EUVIMEDEUVIMED entdecken
Crossref

Three-dimensional automated muscle and adipose analysis to predict hepatic encephalopathy after transjugular intrahepatic portosystemic shunt

Yujie Zhao, Meiyan Huang, Jiawei Huang, Ruilai Yang, Danyan Lin, Shuoling Zhou, Zhanjiang Luo, Weifu Lv, Zhaochen Liu, Xiaoqiong Chen, Jiawen Dai, Jiayi Huang und 8 weitere

European Radiology · 15.11.2026

DOI: 10.1007/s00330-026-13063-9 ↗

Kurz gefasst

Fachartikel in der Zeitschrift European Radiology, online erschienen am 15.11.2026. Verfasst von Yujie Zhao und 19 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Zhuhai Clinical Medical College of Jinan University (Zhuhai People's Hospital, The Affiliated Hospital of Beijing Institute …. Die Arbeit nennt 31 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Guangdong Basic and Applied Basic Research Foundation, Clinical Research Promotion Project of Zhuhai People's Hospital.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

OBJECTIVES: Overt hepatic encephalopathy (HE) limits the use of transjugular intrahepatic portosystemic shunt (TIPS) in portal hypertension. We aimed to determine whether automated three-dimensional (3D) body composition analysis could improve pre-TIPS prediction of overt HE compared with conventional two-dimensional (2D) manual assessment. MATERIALS AND METHODS: This multicenter study included 593 eligible patients from seven institutions (training cohort: n = 308; external validation 1 cohort: n = 151; and external validation 2 cohort: n = 134). After evaluating L3-based 2D measurements, we developed a fully automated 3D muscle and adipose assessment incorporating morphologic and radiomic features. We combined the 3D assessment with selected clinical variables to predict post-TIPS overt HE risk, and benchmarked the final model against published clinical scores. Model performance was evaluated via ROC analysis, DeLong tests, calibration, decision curve analysis, and subgroup analyses. RESULTS: Traditional 2D measurements showed poor reproducibility, with significant variation in 96.2% (25/26) of features. The 3D Model significantly outperformed the 2D Model. The Integrated Model yielded AUCs of 0.78, 0.73, and 0.77 in the training and two external validation cohorts, exceeding those of Child-Pugh, MELD, ALBI, AMMON-OHE, and Lambda (AUCs 0.47-0.64; all p < 0.05). No significant subgroup differences were observed. High-risk patients had 2- to 4-fold higher overt HE incidence. The model reclassified 40.2% of patients with low pre-TIPS ammonia as high risk and 60.2% of those with high ammonia as low risk. CONCLUSION: Automated 3D muscle and adipose analysis enables objective preprocedural prediction of overt HE after TIPS, which may facilitate clinical decision-making and optimize patient selection. KEY POINTS: Question Can automated three-dimensional muscle and adipose analysis improve reproducible preprocedural prediction of overt hepatic encephalopathy after transjugular intrahepatic portosystemic shunt beyond conventional two-dimensional assessment? Findings In 593 patients from seven hospitals, an Integrated Model combining automated three-dimensional body composition and clinical variables outperformed five published clinical comparator scores. Clinical relevance This fully automated three-dimensional body composition (muscle and adipose) analysis enables objective preprocedural risk stratification for overt hepatic encephalopathy after transjugular intrahepatic portosystemic shunt, supporting more consistent patient selection.

Schlagwörter

Stichwörter der Autor:innen

Adipose tissueHepatic encephalopathyMuscle, skeletalPortasystemic shunt, transjugular intrahepaticTomography, X-ray computed

Autor:innen und Institutionen

  • Department of Radiology, Zhuhai Clinical Medical College of Jinan University (Zhuhai People's Hospital, The Affiliated Hospital of Beijing Institute of Technology), Zhuhai, China.
  • Department of Radiology, The First Affiliated Hospital of Jinan University, Guangzhou, China.
  • School of Biomedical Engineering, Southern Medical University, Guangzhou, China.
  • Guangdong Provincial Key Laboratory of Medical Image Processing, Southern Medical University, Guangzhou, China.
  • Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou, China.
  • Department of Biostatistics, School of Public Health, Southern Medical University, Guangzhou, China.

Förderung

  • National Natural Science Foundation of China
  • Guangdong Basic and Applied Basic Research Foundation
  • Clinical Research Promotion Project of Zhuhai People's Hospital

Zitieren

APA

Zhao, Y., Huang, M., Huang, J., Yang, R., Lin, D., Zhou, S., Luo, Z., Lv, W., Liu, Z., Chen, X., Dai, J., Huang, J., Xu, W., Luo, J., Li, Q., Chen, J., Zhang, L., Zhang, S., Duan, C., & Fu, S. (2026). Three-dimensional automated muscle and adipose analysis to predict hepatic encephalopathy after transjugular intrahepatic portosystemic shunt. European Radiology. https://doi.org/10.1007/s00330-026-13063-9

Vancouver

Zhao Y, Huang M, Huang J, Yang R, Lin D, Zhou S, et al. Three-dimensional automated muscle and adipose analysis to predict hepatic encephalopathy after transjugular intrahepatic portosystemic shunt. European Radiology. 2026. doi:10.1007/s00330-026-13063-9

BibTeX
@article{Zhao2026,
  author = {Zhao, Yujie and Huang, Meiyan and Huang, Jiawei and Yang, Ruilai and Lin, Danyan and Zhou, Shuoling and Luo, Zhanjiang and Lv, Weifu and Liu, Zhaochen and Chen, Xiaoqiong and Dai, Jiawen and Huang, Jiayi and Xu, Weiguo and Luo, Junyang and Li, Qiyang and Chen, Jinqiang and Zhang, Lu and Zhang, Shuixing and Duan, Chongyang and Fu, Sirui},
  title = {Three-dimensional automated muscle and adipose analysis to predict hepatic encephalopathy after transjugular intrahepatic portosystemic shunt},
  journal = {European Radiology},
  year = {2026},
  doi = {10.1007/s00330-026-13063-9}
}
Kostenlos starten →