The controversy of dietary fiber and Crohn’s disease

  In 2013, Ananthakrishnan AN et al. published an observation based on the Nurses’ Health Study in the prestigious medical journal Gastroenterology, a prospective study that analyzed data from 170,766 women followed for up to 26 years. The annual incidence of inflammatory bowel disease in the observed population was 8 per 100,000 for Crohn’s disease and 10 per 100,000 for ulcerative colitis. The study divided the population into five equal groups based on dietary fiber intake, and compared the top 1/5 women (median dietary fiber intake of 24.3 g/day) with the bottom 1/5, and found an approximately 40% reduction in the incidence of Crohn’s disease (HR 0.59 95% CI 0.39-0.90) in the former group. Intake of fruit-derived dietary fiber reduced the risk of disease by 43% (HR 0.57 95% CI 0.38-0.85), vegetable-derived dietary fiber showed a trend but no statistical difference in reducing the incidence of Crohn’s disease (HR 0.74 95% CI 0.50-1.07), while cereal, whole-grain, and legume dietary fiber did not appear to alter the risk of disease. Dietary fiber was superior to colonic CD (HR 0.62 95% CI 0.38-1.01) in reducing ileal Crohn’s disease (HR 0.47 95% CI 0.22-1.00) or ileocolonic Crohn’s disease (HR 0.50 95% CI 0.29-0.86). Neither the total dietary fiber intake nor the intake of each fiber category changed the risk of ulcerative colitis. The median intake of fruit dietary fiber ranked in the top 1/5 was 6.4 g/day, which is approximately equivalent to 2 medium-sized apples or bananas (sic). The possible reason it reduces the incidence of Crohn’s disease is that fruit fiber contains more soluble fermentable fiber, which regulates intestinal flora inhibits flora translocation and ferments to SCFA to inhibit transcription of NFkb and pro-inflammatory factors. Certain components of vegetables activate aryl hydrocarbon receptors and modulate the immune response to extrinsic intestinal antigens. This study provided important information on the association between inflammatory bowel disease and dietary fiber as the first prospective long-term observation in a large sample, and attracted attention upon publication.  A review by Kaplan GG Land, published in the same issue, concluded that the retrospective study of vestigial hope has been flawed by recall error, low reliability of single dietary assessment, and small sample size, so that previous studies have had many contradictory results. The present study was conducted as a prospective observation with dietary [assessment every two years or so, using personal medical information as the basis for diagnosis, and its data are more reliable and its results have a high reference value. However, the paper also has some points to note: first, this investigation found fruit fiber to be effective in Crohn’s and prevention, but a variety of other fibers were not counted separately, and the intake of soluble and insoluble fibers could not be well distinguished. Second, this study was conducted on mainly well-educated white adult women, and its findings cannot yet be directly extended to different races, genders, and age groups. The top 1/5 of dietary fiber intake in this study were also those who smoked less, weighed less, and took aspirin less regularly, and although the effects of these factors can be statistically removed, the effects of confounding factors still need to be considered. The dietary fiber intake recommended by the American Medical Association is 14g of dietary fiber per 1000kcal of energy intake, so adult women and men should consume 25 or 38g of dietary fiber per day, respectively, but the subgroup with the highest dietary fiber intake in this case only consumed about 24g, indicating that most women do not consume enough, so the first thing we should do is to educate the public to consume dietary fiber according to the recommended amount. Future studies should also focus on the different effects of dietary fiber on Crohn’s disease and ulcerative colitis, the effects of dietary fiber on Crohn’s disease at different sites and behavioral characteristics (penetration, stricture), and the amount and timing of dietary fiber needed to prevent Crohn’s disease.  Stein AC et al. questioned the cause-and-effect relationship derived from this study. In Crohn’s disease there is a long time interval between the appearance of the first symptoms and the definitive diagnosis, the older the age the more so. Due to the presence of intestinal strictures, dietary fiber intake during this period may lead to worsening of symptoms, so patients with Crohn’s disease automatically reduce their dietary fiber intake. Many of the patients with inflammatory bowel disease in this study mainly counted their dietary fiber intake data from 2 to 4 years prior to their diagnosis, so it is possible that reduced dietary fiber intake due to intestinal stricture was considered as the etiology of the later diagnosis of Crohn’s disease. The diseased ileum is finer than the diseased colon so it is more sensitive to fiber, which could explain the greater effect of dietary fiber on ileal and ileocolic Crohn’s disease in this study. The lack of effect of dietary fiber on ulcerative colitis can also be explained by the fact that ulcerative colitis is less likely to have intestinal luminal strictures. Therefore, Stein AC et al. argue that the causal relationship stated in the article is not necessarily valid and clearly oppose the administration of high-fiber diets for those at risk of CD or with suspected or diagnosed CD.  Lee YY also argued that “a high-fat, high-carbohydrate diet lacking in fruits and vegetables is the culprit of many diseases” has become a dogma that dare not be crossed. Only some of the food questionnaires in the study clearly recorded dietary fiber intake, and there were no reliable statistics on the source. There is a general tendency for women to overestimate their intake of fruits and vegetables and underestimate their intake of meat and milk when they take the questionnaire. The grouping of the population based on dietary fiber intake may be completely wrong for 5% of the population. In addition, some of the women in this study who were initially thought to have inflammatory bowel disease later rejected the diagnosis, suggesting the possibility of overdiagnosis of inflammatory bowel disease in women. lee et al. concluded that there is insufficient evidence on how much fiber to consume in patients with inflammatory bowel disease, much less how much vegetables and fruits to recommend.  The original authors responded to these points. With regard to the question of causality, the intake in this study was not based on a single survey, but on multiple surveys over several decades. Although some of the data are based on 2 to 4 years before diagnosis, this time has exceeded the average interval between the appearance of symptoms and definitive diagnosis of Crohn’s disease; we also specifically studied data from 4 to 8 years before the diagnosis of Crohn’s disease, which also supports the above conclusion. In addition, stenosis is present at the beginning of the disease only in a small number of patients, so the automatic reduction of dietary fiber intake as a result is only a minority. In response to the comments of Lee et al, the original authors have carefully analyzed again in the unpublished literature the negative association of fruit and vegetable intake with the development of CD. As a prospective study, the dietary survey was conducted prior to the diagnosis of CD and multiple dietary assessments were performed, so the bias, if present, should be offset by differences between groups and not exaggerate the role of fruits and vegetables to influence the study conclusions. Because of the low prevalence of inflammatory bowel disease, our study is prospective, long-term, and with a large sample, and is strong evidence to explore IBD and environmental factors. We agree that a low residue diet is appropriate for patients with CD with confirmed or suspected small bowel strictures. However, further studies on the appropriate intake of dietary fiber and the relationship between dietary fiber and intestinal inflammation are needed. Our study is only about the risk factors for the development of Crohn’s disease and cannot be used to give treatment recommendations for patients with suspected or confirmed Crohn’s disease.  Combining the above debate and previous studies, the generally accepted views include: soluble dietary fiber is useful for maintaining intestinal health and promoting remission of intestinal inflammation; soluble and insoluble dietary fiber is useful for promoting intestinal motility, controlling body weight, lowering blood cholesterol, and reducing the risk of diabetes in the general population; the use of crude fiber (including stem and leafy vegetables and whole grains) in Crohn’s disease patients with combined intestinal stenosis may aggravate the obstruction and should be used. The use of coarse fiber (including leafy vegetables and whole grains) in Crohn’s disease with intestinal strictures may aggravate obstruction and should be avoided. The intake of dietary fiber varies between guidelines, ranging from 14 g of dietary fiber per 1000 kcal of energy intake to 40 g per day.  Personally, I agree that for the general healthy population, 250g of vegetables and 250g of fruits (both fresh weight) should be consumed daily, and vegetables and fruits should not be substituted for each other; from the experience of clinical treatment, some Crohn’s disease patients with insoluble fiber from the diet will indeed aggravate the symptoms of obstruction, while the insoluble fiber in the enteral nutrition preparation has been processed into powder, so it has no effect on the obstruction. There is no significant difference between enteral nutrition preparations with or without dietary fiber in maintaining inflammation remission and improving nutritional status.