Beyond MELD Score: Association of Machine Learning-derived CT Body Composition with 90-Day Mortality Post
Tarig Elhakim1,2, Arian Mansur3, Jordan Kondo4
1Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA. Tarigelhakim@gmail.com.
Cardiovascular and Interventional Radiology
|October 30, 2024
Summary
Machine learning analysis of CT body composition predicts 90-day mortality after transjugular intrahepatic portosystemic shunt (TIPS). These metrics enhance the predictive accuracy of the Model for End-Stage Liver Disease (MELD) score for patient outcomes.
Area of Science:
- Radiology and Medical Imaging
- Artificial Intelligence in Medicine
- Hepatology and Gastroenterology
Background:
- Transjugular intrahepatic portosystemic shunt (TIPS) is a critical procedure for managing complications of portal hypertension.
- Accurate prediction of short-term mortality after TIPS is essential for patient selection and management.
- Current risk stratification models, like the Model for End-Stage Liver Disease (MELD) score, may not fully capture all relevant prognostic factors.
Purpose of the Study:
- To investigate the association between computed tomography (CT)-derived body composition metrics and 90-day mortality following TIPS.
- To evaluate the utility of these body composition parameters as a supplement to the MELD score for improving mortality risk prediction.
Main Methods:
- Retrospective multi-center cohort study including 122 patients who underwent TIPS between 1995 and 2018.
- Machine learning algorithms extracted L3 vertebral level CT body composition metrics (e.g., skeletal muscle area, visceral fat area).
- Statistical analyses included logistic regression and ROC curve analysis to assess predictive performance against 90-day mortality.
Main Results:
- Patients with 90-day mortality had significantly higher MELD scores and lower skeletal muscle area (SMA), skeletal muscle index (SMI), subcutaneous fat area (SFA), and visceral fat area (VFA).
- SMA, SMI, SFA, and VFA were independently associated with 90-day mortality after adjusting for MELD score.
- Incorporating SMA, SFA, and VFA into prediction models significantly improved the predictive power beyond the MELD score alone (AUC 0.84).
Conclusions:
- CT-derived body composition metrics, particularly muscle and fat indices, are significant predictors of 90-day mortality after TIPS.
- These quantitative imaging biomarkers offer valuable prognostic information and enhance the predictive accuracy of the MELD score for post-TIPS mortality risk.
Related Concept Videos
01:28Imaging Studies for Cardiovascular System V: CT
Cardiac computed tomography (CT) scanning is an advanced cardiac imaging technique that utilizes CT technology, with or without intravenous (IV) contrast, to produce accurate cross-sectional virtual slices of specific areas of the heart, coronary circulation, and major blood vessels such as the aorta, pulmonary veins, and arteries. The computer processes these slices to generate three-dimensional images. Multidetector CT (MDCT) is a rapid form of CT scanning that captures multiple slices...
01:25Imaging Studies for Cardiovascular System VI: Calcium -Scoring CT
Calcium-Scoring CT ScanA calcium-scoring CT scan, also known as coronary artery calcium (CAC) scan, detects calcium deposits in the coronary arteries. This test assesses the risk of coronary artery disease (CAD), which can lead to cardiovascular events such as angina, heart failure, and sudden cardiac arrest.A calcium-scoring CT scan is generally recommended for individuals at intermediate risk of CAD without symptoms. It includes:Men aged 40-75 and women aged 50-75: Especially those with a...