JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Preoperative blood and CT-image nutritional indicators in short-term outcomes and machine learning survival framework of intrahepatic cholangiocarcinoma

  • 0Department of Ultrasonography, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, 325000, Zhejiang Province, China.
European Journal of Surgical Oncology : the Journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology +

|

February 26, 2025

|

February 26, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

Preoperative nutritional status, including blood markers (ALBI, CONUT, PNI) and imaging indicators (SMI, VSR), impacts intrahepatic cholangiocarcinoma (iCCA) outcomes. Machine learning models effectively predict overall survival in iCCA patients.

Area Of Science

  • Hepatobiliary Surgery
  • Surgical Oncology
  • Nutritional Biomarkers

Background

  • Intrahepatic cholangiocarcinoma (iCCA) presents aggressive behavior with limited therapeutic options and poor prognosis.
  • Preoperative nutritional status is a critical factor for predicting patient outcomes.
  • Current assessment methods require comprehensive evaluation for iCCA patients.

Purpose Of The Study

  • To compare the predictive performance of preoperative blood-based nutritional indicators (albumin-bilirubin [ALBI], controlling nutritional status [CONUT], prognostic nutritional index [PNI]) and CT-imaging nutritional indicators (skeletal muscle index [SMI], visceral adipose tissue [VAT], subcutaneous adipose tissue [SAT], visceral to subcutaneous adipose tissue ratio [VSR]).
  • To evaluate the efficacy of these indicators in predicting short-term outcomes (complications, early recurrence) and long-term outcomes (overall survival) in iCCA patients undergoing curative hepatectomy.
  • To develop and validate machine learning (ML) models for predicting overall survival in iCCA.

Main Methods

  • A cohort of 290 iCCA patients from two centers was analyzed.
  • Preoperative blood and CT-imaging nutritional indicators were assessed.
  • Short-term outcomes (complications, early recurrence) and long-term outcome (overall survival) were evaluated. Six ML models, including Gradient Boosting (GB) survival analysis, were developed to predict overall survival.

Main Results

  • Preoperative blood nutritional indicators (ALBI, CONUT, PNI) were significantly associated with postoperative complications but not early recurrence.
  • CT-imaging nutritional indicators showed insignificant associations with short-term outcomes.
  • ALBI, CONUT, PNI, SMI, and VSR were significantly associated with overall survival. The GB ML model demonstrated the best predictive performance (C-index: 0.755 training, 0.714 validation).

Conclusions

  • Preoperative nutritional status, assessed by ALBI, CONUT, and PNI, correlates with complications but not early recurrence in iCCA.
  • Image-based nutritional indicators were less effective for short-term outcome prediction.
  • Developed ML models effectively predict iCCA prognosis using nutritional and clinicopathological variables.

Keywords: