O077 - MAKING EVERY CT COUNT: ASSESSING WHETHER CONTRAST PHASES CHANGE THE PICTURE OF 3D MUSCLE AND FAT ANALYSIS

Linked sessions

O077

MAKING EVERY CT COUNT: ASSESSING WHETHER CONTRAST PHASES CHANGE THE PICTURE OF 3D MUSCLE AND FAT ANALYSIS

M. S. G. Brath1,2,3,*, L. H. Mohammad1, L. L. L. Kruse1, C. H. Aakjær1, R. B. Henrichsen1, E. B. Mark2,4, J. B. Frøkjær2,5, L. R. Østergaard6, H. H. Rasmussen2,7,8,9,10, K. Popuri11, M. F. Beg12, U. M. Weinreich1,2,3

1Dept. of Respiratory diseases, Aalborg University Hospital, 2Dept. of Clinical Medicine, Aalborg University, 3Respiratory Research Aalborg, 4Mech-Sense, Dept. of Gastroenterology and Hepatology, 5Dept. of Radiology, Aalborg University Hospital, 6Dept. of Health Science and Technology, Aalborg University, 7Dept. of Gastroenterology and Hepatology, 8Danish Nutrition Science Center, 9Center for Nutrition and Intestinal Failure, Aalborg University Hospital, Aalborg, 10Dept. of Dietetic and Nutritional Research, Copenhagen University Hospitals, Herlev and Gentofte Hospitals, Herlev, Denmark, 11Dept. of Computer Science, Memorial University of Newfoundland, St. John, 12Engineering Science, Simon Fraser University, Barnaby, Canada

 

Rationale: Accurate body composition assessment from CT imaging is increasingly used for risk stratification and nutritional evaluation. However, it remains unclear whether different IV contrast phases influence automated 3D quantification of muscle and adipose tissue, which may limit comparability across clinical scans.

Methods: A retrospective study of Danish patients underwent unenhanced (U), arterial (A), venous (V), and delayed (D) phase abdominal CT in a consecutive standardized four-phase protocol (2018–2023, Northern Region Denmark), using weight-adjusted IV contrast. Skeletal muscle (SM) and subcutaneous (SAT) and visceral fat (VAT) compartments were segmented from T11 to L4 with DAFS 3.0. Linear mixed-effects models with random intercepts assessed phase-related differences, including sex and phase-by-sex interactions.

Results: In total, 456 observations from 114 patients (76 males; 38 females) were analyzed. SM volume increased significantly in all contrast phases (A/V/D) vs. (U) (∆SM 55; 64 cm3, p < 0.001). SM density also increased across phases, with a greater effect in females (p for interaction < 0.01). VAT and SAT volumes decreased significantly in V and D (∆ –105; –63 cm3, p < 0.001), while HU values increased (∆ 1.5; 6.4, p<0.001). Males had higher SM volume (+983 cm3, p<0.001) and VAT volume (+1650 cm3, p< 0.001), and lower SAT (–815 cm3, p<0.001), regardless of phase. Contrast-related changes in VAT and SAT density were largely consistent across sexes, except for a small but significant sex interaction in VAT density in D phase (p = 0.03).

Conclusion: IV contrast phase significantly affects 3D body composition on CT. Women showed greater sensitivity in density measures, indicating that lower contrast doses may be sufficient. Future studies should aim to develop phase-conversion algorithms that account for sex-specific responses to improve enable comparability across contrast protocols.

Disclosure of Interest: None declared