Abdom Radiol (NY). 2022 Nov 15. doi: 10.1007/s00261-022-03727-7. Online ahead of print.
ABSTRACT
PURPOSE: To investigate the performance of spleen to non-cancerous liver volume ratio (STnLR) for diagnosing liver cirrhosis in patients with hepatocellular carcinoma (HCC) during preoperative evaluation.
METHODS: Patients were randomly divided into experiment group and validation group. Patients were grouped into cirrhosis group and non-cirrhosis group according to Scheuer staging. Patients’ routine image data were reconstructed using a three-dimensional system. STnLR, spleen to liver volume ratio (STLR), spleen volume, aspartate aminotransferase to platelet ratio index (APRI), and fibrosis index based on the four factors (FIB-4) were calculated. Correlations between indices and cirrhosis were measured by Spearman correlation analysis. Diagnostic performance was assessed and compared using receiver operating characteristic analysis. Accuracies of the models were analyzed in validation group.
RESULTS: No statistical difference in demographic and clinical characteristics was observed between groups. In experiment group, STnLR had the strongest correlation (r = 0.5399, P < 0.0001), and STLR, spleen volume, APRI, and FIB-4 had moderate correlations (r = 0.4583, 0.4123, 0.3648, and 0.3405, P < 0.0001, < 0.0001, < 0.0001, and = 0.0002) with liver cirrhosis stage. AUROC of STnLR (0.8326) was not statistically higher than that for spleen volume (0.7542, P = 0.09832) and STLR (0.8046, P = 0.3034), but was significantly higher than that for APRI (0.7099, P = 0.02046) and FIB-4 (0.7294, P = 0.03987). In validation group, STnLR showed the highest AUROC value (0.8538) and highest Youden index (0.5869) among all models.
CONCLUSION: STnLR is an accurate and stable volumetric model to diagnose hepatic cirrhosis in the HCC population, which is superior to APRI and FIB-4.
PMID:36380210 | DOI:10.1007/s00261-022-03727-7
