Contribution of histological, serological, and genetic factors to the clinical heterogeneity of adult-onset coeliac disease.

OBJECTIVE: Although the factors predisposing to coeliac disease (CD) are largely understood, it remains unclear what determines the clinical heterogeneity of the disease. The aim of this study was to explore the contribution of histological, serological, and genetic factors to disease presentation. MATERIAL AND METHODS: The study was designed as a retrospective chart review of 384 unrelated Caucasian patients diagnosed with CD after the age of 16 at a single UK centre. RESULTS: We found that 8.8% of IgA-competent CD patients were endomysial antibody (EMA)-negative. Compared with the EMA-positive group, EMA-negative CD patients had a lower prevalence of iron deficiency (52.0% versus 72.6%, p=0.03) and Marsh IIIb-c lesions (66.7% versus 85.3%, p=0.03). Histological severity at diagnosis correlated with anaemia (p<0.01), folate deficiency (p<0.01), and iron deficiency (p=0.05), but no other laboratory or clinical features. Compared with human leucocyte antigen (HLA)-DQ2.5-positive patients, those carrying HLA-DQ2.2 were similar in terms of all the characteristics we considered, whereas those carrying HLA-DQ8 had a lower frequency of EMA positivity (62.5% versus 92.6%, p<0.01). The proportion of EMA-positive patients increased with frequency of the HLA-DQB1*0201 allele (76.7% versus 92.3% versus 96.4% for 0 versus 1 versus 2 alleles, p<0.01); no other evidence of a gene-dose effect of HLA-DQB1*0201 was observed. CONCLUSIONS: Histological severity at diagnosis of CD is associated with anaemia and some micronutrient deficiencies, but no other clinical features. The proportion of EMA-positive patients is higher amongst those carrying HLA-DQ2 than in those carrying HLA-DQ8, and is highest in HLA-DQ2 homozygotes. We found no correlation between frequency of the HLA-DQ alleles encoding HLA-DQ2.5 and CD severity.

Pneumococcal infection in patients with coeliac disease.

OBJECTIVES: Some patients with coeliac disease are hyposplenic. Splenectomy is a risk factor for pneumococcal infection. Our objective was to determine the risk of invasive pneumococcal disease - septicaemia, pneumonia or meningitis - in patients with coeliac disease. METHODS: We analysed routinely collected, linked statistical records of hospital admission to study the risk of pneumococcal infection in patients with coeliac disease, in patients who underwent splenectomy and in a comparison cohort. The main outcome measure was the rate ratio for pneumococcal infection in the coeliac and splenectomized cohorts, compared with the comparison cohort. We undertook the study using linked records in two temporally and geographically distinct populations: the Oxford region (1963-1999) and the whole of England (1998-2003). RESULTS: The rate ratio of pneumococcal infection in patients with coeliac disease was 2.06 (95% confidence interval, 1.27-3.15) in the Oxford population and 1.61 (1.36-1.90) in England as a whole. As a comparison, the rate ratios in splenectomized patients were 3.40 (2.44-4.60) and 3.32 (2.80-3.90) in the Oxford and England populations, respectively. The increased rate ratio in coeliac patients persisted beyond the first year after diagnosis of the coeliac disease. The period covered by the Oxford study was mainly before the widespread availability of pneumococcal vaccination; but the availability of pneumococcal vaccine was widespread during the time of the English study. CONCLUSION: Some patients with coeliac disease have an elevated risk of invasive pneumococcal disease.