The choice of diagnostic modality influences the proportion of low muscle strength, low muscle mass, and sarcopenia in colorectal cancer patients

Abstract

Low muscle strength, low muscle mass, and sarcopenia have a negative impact on health outcomes in colorectal cancer (CRC) patients. Different diagnostic modalities are used to identify these conditions but it is unknown how well the modalities agree. The aim of this study was to compare different diagnostic modalities by means of calculating the proportion of low muscle strength, low muscle mass, and sarcopenia in CRC patients, and to investigate the agreement for sarcopenia between the various modalities. Men and women participating in the Norwegian Dietary Guidelines and colorectal cancer Survival (CRC-NORDIET) study were included in the analyses. Cut-off values for low muscle strength, low muscle mass, and sarcopenia were defined according to the second consensus set by the European Working Group on Sarcopenia in Older People (EWGSOP2). The diagnostic modalities used to assess muscle strength were handgrip strength and the sit-to-stand test. For muscle mass, computed tomography, dual-energy X-ray absorptiometry (DXA), multi-frequency bioelectrical impedance analysis (MF-BIA), and single-frequency BIA (SF-BIA) were applied. Cohen's kappa was calculated to determine the agreement for low muscle strength and confirmed sarcopenia between diagnostic modalities. Five hundred and three men and women (54% men, mean age of 66 (range 50-80) years old) were included in the analysis. As much as 99% (n=70) of the population was identified with low muscle mass by MF-BIA, while the other modalities identified 9-49% as having low muscle mass. Handgrip strength identified a lower proportion of low muscle strength as compared with the sit-to-stand test (4% vs. 8%). When applying various combinations of diagnostic modalities for low muscle strength and low muscle mass, the proportion of sarcopenia was found to be between 0.3 and 11.4%. There was relatively poor agreement between the different diagnostic modalities with Cohen's Kappa ranging from 0.0 to 0.55, except for the agreement between SF-BIASergi and MF-BIASergi, which was 1. The proportion of low muscle strength, low muscle mass, and sarcopenia in CRC patients varied considerably depending on the diagnostic modalities used. Further studies are needed to provide modality-specific cut-off values, adjusted to sex, age and body size.