Reliable and robust method for abdominal muscle mass quantification using CT/MRI: An explorative study in healthy subjects

• Published in: PloS one Vol. 14; no. 9; p. e0222042
• Main Authors: Park, Jisuk, Gil, Jea Ryung, Shin, Youngbin, Won, Sang Eun, Huh, Jimi, You, Myung-Won, Park, Hyo Jung, Sung, Yu Sub, Kim, Kyung Won
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 19.09.2019; Public Library of Science (PLoS)
• Subjects: Abdomen; Abdominal muscles; Abdominal Muscles - diagnostic imaging; Adolescent; Adult; Biological markers; Biology and Life Sciences; Biomarkers; Biomedical engineering; Cancer; CAT scans; Coefficient of variation; Computed tomography; Correlation coefficient; Correlation coefficients; Diagnosis; Diagnostic imaging; Engineering; Female; Health aspects; Healthy Volunteers; Humans; Levels; Liver; Magnetic Resonance Imaging; Male; Medical imaging; Medical prognosis; Medical schools; Medicine; Medicine and Health Sciences; Methods; Middle Aged; Mortality; Multimodal Imaging - methods; Muscles; Musculoskeletal system; NMR; Nuclear magnetic resonance; Observer Variation; Organ transplantation; Practice Guidelines as Topic; Psoas muscle; Psoas Muscles - diagnostic imaging; Readers; Reproducibility of Results; Research and Analysis Methods; Retrospective Studies; Sarcopenia; Tissue donation; Tomography, X-Ray Computed; Transplants & implants; Vertebrae; Young Adult
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0222042

Quantification of abdominal muscle mass by cross-sectional imaging has been increasingly used to diagnose sarcopenia; however, the technical method for quantification has not been standardized yet. We aimed to determine an optimal method to measure the abdominal muscle area. Among 50 consecutive subjects who underwent abdominal CT and MRI for possible liver donation, total abdominal muscle area (TAMA) and total psoas muscle area (TPA) at the L3 inferior endplate level were measured by two blinded readers. Inter-scan agreement between CT and MRI and inter-reader agreement between the two readers were evaluated using intraclass correlation coefficient (ICC) and within-subject coefficient of variation (WSCV). To evaluate the effect of measurement level, one reader measured TAMA and TPA at six levels from the L2 to L4 vertebral bodies. TAMA was a more reliable biomarker than TPA in terms of inter-scan agreement (ICC: 0.928 vs. 0.788 for reader 1 and 0.853 vs. 0.821 for reader 2, respectively; WSCV: 8.3% vs. 23.4% for reader 1 and 10.4% vs. 22.3% for reader 2, respectively) and inter-reader agreement (ICC: 0.986 vs. 0.886 for CT and 0.865 vs. 0.669 for MRI, respectively; WSCV: 8.2% vs. 16.0% for CT and 11.6% vs. 29.7% for MRI, respectively). In terms of the measurement level, TAMA did not differ from the L2inf to L4inf levels, whereas TPA increased with a decrease in measurement level. TAMA is a better biomarker than TPA in terms of inter-scan and inter-reader agreement and robustness to the measurement level. CT was a more reliable imaging modality than MRI. Our results support the use of TAMA measured by CT as a standard biomarker for abdominal muscle area measurement.