A high-fat diet has been associated with an increased risk of ulcerative colitis (UC). We studied the effects of a low-fat, high-fiber diet (LFD) vs an improved standard American diet (iSAD, included higher quantities of fruits, vegetables, and fiber than a typical SAD). We collected data on quality of life, markers of inflammation, and fecal markers of intestinal dysbiosis in patients with UC. We analyzed data from a parallel-group, cross-over study of 17 patients with UC in remission or with mild disease (with a flare within the past 18 mo), from February 25, 2015, through September 11, 2018. Participants were assigned randomly to 2 groups and received a LFD (10% of calories from fat) or an iSAD (35%-40% of calories from fat) for the first 4-week period, followed by a 2-week washout period, and then switched to the other diet for 4 weeks. All diets were catered and delivered to patients' homes, and each participant served as her or his own control. Serum and stool samples were collected at baseline and week 4 of each diet and analyzed for markers of inflammation. We performed 16s ribosomal RNA sequencing and untargeted and targeted metabolomic analyses on stool samples. The primary outcome was quality of life, which was measured by the short inflammatory bowel disease (IBD) questionnaire at baseline and week 4 of the diets. Secondary outcomes included changes in the Short-Form 36 health survey, partial Mayo score, markers of inflammation, microbiome and metabolome analysis, and adherence to the diet. Participants' baseline diets were unhealthier than either study diet. All patients remained in remission throughout the study period. Compared with baseline, the iSAD and LFD each increased quality of life, based on the short IBD questionnaire and Short-Form 36 health survey scores (baseline short IBD questionnaire score, 4.98; iSAD, 5.55; LFD, 5.77; baseline vs iSAD, P= .02; baseline vs LFD, P= .001). Serum amyloid A decreased significantly from 7.99 mg/L at baseline to 4.50 mg/L after LFD (P= .02), but did not decrease significantly compared with iSAD (7.20 mg/L; iSAD vs LFD, P= .07). The serum level of C-reactive protein decreased numerically from 3.23 mg/L at baseline to 2.51 mg/L after LFD (P= .07). The relative abundance of Actinobacteria in fecal samples decreased from 13.69% at baseline to 7.82% after LFD (P= .017), whereas the relative abundance of Bacteroidetes increased from 14.6% at baseline to 24.02% on LFD (P= .015). The relative abundance of Faecalibacterium prausnitzii was higher after 4 weeks on the LFD (7.20%) compared with iSAD (5.37%; P= .04). Fecal levels of acetate (an anti-inflammatory metabolite) increased from a relative abundance of 40.37 at baseline to 42.52 on the iSAD and 53.98 on the LFD (baseline vs LFD, P= .05; iSAD vs LFD, P= .09). The fecal level of tryptophan decreased from a relative abundance of 1.33 at baseline to 1.08 on the iSAD (P= .43), but increased to a relative abundance of 2.27 on the LFD (baseline vs LFD, P= .04; iSAD vs LFD, P= .08); fecal levels of lauric acid decreased after LFD (baseline, 203.4; iSAD, 381.4; LFD, 29.91; baseline vs LFD, P= .04; iSAD vs LFD, P= .02). In a cross-over study of patients with UC in remission, we found that a catered LFD or iSAD were each well tolerated and increased quality of life. However, the LFD decreased markers of inflammation and reduced intestinal dysbiosis in fecal samples. Dietary interventions therefore might benefit patients with UC in remission. ClinicalTrials.gov no: NCT04147598.
Read full abstract