Dietary triggers of colonic inflammation following treatment with exclusive enteral nutrition in children with Crohn's disease

Abstract Background Exclusive enteral nutrition (EEN) ameliorates gut inflammation in children with Crohn’s disease (CD). We have previously described the rapid rise in faecal calprotectin levels (FC) when children with CD return to their habitual diet after EEN treatment (1). We performed dietary assessment and measured biomarkers of food consumption in faeces of children with CD during early food reintroduction. Methods A composite outcome of clinical remission post EEN (weighted Paediatric Crohn’s Disease Activity Index <12.5) and a significant drop in FC (FC decrease: >500 mg/kg / >35%) was used to define the patient group. All patients completed 3-day estimated weight food diaries and provided a faecal sample. Patients were divided equally for statistical analysis purposes in two groups; above (Group 1) and below (Group 2) the median FC concentration at food reintroduction. Nutrient analysis was performed with WinDiets. Short-chain fatty acids, the gluten immunogenic peptide (GIP) and starch were measured in stool, as proxy of fibre, gluten and malabsorbed starch, respectively. Food groups were assigned using the National Diet and Nutrition Survey approach and statistics employing taxonomy were utilised in the food group analysis. All data are displayed using medians (Q1, Q3). Results Fourteen children provided an FC sample within 21 (15, 51) days post EEN. Patients were of similar age, weight and height (p = 0.4, p = 0.37, p = 0.79). Total energy intake did not differ between the groups (p = 0.37). Patients in Group 1 consumed more fibre than Group 2 [12.1 g (11.3, 19.9) vs. 9.9 g (7.6, 12.1), p = 0.04]. Protein and phosphorus intakes, expressed as a percentage of reference nutrient intakes, were also higher in Group 1 than Group 2 [protein (%): 262 (195, 291) vs. 211 (88, 215), p = 0.02], [phosphorus (%): 241 (171, 290) vs. 177 (131, 1901), p = 0.04]. Butyrate and valerate levels were higher in Group 1 than Group 2 [butyrate: 62.5 (41.8, 114) vs. 34.1 (29.2, 49.6), p = 0.02; valerate: 13.7 (9.3, 17.9) vs. 4.5 (2.3, 8.1), p < 0.01. There were no differences in GIP and faecal starch levels between the two groups. Red and processed meat intake was higher in Group 1 than Group 2 [151 g (66.7, 190) vs. 63.3 g (21.7, 67), p = 0.02)]. Cereals intake was non-significantly higher in Group 1 [389 (207, 405) vs. 231 (141, 279), p = 0.08]. Overall diet diversity did not differ between the two groups (p = 0.47). Conclusion The current analysis suggests that fibre, protein, phosphorus and red and processed meat may be associated with recurrence of colonic inflammation in children with CD during early food reintroduction. These findings should be confirmed in larger studies. Reference