Food antigens and enteroviruses are possible triggers of type 1 diabetes. Because permeability of the intestinal epithelium may facilitate contact of these antigens with the mucosal immune system, we set out to study intestinal permeability in patients with type 1 diabetes. Children with type 1 diabetes (n = 26, mean age 12 years, mean duration of disease 4 years) and 24 healthy age-matched control children were given mannitol and lactulose orally, and their intestinal permeability was measured as a percentage of this dose recovered in urine. Patients with type 1 diabetes did not differ in their permeability to lactulose, nor was their lactulose/mannitol ratio any different from that of controls. However, patients with type 1 diabetes who had the HLA-DQB 1*02 allele and, therefore, a higher risk for celiac disease (CD) absorbed significantly more mannitol (mean + 95% CI): 17.7% (15.2-20.2) than did those negative for this allele: 12.3% (8.2-16.4), p = 0.04. Their lactulose permeability was also higher: 0.30 (0.16-0.44) and 0.09% (0-0.18), respectively, p = 0.02. Although the differences in permeability reach statistical significance, there was still much overlap between the two groups in terms of actual laboratory values. The higher permeability of patients with the HLA-DQB1*02 allele suggests that these patients may be more prone to develop abnormal immune responses to food antigens.