We screened for celiac disease, by means of IgA class anti-endomysium antibodies (EmA), 383 consecutive adults with insulin-dependent diabetes mellitus (IDDM). Two control populations entered the study as well: 151 adults with biopsy proven celiac disease, as true positives; and 520 controls (healthy and diseased) as true negatives. IgA-EmA positivity was found in 145 of 151 (96%) celiac disease patients but in none of the controls (100% specificity). EmA were positive in 12 of 383 (3.13%) IDDM patients: 10 of these positives underwent intestinal biopsy, which showed either partial or total villous atrophy. Only one patient presented with gastrointestinal complaints, but severe iron deficiency was found in all. The IDDM celiac patients were started on a gluten-free diet: four refused both the diet and the follow-up protocol. Approximately one year after gluten withdrawal no significant change in the degree of diabetes control was observed, while an increased requirement for insulin was observed in three of four patients who strictly complied with the diet. The prevalence of biopsy-proven celiac disease among adult IDDM patients (1:38), eight times higher than that recently estimated for the general Italian population and the absence, except in one case, of gastrointestinal symptoms emphasizes the benefit of screening programs on populations at risk.

No abstract

ysis compared with the former one, and the results are not shown.SIRs for UC were decreased to 0.83 or 0.93 in offspring whose parents were diagnosed with rectal or prostate cancers, respectively (Table 1). The SIRs were increased in offspring whose parents were diagnosed with thyroid cancer (SIR 1.28). The 95% CIs for these SIRs were close to 1.00, indicating that the results were of borderline significance. Similarly, familial risks were examined between cancer and CD. SIRs for CD were decreased to 0.90 in offspring whose parents were diagnosed with prostate cancer; also, there was a borderline (upper 95% CI 1.00) decrease for colon and rectal cancers. The SIRs were increased in offspring whose parents were diagnosed with thyroid (1.34) and other endocrine tumors (1.20).The results of the analyses showed no large familial aggregation between IBD and any type of cancer. However, SIRs for offspring IBD were decreased when their parents were diagnosed with prostate and rectal cancers. The changes were of borderline significance, but the consistency between UC and CD suggests that these are not chance findings. These cancers are not known to share genetic susceptibility, and unidentified environmental factors may be an explanation.The SIRs for IBD were increased when the parents were diagnosed with thyroid cancer. Previous observations have connected thyroid disease, including autoimmune thyroid conditions, and inflammatory bowel disease, but the familial aspects between these diseases need further attention. 7,8 The observed familial association between thyroid cancer, possibly related to thyroid disease, and IBD may be a lead to further genotyping studies testing the possibility of shared susceptibility genes between thyroid conditions and IBD.

The finding in primary IgA nephropathy of increased levels of IgA to food antigens and particularly to gliadin prompted the hypothesis that a subgroup of these patients may have latent coeliac disease.