Background Small intestinal biopsy remains the reference standard for diagnosing coeliac disease (CD) in children. Recent international guidelines suggest that for select patients, biopsy may not be necessary.

Aims The aim of this study were to: (a) correlate coeliac serology with histology findings in Irish children with a high pre test probability of the disease (b) determine if proposed changes to the CD diagnostic algorithm hold true in an Irish paediatric cohort.

Methods A retrospective chart review of all cases of suspected CD that underwent endoscopy at the National Centre for Paediatric Gastroenterology, Our Lady’s Children’s Hospital from January 2004 to December 2011 was undertaken. Data retrieved included endoscopic and histologic findings, co-morbidities and serology results (where available).

Results 596 patients (57% female) underwent endoscopy for suspected CD. Serology data were available for 544 (91%) patients. CD was confirmed on histology in 304 (51%) children. tTG (tissue transglutaminase) was positive in 405 (79%) patients, of whom 288 (71%) also had a positive biopsy and 7 patients (5%) with negative tTG had a positive biopsy. This result was highly significant with a p value of 0.0001. Of the available 195 positive EMA (endomysial antibody) results, 184 (94%) had a positive tTG as well. 110 of 117 children with tTG values > 20 μg/ml plus a + EmA had CD on histology. 3 of 105 (2.8%) patients with tTG >100 μg/ml had a normal histology. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value of tTG levels ranging from 2–99 were 97% (CI 95–98), 53% (CI 46–59), 71% (CI 63–75) and 95% (CI 90–98) respectively. The corresponding values for tTG levels >100 were 94% (CI 87–97), 98% (CI 93–99), 97% (CI 92–99) and 95% (CI 90–98). The PPV of tTG 2–10 μg/ml was 37%, but at 100 μg/ml, this was 97%.

Conclusions Intestinal biopsy remains the reference standard for CD diagnosis. High titre serology correlates well but not perfectly with histology findings. Diagnostic accuracy rather than screening convenience is essential before prescribing a life-long gluten free diet.