Skeletal Muscle Texture Assessment Using Ultrasonography: Comparison with Magnetic Resonance Imaging in Chronic Kidney Disease

Abstract

Skeletal muscle dysfunction is common in chronic kidney disease (CKD). Of interest is the concept of “muscle quality,” of which measures include ultrasound-derived echo intensity (EI). Alternative parameters of muscle texture, for example, gray level of co-occurrence matrix (GCLM), are available and may circumvent limitations in EI. The validity of EI is limited in humans, particularly in chronic diseases. This study aimed to investigate the associations between ultrasound-derived parameters of muscle texture with MRI. Images of the thigh were acquired using a 3 Tesla MRI scanner. Quantification of muscle (contractile), fat (non-contractile), and miscellaneous (connective tissue, fascia) components were estimated. Anatomical rectus femoris cross-sectional area was measured using B-mode 2D ultrasonography. To assess muscle texture, first (i.e., EI)- and second (i.e., GLCM)-order statistical analyses were performed. Fourteen participants with CKD were included (age: 58.0 ± 11.9 years, 50% male, eGFR: 27.0 ± 7.4 ml/min/1.73m2, 55% Stage 4). Higher EI was associated with lower muscle % (quadriceps: β = −.568, p = .034; hamstrings: β = −.644, p = .010). Higher EI was associated with a higher fat % in the hamstrings (β = −.626, p = .017). A higher angular second moment from GLCM analysis was associated with greater muscle % (β = .570, p = .033) and lower fat % (β = −.534, p = .049). A higher inverse difference moment was associated with greater muscle % (β = .610, p = .021 and lower fat % (β = −.599, p = .024). This is the first study to investigate the associations between ultrasound-derived parameters of muscle texture with MRI. Our preliminary findings suggest ultrasound-derived texture analysis provides a novel indicator of reduced skeletal muscle % and thus increased intramuscular fat.