Assessment of disease severity with magnetic resonance cholangiography in pediatric-onset primary sclerosing cholangitis.

Abstract

Magnetic resonance cholangiopancreaticography (MRCP) has supplanted endoscopic retrograde cholangiopancreaticography (ERCP) as the preferred imaging modality for primary sclerosing cholangitis (PSC). However, data about the accuracy of MRCP in assessing disease severity are limited, particularly in children. We assessed the accuracy of MRCP in disease severity evaluation and investigated the correlation between imaging findings and biochemical parameters (including the multivariate risk index SCOPE) in patients with pediatric-onset PSC. We included 36 patients with PSC (median age: 16) who had MRCP and ERCP performed within 4-month intervals. Two experts, blinded to ERCP findings, evaluated the bile duct changes in consensus using the Modified Amsterdam PSC Score. The agreement between MRCP and ERCP evaluations was tested with weighted kappa statistics and the correlation between disease severity and biochemical parameters with Spearman's rank correlation. The agreement between MRCP and ERCP was good for extrahepatic (weighted kappa 0.69; 95% CI 0.53-0.84) but fair for intrahepatic (weighted kappa 0.35; 95% CI 0.14-0.56) bile ducts. Intrahepatic and extrahepatic MRCP scores correlated with APRI (ρ = 0.42, p = 0.020 and ρ = 0.39, p = 0.033, respectively), while extrahepatic MRCP score also correlated with biliary neutrophils (ρ = 0.36, p = 0.035). We found a good correlation between the SCOPE index and intrahepatic MRCP score (ρ = 0.53, p = 0.004), and extrahepatic MRCP score (ρ = 0.57, p = 0.001). MRCP is accurate at evaluating the severity of extrahepatic bile duct changes in pediatric-onset PSC but tends to underestimate intrahepatic changes. The SCOPE index's robust correlation with imaging scores supports its role as a comprehensive diagnostic tool, outperforming individual laboratory metrics.