Age related clinical features of childhood Coeliac disease in Australia

There has been an apparent increase in the incidence of coeliac disease (coeliac disease) over the past 30 years [1]. For instance, the number of reported cases in the Netherlands increased from 0.18 per 1000 live births from 1975–1990, to 0.54 per 1000 live births 1993–1994 [1]. There is also increasing recognition that symptomatic coeliac disease may be the tip of the iceberg, with many more asymptomatic cases in the community [2‐4].

It is unclear whether the changing prevalence of reported cases of coeliac disease is due to a real increase in the number of cases, enhanced awareness of disease or more reliable serological testing. Dietary changes, such as earlier introduction of gluten, have been suggested as contributing to the changing presentation of coeliac disease in some populations [5, 6]. Breast feeding, in particular breast feeding after the introduction of gluten, is protective against the development of coeliac disease [7].

There are few data on the clinical features, incidence and prevalence of coeliac disease in Australian populations. In a population based screening study from a rural community of Western Australia, 10/3001 were positive for anti-endomysial antibodies. This community was of predominantly Anglo-Celtic origin [4]. Australians come from diverse ethnic backgrounds, thus the prevalence reported in this community cannot be generalised to the Australian population.

The primary aim of this study was to describe the clinical features of children diagnosed with coeliac disease at one tertiary paediatric centre in Australia. Our secondary aim was to determine the presentation patterns of coeliac disease in the local community at different ages.

A retrospective cross sectional study was conducted at Sydney Children's Hospital, one of two tertiary referral paediatric hospitals in Sydney, Australia. The study population consisted of infants and children who had been referred to the Gastroenterology Department at Sydney Children's Hospital for investigation of possible coeliac disease between 1997 and 2002. Subjects were identified through records kept by the Departments of Nutrition and Dietetics, Gastroenterology, and Anatomical Pathology. Ethics approval was granted from the South East Sydney Area Health Service Research Ethics Committee-Eastern Section.

This study describes the clinical features at presentation in a group of Australian children diagnosed with coeliac disease during a time of increasing awareness of this disease and widespread availability of reliable laboratory screening tests. The children included in this retrospective review presented with a wide variety of clinical features. The most common clinical features of coeliac disease in younger children were the so-called classical symptoms of diarrhoea, irritability and weight loss. However, older children more commonly presented with abdominal pain. There were few cases without gastrointestinal symptoms in our cohort.

The clinical features of the children in this study were similar to the clinical features described from populations in New Zealand [11], South Yorkshire [12] and Sweden [13]. Although abdominal pain is a common symptom of older children with coeliac disease, there is no association between classical recurrent abdominal pain and coeliac disease [14].

Anaemia was identified in approximately one quarter of the cases. Furthermore, many children had laboratory evidence of iron deficiency but were not anaemic at the time of diagnosis. Unfortunately only 86% of subjects had a full blood count taken, and 50% had blood taken for iron studies. Thus the true prevalence of anaemia and iron deficiency is uncertain. Cross sectional studies have identified a high prevalence of coeliac disease in patients with iron or folate deficiency [15, 16]. The incidence of coeliac disease in patients in the community presenting only with iron or folate deficiency has been reported at 4.7% [15]. Anaemia was reported in only 5% of children with coeliac disease from the Netherlands [1]. In the United Kingdom, iron deficiency anaemia was reported in approximately 20% of adults with coeliac disease. Although the current data may have underestimated the true incidence of anaemia and iron deficiency, it is likely higher than that reported in these previous cohorts.

Growth failure in height and weight was not as common in this cohort as was described in coeliac children from the Netherlands, where 49.7% of children had poor growth [1]. In our study, 5% of children had a weight more than 2 SD below the mean, and 12% of children had a height more than 2 SD below the mean. Growth failure is more commonly seen in children diagnosed at a younger age than in our study [17]. The discrepancy in reported growth failure may also be due to different definitions of growth failure. In addition, because this was a retrospective study, it was not possible to obtain adequate information about growth velocity or of the children's genetic growth potential.

The children in our study were older at the time of diagnosis than previous surveys of children with coeliac disease. In a cohort of patients diagnosed between 1950 and 1969, 70 of 91 were less than 2 years at the time of diagnosis and only 3 were more than 6 years. In a smaller cohort of children diagnosed between 1972 and 1975, 32 of 42 were less than 2 years at the time of diagnosis and 6 were greater than 6 years [18]. In the Netherlands, the proportion of children with coeliac disease aged < 2 yrs has remained at about 60% from 1975 to 1994 [1]. An epidemic in coeliac disease in Swedish Children in the 1980's paralleled the change in feeding practices at the time, in particular increased exposure to wheat, rye and barley and decreased duration of breast feeding [5]. A prospective study of coeliac disease in infants showed that a longer duration of breast feeding, smaller amounts of gluten in the diet in infants less than 12 months [17] and breast feeding after gluten is introduced to the diet reduces the risk of coeliac disease in infants less than 2 years [7]. Thus, the later age of acquiring coeliac disease in the Australian population may be due to the delayed introduction of solids in the diet of Australian children. The current recommendations in Australia are to introduce solids at 6 months (NHMRC Dietary Guidelines for Children and Adolescents in Australia 1993).

In this study, there were relatively few cases of coeliac disease detected by the screening of at risk groups. This contrasts with the worldwide trend of an increased prevalence of coeliac disease detected by screening [2, 19‐22]. A large multi-centered study of children and adults in the United States of America showed that the prevalence of celiac disease was 1:22 in first degree relatives, 1:56 in symptomatic individuals or a disorder known to be associated with celiac disease (type 1 diabetes, down syndrome, anemia, arthritis, osteoporosis, infertility and short stature), and 1:133 in the not-at-risk population [23].

In our study, the best screening test in children aged over 2 years with symptoms of, or who are at risk of, coeliac disease was IgA EMA. Other studies have shown that this test has a sensitivity of 95–98%, specificity of 94–95%, positive predictive value of 91–95% and negative predictive value of 96–98% [24]. For children less than 2 years of age, IgA AGA may be more reliable [24]. Increasingly over the last few years, reports suggest that tissue-transglutaminase (tTG) antibody may be even more reliable than EMA as a screening test [25, 26], and in addition is more reproducible and reliable. This test was not available during the period of this review. Although the utility of these newer tests is improved compared to AGA antibodies, they are not yet accepted as diagnostic tests. Any child with symptoms suspicious of coeliac disease, even if negative serology, warrants further review by a paediatric gastroenterologist, as a small bowel biopsy remains the gold standard test for diagnosis of coeliac disease.

Coeliac disease in children does not always present with classical clinical features, especially in older children. The possibility of coeliac disease should be considered in any child presenting with diarrhoea, irritability and weight loss, and particularly in older children with abdominal pain. With the widespread availability of minimally invasive screening tests for coeliac disease, identifying cases in at risk groups by screening should also be considered.