Background: The increased prevalence of UC has been paralleled by significant changes in diet over recent decades. These changes, coupled with the intimate interaction food has with both the gastrointestinal microbiome and mucosa, has lead to the hypothesis that dietary components play a role in ulcerative colitis (UC) pathogenesis. The aim of this study was to characterise the nutrient and food additive intake of patients with active UC and determine whether these differ between patients and healthy stool donors or the general Australian population.
Methods: Participants with mild to moderately active UC (Mayo score 3–10, endoscopic subscore ≥2) were recruited for a randomised control trial of faecal microbiota transplant for active UC. Faecal donors were healthy volunteers with no active medical problems as assessed by medical history, blood and stool screening. Patients with UC and faecal donors each completed a 3-day diet diary, which was analysed using FoodWorks 7 software (Xyris, Australia) to yield macro and micronutrient intake as well as dietary emulsifier and sulphate content for both groups. These were compared to sex-matched Australian population data on nutrient intake sourced from the Australian bureau of statistics (2014).
Results: 82 diet diaries for 65 patients and 17 donors were analysed. Patients were significantly older than donors (mean age 39.8 vs 30.6 years p=0.01); however the groups were well matched for gender distribution (UC: 58% male; donor: 47% male, p=0.42). The mean daily protein intake was significantly higher in those with UC compared to donors (103.2g, 93–113g vs. 81.0g, 68–93g; p=0.04) as was the mean energy intake (8915kj, 8226–9604kj vs. 7453kj, 6377–8529kj; p=0.04) and iron intake (UC 10.9g, 9.5–12.2g vs Donors 7.5g, 5.9–9.0g; p=0.01). Compared to donors, patients had a numerically higher mean intake of saturated fats (30.5g, 26–34g vs 23.6g, 20–27g, p=0.09) sugar (96.3g, 82–110g vs 76.6, 67–88g p=0.17) and sulphates (1923g, 1267–2578g vs 825g, 551–1099g, p=0.10) however these were not statistically significant. There were no significant differences between the groups for mean intake of: fibre (p=0.42), starch (p=0.39), total carbohydrate (p=0.16) or emulsifier (p=0.50). The mean intake of tested macronutrients for the general Australian population fell between the 95% confidence intervals (CI) for both male and female UC patients. The mean intake of sugar for the Australian population was greater than the 95% CI of sugar intake for donors.
Conclusions: Patients with active UC had significantly higher intake of protein and energy intake than healthy faecal donors. However, they had similar intake of these foods to the general Australian population. Donors consumed less sugar than the general Australian population.