Disease severity prognostication in primary sclerosing cholangitis: a validation of the Anali scores and comparison with the potential functional stricture Sarah Poetter-Lang, Ahmed Ba-Ssalamah, Alina Messner, Nina Bastati, Raphael Ambros, Antonia Kristic, Jakob Kittinger, Svitlana Pochepnia, Sami A. Ba-Ssalamah, Jacqueline. C. Hodge, Emina Halilbasic, Sudhakar K. Venkatesh, Nikolaos Kartalis, Kristina Ringe, Lionel Arrivé and Michael Trauner European Radiology (2024) 34:7632–7644 doi.org/10.1007/s00330-024-10787-4 The study focuses on primary sclerosing cholangitis (PSC), a rare and chronic progressive inflammatory liver disease. PSC is characterized by bile duct strictures and fibrosis, which can lead to severe complications, including liver cirrhosis, liver failure, and an increased risk of cholangiocarcinoma. Therefore, a non-invasive test is needed to monitor PSC activity or predict outcomes over its typically long disease course (median duration: 12–20 years). Ruiz et al. developed the Anali scoring system with and without extracellular Gd-chelate to predict the most important factors determining PSC survival, including liver decompensation, liver-related death and/or the need for orthotopic liver transplant (OLT) [1]. Poetter-Lang et al. [2] introduced a new imaging marker, potential functional stricture (PFS) in PSC patients, which can be diagnosed on gadoxetic acid-enhanced MRI (GA-MRI). It showed promise in diagnosing functional strictures requiring ERCP dilation and/or stenting, and in predicting OLT and liver-related death by diagnosing advanced PSC stages. The purpose of this study was to (a) validate Anali scores with and without gadolinium (Gd) using gadoxetic acid instead of extracellular Gd-chelates (ANALIGdAP, ANALIGdHBP and ANALINoGd) and (b) to compare their prognostic ability with the PFS. In this retrospective, single-center study 123 patients diagnosed with large-duct PSC according to EASL guidelines, who underwent gadoxetic-acid enhanced MRI including T2-weighted MRCP between October 2007 and March 2022 were included. Liver-related events including orthotopic liver transplantation (OLT), death, and decompensation were recorded. MRI exams were evaluated by 5 readers independently for ANALI scores including: intrahepatic bile duct dilatation (IHBD), liver dysmorphia (LD), portal hypertension (PH), and arterial and hepatobiliary phase liver parenchymal enhancement. ANALINoGd included IHBD, LD and PH and ranged from 0 to 5 (low risk £2, high risk >2). ANALIGd included LD and parenchymal enhancement and ranged from 0-6 (low risk £1, high risk >1). On 20-minutes hepatobiliary phase a PFS was present if no contrast excretion was seen in the right, left or common hepatic bile duct [2]. Interreader agreement was analysed with Fleiss-Kappa statistics. The association between outcomes and binary data for all four imaging parameters was analysed with univariate Cox proportional hazards analysis. PFS demonstrated almost perfect inter- and intra-reader agreement, while Anali scores showed moderate to substantial agreement. | All imaging scores differed significantly between patients with and without liver-related events. Patients with high-risk imaging scores showed significantly higher probability of liver-related events, which was especially true for the ANALINoGd score. The mean negative predictive values (NPV) were around 90% for the ANALI scores and around 80% for the PFS. All three ANALI scores and the PFS correlated well with all clinical scores and laboratory tests in assessing PSC severity and can indicate whether patients are at high-risk due to hepatic dysfunction (i.e. liver cirrhosis) or due to functional strictures. PFS and ANALI scores can be considered complementary. PFS is particularly effective for diagnosing functional strictures that might benefit from timely therapeutic interventions, such as ERCP. The Anali scores are better suited for long-term prognostication of PSC severity and outcomes. This means that the whole gadoxetic acid enhanced MRI should be analysed including not only morphologic information of MRCP images, but also functional information of gadoxetic-acid enhanced MRI in patients with PSC. The high NPVs for all scores suggest that low-risk patients identified by MRI are unlikely to experience adverse outcomes. This could reduce unnecessary invasive procedures for these individuals. The limitations of this study include the retrospective study design, the single-center cohort, and a relatively short follow-up period of only 3.9 years given the typically prolonged course of PSC. In addition, the cohort primarily included patients with moderate to advanced disease. This may have led to an overrepresentation of adverse events and could overestimate the predictive power of the evaluated models for milder cases. In conclusion, the combination of Anali scores and PFS provides a non-invasive tool for risk assessment in PSC. While PFS is particularly useful for short-term decisions, such as determining the need for ERCP, the Anali scores seem to be better suited for long-term prognostication. References:
Cäcilia Reiner, Prof. Dr. med., is an abdominal imaging specialist, ESGAR Fellow and Head of the Abdominal Imaging Section at the Radiology Department at Hirslanden Clinic Zurich, Switzerland. Comments may be sent to: caecilia.reiner@gmail.com |