Abstract Background Ultra-processed food (UPF) intake, which often contains emulsifiers that might affect the intestinal barrier, has been linked to inflammatory bowel disease (IBD) incidence and progression.1 Here, we assessed intake of UPFs, emulsifiers and macronutrients in patients with IBD and healthy controls. Methods Dietary intake was cross-sectionally calculated from Food Frequency Questionnaires (FFQs) and food logs. Two 24-hour dietary recalls in IBD patients attending the infusion unit were compared to two prospective logs from healthy controls from the FOAM study.2 Kruskal Wallis with post-hoc Dunn test was used for continuous data, and Fisher exact test for proportions, with Bonferroni correction for multiple testing. Results We included 186 subjects (Table 1). Assessing food logs, UPF intake was similar in all groups, yet emulsifier intake was higher in UC and CD compared to controls (p=0.03 and p=0.0004, resp). Compared to controls, higher intakes of E402 (UC p=0.02, CD p=0.002), E411 (both p=0.01), E471 (UC p=0.02, CD p=0.0009), as well as carrageenan E407 (UC p=0.001, CD p=0.002), and xanthan gum E415 (UC p=0.03, CD p=0.01) were seen (Fig 1). Intake of E476 was higher in CD compared to controls (p=0.02), while intake of guar gum E412 was higher in UC compared to controls (p=0.02). The proportion of non-consumers of E402 was higher in controls (100%) compared to IBD (CD 83%, UC 67%, p=0.03). Sugar intake was higher in UC (p=0.004) compared to controls. When comparing FFQs, a higher intake of deli meat in CD was seen, compared to UC (p=0.04). Compared to controls, soft drink (UC p=0.0005, CD p=0.001), meat (UC p=0.02, CD p=0.04), and butter intake (UC p=0.03, CD p= 0.005) was higher in IBD. However, intake of breakfast cereals (UC p=6e-13, CD p=1e-10), legumes (UC p=2e-04, CD p=1e-05), meat substitutes (UC 6e-04, CD p=0.02), whole grain pasta and rice (UC p=2e-6, CD p=4e-5), and fibre (UC p=0.01, CD p=0.002) was lower in UC and CD. Coffee (p=0.03), potato (p=0.01), deli meat (p=3e-04), and sodium intake (p=0.02) was higher in CD patients than controls, while nut and seed intake was lower (p=3e-04). For UC, polysaccharide intake was lower in UC than controls (p=0.007). There was no correlation between C-reactive protein and intake of UPFs, emulsifiers, macronutrients or total energy intake (p>0.1). Conclusion We show higher intakes of emulsifiers in patients with IBD compared to controls, while UPF intake was comparable. The concomitant increased intake of specific emulsifiers such as carrageenan and guar gum that have previously been associated with disruption of the intestinal barrier, warrant further investigation into their role in the propagation and maintenance of gut inflammation.
Read full abstract