Comparison between CT volumetry, technetium99m galactosyl-serum-albumin scintigraphy, and gadoxetic-acid-enhanced MRI to estimate the liver fibrosis stage in preoperative patients

• Published in: European radiology 2024-04, Vol.34 (4), p.2212-2222
• Main Authors: Nakazawa, Yujiro, Okada, Masahiro, Hyodo, Tomoko, Tago, Kenichiro, Shibutani, Kazu, Mizuno, Mariko, Yoshikawa, Hiroki, Abe, Hayato, Higaki, Tokio, Okamura, Yukiyasu, Takayama, Tadatoshi
• Format: Article
• Language: English
• Subjects: Albumin; Albumins; Body measurements; Collagen (type I); Computed tomography; Confidence intervals; Diagnostic Radiology; Diethylenetriamine pentaacetic acid; Fibrosis; Gadolinium; Gastrointestinal; Hepatectomy; Hepatocellular carcinoma; Hyaluronic acid; Imaging; Independent variables; Internal Medicine; Interventional Radiology; Liver; Liver cirrhosis; Magnetic resonance imaging; Medicine; Medicine & Public Health; Neuroradiology; Parameter estimation; Parameter identification; Radiology; Regression analysis; Scintigraphy; Spleen; Statistical analysis; Surface area; Ultrasound
• ISSN: 1432-1084; 0938-7994; 1432-1084
• DOI: 10.1007/s00330-023-10219-9

Objectives To compare the efficacy of computed tomography volumetry (CTV), technetium 99m galactosyl-serum-albumin ( 99m Tc-GSA) scintigraphy, and gadolinium-ethoxybenzyl-diethylenetriamine-pentaacetic-acid-enhanced MRI (EOB-MRI) in estimating the liver fibrosis (LF) stage in patients undergoing liver resection. Methods This retrospective study included 91 consecutive patients who had undergone preoperative dynamic CT and 99m Tc-GSA scintigraphy. EOB-MRI was performed in 76 patients. CTV was used to measure the total liver volume (TLV), spleen volume (SV), normalised to the body surface area (BSA), and liver-to-spleen volume ratio (TLV/SV). 99m Tc-GSA scintigraphy provided LHL15, HH15, and GSA indices. The liver-to-spleen ratio (LSR) was calculated in the hepatobiliary phase of EOB-MRI. Hyaluronic acid and type 4 collagen levels were measured in 65 patients. Logistic regression and receiver operating characteristic (ROC) analyses were performed to identify useful parameters for estimating the LF stage and laboratory data. Results According to the multivariable logistic regression analysis, SV/BSA (odds ratio [OR], 1.01; 95% confidence interval [CI], 1.003–1.02; p  = 0.011), LSR (OR, 0.06; 95%CI, 0.004–0.70; p  = 0.026), and hyaluronic acid (OR, 1.01; 95%CI, 1.001–1.02; p  = 0.024) were independent variables for severe LF (F3–4). Combined SV/BSA, LSR, and hyaluronic acid correctly estimated severe LF, with an AUC of 0.91, which was significantly larger than the AUCs of the GSA index (AUC = 0.84), SV/BSA (AUC = 0.83), or LSR (AUC = 0.75) alone. Conclusions Combined CTV, EOB-MRI, and hyaluronic acid analyses improved the estimation accuracy of severe LF compared to CTV, EOB-MRI, or 99m Tc-GSA scintigraphy individually. Clinical relevance statement The combined analysis of spleen volume on CT volumetry, liver-to-spleen ratio on gadolinium-ethoxybenzyl-diethylenetriamine-pentaacetic-acid-enhanced MRI, and hyaluronic acid can identify severe liver fibrosis associated with a high risk of liver failure after hepatectomy and recurrence in patients with hepatocellular carcinoma. Key Points • Spleen volume of CT volumetry normalised to the body surface area, liver-to-spleen ratio of EOB-MRI, and hyaluronic acid were independent variables for liver fibrosis. • CT volumetry and EOB-MRI enable the detection of severe liver fibrosis, which may correlate with post-hepatectomy liver failure and complications. • Combined CT volumetry, gadolinium-ethoxybenzyl-diethylenetriam