点击显示 收起
Division of Gastroenterology, Hepatology, and Nutrition
Department of Pediatrics
Lille University Hospital and Faculty of Medicine
Lille
France
E-mail: p.jantchou@chru-lille.fr
Department of Epidemiology and Public Health
Lille University Hospital and Faculty of Medicine
Lille
France
Dear Sir:
We read with great interest the meta-analysis recently published in the Journal by Klement et al (1). A protective effect of breastfeeding on the risk of inflammatory bowel disease (IBD) was shown; the risk of Crohn disease (CD) decreased by 33%, and the risk of ulcerative colitis (UC) decreased by 23%. Klement et al emphasized that out of a total of 17 published studies, only 4 studies of CD and 4 studies of UC were of high methodologic quality (ie, the diagnosis of IBD was confirmed by a physician, information on breastfeeding was confirmed by the patients' mothers or another older close relative, and the response rate was 80% for both the case and control subjects). Klement et al suggested that additional research be conducted to strengthen the validity of their observations.
We recently conducted a pediatric, population-based, case-control study to examine the environmental risk factors associated with the development of IBD (2). The study area was the northern part of France, which has 5 790 526 inhabitants, 1 312 141 of whom are aged <17 y. Data on 222 incident cases of CD and 60 incident cases of UC occurring before 17 January 1988 and December 1997 were extracted from our IBD registry (EPIMAD), which has been in existence in northern France since 1988. The methods of the EPIMAD registry were published in detail previously (3). Briefly, interviewer practitioners collected the data for all patients who received a diagnosis of IBD between 1 January 1988 and 31 December 1997 from all of the gastroenterologists and pediatric gastroenterologists in the area. A final diagnosis of CD or UC was made by 2 expert gastroenterologists according to previously published criteria (3). The control subjects were randomly selected from telephone number lists (random digit dialing) and matched 1:1 to each case by age (±2 y), sex, and area of residence.
A questionnaire was devised that comprised 140 questions, including questions about the infants' and children's diets. Trained investigators personally interviewed both the study subjects and their mothers at home. The pediatric health booklet—which is mandatory in France for infants, children, and adolescents and contains information on pregnancy, delivery, childhood growth, vaccinations, and childhood infections—was also required to validate questions about the children's health. The response rate was 100% for case and control subjects. In a multivariate model adjusted for the mother's education level, breastfeeding (partial or exclusive) was a risk factor for CD (odds ratio: 2.1; 95% CI: 1.3, 3.4; P = 0.003). Children with CD were breastfed exclusively or nonexclusively for an average of 2 wk more than were the control subjects, but the difference was not significant (10 wk compared with 8 wk and 8 wk compared with 6 wk, respectively; P = 0.08). Familial history of IBD, Bacille Calmette-Guerin vaccination, and history of eczema were other significant risk factors for CD, whereas regular drinking of tap water was a protective factor. In the univariate analysis, history of breastfeeding had no effect on the risk of UC (OR: 1.07; 95% CI: 0.52, 2.22; P = 0.85). In a multivariate model, familial history of IBD, disease during pregnancy, and bedroom sharing were risk factors for UC, whereas appendectomy was a protective factor.
On the basis of Klement et al's criteria, the quality of our study methods was high. The diagnosis of IBD was always confirmed by 2 experienced gastroenterologists, the information on breastfeeding was always confirmed by the patients' mothers, and the response rate to the questions related to breastfeeding was very high (80% for both case and control subjects). Moreover, the availability of the pediatric health booklet enabled us to confirm the information given by the family. The association of breastfeeding with an increased risk of CD was an unexpected finding of our study. There is strong evidence that breastfeeding is associated with a decreased incidence and severity of a wide range of infectious diseases (4). Delayed infections occurring after weaning may lead to an inappropriate immune response and persistence of intestinal inflammation. The high level of pollution in our highly industrialized region may also have played a role, because industrial chemicals and environmental contaminants have been found in breast milk (5). Ultrafine and fine particles are potent adjuvants in antigen-mediated immune responses and cause inflammation in susceptible persons; a relation between microparticles and CD was recently considered (6, 7). Additional studies in other populations are needed to confirm or refute our results. From a public health perspective, the short and long-term benefits of breastfeeding overrule by far the increased risk of CD that we observed in our study (4).
ACKNOWLEDGMENTS
None of the authors had a conflict of interest related to the topic of this letter.
REFERENCES