Introduction The current gold standard diagnostic test for CD is oesophogastroduodenoscopy (OGD) and duodenal biopsies, aiming to demonstrate the presence of villous atrophy. Given the low sensitivity of endoscopic markers and the patchy distribution of CD within the small bowel, a duodenal bulb biopsy is currently recommended. However, recent studies have raised questions about the need for biopsies in CD, particularly within the paedatric literature, with high anti-tissue transglutaminase antibody (TTG) levels believed to be sufficient. The hypothesis being that high TTG levels correlate closely with definitive CD histology (Marsh 3a-c) giving a high positive predicative value (PPV). This study evaluates TTG levels and histology (inclusive of a bulb biopsy) in adult patients suspected of having CD, with the aim of defining a cut off value for TTG when biopsy may be unnecessary.

Methods Recruitment occurred between Nov 2008 and July 2012, 523 adult patients (>16 years). All patients had a minimum of 5 duodenal biopsies taken whilst on a gluten containing diet. Furthermore, all patients were tested for IgA TTG, Endomysial Antibody (EMA) and total IgA immunoglobulin at the time of their endoscopy. This study retrospectively reviews the correlation between TTG levels (Aesku Diagnostics, Wendelsheim, Germany) and histological outcomes in these patients, with CD being defined as villous atrophy (Marsh 3a-3c) and a clinical and serological response to a gluten free diet.

Results Of the 523 adult patients (median age 51 years, range 16–91) recruited, 212 (41%) had positive TTG serology (>15 U/ml). EMA positivity was identified in 31% of the cohort (163/523) with CD diagnosed in 32% (169/523). The sensitivity, specificity, PPV and NPV for TTG was 93.3%, 83.3%, 71.7%, 96.4% respectively and for EMA 93.8%, 98.9%, 97.4% 97.2%. Table 1 shows the PPV for diagnosing CD at differing TTG levels. No cut off level was associated with a PPV of 100%, with the highest PPV value of 97.1% seen when the TTG level was set at 300U/ml (20 × upper normal limit).

Conclusion Contrary to ESPHGAN guidelines our findings would not support a biopsy avoidance strategy in those with high TTG levels, with patient potentially being wrongly diagnosed. We advocate that duodenal biopsies including a bulb biopsy remain the gold standard diagnostic test in those with suspected CD.

Disclosure of Interest None Declared