OBJECTIVES: To assess the prevalence and clinical presentation of celiac disease (CD) in a cohort of children with HLA-DQ2 positive and evaluate the risk factors in the development of CD. METHODS: Between July 2004 and July 2005, parents of all healthy full-term newborns in our hospital were invited to participate. HLA-DQ2 was tested in blood sample of the umbilical cord. A point of contact serological test was performed on children between 2 and 3 years of age. Positive results were confirmed by serum anti-transglutaminase 2 and endomysial antibodies. Children with high autoantibody titers underwent an intestinal biopsy. Children of the cohort diagnosed with CD before the screening study were included. Sex, mode of delivery, breast-feeding duration, and age of gluten introduction were studied. RESULTS: Of 1291 children, 362 were HLA-DQ2 positive and 262 participated in the study. CD was diagnosed in 4.1% (95% confidence interval (CI) 1.9-6.3). In the whole cohort, 60% had gastrointestinal symptoms, 7% poor weight gain, and 33% were asymptomatic. Five children with potential CD and 6 with CD autoimmunity became negative (42.3%) and are still negative after 5 to 7 years. Female sex was at-risk factor odds ratio 5.7 (95% CI 1.5-20.9), whereas breast-feeding during gluten introduction had a protective effect odds ratio 0.1 (95% CI 0.01-0.8). CONCLUSIONS: Prevalence of CD in this cohort was 4%, half of whom had digestive symptoms. Because a high proportion of children showed a spontaneous disappearance of antibodies, prevalence studies of CD in young children should be based on intestinal damage so as not to overestimate results.
OBJECTIVES: To assess the prevalence and clinical presentation of celiac disease (CD) in a cohort of children with HLA-DQ2 positive and evaluate the risk factors in the development of CD. METHODS: Between July 2004 and July 2005, parents of all healthy full-term newborns in our hospital were invited to participate. HLA-DQ2 was tested in blood sample of the umbilical cord. A point of contact serological test was performed on children between 2 and 3 years of age. Positive results were confirmed by serum anti-transglutaminase 2 and endomysial antibodies. Children with high autoantibody titers underwent an intestinal biopsy. Children of the cohort diagnosed with CD before the screening study were included. Sex, mode of delivery, breast-feeding duration, and age of gluten introduction were studied. RESULTS: Of 1291 children, 362 were HLA-DQ2 positive and 262 participated in the study. CD was diagnosed in 4.1% (95% confidence interval (CI) 1.9-6.3). In the whole cohort, 60% had gastrointestinal symptoms, 7% poor weight gain, and 33% were asymptomatic. Five children with potential CD and 6 with CD autoimmunity became negative (42.3%) and are still negative after 5 to 7 years. Female sex was at-risk factor odds ratio 5.7 (95% CI 1.5-20.9), whereas breast-feeding during gluten introduction had a protective effect odds ratio 0.1 (95% CI 0.01-0.8). CONCLUSIONS: Prevalence of CD in this cohort was 4%, half of whom had digestive symptoms. Because a high proportion of children showed a spontaneous disappearance of antibodies, prevalence studies of CD in young children should be based on intestinal damage so as not to overestimate results.