreased microbial bacterial density | Xanthan gum: increased lipopolysaccharide and flagellin levels Carrageenans: increased flagellin levels |
| Sandall, 20202 | All dietary emulsifiers (n = 65) | Uncontrolled human feasibility study (n = 20) | Not measured | Not measured |
| Shang, 201716 | Carrageenan | Mice | Decreased abundance of anti-inflammatory bacterium (A. muciniphila) in the gut | Induced low-grade colitis |
| Viennois, 20207 | CMC P80 | Mice | Not measured | Promoted chronic inflammation and metabolism dysregulation |
| Zangara, 20218 | CMC | Mice | Elevated flagella expression by microbes | Accelerated onset of colitis |
| Author | Emulsifier/s | Model | Effect on Microbiota | Inflammatory Effects |
|---|---|---|---|---|
| Bhattacharyya, 201711 | Carrageenan | Human RCT (n = 12) | Not measured | Increase in faecal calprotectin and interleukin-6 in the carrageenan-containing diet group (n = 5) but not in the carrageenan-free group (n = 7) |
| Chassaing, 20154 | CMC P80 | Mice | Reduction in microbial diversity and reduced mucus thickness | Induced low-grade intestinal inflammation and colitis, increased levels of inflammatory markers (bioactive lipopolysaccharide and flagellin) |
| Chassaing, 20175 | CMC P80 | M-SHIME (mucosal simulator of the human microbiota) | Altered intestinal microbiota composition and reduced diversity | Increased proinflammatory potential, evidence by elevated bioactive flagellin |
| Chassaing, 202212 | CMC | Human RCT (n = 16) | Altered intestinal microbiota composition and reduced diversity, increased microbiota encroachment of the mucus layer | Not measured |
| Furuhashi, 20206 | P80 | Mice | Decreased the α-diversity in the small intestine | Increased expression of pro-inflammatory cytokine (interleukin-1β) |
| Mi, 202013 | Carrageenan | Mice | Altered microbial composition | Increased abundance of inflammatory bacteria, aggravated intestinal inflammation |
| Munyaka, 201614 | Carrageenan | Mice | Decreased bacterial richness and composition | Induced colitis in mice. |
| Naimi, 202115 | A total of 20 dietary emulsifiers | Human microbiota-maintained ex vivo | P80, CMC, carrageenans and gums: Induced dysbiosis and decreased microbial bacterial density | Xanthan gum: increased lipopolysaccharide and flagellin levels Carrageenans: increased flagellin levels |
| Sandall, 20202 | All dietary emulsifiers (n = 65) | Uncontrolled human feasibility study (n = 20) | Not measured | Not measured |
| Shang, 201716 | Carrageenan | Mice | Decreased abundance of anti-inflammatory bacterium (A. muciniphila) in the gut | Induced low-grade colitis |
| Viennois, 20207 | CMC P80 | Mice | Not measured | Promoted chronic inflammation and metabolism dysregulation |
| Zangara, 20218 | CMC | Mice | Elevated flagella expression by microbes | Accelerated onset of colitis |
| Author | Emulsifier/s | Model | Effect on Microbiota | Inflammatory Effects |
|---|---|---|---|---|
| Bhattacharyya, 201711 | Carrageenan | Human RCT (n = 12) | Not measured | Increase in faecal calprotectin and interleukin-6 in the carrageenan-containing diet group (n = 5) but not in the carrageenan-free group (n = 7) |
| Chassaing, 20154 | CMC P80 | Mice | Reduction in microbial diversity and reduced mucus thickness | Induced low-grade intestinal inflammation and colitis, increased levels of inflammatory markers (bioactive lipopolysaccharide and flagellin) |
| Chassaing, 20175 | CMC P80 | M-SHIME (mucosal simulator of the human microbiota) | Altered intestinal microbiota composition and reduced diversity | Increased proinflammatory potential, evidence by elevated bioactive flagellin |
| Chassaing, 202212 | CMC | Human RCT (n = 16) | Altered intestinal microbiota composition and reduced diversity, increased microbiota encroachment of the mucus layer | Not measured |
| Furuhashi, 20206 | P80 | Mice | Decreased the α-diversity in the small intestine | Increased expression of pro-inflammatory cytokine (interleukin-1β) |
| Mi, 202013 | Carrageenan | Mice | Altered microbial composition | Increased abundance of inflammatory bacteria, aggravated intestinal inflammation |
| Munyaka, 201614 | Carrageenan | Mice | Decreased bacterial richness and composition | Induced colitis in mice. |
| Naimi, 202115 | A total of 20 dietary emulsifiers | Human microbiota-maintained ex vivo | P80, CMC, carrageenans and gums: Induced dysbiosis and decreased microbial bacterial density | Xanthan gum: increased lipopolysaccharide and flagellin levels Carrageenans: increased flagellin levels |
| Sandall, 20202 | All dietary emulsifiers (n = 65) | Uncontrolled human feasibility study (n = 20) | Not measured | Not measured |
| Shang, 201716 | Carrageenan | Mice | Decreased abundance of anti-inflammatory bacterium (A. muciniphila) in the gut | Induced low-grade colitis |
| Viennois, 20207 | CMC P80 | Mice | Not measured | Promoted chronic inflammation and metabolism dysregulation |
| Zangara, 20218 | CMC | Mice | Elevated flagella expression by microbes | Accelerated onset of colitis |
In terms of human studies, trials to date have shown that emulsifier intake can negatively contribute to gut microbiota changes in healthy individuals and earlier disease relapse (in those with ulcerative colitis [UC] and Crohn’s disease [CD]).2,11,12 These studies, however, are small and require further validation. Other human trials have involved variations of a CD exclusion diet, which by nature excludes all emulsifiers.17–19 These have been observed to induce symptomatic and clinical remission in patients with CD; however, a combination of other dietary factors may be involved other than emulsifiers.
Few published studies have explored the emulsifier exposure in people with existing IBD. One study by Lee at al20 observed the frequency of exposure to specific food additives (including emulsifiers) in a cohort of 138 children with CD by identifying the presence or absence of these additives in consumed food products. The results identified that the study cohort frequently consumed food additives. However, this study lacked a healthy control group, meaning it was not possible to compare findings with a general population. Another recent study by Trakman et al found adults with CD had a significantly higher intake of emulsifiers P80, CMC, and carrageenan compared with healthy controls.21 However, information on the emulsifier exposure of people with UC and comparison between children and adults with IBD as well as active disease and remission is unknown.
There are significant challenges to quantifying absolute emulsifier intake (eg, in milligrams/day).1 For example, food regulation in Australia defines the maximum level of emulsifiers added to food products. However, food companies are only required to list the presence, not quantity, of emulsifiers on food labels. Trakman et al21 attempted to quantify emulsifier intake in people with CD using the “maximum permissible limit” approach. However, this approach overestimates intake because not all manufacturers will use additives at the maximum allowed level, and thus it is difficult to accurately quantify emulsifiers in food.
Data on the current levels of emulsifier exposure in people with existing IBD are minimal, and information is also lacking for people with UC as well as children in comparison to a healthy control group. This knowledge is important, as childhood has been identified as a critical time for microbiota formation, and many dietary emulsifiers have been linked to microbial dysbiosis.22 Hence, in the absence of validated methodologies to measure emulsifier intake accurately, the aim of this study is to describe the frequency of exposure to 6 selected emulsifiers in a contemporary cohort of adults and children with IBD and compare intake with those of healthy controls.
Materials and Methods
This study is a secondary analysis of baseline data collected from participants enrolled in a large prospective longitudinal cohort study (the Australian Inflammatory Bowel Disease Microbiome “AIM” study).23 Participants eligible for the AIM study included children (6-17 years) and adults (18-80 years) with a diagnosis of CD or UC, and healthy controls. The current study involved the analysis of data from participants recruited between 2019 and 2022 who had completed a 3-day food record at baseline. Participants with IBD were characterized as having active disease or in remission based on validated clinical disease activity scores (Table 2). Healthy controls were included if they had no history of autoimmune disease, irritable bowel syndrome, or bowel surgery. Participants were required to be antibiotic-free 3 months prior to joining the study and probiotic-free for 1 month. Female participants who were pregnant or breastfeeding were excluded due to the influence on the gut microbiota.24 Individuals who recorded implausible dietary intake were excluded from this study using the Goldberg cutoff method. This was determined using estimated energy (EI) cut-offs of <500 kcal/day or >3500 kcal/day for adult participants25 and EI/Basal Metabolic Rate (BMR) <0.90 or EI/BMR >2.93 for pediatric participants26
Classification of disease activity.23
| Disease and Patient Type | Remission | Active Disease |
|---|---|---|
| CD (children) | PCDAI score < 12.5 | PCDAI ≥ 12.5 |
| CD (adult) | CDAI score < 150 | CDAI score ≥ 150 |
| UC (children) | PUCAI < 10 | PUCAI ≥ 10 |
| UC (adult) | Partial Mayo Score < 2 | Partial Mayo Score ≥ 2 |
| Disease and Patient Type | Remission | Active Disease |
|---|---|---|
| CD (children) | PCDAI score < 12.5 | PCDAI ≥ 12.5 |
| CD (adult) | CDAI score < 150 | CDAI score ≥ 150 |
| UC (children) | PUCAI < 10 | PUCAI ≥ 10 |
| UC (adult) | Partial Mayo Score < 2 | Partial Mayo Score ≥ 2 |
Classification of disease activity.23
| Disease and Patient Type | Remission | Active Disease |
|---|---|---|
| CD (children) | PCDAI score < 12.5 | PCDAI ≥ 12.5 |
| CD (adult) | CDAI score < 150 | CDAI score ≥ 150 |
| UC (children) | PUCAI < 10 | PUCAI ≥ 10 |
| UC (adult) | Partial Mayo Score < 2 | Partial Mayo Score ≥ 2 |
| Disease and Patient Type | Remission | Active Disease |
|---|---|---|
| CD (children) | PCDAI score < 12.5 | PCDAI ≥ 12.5 |
| CD (adult) | CDAI score < 150 | CDAI score ≥ 150 |
| UC (children) | PUCAI < 10 | PUCAI ≥ 10 |
| UC (adult) | Partial Mayo Score < 2 | Partial Mayo Score ≥ 2 |
Participants were recruited to the AIM study via a convenience and snowballing sampling method. Multiple recruitment strategies were used. This included recruitment of people with IBD through 8 hospital and private practice databases across New South Wales (NSW). Additional participants were recruited via attendance at routine clinics or endoscopy visits. Poster advertisements in the waiting rooms of hospital clinics and private practices were also used in addition to social media. Healthy controls were recruited using advertisements or by word of mouth.
Information on participants’ clinical and demographic characteristics were collected by a clinician at the AIM research clinic during the baseline visit. This included gender, age, height, weight, body mass index (BMI), smoking history, and alcohol intake. Participants with IBD were assessed for disease type and activity status using the validated clinical indices, such as the CD Activity Index (CDAI), Pediatric CDAI (PCDAI), Partial Mayo Score, and Pediatric UC Activity Index (PUCAI). Disease activity cutoff scores are shown in Table 2.
The current study utilized secondary analysis of prospective 3-day food diaries collected from participants in the AIM study at baseline. Food records were completed using the Easy Diet Dairy (EDD) smartphone application (Xyris Software Pty Ltd, Australia) or by hard copy. The EDD uses a large database of simplified food descriptors and brand name commercial foods commonly sold in Australia (AusBrands2019).27 Portion sizes can be entered using household measures or typical serving size. Completed EDD food records were emailed by participants to the research coordinator and uploaded to FoodWorks 10 Professional nutrient analysis software (Version 10; Xyris Pty Ltd, Brisbane, QLD, Australia, 2019) using the AusFoods 2019 database (derived from AUSNUT 2011-13 and the Australian Food Composition Database). Alternatively, hard copies of food records were entered manually into FoodWorks.
All food items captured by the 3-day food records were downloaded from FoodWorks onto a Microsoft Excel spreadsheet with patient ID code next to each food item. Subsequent columns were added to the spreadsheet to denote presence of the emulsifiers of interest. The 6 emulsifiers of interest examined in this study included polysorbate-80 (P80; E433), carboxymethylcellulose (CMC; E466), carrageenan (E407), xanthan gum (E415), lecithin (soy and sunflower; E322), and mono- and diglycerides of fatty acids (MDGs; E471). The emulsifiers P80, CMC, carrageenan, and xanthan gum are of interest, as they have been identified by ex vivo and animal studies to have pro-inflammatory effects. Lecithin and MDGs were included as they are reported in the literature to be 2 of the most widely used emulsifiers and have yet to be tested for their inflammatory properties.3,28
The AUSNUT 2011-13 food and dietary supplement classification system was used to group foods items into 21 food groups (Table 3). Food items were evaluated for the presence of the 6 emulsifiers using ingredients labels provided online by 2 of Australia’s largest supermarket chains, Woolworths Group Limited and Coles Group Limited,29 as well as product websites. In-person supermarket scans were performed for supermarkets that did not provide ingredients lists online (eg, ALDI). In cases where specific food brands were not provided in food diaries, an assumption list was made using multiple brand varieties of the same food product. Consistent with prior studies,2,20 mean emulsifier exposure per day was used as the primary outcome of this study. The AUSNUT food groups were used to identify the most common source of emulsifiers. Methods are outlined in Figure 1.
AUSNUT 2011-13 food and dietary supplement classification system.
| Food Group Code | AUSNUT Food Group Name | AUSNUT Subgroups |
|---|---|---|
| 11 | Non-alcoholic beverages | Tea, coffee, fruit and vegetable juices, cordials, soft drinks, flavored mineral waters |
| 12 | Cereal and cereal products | Rice and other grains, regular breads, bread rolls, English-style muffins, savory and sweet breads, pasta (without sauce), noodles, breakfast cereals, porridge |
| 13 | Cereal-based products and dishes | Sweet biscuits, savory biscuits, cakes, muffins, scones, cake-type desserts, pastries, mixed dishes where cereal is the major ingredient, batter-based products |
| 14 | Fats and oils | Butter, dairy blends, margarine and table spreads, plant oils |
| 15 | Fish and seafood products | Fin fish, crustacea and molluscks, other sea and freshwater foods, packed fish and seafood, fish and seafood products (homemade and takeaway), mixed dishes with fish or seafood as the major component |
| 16 | Fruit products and dishes | Fresh fruit, dried fruit, mixed dishes where fruit is the major component |
| 17 | Egg products and dishes | Eggs, dishes where egg is the major ingredient |
| 18 | Meat, poultry and game products and dishes | Beef, lamb, pork, kangaroo, poultry, sausages, processed meats, mixed dishes where meat is the major component |
| 19 | Milk products and dishes | Dairy milk, yoghurt, cream, cheese, ice cream, custard, flavored milk, other dishes where milk or a milk product is the major component |
| 20 | Dairy and meat substitutes | Dairy milk substitutes, cheese substitutes, soy-based ice cream and yoghurt, meat substitutes |
| 21 | Soups | Homemade soup, dry soup mix, canned soup, ready to heat soup |
| 22 | Seed and nut products and dishes | Seeds, nuts, seed and nut products, coconut cream |
| 23 | Savory sauces and condiments | Gravies and savory sauces, pickles, chutneys and relishes, salad dressings, dips |
| 24 | Vegetable products and dishes | Vegetables, dishes where vegetable is the major component |
| 25 | Legume and pulse products and dishes | Legumes, pulses, legume and pulse products and dishes |
| 26 | Snack foods | Potato snacks, corn snacks, pretzels |
| 27 | Sugar products and dishes | Sugar, honeys, syrups, jam, chocolate spreads |
| 28 | Confectionary and cereal/nut/seed bars | Chocolate and chocolate-based confectionery, lollies, fruit, nut and seed-bars, muesli or cereal style bars |
| 29 | Alcoholic beverages | Beers, wine, spirits, cider |
| 30 | Special dietary food | Shake and bar meal replacements, protein shakes, nutrition supplements |
| 31 | Miscellaneous | Yeast extracts, herbs, spices, stock cubes, essences |
| Food Group Code | AUSNUT Food Group Name | AUSNUT Subgroups |
|---|---|---|
| 11 | Non-alcoholic beverages | Tea, coffee, fruit and vegetable juices, cordials, soft drinks, flavored mineral waters |
| 12 | Cereal and cereal products | Rice and other grains, regular breads, bread rolls, English-style muffins, savory and sweet breads, pasta (without sauce), noodles, breakfast cereals, porridge |
| 13 | Cereal-based products and dishes | Sweet biscuits, savory biscuits, cakes, muffins, scones, cake-type desserts, pastries, mixed dishes where cereal is the major ingredient, batter-based products |
| 14 | Fats and oils | Butter, dairy blends, margarine and table spreads, plant oils |
| 15 | Fish and seafood products | Fin fish, crustacea and molluscks, other sea and freshwater foods, packed fish and seafood, fish and seafood products (homemade and takeaway), mixed dishes with fish or seafood as the major component |
| 16 | Fruit products and dishes | Fresh fruit, dried fruit, mixed dishes where fruit is the major component |
| 17 | Egg products and dishes | Eggs, dishes where egg is the major ingredient |
| 18 | Meat, poultry and game products and dishes | Beef, lamb, pork, kangaroo, poultry, sausages, processed meats, mixed dishes where meat is the major component |
| 19 | Milk products and dishes | Dairy milk, yoghurt, cream, cheese, ice cream, custard, flavored milk, other dishes where milk or a milk product is the major component |
| 20 | Dairy and meat substitutes | Dairy milk substitutes, cheese substitutes, soy-based ice cream and yoghurt, meat substitutes |
| 21 | Soups | Homemade soup, dry soup mix, canned soup, ready to heat soup |
| 22 | Seed and nut products and dishes | Seeds, nuts, seed and nut products, coconut cream |
| 23 | Savory sauces and condiments | Gravies and savory sauces, pickles, chutneys and relishes, salad dressings, dips |
| 24 | Vegetable products and dishes | Vegetables, dishes where vegetable is the major component |
| 25 | Legume and pulse products and dishes | Legumes, pulses, legume and pulse products and dishes |
| 26 | Snack foods | Potato snacks, corn snacks, pretzels |
| 27 | Sugar products and dishes | Sugar, honeys, syrups, jam, chocolate spreads |
| 28 | Confectionary and cereal/nut/seed bars | Chocolate and chocolate-based confectionery, lollies, fruit, nut and seed-bars, muesli or cereal style bars |
| 29 | Alcoholic beverages | Beers, wine, spirits, cider |
| 30 | Special dietary food | Shake and bar meal replacements, protein shakes, nutrition supplements |
| 31 | Miscellaneous | Yeast extracts, herbs, spices, stock cubes, essences |
AUSNUT 2011-13 food and dietary supplement classification system.
| Food Group Code | AUSNUT Food Group Name | AUSNUT Subgroups |
|---|---|---|
| 11 | Non-alcoholic beverages | Tea, coffee, fruit and vegetable juices, cordials, soft drinks, flavored mineral waters |
| 12 | Cereal and cereal products | Rice and other grains, regular breads, bread rolls, English-style muffins, savory and sweet breads, pasta (without sauce), noodles, breakfast cereals, porridge |
| 13 | Cereal-based products and dishes | Sweet biscuits, savory biscuits, cakes, muffins, scones, cake-type desserts, pastries, mixed dishes where cereal is the major ingredient, batter-based products |
| 14 | Fats and oils | Butter, dairy blends, margarine and table spreads, plant oils |
| 15 | Fish and seafood products | Fin fish, crustacea and molluscks, other sea and freshwater foods, packed fish and seafood, fish and seafood products (homemade and takeaway), mixed dishes with fish or seafood as the major component |
| 16 | Fruit products and dishes | Fresh fruit, dried fruit, mixed dishes where fruit is the major component |
| 17 | Egg products and dishes | Eggs, dishes where egg is the major ingredient |
| 18 | Meat, poultry and game products and dishes | Beef, lamb, pork, kangaroo, poultry, sausages, processed meats, mixed dishes where meat is the major component |
| 19 | Milk products and dishes | Dairy milk, yoghurt, cream, cheese, ice cream, custard, flavored milk, other dishes where milk or a milk product is the major component |
| 20 | Dairy and meat substitutes | Dairy milk substitutes, cheese substitutes, soy-based ice cream and yoghurt, meat substitutes |
| 21 | Soups | Homemade soup, dry soup mix, canned soup, ready to heat soup |
| 22 | Seed and nut products and dishes | Seeds, nuts, seed and nut products, coconut cream |
| 23 | Savory sauces and condiments | Gravies and savory sauces, pickles, chutneys and relishes, salad dressings, dips |
| 24 | Vegetable products and dishes | Vegetables, dishes where vegetable is the major component |
| 25 | Legume and pulse products and dishes | Legumes, pulses, legume and pulse products and dishes |
| 26 | Snack foods | Potato snacks, corn snacks, pretzels |
| 27 | Sugar products and dishes | Sugar, honeys, syrups, jam, chocolate spreads |
| 28 | Confectionary and cereal/nut/seed bars | Chocolate and chocolate-based confectionery, lollies, fruit, nut and seed-bars, muesli or cereal style bars |
| 29 | Alcoholic beverages | Beers, wine, spirits, cider |
| 30 | Special dietary food | Shake and bar meal replacements, protein shakes, nutrition supplements |
| 31 | Miscellaneous | Yeast extracts, herbs, spices, stock cubes, essences |
| Food Group Code | AUSNUT Food Group Name | AUSNUT Subgroups |
|---|---|---|
| 11 | Non-alcoholic beverages | Tea, coffee, fruit and vegetable juices, cordials, soft drinks, flavored mineral waters |
| 12 | Cereal and cereal products | Rice and other grains, regular breads, bread rolls, English-style muffins, savory and sweet breads, pasta (without sauce), noodles, breakfast cereals, porridge |
| 13 | Cereal-based products and dishes | Sweet biscuits, savory biscuits, cakes, muffins, scones, cake-type desserts, pastries, mixed dishes where cereal is the major ingredient, batter-based products |
| 14 | Fats and oils | Butter, dairy blends, margarine and table spreads, plant oils |
| 15 | Fish and seafood products | Fin fish, crustacea and molluscks, other sea and freshwater foods, packed fish and seafood, fish and seafood products (homemade and takeaway), mixed dishes with fish or seafood as the major component |
| 16 | Fruit products and dishes | Fresh fruit, dried fruit, mixed dishes where fruit is the major component |
| 17 | Egg products and dishes | Eggs, dishes where egg is the major ingredient |
| 18 | Meat, poultry and game products and dishes | Beef, lamb, pork, kangaroo, poultry, sausages, processed meats, mixed dishes where meat is the major component |
| 19 | Milk products and dishes | Dairy milk, yoghurt, cream, cheese, ice cream, custard, flavored milk, other dishes where milk or a milk product is the major component |
| 20 | Dairy and meat substitutes | Dairy milk substitutes, cheese substitutes, soy-based ice cream and yoghurt, meat substitutes |
| 21 | Soups | Homemade soup, dry soup mix, canned soup, ready to heat soup |
| 22 | Seed and nut products and dishes | Seeds, nuts, seed and nut products, coconut cream |
| 23 | Savory sauces and condiments | Gravies and savory sauces, pickles, chutneys and relishes, salad dressings, dips |
| 24 | Vegetable products and dishes | Vegetables, dishes where vegetable is the major component |
| 25 | Legume and pulse products and dishes | Legumes, pulses, legume and pulse products and dishes |
| 26 | Snack foods | Potato snacks, corn snacks, pretzels |
| 27 | Sugar products and dishes | Sugar, honeys, syrups, jam, chocolate spreads |
| 28 | Confectionary and cereal/nut/seed bars | Chocolate and chocolate-based confectionery, lollies, fruit, nut and seed-bars, muesli or cereal style bars |
| 29 | Alcoholic beverages | Beers, wine, spirits, cider |
| 30 | Special dietary food | Shake and bar meal replacements, protein shakes, nutrition supplements |
| 31 | Miscellaneous | Yeast extracts, herbs, spices, stock cubes, essences |
Methods used to categorize and describe frequency of emulsifier intake.
All data analysis was performed using IBM SPSS Statistics IOS Version 25 (SPSS, Chicago, IL, USA). The Shapiro Wilk test was used to determine normality. Continuous data was reported as mean and standard deviation (SD) or median with interquartile range (IQR) for normal and non-normal data, respectively. Categorical data were reported as frequencies with percentages. Demographic and clinical characteristics were compared between population groups using independent samples t tests for normally distributed data, Mann-Whitney test for non-normal data, or χ2 tests for categorical data.
Data from the 3-day food diaries were averaged for each study participant to obtain a mean emulsifier exposure per day for each of the 6 emulsifiers of interest and total emulsifier exposure. Analysis of covariance was used to compare the frequency of emulsifier exposure between groups (IBD vs healthy controls, adults with CD vs UC, children vs adults with IBD, active disease vs remission) controlling for confounders of BMI, total energy intake, gender, and smoking status. Linear mixed models were used to confirm the results of the ANCOVA and were found to be identical. Food sources of emulsifiers were analyzed to identify which food groups contributed the most to total and individual emulsifier exposure. The share of percentage contribution from the food groups was calculated and compared between people with IBD and HC using χ2 tests. All statistical tests were 2-sided, with a P value < .05 considered statistically significant.
Ethical Considerations
Ethics approval was received for the AIM study from the South Eastern Sydney Local Health District Human Research Ethics Committee (2019 ETH11443). Informed consent was received from all participants over 18 years and from parents or guardians of those younger than 18 years. All participant data was de-identified using study ID codes and uploaded into the secure REDCap (Research Electronic Data Capture) software.
Results
Of the 379 enrolled participants, 7 (2%) were excluded due to implausible dietary intake as determined by the Goldberg cutoff method for adults and children,25,26 and 5 (1%) were excluded due to incomplete 3-day food diaries. A total of 367 participants were eligible for the study with baseline data shown in Table 4.
Clinical and demographic characteristics of study participants (n = 367).
| Disease Status | Disease Phenotype | |||||
|---|---|---|---|---|---|---|
| IBD n = 176 | Healthy controls n = 191 | P | CD n = 101 | UC n = 75 | P | |
| Age, median (IQR) | 40 (25-52) | 36 (27-49) | .08 | 38 (21-52) | 41 (31-52) | .22 |
| Children (6-18 yrs), n (%) | 20 (11) | 12 (6) | .06 | 18 (18) | 2 (3) | .001* |
| Adults (≥ 19 yrs), n (%) | 156 (89) | 179 (94) | 83 (82) | 73 (97) | ||
| Adults, n (%) | .51 | |||||
| 19-50 yrs | 109 (70) | 143 (80) | .01* | 57 (69) | 52 (71) | |
| 51-70 yrs | 39 (25) | 35 (19) | 23 (28) | 16 (22) | ||
| >70 yrs | 8 (5) | 1 (1) | 3 (4) | 6 (7) | ||
| Sex, n (%) | .78 | |||||
| Female | 97 (55) | 114 (60) | .34 | 55 (54) | 42 (56) | |
| Male | 79 (45) | 77 (40) | 47 (46) | 33 (44) | ||
| Children | .36 | |||||
| Active disease, n (%) | 4 (20) | - | - | 4 (21) | 1 (50) | |
| Remission, n (%) | 16 (80) | - | 15 (79) | 1 (50) | ||
| Adults | .20 | |||||
| Active disease, n (%) | 53 (34) | - | - | 51 (61) | 52 (71) | |
| Remission, n (%) | 103 (66) | - | 32 (39) | 21 (29) | ||
| BMI—Adult (kg/m2) median (IQR) | 24.9 (22.9-28.1) | 24.0 (21.9-26.4) | .12 | 24.7 (22.0-28.1) | 25.1 (21.6-28.2) | .70 |
| BMI—Adult (kg/m2), n (%) | .16 | .95 | ||||
| <18.5 | 6 (4) | 5 (3) | 4 (5) | 2 (3) | ||
| 18.5-24.9 | 73 (47) | 107 (60) | 40 (48) | 33 (46) | ||
| 25-29.9 | 50 (32) | 46 (26) | 25 (30) | 25 (35) | ||
| 30-39.9 | 24 (16) | 17 (10) | 13 (16) | 11 (15) | ||
| ≥40 | 2 (1) | 3 (2) | 1 (1) | 1 (1) | ||
| Current smoker—Adult n (%) | 9 (6) | 5 (3) | .18 | 6 (7) | 3 (4) | .40 |
| Disease Status | Disease Phenotype | |||||
|---|---|---|---|---|---|---|
| IBD n = 176 | Healthy controls n = 191 | P | CD n = 101 | UC n = 75 | P | |
| Age, median (IQR) | 40 (25-52) | 36 (27-49) | .08 | 38 (21-52) | 41 (31-52) | .22 |
| Children (6-18 yrs), n (%) | 20 (11) | 12 (6) | .06 | 18 (18) | 2 (3) | .001* |
| Adults (≥ 19 yrs), n (%) | 156 (89) | 179 (94) | 83 (82) | 73 (97) | ||
| Adults, n (%) | .51 | |||||
| 19-50 yrs | 109 (70) | 143 (80) | .01* | 57 (69) | 52 (71) | |
| 51-70 yrs | 39 (25) | 35 (19) | 23 (28) | 16 (22) | ||
| >70 yrs | 8 (5) | 1 (1) | 3 (4) | 6 (7) | ||
| Sex, n (%) | .78 | |||||
| Female | 97 (55) | 114 (60) | .34 | 55 (54) | 42 (56) | |
| Male | 79 (45) | 77 (40) | 47 (46) | 33 (44) | ||
| Children | .36 | |||||
| Active disease, n (%) | 4 (20) | - | - | 4 (21) | 1 (50) | |
| Remission, n (%) | 16 (80) | - | 15 (79) | 1 (50) | ||
| Adults | .20 | |||||
| Active disease, n (%) | 53 (34) | - | - | 51 (61) | 52 (71) | |
| Remission, n (%) | 103 (66) | - | 32 (39) | 21 (29) | ||
| BMI—Adult (kg/m2) median (IQR) | 24.9 (22.9-28.1) | 24.0 (21.9-26.4) | .12 | 24.7 (22.0-28.1) | 25.1 (21.6-28.2) | .70 |
| BMI—Adult (kg/m2), n (%) | .16 | .95 | ||||
| <18.5 | 6 (4) | 5 (3) | 4 (5) | 2 (3) | ||
| 18.5-24.9 | 73 (47) | 107 (60) | 40 (48) | 33 (46) | ||
| 25-29.9 | 50 (32) | 46 (26) | 25 (30) | 25 (35) | ||
| 30-39.9 | 24 (16) | 17 (10) | 13 (16) | 11 (15) | ||
| ≥40 | 2 (1) | 3 (2) | 1 (1) | 1 (1) | ||
| Current smoker—Adult n (%) | 9 (6) | 5 (3) | .18 | 6 (7) | 3 (4) | .40 |
n (%), Expressed as number (percentage) *Indicates significant at P < .05.
Clinical and demographic characteristics of study participants (n = 367).
| Disease Status | Disease Phenotype | |||||
|---|---|---|---|---|---|---|
| IBD n = 176 | Healthy controls n = 191 | P | CD n = 101 | UC n = 75 | P | |
| Age, median (IQR) | 40 (25-52) | 36 (27-49) | .08 | 38 (21-52) | 41 (31-52) | .22 |
| Children (6-18 yrs), n (%) | 20 (11) | 12 (6) | .06 | 18 (18) | 2 (3) | .001* |
| Adults (≥ 19 yrs), n (%) | 156 (89) | 179 (94) | 83 (82) | 73 (97) | ||
| Adults, n (%) | .51 | |||||
| 19-50 yrs | 109 (70) | 143 (80) | .01* | 57 (69) | 52 (71) | |
| 51-70 yrs | 39 (25) | 35 (19) | 23 (28) | 16 (22) | ||
| >70 yrs | 8 (5) | 1 (1) | 3 (4) | 6 (7) | ||
| Sex, n (%) | .78 | |||||
| Female | 97 (55) | 114 (60) | .34 | 55 (54) | 42 (56) | |
| Male | 79 (45) | 77 (40) | 47 (46) | 33 (44) | ||
| Children | .36 | |||||
| Active disease, n (%) | 4 (20) | - | - | 4 (21) | 1 (50) | |
| Remission, n (%) | 16 (80) | - | 15 (79) | 1 (50) | ||
| Adults | .20 | |||||
| Active disease, n (%) | 53 (34) | - | - | 51 (61) | 52 (71) | |
| Remission, n (%) | 103 (66) | - | 32 (39) | 21 (29) | ||
| BMI—Adult (kg/m2) median (IQR) | 24.9 (22.9-28.1) | 24.0 (21.9-26.4) | .12 | 24.7 (22.0-28.1) | 25.1 (21.6-28.2) | .70 |
| BMI—Adult (kg/m2), n (%) | .16 | .95 | ||||
| <18.5 | 6 (4) | 5 (3) | 4 (5) | 2 (3) | ||
| 18.5-24.9 | 73 (47) | 107 (60) | 40 (48) | 33 (46) | ||
| 25-29.9 | 50 (32) | 46 (26) | 25 (30) | 25 (35) | ||
| 30-39.9 | 24 (16) | 17 (10) | 13 (16) | 11 (15) | ||
| ≥40 | 2 (1) | 3 (2) | 1 (1) | 1 (1) | ||
| Current smoker—Adult n (%) | 9 (6) | 5 (3) | .18 | 6 (7) | 3 (4) | .40 |
| Disease Status | Disease Phenotype | |||||
|---|---|---|---|---|---|---|
| IBD n = 176 | Healthy controls n = 191 | P | CD n = 101 | UC n = 75 | P | |
| Age, median (IQR) | 40 (25-52) | 36 (27-49) | .08 | 38 (21-52) | 41 (31-52) | .22 |
| Children (6-18 yrs), n (%) | 20 (11) | 12 (6) | .06 | 18 (18) | 2 (3) | .001* |
| Adults (≥ 19 yrs), n (%) | 156 (89) | 179 (94) | 83 (82) | 73 (97) | ||
| Adults, n (%) | .51 | |||||
| 19-50 yrs | 109 (70) | 143 (80) | .01* | 57 (69) | 52 (71) | |
| 51-70 yrs | 39 (25) | 35 (19) | 23 (28) | 16 (22) | ||
| >70 yrs | 8 (5) | 1 (1) | 3 (4) | 6 (7) | ||
| Sex, n (%) | .78 | |||||
| Female | 97 (55) | 114 (60) | .34 | 55 (54) | 42 (56) | |
| Male | 79 (45) | 77 (40) | 47 (46) | 33 (44) | ||
| Children | .36 | |||||
| Active disease, n (%) | 4 (20) | - | - | 4 (21) | 1 (50) | |
| Remission, n (%) | 16 (80) | - | 15 (79) | 1 (50) | ||
| Adults | .20 | |||||
| Active disease, n (%) | 53 (34) | - | - | 51 (61) | 52 (71) | |
| Remission, n (%) | 103 (66) | - | 32 (39) | 21 (29) | ||
| BMI—Adult (kg/m2) median (IQR) | 24.9 (22.9-28.1) | 24.0 (21.9-26.4) | .12 | 24.7 (22.0-28.1) | 25.1 (21.6-28.2) | .70 |
| BMI—Adult (kg/m2), n (%) | .16 | .95 | ||||
| <18.5 | 6 (4) | 5 (3) | 4 (5) | 2 (3) | ||
| 18.5-24.9 | 73 (47) | 107 (60) | 40 (48) | 33 (46) | ||
| 25-29.9 | 50 (32) | 46 (26) | 25 (30) | 25 (35) | ||
| 30-39.9 | 24 (16) | 17 (10) | 13 (16) | 11 (15) | ||
| ≥40 | 2 (1) | 3 (2) | 1 (1) | 1 (1) | ||
| Current smoker—Adult n (%) | 9 (6) | 5 (3) | .18 | 6 (7) | 3 (4) | .40 |
n (%), Expressed as number (percentage) *Indicates significant at P < .05.
The study cohort included 176 participants with IBD (101 with CD and 75 with UC) and 191 HCs. There were more females than males in the study cohort, comprising 55% of the IBD group and 60% of the HC group. The median ages for participants with IBD and HC were 40 (25-52) years and 36 (27-49) years, respectively. Post hoc analysis indicated a significantly higher number of participants aged >70 years in the IBD group compared with HC (5% vs 1%, P = .01). The median ages for participants with UC and CD were 41 (31-52) years and 38 (21-52), respectively.
The IBD group was mostly composed of adults (89%) with only a small proportion of children (11%; Table 4). Similarly, the control group mostly comprised adults (94%) with a small number of children (6%). The CD group was composed of 82% adults and 18% children, and the UC group contained 97% adults and 3% children. Post hoc testing revealed there were significantly more children with CD compared with UC (P = .001). Of the adults with IBD, a larger proportion were in remission (66%) compared with those having active disease (34%) at the point of analysis. Similarly, a greater proportion of children with IBD were in remission (80%) compared with having active disease (20%).
Emulsifier Exposure
A total of 5022 unique food items were reported in the 3-day dietary recalls of all participants. Of all unique food items, 902 (18%) contained ≥1 emulsifier of interest. When evaluating the 6 emulsifiers of interest (polysorbate-80 [P80; E433]; carboxymethylcellulose [CMC; E466]; carrageenan [E407]; xanthan gum [E415]; lecithin [soy and sunflower; E322]; mono- and diglycerides of fatty acids [MDGs; E471]), participants with IBD had a significantly higher total emulsifier exposure per day compared with HCs (2.7 ± 1.8 vs 2.3 ± 1.6 exposures per day, P = .02; Table 5). Similarly, participants with CD had a higher exposure of 2.7 ± 1.9 compared with HCs (P = .04). However, no differences were observed between UC and HCs (P = .10).
Summary of mean daily emulsifier exposure.
| IBD n = 176 | HC n = 191 | UC n = 75 | CD n = 101 | IBD (adults) n = 156 | HC (adults) n = 179 | UC (adults) n = 73 | CD (adults) n = 83 | IBD (children) n = 20 | HC (children) n = 12 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Total | 2.7 ± 1.8a | 2.3 ± 1.6 | 2.6 ± 1.6 | 2.7 ± 1.9a | 2.6 ± 1.8b | 2.2 ± 1.5 | 2.7 ± 1.7b | 2.6 ± 1.9 | 3.3 ± 2.0c | 3.9 ± 1.6 |
| MDGs | 1.2 ± 0.93 | 1.1 ± 0.84 | 1.2 ± 0.93 | 1.1 ± 0.93 | 1.2 ± 0.94 | 1.1 ± 0.82 | 1.2 ± 0.94 | 1.2 ± 0.95 | 1.1 ± 0.84 | 1.5 ± 0.93 |
| Lecithin | 0.85 ± 0.93 | 0.75 ± 0.73 | 0.88 ± 0.86 | 0.83 ± 0.97 | 0.79 ± 0.85 | 0.71 ± 0.68 | 0.87 ± 0.86 | 0.72 ± 0.84 | 1.3 ± 1.3c | 1.3 ± 1.1 |
| Xanthan gum | 0.38 ± 0.42 | 0.29 ± 0.43 | 0.37 ± 0.45 | 0.38 ± 0.41 | 0.39 ± 0.42b | 0.28 ± 0.42 | 0.38 ± 0.45 | 0.40 ± 0.39b | 0.28 ± 0.46 | 0.58 ± 0.47 |
| CGN | 0.20 ± 0.33 | 0.16 ± 0.28 | 0.19 ± 0.32 | 0.21 ± 0.35 | 0.18 ± 0.32 | 0.14 ± 0.27 | 0.19 ± 0.32 | 0.18 ± 0.32 | 0.33 ± 0.43c | 0.39 ± 0.40 |
| CMC | 0.12 ± 0.36a | 0.06 ± 0.17 | 0.07 ± 0.22 | 0.16 ± 0.44a | 0.09 ± 0.28 | 0.05 ± 0.17 | 0.06 ± 0.22 | 0.12 ± 0.32b | 0.32 ± 0.74c | 0.08 ± 0.15 |
| P-80 | 0a | 0.01 ± 0.05 | 0 | 0 | 0 | 0.01 ± 0.05 | 0 | 0 | 0 | 0.08 ± 0.15 |
| IBD n = 176 | HC n = 191 | UC n = 75 | CD n = 101 | IBD (adults) n = 156 | HC (adults) n = 179 | UC (adults) n = 73 | CD (adults) n = 83 | IBD (children) n = 20 | HC (children) n = 12 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Total | 2.7 ± 1.8a | 2.3 ± 1.6 | 2.6 ± 1.6 | 2.7 ± 1.9a | 2.6 ± 1.8b | 2.2 ± 1.5 | 2.7 ± 1.7b | 2.6 ± 1.9 | 3.3 ± 2.0c | 3.9 ± 1.6 |
| MDGs | 1.2 ± 0.93 | 1.1 ± 0.84 | 1.2 ± 0.93 | 1.1 ± 0.93 | 1.2 ± 0.94 | 1.1 ± 0.82 | 1.2 ± 0.94 | 1.2 ± 0.95 | 1.1 ± 0.84 | 1.5 ± 0.93 |
| Lecithin | 0.85 ± 0.93 | 0.75 ± 0.73 | 0.88 ± 0.86 | 0.83 ± 0.97 | 0.79 ± 0.85 | 0.71 ± 0.68 | 0.87 ± 0.86 | 0.72 ± 0.84 | 1.3 ± 1.3c | 1.3 ± 1.1 |
| Xanthan gum | 0.38 ± 0.42 | 0.29 ± 0.43 | 0.37 ± 0.45 | 0.38 ± 0.41 | 0.39 ± 0.42b | 0.28 ± 0.42 | 0.38 ± 0.45 | 0.40 ± 0.39b | 0.28 ± 0.46 | 0.58 ± 0.47 |
| CGN | 0.20 ± 0.33 | 0.16 ± 0.28 | 0.19 ± 0.32 | 0.21 ± 0.35 | 0.18 ± 0.32 | 0.14 ± 0.27 | 0.19 ± 0.32 | 0.18 ± 0.32 | 0.33 ± 0.43c | 0.39 ± 0.40 |
| CMC | 0.12 ± 0.36a | 0.06 ± 0.17 | 0.07 ± 0.22 | 0.16 ± 0.44a | 0.09 ± 0.28 | 0.05 ± 0.17 | 0.06 ± 0.22 | 0.12 ± 0.32b | 0.32 ± 0.74c | 0.08 ± 0.15 |
| P-80 | 0a | 0.01 ± 0.05 | 0 | 0 | 0 | 0.01 ± 0.05 | 0 | 0 | 0 | 0.08 ± 0.15 |
aSignificantly different to HC at P < .05
bSignificantly different to adult HC at P < .05
cSignificantly different to adults with IBD at P < .05
Abbreviations: CGN, Carrageenan; CMC, carboxymethylcellulose; P-80 polysorbate-80
Total refers to the total of the 6 emulsifiers of interest
Children with UC not reported separately due to low numbers (n = 2)
Summary of mean daily emulsifier exposure.
| IBD n = 176 | HC n = 191 | UC n = 75 | CD n = 101 | IBD (adults) n = 156 | HC (adults) n = 179 | UC (adults) n = 73 | CD (adults) n = 83 | IBD (children) n = 20 | HC (children) n = 12 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Total | 2.7 ± 1.8a | 2.3 ± 1.6 | 2.6 ± 1.6 | 2.7 ± 1.9a | 2.6 ± 1.8b | 2.2 ± 1.5 | 2.7 ± 1.7b | 2.6 ± 1.9 | 3.3 ± 2.0c | 3.9 ± 1.6 |
| MDGs | 1.2 ± 0.93 | 1.1 ± 0.84 | 1.2 ± 0.93 | 1.1 ± 0.93 | 1.2 ± 0.94 | 1.1 ± 0.82 | 1.2 ± 0.94 | 1.2 ± 0.95 | 1.1 ± 0.84 | 1.5 ± 0.93 |
| Lecithin | 0.85 ± 0.93 | 0.75 ± 0.73 | 0.88 ± 0.86 | 0.83 ± 0.97 | 0.79 ± 0.85 | 0.71 ± 0.68 | 0.87 ± 0.86 | 0.72 ± 0.84 | 1.3 ± 1.3c | 1.3 ± 1.1 |
| Xanthan gum | 0.38 ± 0.42 | 0.29 ± 0.43 | 0.37 ± 0.45 | 0.38 ± 0.41 | 0.39 ± 0.42b | 0.28 ± 0.42 | 0.38 ± 0.45 | 0.40 ± 0.39b | 0.28 ± 0.46 | 0.58 ± 0.47 |
| CGN | 0.20 ± 0.33 | 0.16 ± 0.28 | 0.19 ± 0.32 | 0.21 ± 0.35 | 0.18 ± 0.32 | 0.14 ± 0.27 | 0.19 ± 0.32 | 0.18 ± 0.32 | 0.33 ± 0.43c | 0.39 ± 0.40 |
| CMC | 0.12 ± 0.36a | 0.06 ± 0.17 | 0.07 ± 0.22 | 0.16 ± 0.44a | 0.09 ± 0.28 | 0.05 ± 0.17 | 0.06 ± 0.22 | 0.12 ± 0.32b | 0.32 ± 0.74c | 0.08 ± 0.15 |
| P-80 | 0a | 0.01 ± 0.05 | 0 | 0 | 0 | 0.01 ± 0.05 | 0 | 0 | 0 | 0.08 ± 0.15 |
| IBD n = 176 | HC n = 191 | UC n = 75 | CD n = 101 | IBD (adults) n = 156 | HC (adults) n = 179 | UC (adults) n = 73 | CD (adults) n = 83 | IBD (children) n = 20 | HC (children) n = 12 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Total | 2.7 ± 1.8a | 2.3 ± 1.6 | 2.6 ± 1.6 | 2.7 ± 1.9a | 2.6 ± 1.8b | 2.2 ± 1.5 | 2.7 ± 1.7b | 2.6 ± 1.9 | 3.3 ± 2.0c | 3.9 ± 1.6 |
| MDGs | 1.2 ± 0.93 | 1.1 ± 0.84 | 1.2 ± 0.93 | 1.1 ± 0.93 | 1.2 ± 0.94 | 1.1 ± 0.82 | 1.2 ± 0.94 | 1.2 ± 0.95 | 1.1 ± 0.84 | 1.5 ± 0.93 |
| Lecithin | 0.85 ± 0.93 | 0.75 ± 0.73 | 0.88 ± 0.86 | 0.83 ± 0.97 | 0.79 ± 0.85 | 0.71 ± 0.68 | 0.87 ± 0.86 | 0.72 ± 0.84 | 1.3 ± 1.3c | 1.3 ± 1.1 |
| Xanthan gum | 0.38 ± 0.42 | 0.29 ± 0.43 | 0.37 ± 0.45 | 0.38 ± 0.41 | 0.39 ± 0.42b | 0.28 ± 0.42 | 0.38 ± 0.45 | 0.40 ± 0.39b | 0.28 ± 0.46 | 0.58 ± 0.47 |
| CGN | 0.20 ± 0.33 | 0.16 ± 0.28 | 0.19 ± 0.32 | 0.21 ± 0.35 | 0.18 ± 0.32 | 0.14 ± 0.27 | 0.19 ± 0.32 | 0.18 ± 0.32 | 0.33 ± 0.43c | 0.39 ± 0.40 |
| CMC | 0.12 ± 0.36a | 0.06 ± 0.17 | 0.07 ± 0.22 | 0.16 ± 0.44a | 0.09 ± 0.28 | 0.05 ± 0.17 | 0.06 ± 0.22 | 0.12 ± 0.32b | 0.32 ± 0.74c | 0.08 ± 0.15 |
| P-80 | 0a | 0.01 ± 0.05 | 0 | 0 | 0 | 0.01 ± 0.05 | 0 | 0 | 0 | 0.08 ± 0.15 |
aSignificantly different to HC at P < .05
bSignificantly different to adult HC at P < .05
cSignificantly different to adults with IBD at P < .05
Abbreviations: CGN, Carrageenan; CMC, carboxymethylcellulose; P-80 polysorbate-80
Total refers to the total of the 6 emulsifiers of interest
Children with UC not reported separately due to low numbers (n = 2)
When comparing adults only, participants with IBD had a significantly higher total emulsifier exposure per day compared with HCs (2.6 ± 1.8 vs 2.2 ± 1.5, P = .04). A significant difference was observed between adult participants with UC (2.7 ± 1.7 exposures per day) compared with HCs (P = .04). No differences were observed between adults with CD compared with HCs (P = .14). There were no significant differences in total emulsifier exposure between CD and UC (P = .72) or those with active disease compared with those in remission (P = .12).
Children with IBD had a significantly higher mean emulsifier exposure compared with adults with IBD (3.3 ± 2.0 vs 2.6 ± 1.8, P = .04). Total emulsifier exposure per day did not differ between children with IBD and HC children (P = .55). A comparison between CD and UC was not undertaken in the pediatric cohort due to the low number of children with UC in the study population (n = 2).
For participants with IBD, the emulsifiers with the highest exposure per day were mono- and diglycerides of fatty acids (1.2 ± 0.93), followed by lecithin (0.85 ± 0.93) and xanthan gum (0.38 ± 0.42; Table 5). On average, participants with IBD were less frequently exposed to carrageenan (0.20 ± 0.33) and carboxymethylcellulose (0.12 ± 0.36 exposures per day), and there were no recorded exposures to polysorbate-80. Participants with IBD had a significantly higher exposure to carboxymethylcellulose per day compared with HCs (0.12 ± 0.36 vs 0.06 ± 0.17, P = .02). Conversely, HCs had a higher exposure to polysorbate-80 compared with participants with IBD (0.01 ± 0.05 vs 0, P = .03). No other differences in emulsifier exposure were found when comparing all participants with IBD to HCs or when comparing children with IBD to HCs.
When comparing adults only, people with IBD had a significantly higher exposure to xanthan gum per day compared with HCs (0.39 ± 0.42 vs 0.28 ± 0.42, P = .03; Table 5). Similarly, adults with CD had a higher exposure to xanthan gum compared with HCs (P = .05). Adults with CD also had a higher exposure to carboxymethylcellulose compared with HCs (0.12 ± 0.32 vs 0.05 ± 0.17, P = .02). No differences in emulsifier exposures were observed when comparing adults with UC to those with CD.
Amongst the IBD cases, children had a significantly higher exposure to lecithin (1.3 ± 1.3 vs 0.79 ± 0.85, P = .01), carrageenan (0.33 ± 0.43 vs 0.18 ± 0.32, P = .02), and carboxymethylcellulose (0.32 ± 0.74 vs 0.09 ± 0.28, P = .05) compared with adults. No differences were seen for emulsifier exposures when comparing people with active IBD with those in remission.
Food Sources of Emulsifiers
For participants with IBD, the most common food sources of the 6 emulsifiers of interest were “cereal-based products and dishes” (26%), closely followed by “cereal and cereal products” (25%), and “milk products and dishes” (12%; Table 6). The most common food sources of mono- and diglycerides of fatty acids were “cereal and cereal products” (48%), followed by “cereal-based products and dishes” (24%), and “fats and oils” (14%). Lecithin was most commonly found in “cereal-based products and dishes” (34%), “confectionary and cereal/nut/seed bars” (25%), and “special dietary food” (10%). The most common food sources of xanthan gum were “cereal-based products and dishes” (31%), “savory sauces and condiments” (30%), and “meat, poultry and game products and dishes” (18%). Carrageenan was mainly found in “milk products and dishes” (71%), while carboxymethylcellulose was mainly found in “cereal and cereal products” (53%) and “special dietary food” (21%). Polysorbate-80 was not found in any food items consumed by people with IBD. However, polysorbate-80 was identified in 4 unique food items consumed by healthy controls including McDonalds Big Mac, McDonalds McFlurry, Hungry Jacks Cheeseburger, and Hungry Jacks Whopper Junior. The following food groups did not contribute to emulsifier intake: “sugar products and dishes,” “legume and pulse products and dishes,” “soups,” and “fish and seafood products.”
Food group contribution to emulsifier exposure in participants baseline 3-day food diaries.
| Total* n (%) | MDGs n (%) | Lecithin n (%) | XG n (%) | CGN n (%) | CMC n (%) | P-80 n (%) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Food group | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC |
| Cereal-based products and dishes (eg, sweet biscuits, cakes, pastries) | 374 (26) | 385 (29) | 146 (24) | 179 (29) | 153 (34) | 142 (33) | 63 (31) | 50 (30) | 10 (9) | 7 (8) | 2 (3) | 3 (9) | 0 (0) | 4 (80) |
| Cereal and cereal products (eg, cereal, bread, rice) | 353 (25) | 356 (27) | 294 (48) | 306 (50) | 18 (4) | 18 (4) | 7 (3) | 14 (8) | 1 (1) | 4 (5) | 33 (53) | 14 (44) | 0 (0) | 0 (0) |
| Milk products and dishes (eg, ice-cream, custard, flavored milk) | 179 (12) | 149 (11) | 60 (10) | 63 (10) | 19 (4) | 22 (5) | 16 (8) | 1 (1) | 77 (71) | 57 (65) | 7 (11) | 5 (16) | 0 (0) | 1 (20) |
| Fats and Oils (eg, butter, margarine) | 153 (11) | 84 (6) | 86 (14) | 34 (6) | 67 (15) | 50 (12) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Confectionary and cereal/nut/seed bars (eg, chocolate, muesli bars) | 128 (9) | 152 (11) | 8 (1) | 9 (1) | 112 (25) | 143 (33) | 4 (2) | 0 (0) | 2 (2) | 0 (0) | 2 (3) | 0 (0) | 0 (0) | 0 (0) |
| Savory sauces and condiments (eg, gravy, dips, dressings) | 64 (4) | 80 (6) | 2 (0) | 1 (0) | 0 (0) | 0 (0) | 61 (30) | 71 (42) | 1 (1) | 7 (8) | 0 (0) | 1 (3) | 0 (0) | 0 (0) |
| Meat, poultry and game products and dishes (eg, sausages, processed meats) | 53 (4) | 39 (3) | 7 (1) | 4 (1) | 1 (0.2) | 2 (0.5) | 37 (18) | 25 (15) | 8 (7) | 8 (9) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Special dietary food (eg, protein shakes, nutrition supplements) | 74 (5) | 36 (3) | 6 (1) | 6 (1) | 46 (10) | 26 (6) | 4 (2) | 3 (2) | 5 (5) | 0 (0) | 13 (21) | 1 (3) | 0 (0) | 0 (0) |
| Dairy and meat substitutes (eg, almond milk, vegan cheese) | 44 (3) | 23 (2) | 1 (0.2) | 0 (0) | 35 (8) | 18 (4) | 6 (3) | 1 (1) | 2 (2) | 4 (5) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Other^ (eg, cordial, coconut cream) | 18 (1) | 31 (2) | 6 (1) | 12 (2) | 1 (0.2) | 6 (1) | 3 (1) | 4 (2) | 3 (3) | 1 (1) | 5 (8) | 8 (25) | 0 (0) | 0 (0) |
| Total* n (%) | MDGs n (%) | Lecithin n (%) | XG n (%) | CGN n (%) | CMC n (%) | P-80 n (%) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Food group | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC |
| Cereal-based products and dishes (eg, sweet biscuits, cakes, pastries) | 374 (26) | 385 (29) | 146 (24) | 179 (29) | 153 (34) | 142 (33) | 63 (31) | 50 (30) | 10 (9) | 7 (8) | 2 (3) | 3 (9) | 0 (0) | 4 (80) |
| Cereal and cereal products (eg, cereal, bread, rice) | 353 (25) | 356 (27) | 294 (48) | 306 (50) | 18 (4) | 18 (4) | 7 (3) | 14 (8) | 1 (1) | 4 (5) | 33 (53) | 14 (44) | 0 (0) | 0 (0) |
| Milk products and dishes (eg, ice-cream, custard, flavored milk) | 179 (12) | 149 (11) | 60 (10) | 63 (10) | 19 (4) | 22 (5) | 16 (8) | 1 (1) | 77 (71) | 57 (65) | 7 (11) | 5 (16) | 0 (0) | 1 (20) |
| Fats and Oils (eg, butter, margarine) | 153 (11) | 84 (6) | 86 (14) | 34 (6) | 67 (15) | 50 (12) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Confectionary and cereal/nut/seed bars (eg, chocolate, muesli bars) | 128 (9) | 152 (11) | 8 (1) | 9 (1) | 112 (25) | 143 (33) | 4 (2) | 0 (0) | 2 (2) | 0 (0) | 2 (3) | 0 (0) | 0 (0) | 0 (0) |
| Savory sauces and condiments (eg, gravy, dips, dressings) | 64 (4) | 80 (6) | 2 (0) | 1 (0) | 0 (0) | 0 (0) | 61 (30) | 71 (42) | 1 (1) | 7 (8) | 0 (0) | 1 (3) | 0 (0) | 0 (0) |
| Meat, poultry and game products and dishes (eg, sausages, processed meats) | 53 (4) | 39 (3) | 7 (1) | 4 (1) | 1 (0.2) | 2 (0.5) | 37 (18) | 25 (15) | 8 (7) | 8 (9) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Special dietary food (eg, protein shakes, nutrition supplements) | 74 (5) | 36 (3) | 6 (1) | 6 (1) | 46 (10) | 26 (6) | 4 (2) | 3 (2) | 5 (5) | 0 (0) | 13 (21) | 1 (3) | 0 (0) | 0 (0) |
| Dairy and meat substitutes (eg, almond milk, vegan cheese) | 44 (3) | 23 (2) | 1 (0.2) | 0 (0) | 35 (8) | 18 (4) | 6 (3) | 1 (1) | 2 (2) | 4 (5) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Other^ (eg, cordial, coconut cream) | 18 (1) | 31 (2) | 6 (1) | 12 (2) | 1 (0.2) | 6 (1) | 3 (1) | 4 (2) | 3 (3) | 1 (1) | 5 (8) | 8 (25) | 0 (0) | 0 (0) |
Significant difference between IBD and HC P < .05 (indicated in orange) *Refers to total of the 6 emulsifiers of interest. Abbreviations: MDGs, mono- and diglycerides of fatty acids; XG, xanthan gum; CGN, carrageenan; CMC, carboxymethylcellulose; P-80, polysorbate-80. n (%): number of emulsifier exposures in food group, and percentage of food group contribution to emulsifier exposure. ^Other includes seed and nut products and dishes, snack foods, egg products and dishes, vegetable products and dishes, fruit products and dishes, alcoholic beverages, and nonalcoholic beverage.
Food group contribution to emulsifier exposure in participants baseline 3-day food diaries.
| Total* n (%) | MDGs n (%) | Lecithin n (%) | XG n (%) | CGN n (%) | CMC n (%) | P-80 n (%) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Food group | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC |
| Cereal-based products and dishes (eg, sweet biscuits, cakes, pastries) | 374 (26) | 385 (29) | 146 (24) | 179 (29) | 153 (34) | 142 (33) | 63 (31) | 50 (30) | 10 (9) | 7 (8) | 2 (3) | 3 (9) | 0 (0) | 4 (80) |
| Cereal and cereal products (eg, cereal, bread, rice) | 353 (25) | 356 (27) | 294 (48) | 306 (50) | 18 (4) | 18 (4) | 7 (3) | 14 (8) | 1 (1) | 4 (5) | 33 (53) | 14 (44) | 0 (0) | 0 (0) |
| Milk products and dishes (eg, ice-cream, custard, flavored milk) | 179 (12) | 149 (11) | 60 (10) | 63 (10) | 19 (4) | 22 (5) | 16 (8) | 1 (1) | 77 (71) | 57 (65) | 7 (11) | 5 (16) | 0 (0) | 1 (20) |
| Fats and Oils (eg, butter, margarine) | 153 (11) | 84 (6) | 86 (14) | 34 (6) | 67 (15) | 50 (12) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Confectionary and cereal/nut/seed bars (eg, chocolate, muesli bars) | 128 (9) | 152 (11) | 8 (1) | 9 (1) | 112 (25) | 143 (33) | 4 (2) | 0 (0) | 2 (2) | 0 (0) | 2 (3) | 0 (0) | 0 (0) | 0 (0) |
| Savory sauces and condiments (eg, gravy, dips, dressings) | 64 (4) | 80 (6) | 2 (0) | 1 (0) | 0 (0) | 0 (0) | 61 (30) | 71 (42) | 1 (1) | 7 (8) | 0 (0) | 1 (3) | 0 (0) | 0 (0) |
| Meat, poultry and game products and dishes (eg, sausages, processed meats) | 53 (4) | 39 (3) | 7 (1) | 4 (1) | 1 (0.2) | 2 (0.5) | 37 (18) | 25 (15) | 8 (7) | 8 (9) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Special dietary food (eg, protein shakes, nutrition supplements) | 74 (5) | 36 (3) | 6 (1) | 6 (1) | 46 (10) | 26 (6) | 4 (2) | 3 (2) | 5 (5) | 0 (0) | 13 (21) | 1 (3) | 0 (0) | 0 (0) |
| Dairy and meat substitutes (eg, almond milk, vegan cheese) | 44 (3) | 23 (2) | 1 (0.2) | 0 (0) | 35 (8) | 18 (4) | 6 (3) | 1 (1) | 2 (2) | 4 (5) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Other^ (eg, cordial, coconut cream) | 18 (1) | 31 (2) | 6 (1) | 12 (2) | 1 (0.2) | 6 (1) | 3 (1) | 4 (2) | 3 (3) | 1 (1) | 5 (8) | 8 (25) | 0 (0) | 0 (0) |
| Total* n (%) | MDGs n (%) | Lecithin n (%) | XG n (%) | CGN n (%) | CMC n (%) | P-80 n (%) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Food group | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC | IBD | HC |
| Cereal-based products and dishes (eg, sweet biscuits, cakes, pastries) | 374 (26) | 385 (29) | 146 (24) | 179 (29) | 153 (34) | 142 (33) | 63 (31) | 50 (30) | 10 (9) | 7 (8) | 2 (3) | 3 (9) | 0 (0) | 4 (80) |
| Cereal and cereal products (eg, cereal, bread, rice) | 353 (25) | 356 (27) | 294 (48) | 306 (50) | 18 (4) | 18 (4) | 7 (3) | 14 (8) | 1 (1) | 4 (5) | 33 (53) | 14 (44) | 0 (0) | 0 (0) |
| Milk products and dishes (eg, ice-cream, custard, flavored milk) | 179 (12) | 149 (11) | 60 (10) | 63 (10) | 19 (4) | 22 (5) | 16 (8) | 1 (1) | 77 (71) | 57 (65) | 7 (11) | 5 (16) | 0 (0) | 1 (20) |
| Fats and Oils (eg, butter, margarine) | 153 (11) | 84 (6) | 86 (14) | 34 (6) | 67 (15) | 50 (12) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Confectionary and cereal/nut/seed bars (eg, chocolate, muesli bars) | 128 (9) | 152 (11) | 8 (1) | 9 (1) | 112 (25) | 143 (33) | 4 (2) | 0 (0) | 2 (2) | 0 (0) | 2 (3) | 0 (0) | 0 (0) | 0 (0) |
| Savory sauces and condiments (eg, gravy, dips, dressings) | 64 (4) | 80 (6) | 2 (0) | 1 (0) | 0 (0) | 0 (0) | 61 (30) | 71 (42) | 1 (1) | 7 (8) | 0 (0) | 1 (3) | 0 (0) | 0 (0) |
| Meat, poultry and game products and dishes (eg, sausages, processed meats) | 53 (4) | 39 (3) | 7 (1) | 4 (1) | 1 (0.2) | 2 (0.5) | 37 (18) | 25 (15) | 8 (7) | 8 (9) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Special dietary food (eg, protein shakes, nutrition supplements) | 74 (5) | 36 (3) | 6 (1) | 6 (1) | 46 (10) | 26 (6) | 4 (2) | 3 (2) | 5 (5) | 0 (0) | 13 (21) | 1 (3) | 0 (0) | 0 (0) |
| Dairy and meat substitutes (eg, almond milk, vegan cheese) | 44 (3) | 23 (2) | 1 (0.2) | 0 (0) | 35 (8) | 18 (4) | 6 (3) | 1 (1) | 2 (2) | 4 (5) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Other^ (eg, cordial, coconut cream) | 18 (1) | 31 (2) | 6 (1) | 12 (2) | 1 (0.2) | 6 (1) | 3 (1) | 4 (2) | 3 (3) | 1 (1) | 5 (8) | 8 (25) | 0 (0) | 0 (0) |
Significant difference between IBD and HC P < .05 (indicated in orange) *Refers to total of the 6 emulsifiers of interest. Abbreviations: MDGs, mono- and diglycerides of fatty acids; XG, xanthan gum; CGN, carrageenan; CMC, carboxymethylcellulose; P-80, polysorbate-80. n (%): number of emulsifier exposures in food group, and percentage of food group contribution to emulsifier exposure. ^Other includes seed and nut products and dishes, snack foods, egg products and dishes, vegetable products and dishes, fruit products and dishes, alcoholic beverages, and nonalcoholic beverage.
For total emulsifier exposure, people with IBD were exposed to significantly more emulsifiers from “fats and oils” (11% vs 6%, P < .001), “special dietary foods (Shake and bar meal replacements, protein shakes, nutrition supplements)” (5% vs 3%, P = .001), and “dairy and meat substitutes” (3% vs 2%, P = .02) compared with HCs (Table 6). Conversely, HCs had a higher total emulsifier exposure from “confectionary and cereal/nut/seed bars” (11% vs 9%, P = .03) compared with people with IBD. People with IBD were exposed to significantly more lecithin (10% vs 6%, P = .03) and carboxymethylcellulose (21% vs 3%, P = .02) from “special dietary food” compared with HC.
Discussion
This study quantified the frequency of emulsifier exposure in a cohort of Australian adults and children with IBD and compared intake to healthy controls. The key finding of this analysis was that people with IBD had a higher total emulsifier exposure per day compared with HCs. People with CD also had a higher exposure to emulsifiers than HCs.
The finding that emulsifier exposure in participants with CD differed from HC is consistent with recent literature. The ENIGMA study,21 a large international case-control study, found people with CD (aged ≥18 years) had a higher emulsifier intake (CMC, P80, carrageenan) compared with controls across Australia, Hong Kon,g and China (1 100 000 milligrams/year vs 770 000 milligrams/year). However, when the participant age in the present analysis was refined to adults only (≥19 years), the level of exposure of people with CD did not differ from HCs. This may be due to the different methods used between studies. First, the ENIGMA study calculated the total of 3 emulsifiers while this study determined the total of 6 selected emulsifiers. Second, the ENIGMA study quantified emulsifier intake (in milligrams) using the “maximum permissible limit” allowed within foods, while this study examined emulsifier exposure. Uncertainty will remain in this area, as there are currently no validated tools to measure emulsifier intake accurately, and it is likely that inconsistencies in measurement methods will continue.
Another important finding from this study was that despite small numbers of participants, children with IBD had a higher total emulsifier exposure compared with adults with IBD. They also had higher exposure to specific emulsifiers including lecithin, carrageenan, and carboxymethylcellulose. Children with IBD did not differ from HC children in this analysis. A study in France, Italy, the United Kingdom, and Ireland evaluated the intake of 13 food additives (including 5 emulsifiers) across age groups and observed the highest intake amongst children (ranging 1-17 years) across all regions.30 Therefore, it is likely that children in general have a higher ultra-processed food and emulsifier intake.
Other studies have also found MDGs and lecithin to be major contributors to emulsifier exposure in people with IBD. A study involving 20 adults with CD in the United Kingdom found the highest emulsifier exposures from lecithin (1.31/day) and MDGs (0.99/day) from 7-day food diaries.2 Similarly, this study found the highest exposures from MDGs (1.2/day) and lecithin (0.72/day) in adults with CD. Another study in the United States20 reported lecithin exposure to be high in children with CD aged 8 to 12 years (0.90/day), which is similar to the lecithin exposure (1.3/day) reported in children with IBD in this study (composed of 90% CD). Exposure to MDGs and lecithin did not differ between people with IBD and HCs in this study. This is likely because in general, people have high exposures to MDGs and lecithin, as they are reported to be 2 of the most widely used emulsifiers with respective estimated intakes of 80 and 55 mg/kg/body weight/day.28 The clinical implications of exposure to MDGs and lecithin in IBD are unclear and, at this stage, are regarded as safe.1
A surprising finding from this study was that CMC and P80 exposures were very low and were rarely found in foods consumed by study participants. Many previous authors have expressed concerns about the safety of CMC and P80 due to induction of intestinal inflammation and dysbiosis in preclinical studies. Despite these additives frequently referred to in the literature as commonly used dietary emulsifiers,5,15,31,32 people with IBD had no exposure to P80 and 0.12 exposures/day to CMC (equivalent to <1 exposure/week). Polysorbate-80 was only found in 4 fast-food items consumed by HCs, which equates to 0.08% of all unique food items consumed by participants. Similarly, Sandall et al2 found 0.01 exposures to P80/day and 0.14 exposures to CMC/day in people with CD. A study by Lee at al20 also found children with CD to be exposed to low levels of P80 (0.07/day) and CMC (0.05/day). It is important to acknowledge that people with IBD were exposed to more CMC from “special dietary food” than HCs in this study. However, this may be accounted for by 3 children with CD who were consuming oral nutrition supplements. Despite exposure to CMC being higher in people with IBD compared with HCs, it is unlikely that the levels consumed are reflective of the concentrations used in preclinical mice models.
The results of this study found that emulsifier exposure did not differ between active disease and remission. Similarly, a study by Logan et al found remission rates in children with CD did not differ between those on exclusive enteral nutrition (EEN) formulas containing soy lecithin, carboxymethylcellulose, polysorbate 80, and carrageenan compared with those that did not contain these emulsifiers. It is interesting to note that EEN, an effective nutrition intervention used in children with CD to achieve clinical remission, contains emulsifiers.33 Logan et al found that emulsifiers used in EEN formulas did not impact remission rates and thereby challenges the theory that emulsifiers are drivers of inflammation. However, the short duration of EEN (6-8 weeks) may not be reflective of the accumulative effect of emulsifier exposure over the long-term through dietary intake.
Carrageenan exposure is of particular interest to those with IBD, as there is evidence that consumption may induce IBD activity.11 A small randomized controlled trial in 12 people with UC found that carrageenan intake contributed to earlier relapse over a period of 1 year.11 Other preclinical trials have demonstrated carrageenan-induced inflammation in mice appears similar to the histopathology of UC.34 However, the results of this study found that emulsifier exposure did not differ between active disease and remission.11 A reason for this difference may be attributed to a type 1 error due to small sample size. Moreover, the food matrix may play a role in the digestive fate of emulsifiers, which may differ when administration of emulsifiers is in capsule form or via the drinking water of mice. No prior research exists exploring emulsifier exposure in a population of people with UC. Therefore, further research is required in this population before dietary recommendations can be made.
Cereal and cereal-based products have also been reported as a large contributor to emulsifier exposure in other studies. Sandall et al2 found bread, cereal, and rice products contributed 25%, and biscuits, cakes, and pastries contributed 17% of emulsifier exposure in people with CD. Similarly, the present study found “cereal and cereal products” (eg, bread, cereal, rice) contributed 25% and “cereal-based products and dishes” (eg, sweet biscuits, cakes, pastries) contributed 26% to emulsifier exposure in people with IBD. This is unsurprising as emulsifying agents are added to many Ultra Processed Foods (UPFs), with one of the main food categories reported to be breads and other baked goods.1 Emulsifiers are used in baked goods to strengthen the dough, increase volume, improve texture, and extend shelf life.35 However, there are some inconsistencies across studies about the most common emulsifiers reported in baked goods. For example, lecithin was most found in “cereal-based products and dishes” (34%), “confectionary and cereal/nut/seed bars,” (25%) and “special dietary food” (that is shake and bar meal replacements, protein shakes, nutrition supplements; 10%) in this study. Another study in the United States reported lecithin to be most common in “meal replacement beverages” (58%), “cookies,” (40%) and “savory snacks” (23%).20 The inconsistencies may be due to differences in the food supply and the variability of emulsifiers used across countries and manufactures.35 Thus, this makes it challenging to generalize findings across different countries.
A major strength of this research study is that it explored 5022 unique food items within the Australian food supply and identified the major food sources contributing to emulsifier intake, a previously identified research gap. As this information was derived from 3-day food diaries, it is reflective and relevant to what Australians are eating and thus can be used to inform future clinical trials and clinical advice. There was also a very small number (<2%) of misreports, indicating unbiased estimates of exposure. Furthermore, this is the first study to compare emulsifier intake between children and adults with IBD, people with active disease and remission, and between IBD types.
This study has limitations that should be acknowledged. First, this study is limited in the ability to capture the actual quantity of emulsifiers consumed (eg, in milligrams/day) due to a lack of information provided on food labels. Currently, there are no established methodologies to measure the exact food content of emulsifiers in foods, and thus individual intake of these foods cannot be quantified.1 Second, only soy lecithin and sunflower lecithin are listed as additive E322 on product labels; however, lecithin is also naturally found in egg yolk.36 As such, it is likely lecithin exposure was underestimated. Third, where food items were recorded in food diaries using simplified food descriptors (eg, Greek yogurt) as opposed to specific commercial brand names (eg, Peters Greek style yogurt), assumptions were made about how to evaluate exposure. This will affect the internal validity, as there may be considerable variation in the type and presence of emulsifiers used in food products across manufacturers.35 Fourth, insufficient detail is present for homemade and takeaway meals, and thus presence of emulsifiers cannot be confidently confirmed in these foods. In these cases, products were determined to have “no exposure,” and thus findings are likely to underestimate emulsifier exposure. This study only examined emulsifiers with established harm in preclinical models. However, there are around 65 emulsifiers permitted for use in the global food supply, and most are yet to be tested for inflammatory potential.2 Finally, this study is not able to determine whether current emulsifier intake reflects intake prior to diagnosis.
Current research investigating the effect of emulsifiers in IBD is limited to experimental models and small, acute human studies. Future research is needed to explore the impact of emulsifiers on the disease course, gut microbiota, and inflammation biomarkers (eg, CRP, fecal calprotectin) in controlled human trials with longer duration. There are several human clinical trials underway that will provide important insights on the efficacy of restricting emulsifier intake in people with IBD. This is of great interest, as there is some evidence that shifts to a gut microbiota with higher pro-inflammatory potential are proportional to the amount of emulsifiers and not type or origin.37 Furthermore, attaining an accurate quantitative measure of emulsifier intake is near impossible due to challenges associated with their quantification and extraction from food manufacturers as well as variability in use within the food supply.1 Thus, there is a need for methodologies to be developed and applied to precisely measure emulsifiers in food. Future research in IBD should extend beyond emulsifiers to explore exposure to other dietary constituents in UPF, such as advanced glycation end products (AGEs). Advanced glycation end products are formed when foods are heat-processed and are widely prevalent in modern Westernized diets. Excessively high intakes of AGEs have shown to cause oxidative stress and inflammation;38 however they are yet to be measured or explored in the context of IBD. Qualitative research exploring food avoidance practices and knowledge of emulsifiers in people with IBD would also be useful.
Overall, people with IBD were exposed to more emulsifiers than healthy controls. However, the largest contributors to emulsifier exposure were MDGs and lecithin, which are yet to demonstrate adverse inflammatory effects in preclinical or human studies. Intake of the inflammatory emulsifiers such as carrageenan, CMC, and P80 were shown to have the lowest mean exposures in this population. This may suggest their presence in the food supply is not as common as frequently stated.
Acknowledgments
University of New South Wales is the primary sponsor for the AIM study. This work was supported by Gastroenterology Society of Australia (GESA; Research Collaboration Award 2018), Sydney Children’s Hospital Randwick, St George and Sutherland Medical Research Foundation (SSMRF; Research Grant 2019), and Crohn’s Colitis Australia.
Funding
No funding was received for this study.
Competing Interests
None to declare.
Data Availability
Available on reasonable request to the authors.
