Article Text

G62 A questionnaire survey of adherence to European guidelines for the diagnosis of coeliac disease in asymptomatic children with type 1 diabetes mellitus
  1. RR Atherton1,
  2. M Zilbauer2,
  3. R Heuschkel2
  1. 1School of Clinical Medicine, University of Cambridge, Cambridge, UK
  2. 2Department of Paediatric Gastroenterology, University of Cambridge, Cambridge, UK


Aims Updated guidelines were released by ESPGHAN in January 2012, outlining separate diagnostic algorithms for asymptomatic diabetic patients to those presenting symptomatically. Recent epidemiological analysis at our tertiary centre revealed an average period prevalence of coeliac disease (CD) in the type 1 diabetes mellitus (T1DM) population of 5.8%, highlighting the size of the population this concerns. We aimed to assess current practice in the diagnosis of CD in children and adolescents with T1DM in the UK and Germany 16 months following guideline publication.

Methods A short questionnaire was distributed to healthcare practitioners, with replies received from 17 regional centres in the UK and 19 in Germany. Responses were compared using chi-squared testing.

Results The guidelines recommend HLA typing of all T1DM patients at point of diagnosis, with the rationale that future CD testing (including anti-TG2) would only prove necessary for those who were genetically susceptible. Our results showed anti-TG2 to be the most prevalent first-line test in both the UK and Germany, with 81% of respondents choosing this option to screen asymptomatic diabetic children for CD. Only 14% use HLA typing for first-line testing (Figure 1).

Abstract G62 Figure 1

Stacked bar chart displaying responses to the question 'Which of the following tests do you currently use as first line (screen) in asymptomatic patients with T1DM?' Results are presented for both questionnaires distributed via the British Society of Paediatric Gastroenterology, Hepatology and Nutrition (BSPGHAN) and the German Society of Paediatric Gastroenterology (GPGE), as well as the combined result.

The guidelines also recommend serological testing for CD in HLA-DQ2/DQ8-positive (but asymptomatic) T1DM patients. However, only 58% of respondents would test these patients, compared to 86% who would test symptomatic children (p < 0.01). This was due mainly to non-compliance with the guidelines in UK centres where 35% would test asymptomatic children, compared to 82% for children with gastrointestinal symptoms (p <0.01) (Figure 2).

Abstract G62 Figure 2

Clustered bar chart displaying percentage of postive rsponses to the question 'Which of the following patients with T1DM should be tested for CD?’ when presented with the options (a) T1DM + gastrointestinal symptoms, and (b) T1DM + known HLA-DQ2/DQ8 positivity. Results are presented for BSPGHAN, GPGE and combined, as for Figure 1.

Finally, we created five clinical scenarios in which respondents were asked for their choice of management in asymptomatic patients with T1DM. Variable understanding of the ESPGHAN diagnostic algorithm was apparent, with correct responses ranging from 34–83% (Figure 3).

Abstract G62 Figure 3

Clustered bar chart displaying percentage of respondents managing patient scenarios A-E according to the ESPGHAN diagnostic algorithm for managing asymptomatic patients with T1DM. A = HLA negative patients; B = HLA positive, anti-TG2 lgA negative; C = HLA positive, anti-TG2 lgA <3 ULN, anti-EMA nagative; D = HLA positive, anti-TG2 lgA <10 ULN, anti-EMA negative; E = HLA, anti-TG2 lgA >10 ULN and anti-EMA positive, biopsy negative. Options provided were as follows: (1) recommend gluten-free diet; (2) follow up with regular repeat anti-TG2 lgA; (3) additional CD-specific testing; (4) refer; (5) no further CD-specific action. Results are presented for BSPGHAN, GPGE and combined, as for previous figures.

Conclusions There remains a significant lack of awareness of and/or adherence to ESPGHAN guidelines on the diagnosis of CD in the T1DM population, both in the UK and Germany. Clearer, potentially more simplified guidance and further education are required to optimise the identification and subsequent management of this significant patient population.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.