Aliment Pharmacol Ther. 2026 Mar 25. doi: 10.1111/apt.70629. Online ahead of print.
ABSTRACT
BACKGROUND: Cirrhosis is the irreversible architectural endpoint of chronic liver disease and is the greatest risk factor for hepatocellular carcinoma (HCC). Despite its centrality in HCC management algorithms, there are no standardised morphologic criteria for diagnosing cirrhosis on cross-sectional imaging. Criteria used in clinical studies and clinical practice guidelines have not been reviewed systematically. We aim to assess cross-sectional imaging criteria in clinical practice guidelines and clinical studies.
METHOD: We appraised clinical practice guidelines from major hepatology or radiological associations and conducted a systematic review of MEDLINE and EMBASE to identify original studies using ultrasound, CT, and MRI to diagnose liver cirrhosis from 1 January 2015 to 5 November 2025. Studies were classified as having clear, indirect, or no definition. Bias for diagnostic accuracy studies was assessed using QUADAS-2.
RESULTS: Among 18 clinical practice guidelines, only 3 specified explicit imaging-based criteria for diagnosing cirrhosis in at-risk populations. From 6859 records screened, 73 studies (n = 41,417 patients) met inclusion criteria. Only 4 (5%) studies provided a clear definition, 55 (75%) provided indirect definition, and 14 (19%) provided no imaging criteria. Diagnostic performance of imaging features for cirrhosis varied across modalities. While CT and MRI offered higher specificity for certain features (e.g., regenerative nodules), no single modality or feature was sufficient, highlighting the need for standardised morphological-based criteria applicable for several modalities.
CONCLUSION: Despite the widespread use of cross-sectional imaging in cirrhosis, there is no consensus-based, standardised imaging definition. This highlights the urgent need for a standardised, morphology-based definition of cirrhosis.
PMID:41882492 | DOI:10.1111/apt.70629