Prevalence of celiac disease: Before and after a national change in feeding recommendations

Objective. A national change in infant feeding recommendations was proposed in 1996 in Sweden: a slow introduction to gluten during weaning was stressed, the recommendation being introduction at 4 instead of 6 months of age. The aim of the present study was to compare the prevalence of celiac disease in healthy young children born before and after the new feeding recommendations in 1996. Material and methods. Sera from 679 children at a median age of 2.9 years ( range 2.5 - 4.2 years) born between January 1996 and November 1997 were investigated with IgA-antigliadin antibodies (AGA) and IgA-endomysial autoantibodies (EMA) and compared with 690 age-matched children born between July 1992 and June 1993. Children with a positive test for EMA and AGA or EMA only were re-tested, and if positive at follow up, investigated with intestinal biopsy. Results. At baseline, 2.2% (15/679) children were positive for EMA and another 0.6% (4/679) for both EMA and AGA. One child refused to be re-tested and eight children were still EMA positive at follow-up. Intestinal biopsy was performed in seven children ( one declined biopsy), of whom three showed total villous atrophy. Two children with EMA titers 1: 640, respectively, refused further participation in the study, but were strongly suspected to have celiac disease. In total, 0.7% (5/679) (95% confidence interval (CI) = 0.1 - 1.4%) were considered to have celiac disease compared with 1.3% (9/690) ( 95% CI = 0.4 - 2.2%) in the control group ( p = 0.4217). In addition, 0.3% of the children were diagnosed with symptomatic celiac disease compared with 0.7% in controls ( p = 0.0134). Conclusions. The prevalence of symptomatic celiac disease declined after the infant dietary recommendations were introduced in 1996, but we could not find any difference in undiagnosed celiac disease between the screened children born before and those born after 1996.