Introduction Coeliac disease is associated with significant morbidity, and is an underdiagnosed condition with an estimated prevalence of 1% in the UK population. Current British Society of Gastroenterology Society guidelines classify individuals with positive coeliac serology and negative duodenal biopsy to latent coeliac disease and do not recommend further investigation or the initiation of a gluten free diet.

In our recent practise patients at our institution with a strong clinical suspicion of coeliac disease and negative duodenal biopsies following a gluten containing diet undergo a push enteroscopy with biopsies of the jejunum and duodenum and histological reassessment of the original duodenal biopsies before a diagnosis of latent coeliac disease is made.

We aimed to investigate whether push enteroscopy and histological reassessment increased the diagnostic yield of coeliac disease.

Methods We searched our prospective endoscopy database for all patients with positive EMA or TTG results but negative D2 biopsies, who had undergone enteroscopy for a possible diagnosis of coeliac disease since 2007. Data including serology, result of histological reassessment of duodenal biopsies and assessment of jejunal biopsies were recorded.

Results 13 patients were identified; 4 (31%) were male, the median age was 45 (range 19–78). In all 13, EMA testing had been performed, of which 9 were positive (69%). In 9, TTG results were available, of which 9 were positive (100%). Following review of the original duodenal biopsies by a second pathologist, 7 (54%) of the 13 patients were re-classified to active coeliac disease; median Marsh grade of 1 (range 0–2)

Of the remaining 6 patients, 5 (83%) had evidence of active coeliac disease on jejunal biopsy; median Marsh grade of 2 (range 0–3b). Review of the jejunal biopsies alone, without review of original duodenal biopsies, would have led to active coeliac disease being diagnosed in 12 (92%) of 13 patients.

Conclusion In this small retrospective series of patients with positive coeliac serology and negative duodenal biopsies, repeat histological assessment and jejunal biopsy led to a change in diagnosis in 92% of patients. These findings are unlikely to be limited to our institution. Our findings support the recommendation that a diagnosis of latent coeliac disease should only be made following repeat histological assessment and enteroscopy with jejunal biopsy.

Disclosure of Interest None Declared.