Increased serum levels of antigliadin antibodies in cystic fibrosis
M. Cipolli,EA Valletta,C. Zampieri,G. Mastella
DOI: https://doi.org/10.1111/j.1651-2227.1993.tb12526.x
1993-01-01
Acta Paediatrica
Abstract:Prevalence of celiac disease (CD) in cystic fibrosis (CF) was reported to be at least 1 : 220 (l), which is a value greater than that expected by chance. Gut permeability in C F patients is significantly increased, probably because of changes in mucus characteristics, pancreatic insufficiency and gastric acid hypersecretion (2, 3). It is not known if the permeability defect, along with increased antigenic load caused by pancreatic insufficiency, plays a role in favoring gluten sensitive enteropathy. In this study serum levels of antigliadin antibodies (AGA) were evaluated in a group of children with C F and in healthy age-matched subjects. Serum IgA and IgG AGA levels were evaluated in 64 CF patients and 54 healthy children aged 0.5-10 years on a gluten-containing diet, without any sign suggesting CD. A micro-ELISA method (Eurospital-Pharma, Trieste, Italy) was used as follows: 100 pi of diluted serum (1 : 200) was incubated in duplicate on gliadincoated micro-wells for 1 h at 37°C. After repeated washings, alkaline phosphatase-conjugated anti-IgA and anti-IgG antibodies were incubated for 1 h at 37°C. Then, a fluorogenic substrate (4-methylumbelliferyl phosphate) was added; the reaction was read after 30 min at 37°C in a 3M Fluorofast 24 micro-reader and the result expressed as fluorescence units. Standard sera with low-range (negative serum from healthy subjects) and high-range (positive serum from pooled sera of CD patients with high AGA values) AGA levels were always included in the assay. AGA values were calculated as percent ratio between sample and positive serum readings. Kits from the same lot (IgA AGA: #2667; IgG AGA: #2679) were used throughout the study. Withinassay coefficient of variation (CV) for low-range and high-range sera were 7% and 2.7%, respectively, for IgA AGA, and 6.6% and 3.3% for IgG AGA; between-assay CV were 16% and 4.8% for IgA AGA and 13.5% and 4% for IgG AGA (data from the producer). Serum levels of total IgA and IgG were measured both in CF and in healthy children. No patient underwent jejunal biopsy. In C F patients, pancreatic function (sufficiency/insufficiency) was also recorded, on the basis of either three-day fat balance (normal fecal fat loss < 10% of fat intake), fecal chymotrypsin assay (Boeringwerke Biochemia, Mannheim; normal value > 5 IU/g stool) or duodenal intubation with secretinpancreozymin stimulation (output over 30 min/kg body weight of bicarbonate, trypsin, chymotrypsin and lipase) (4,5). Subjects with pancreatic insufficiency were treated with pancreatic supplements and their digestive function assessed regularly. Enzyme supplements had been administered during the first months of life in 95% of the children with pancreatic insufficiency. Mann-Whitney's U-test and two-tailed Student's ttest were used to evaluate differences between the two groups. Spearman's rank test was used to test the correlation between AGA levels and other variables. A statistical probability < 5% was accepted as significant. Clinical features, mean serum IgA, TgG and IgA, and IgG AGA levels in C F and healthy children are summarized in Table 1. Ten CF subjects with pancreatic insufficiency had abnormal fecal fat loss (> 10% of fat intake) despite pancreatic sufiplements. Total IgA, IgA and IgG AGA levels were significantly higher in CF patients than in healthy children. There was no correlation between IgA and IgG AGA and total IgA and IgG levels in both CF and healthy subjects. Age-dependent distribution of IgA and IgG AGA titers in both groups is shown in Fig. 1 (a-d); AGA in CF patients had a trend to higher values in the first years of age. This behaviour was confirmed when each group (CF and healthy subjects) was divided into two age subgroups-younger and older than four years. Significantly higher IgA (p<O.O5) and IgG AGA (p<O.O2) levels were found in C F patients aged less than 4 years (n=34) compared with older children (n=30). AGA levels were comparable in healthy subjects younger (n = 26) and older (n = 28) than four years. In healthy children, the upper normal IgA and IgG AGA limits (mean+3 SD) were 7% and 25%, respectively, of the