Longitudinal screening of HLA-risk and HLA-nonrisk children for celiac disease to age 15 years: CiPiS study.

OBJECTIVES: Autoantibodies against tissue transglutaminase (tTG) are serological markers of celiac disease. The aim was to study the applicability of human leukocyte antigen (HLA)-genotyping and tTG autoantibodies in the screening of celiac disease in a longitudinal birth cohort followed to age 15 years. METHODS: Included were 13,860 HLA-DQ-genotyped children at birth and previously invited to a screening at age 3 and 9 years, respectively. HLA-DQB1*02 and/or DQB1*03:02 (HLA-risk) children were compared with non-HLA-DQB1*02 and non-DQB1*03:02 (HLA-nonrisk) children. The present study reinvited 12,948/13,860 (93.4%) children at age 15 years of whom 1056/2374 (44.5%) participated in screening at both age 3 and 9 years. Both immunoglobulin A (IgA) and G (IgG) autoantibodies against tTG were analyzed separately in radiobinding assays. Persistently tTG autoantibody-positive children were examined with intestinal biopsy to confirm the diagnosis of celiac disease. RESULTS: At age 3 years, celiac disease was diagnosed in 56/1635 (3.4%) HLA-risk children compared with 0/1824 HLA-nonrisk children (p < 0.001). By age 9 years, celiac disease was diagnosed in 72/1910 (3.8%) HLA-risk children compared with 0/2167 HLA-nonrisk children (p < 0.001). Screening at age 15 years detected 14/1071 (1.3%) HLA-risk children positive for IgA-tTG and/or IgG-tTG of whom 12/1071 (1.1%) remained persistently positive. Among those, 10/1071 (0.9%, 95% confidence interval: 0.4%-1.7%) HLA-risk children were diagnosed with celiac disease compared with 0/1303 HLA-nonrisk children (p < 0.001) and 5/491 (1.0%) were negative in screenings at both 3 and 9 years of age. CONCLUSIONS: Screening for celiac disease needs to be performed at multiple timepoints to detect all cases but can be restricted to children at HLA-risk.

Reduced Bone Mineral Density in Children With Screening-detected Celiac Disease.

OBJECTIVES: The aim of the study was to assess whether bone mass and metabolism are impaired in genetically at-risk children with screening-detected celiac disease. METHODS: Included were 71 children with screening-detected celiac disease diagnosed at 10.0 ±â€Š0.7 (mean ±â€Šstandard deviation) years and 142 matched controls and 30 children with screening-detected celiac disease diagnosed at 3.3 ±â€Š0.4 years of age presently on a gluten-free diet for 6.9 ±â€Š1.1 years and 60 matched controls. All participants were assessed for bone mineral density (BMD) of total body and spine by dual x-ray absorptiometry, serum 25(OH) vitamin D3, parathyroid hormone (PTH), interleukin (IL)-1ß, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-13, IL-15, interferon gamma, and tumor necrosis factor alpha. RESULTS: At diagnosis, screening-detected celiac disease children as compared to controls had a mean -0.03 g/cm reduced BMD of both total body and spine (P = 0.009 and P = 0.005, respectively), a mean -11.4 nmol/L lower level of 25(OH) vitamin D3 (P < 0.001), and a mean +1.0 pmol/L higher PTH level (P < 0.001). Systemic levels of the cytokines IL-1ß, IL-6, IL-8, IL-10, IL-12p70, IL-13, and tumor necrosis factor alpha were all increased in screening-detected celiac disease as compared to controls (P < 0.001). No difference in BMD, 25(OH) vitamin D3, PTH, and cytokine levels were detected in children on a gluten-free diet compared with controls. CONCLUSIONS: Children with screening-detected celiac disease have reduced BMD, lower levels of vitamin D3, higher levels of PTH, and signs of systemic inflammation compared with controls. These differences were not found in celiac disease children on a gluten-free diet, indicating that children with screening-detected celiac disease benefit from an early diagnosis and treatment.

Repeated Screening Can Be Restricted to At-Genetic-Risk Birth Cohorts.

OBJECTIVES: Celiac disease (CD) is associated with tissue transglutaminase autoantibodies (tTGAs) in individuals carrying the human leukocyte antigen (HLA) risk haplotypes DQA1*05:01-DQB1*02:01 (DQ2) and/or DQA1*03:01-DQB1*03:02 (DQ8). The aim of the study was to identify CD in an HLA-genotyped birth cohort prospectively screened for CD. METHODS: In the initial screening, 13,860 HLA-DQ-genotyped children were invited, of whom 3435/13,860 (25%) accepted participation. Of the 3435, 1620 (47%) carried DQ2 and/or DQ8, of whom 73 (4.5%) were tTGA positive assessed in radioligand-binding assays and 56 (3.5%) developed CD. At age 9 years, 13,024 children from the original cohort were re-invited to follow-up screening using the same study protocol and tTGA assays as in the first screening. Diagnosis of CD was confirmed by intestinal biopsy in children with persistent tTGA. RESULTS: In the follow-up screening, 1910/4077 (46.8%) carried DQ2 and/or DQ8, of whom 79/1910 (4.1%) were persistently tTGA positive and 72/1907 (3.8%) developed CD. Only 1/2167 (0.05%) child without HLA risk was IgG-tTGA positive, but did not have CD. Of the 980/1910 (51%) children carrying DQ2 and/or DQ8 who were already screened at 3 years of age, 30/979 (3.1%) were diagnosed as new patients at 9 years of age, compared with 42/928 (4.5%) children who did not participate in the initial screening (P = 0.094). CONCLUSIONS: Screening for CD can be restricted to children carrying HLA-DQ2 and/or DQ8. Repeated screening using tTGA is necessary to identify new patients by 9 years of age. These findings may be relevant when considering implementing screening of the general population.

Coeliac disease in children: a social epidemiological study in Sweden.

AIM: Little is known on the possible existence of socioeconomic and geographical differences in early coeliac disease (CD) risk. Therefore, we investigated these aspects in children before age two. METHODS: Linking the Swedish Medical Birth Registry to several other national registries, we identified all singletons born in Sweden from 1987 to 1993 (n = 792,401) and followed them until 2 years of age to identify cases of CD. Applying multilevel logistic regression analysis, we investigated the association between socioeconomic position (SEP) and CD in children and also whether a possible geographical variation in CD risk was explained by individual characteristics. RESULTS: Low SEP was associated with CD in boys OR 1.37 (95% CI 1.03-1.82), but not in girls OR 0.87 (95% CI 0.68-1.12). We found a considerable geographical variation in disease risk (i.e. intra-municipality correlation ≈ 10%) that was not explained by individual characteristics. CONCLUSIONS: Low SEP is associated with CD in boys but not in girls. Also, CD appears to be conditioned by geographical area of residence. While our study represents an innovative contribution to the epidemiology of CD in children, the reasons for the observed geographical and socioeconomic differences could be speculated but are still unknown.

Early human pregnancy serum cytokine levels predict autoimmunity in offspring.

It is generally believed that pregnancy is mediated by a Th2 response, which includes cytokines that promote placental growth and are involved in inducing tolerance to the foetus. If the balance between Th1/and Th2-mediated cytokines is disrupted, systemic and local changes could predispose the foetus to future disease. Therefore, a shift in the Th1/Th2 balance during pregnancy, possibly caused by underlying environmental factors, could be associated with post-partum autoimmune disease in the offspring. Based on this presumption, we used celiac disease as a model to investigate whether autoimmunity is triggered in the foetus during early pregnancy, observed as changes in the mother's cytokine profile. Ten cytokines were measured by electro-chemi-luminescent multiplex ELISA in serum samples obtained from mothers during early pregnancy. Cases included women with children who had developed verified celiac disease before the age of 5, who were compared with other women as matched controls. We observed that 7 out of 10 cytokine levels were significantly increased in our case mothers when compared to controls. Five of these belonged to what is generally known as a Th1-mediated response (TNFα, IFNγ, IL-2, IL-1ß and IL-12) and two were Th2 cytokines (IL-13 and IL-10). However, the IL-10 cytokine is known to have features from both arms of the immune system. These results were confirmed in a logistic regression model where five out of the initial seven cytokines remained. This study suggests that increase in Th1 serum cytokines may be associated with celiac disease in offspring.

Screening detects a high proportion of celiac disease in young HLA-genotyped children.

BACKGROUND AND AIMS: Celiac disease is associated with tissue transglutaminase autoantibodies (tTGAb) and the human leukocyte antigen (HLA)-risk alleles DQB1*02 and DQB1*0302. The aim was to estimate the proportion of undiagnosed celiac disease in children with HLA risk at 3 years of age. PATIENTS AND METHODS: From a population-based HLA-DQ screening study of newborns born between June 2001 and August 2004 in the southern part of Sweden, 6206 children with HLA-risk alleles were identified and asked to participate at a mean 3.3 +/- 0.4 years of age. As controls, 7654 children with HLA-nonrisk alleles were asked to participate. In all, 1620 (26.1%) children with HLA risk and 1815 (23.7%) controls were screened for tTGAb using radioligand-binding assays. Celiac disease was established by intestinal biopsy in children with a confirmed positive tTGAb test. RESULTS: Twenty-three children reported already having clinically diagnosed celiac disease and did not participate further. In children with HLA-risk genotypes, 73 of 1620 (4.5%, 95% CI 3.5%-5.5%) were tTGAb-positive compared with none of 1815 from the controls (P < 0.0001). Seventy-one children underwent biopsy (1 refused biopsy and 1 biopsy failed), of whom 56 of 1618 (3.5%, 95% CI 2.6%-4.4%) had damaged intestinal mucosa classified as celiac disease. The ratio between clinically and screening detected celiac disease in this study was 1:2.4 (23:56). CONCLUSIONS: The proportion of clinically undetected celiac disease may be particularly high among 3-year-old children with HLA-DQB1*02 and DQB1*0302 in Sweden, where these 2 HLA-risk alleles frequently occur.

Coeliac disease-specific tissue transglutaminase autoantibodies are associated with osteoporosis and related fractures in middle-aged women.

OBJECTIVE: To investigate whether the serological marker for coeliac disease, tissue transglutaminase autoantibody (tTGAb), is associated with decreased bone mass density (BMD) and increased frequency of fractures in middle-aged women screened for osteoporosis. MATERIAL AND METHODS: The study comprised 6480 women (mean age 56 years, range 50-64) who answered a number of questionnaires and who underwent dual X-ray absorptiometry of the wrist bone. Serum samples were analysed for tTGAb using radioligand binding assays. A tTGAb level of >4 U/ml was used to determine a positive value and a level of >17 U/ml was used as an alternative discrimination of high levels. RESULTS: A tTGAb level >4 U/ml was found among 90/6480 (1.4%) women and correlated with lower BMD (multiple linear regression coefficient -382.1; 95% CI = - 673.6-90.7, p=0.011) and with fracture frequency (r=0.18, p=0.023). The 59 women with tTGAb levels >or=17 U/ml had a lower BMD (0.41+/-0.08 g/cm(2) versus 0.44+/-0.08 g/cm(2), p=0.001) and a lower T-score (-1.40+/-1.28 versus -0.90+/-1.40, p=0.003) as well as a higher prevalence of osteoporosis (13.4% versus 6.5%, p=0.008) compared with the remaining 6421 women with tTGAb levels <17 U/ml. Furthermore, fracture frequency was more pronounced in women with tTGAb levels >or=17 U/ml, among whom 19/59 (32.2%) had fractures during the study period compared with 1204/6421 (18.8%) among women with tTGAb levels <17 U/ml (p=0.009). CONCLUSIONS: High levels of tTGAb indicating coeliac disease are associated with lower BMD and higher fracture frequency in women between 50 and 64 years of age. Osteometry is therefore warranted in middle-aged women detected with tTGAb.