Are current clinical, serological, endoscopic and histological methods to diagnose coeliac disease adequate?

Are current clinical, serological, endoscopic and histological methods to diagnose coeliac disease adequate?

Background: Coeliac disease is a common condition caused by a heightened immunological response to dietary gluten. Coeliac disease is now recognised to affect up to 1 % of western populations and is increasingly being recognised in global populations. However for every patient diagnosed with coeliac disease 3 to 7 remain undiagnosed. Undiagnosed coeliac disease can have potentially serious consequences including anaemia, osteoporosis and malignancy. Patients may also have unnecessary investigations and treatment for symptoms that may be effectively treated with a gluten free diet. The aim of this thesis was to identify methods by which the diagnosis of coeliac disease could be improved leading to fewer missed diagnoses.

Methods: To ascertain the importance of presenting characteristics to secondary care a prospective case finding study was undertaken. Patients were prospectively recruited from outpatient clinics from 4 UK hospitals and were invited to undergo serological testing for coeliac disease. Thereafter 3 commercially available point of care tests (Biocard, Coeliac Quick Test, and Simtomax) were evaluated in endoscopy and compared to the gold standard of villous atrophy on duodenal biopsy. Furthermore to evaluate the endoscopic and histological methods of diagnosing coeliac disease biopsy practices were assessed in a retrospective study of 4 UK hospitals. The diagnostic role of an additional duodenal bulb biopsy was then determined in a prospective study of patients attending for upper GI endoscopy who underwent routine duodenal biopsy.
Results: 4089 patients were recruited to the case finding study across 4 sites. Multivariate analysis of patients referred to secondary care showed family history of coeliac disease (Adjusted Odds Ratio (AOR) 1.26 p<0.0001), anaemia (AOR 1.03 p<0.0001) and osteoporosis (AOR 1.1 p0.006) were independent risk factors for a diagnosis of coeliac disease.
Evaluation of the Biocard point of care test in 523 patients (63.5% female, mean age 49.7) with no prior diagnosis of coeliac disease revealed sensitivity, specificity, positive predictive value and negative predictive values were 70.1%, 96.6%, 85.4%, and 91.8% respectively. This compared poorly with standard serology. However in a head to head comparison of 3 available point of care tests, Simtomax performed the best with the highest sensitivity (96.6%) for predicting villous atrophy.
During a 3 month period across 4 hospital sites 1423 patients underwent duodenal biopsy for possible coeliac disease. Guidelines to take at least 4 biopsies were only met in 40% of patients. Coeliac disease was more likely to be diagnosed if guidelines were followed (10.1% versus 4.6%, P<0.0001).
1378 patients underwent routine duodenal biopsy with 268 (19.4%) new coeliac disease cases. 9.7% of coeliac disease patients had villous atrophy confined to duodenal bulb (D1) (Ultra-Short Coeliac Disease (USCD))(P<0.0001). In 171 (mean age 46.5, 64% female) patients quadrantic D1 biopsies were taken. A single additional D1 biopsy from any site increased sensitivity by 9.3-10.8% (P<0.0001). The 26 patients with USCD were compared to conventional coeliac disease and controls. Patients with USCD were younger (p=0.03), with lower tissue-transglutaminase antibody (tTG) titres (p=0.001) and were less likely to present with diarrhoea (P=0.001). The prevalence of ferritin deficiency (P=0.007) and folate deficiency (P=0.003) was higher in conventional coeliac disease than those with USCD and controls.

Conclusions: I have demonstrated that case finding with serological testing of all patients attending gastroenterology services increased the detection rates for coeliac disease. Furthermore adherence to biopsy guidelines and taking an additional duodenal bulb biopsy also led to increased diagnostic rates in the endoscopic setting. A point of care test for coeliac disease used in endoscopy could be used to direct duodenal biopsy however my results would not support the use of the Biocard test in this role but Simtomax appears to be promising as a potential candidate. I believe that this body of work has significantly contributed to the field of coeliac diagnosis and has resulted in significant changes to clinical practice.