Comparison Between Single- and Multi-slice Computed Tomography Body Composition Analysis in Patients With...

Area of Science:

• Radiology
• Oncology
• Body Composition Analysis

Background:

• Single-slice CT body composition is established for cancer prognostication.
• Multi-slice CT offers volumetric measurements, but clinical utility is underexplored.

Purpose of the Study:

• Evaluate agreement between single- and multi-slice CT body composition analyses in oesophagogastric cancer patients.
• Explore associations between these body composition measures and overall survival.

Main Methods:

• Compared single-slice (L3 cross-section) and multi-slice (T12-L4 volume) CT measurements for skeletal muscle (SKM), subcutaneous adipose (SAT), visceral adipose (VAT), and intermuscular adipose (IMAT).
• Utilized Pearson correlation and Bland-Altman analyses for agreement quantification.
• Employed Cox proportional hazard modeling to assess overall survival associations.

Main Results:

• High correlations observed between single- and multi-slice measurements for SKM (R: 0.97), SAT (R: 0.98), VAT (R: 0.97), SKM radiodensity (R: 0.93), and IMAT (R: 0.88).
• Narrow limits of agreement (LoA) for SAT (±0.43), VAT (±0.46), and SKM (±0.48).
• Poorer agreement for adiposity in patients with obesity (BMI > 30).
• Comparable effect sizes for overall survival between single- and multi-slice analyses.

Conclusions:

• Single- and multi-slice CT body composition assessments demonstrate high correlation and comparable survival associations in oesophagogastric cancer.
• Agreement for adiposity measurements is reduced in obese patients, potentially limiting single-slice accuracy.
• Both methods are valuable for prognostication, with considerations for obesity.