Gamma-glutamyltransferase testing in paediatric inflammatory bowel disease to screen for primary sclerosing cholangitis: a diagnostic study based on routinely collected electronic healthcare data.

Objective: To investigate, among children with inflammatory bowel disease (IBD) and elevated liver enzymes, what threshold of gamma-glutamyltransferase (GGT) best distinguishes those with and without primary sclerosing cholangitis (PSC).

Methods: Delayed-type diagnostic study with a paired design. Children with IBD were regularly screened with GGT (index test). Confirmation of PSC was based on magnetic resonance cholangiopancreatography (MRCP) and/or liver histology (preferred reference standard). Children at low risk of PSC continued regular GGT testing for latent PSC to become visible (alternative reference test). The primary outcome was the negative predictive value (NPV) using three predefined test thresholds, respectively, 1, 2 and 5× the upper limit of normal (ULN). The secondary outcome was the GGT threshold based on receiver operating characteristic analysis.

Results: 132 of 469 children (28.1%) had elevated GGT levels at their first colonoscopy or during follow-up. Eventually, 34 children (7.2%) were diagnosed with PSC. Median GGT (IQR) for children with and without PSC was 227 (127-345) and 77 (59-138) U/L, respectively. Of the predefined GGT thresholds, 2× ULN (ie, 100 U/L) had the best test characteristics, including an NPV of 98% and a negative likelihood ratio of 0.04 (95% CI 0.01 to 0.31). The area under the curve was 0.83 (95% CI: 0.75 to 0.90) and the optimal GGT threshold was 103.5 U/L.

Conclusions: In children with IBD who have GGT elevations less than 2× ULN, the likelihood of PSC is extremely low. In such cases, MRCP and liver biopsy can be omitted. Regular GGT monitoring is advised, as PSC may develop over time.