Coeliac disease and chronic fatigue syndrome

Abstract

Chronic fatigue syndrome (CFS) is defined as a medically unexplained fatigue, resulting in impairment of daily activities of at least 50%.1 In the past few years it has been shown that coeliac disease is a common disorder, present in 1 in 200 in the general population,2-4 Serological testing is used to select patients for biopsy of the small intestine, which is required for histological confirmation of the diagnosis of coeliac disease. Assessment of IgA anti-endomysium (EmA) (endomysial antibody) or anti-tissue transglutaminase is considered to be sensitive and specific serological tests for coeliac disease, while IgA antigliadine (AGA-IgA) and IgG antigliadine (AGA-IgG) are less specific.3 Fatigue has been shown to be one of the presenting symptoms of coeliac disease,2 but an association between CFS and coeliac disease has not yet been reported. As an immunological mechanism may underlie both CFS and coeliac disease, a connection between both conditions should not be excluded. The aim of this report was to assess the prevalence of serological antibodies associated with coeliac disease in a population of CFS patients compared with a control group. Because of a study on immunological and clinical correlates of CFS,5 we had available serum of 56 adult patients and 56 age and sex matched controls. Both groups originated from a primary care population. Patients were identified and invited to participate by their general practitioner; healthy controls were recruited from the same environment as the patients. AGA-IgA. AGA-IgG and EmA were assessed in a routine setting using an ELISA for AGA evaluation and monkey oesophagus as substrate for EmA in an indirect immunofluorescence test as described by Lerner et al.6 Mean age of the total group of 112 persons was 37.6 years (range 18–51), of whom 82 (73%) were women. EmA – the primary parameter – was not found in any of the CFS patients or controls. A positive AGA-IgA (>4 AU/ml) was present in one patient (7.3 AU/ml) and one control (4.1 AU/ml). AGA-IgG was positive (>12 AU/ml) in one control (82 AU/ml) but in none of the patients. There was no positive EmA in the patient group. With a sample of 56 persons, the corresponding estimate of population prevalence of zero has a 95% confidence interval of 0–0.054 (Poisson distribution). The prevalence of the serological markers (EmA, AGA-IgA and AGA-IgG) has been reported to be 6.2% in a population-based study.7 McMillan et al8 found a serological prevalence of 1.2% of EmA in a population survey. Hin et al2 have demonstrated a positive EmA test in 30 out of 1000 tested subjects, when they used a case-finding approach in a primary care setting. Coeliac disease was confirmed histologically in all 30 individuals. In our total group of 112, there was no positive test for EmA. AGA-IgA and AGA-IgG, the less specific tests for coeliac disease, were shown to be sporadically positive. The results showed no positive association between CFS and serological markers for coeliac disease. In our study, the patient and control groups were relatively modest in size. The power calculation showed that a prevalence of EmA of less than 5–6% should be expected in a population of CFS patients. Though the prevalence may still be higher than that found in the general population, a strong relationship between CFS and coeliac disease is unlikely to exist.