Introduction Coeliac disease (CD) has a prevalence of up to 1% and yet only 10–15% is diagnosed. The need for a screening test that has a high diagnostic performance is essential. Current NICE guidelines recommend the use of Tissue trans-glutaminase IgA antibody (TTG) testing in both primary and secondary care; however the efficacy and cost effectiveness of TTG has been controversial. We aimed to assess the current role of TTG and its utilisation in the primary and secondary care setting against gold standard histology testing.
Method We retrospectively analysed 502 patients referred to secondary care for gastrointestinal (GI) investigations with potential symptoms of coeliac disease (Anaemia, Diarrhoea, and weight loss). All patients had an upper GI endoscopy at Whipps Cross University Hospital, a large district general hospital in London between July 2013 and August 2014. The endoscopy list is open to primary care and secondary care referrals for patients with suspected CD or reassessment of known CD and also includes patients referred for investigation of other upper GI symptoms such as dyspepsia. Patients were categorised for analysis according to the presence of TTG testing prior to endoscopy, where TTG was initially ordered, D2 biopsy results and gender and age. SPSS v22 was used to estimate the frequency, sensitivity, specificity, positive predictive Values and negative predictive values of TTG and a Pearson Chi square test to calculate significance.
Results A total of 502 patients with a mean age of 59.3 (SD+/- 19.019) at the time of endoscopy were identified (58.4% female vs. 41.6% male). Prior to endoscopy only 16.1% (n = 81) had TTG performed, with an additional 4.9% (n = 24) tested post endoscopy. Of these, 84% (n = 68) were requested in secondary care and 16% (n = 13) in primary care (P < 0.001). 72.3% (n = 363) had one or more key CD symptoms such as anaemia, weight loss, or diarrhoea. However of these only 12.4% (n = 45) had TTG serology tests prior to endoscopy. Comparing TTG with histological evaluation for villous atrophy showed 10 true positives, 3 false negatives, 84 true negatives and 0 false negatives. Sensitivity of TTG was 100% (95% CI 69.15 to 100.00), specificity 95.88% (95% CI 89.78 to 98.7), positive predictive value (PPV) was 71.43% (95% CI 41.90 to 91.61) and negative predictive value was 100% (95% CI 96.11 to 100.00).
Conclusion Our results suggest TTG is underutilised both in primary and secondary care when addressing key CD symptoms. TTG testing has a low PPV in patients being investigated for CD. Our findings support existing reports for the introduction of other antibodies such as deaminated glidine rapid point of care test as screening tools prior to endoscopy in CD.
Disclosure of interest None Declared.