HR-pQCT for the Evaluation of Muscle Quality and Intramuscular Fat Infiltration in Ageing Skeletal Muscle

Abstract

Myosteatosis is the infiltration of fat in skeletal muscle during the onset of sarcopenia. The quantification of intramuscular adipose tissue (IMAT) can be a feasible imaging modality for the clinical assessment of myosteatosis, important for the early identification of sarcopenia patients and timely intervention decisions. There is currently no standardized method or consensus for such an application. The aim of this study was to develop a method for the detection and analysis of IMAT in clinical HR-pQCT images of the distal tibia to evaluate skeletal muscle during the ageing process, validated with animal and clinical experimentation. A pre-clinical model of ovariectomized (OVX) rats with known intramuscular fat infiltration was used, where gastrocnemii were scanned by micro-computed tomography (micro-CT) at an 8.4 μm isotropic voxel size, and the images were analyzed using our modified IMAT analysis protocol. IMAT, muscle density (MD), and muscle volume (MV) were compared with SHAM controls validated with Oil-red-O (ORO) staining. Furthermore, the segmentation and IMAT evaluation method was applied to 30 human subjects at ages from 18 to 81 (mean = 47.3 ± 19.2). Muscle-related parameters were analyzed with functional outcomes. In the animal model, the micro-CT adipose tissue-related parameter of IMAT% segmented at −600 HU to 100 HU was shown to strongly associate with the ORO-positively stained area (r = 0.898, p = 0.002). For the human subjects, at an adjusted threshold of −600 to −20 HU, moderate positive correlations were found between MV and MD (r = 0.642, p < 0.001), and between MV and IMAT volume (r = 0.618, p < 0.01). Moderate negative correlations were detected between MD and IMAT% (r = −0.640, p < 0.001). Strong and moderate associations were found between age and MD (r = −0.763, p < 0.01), and age and IMAT (r = 0.559, p < 0.01). There was also a strong correlation between IMAT% and chair rise time (r = 0.671, p < 0.01). The proposed HR-pQCT evaluation protocol for intramuscular adipose-tissue produced MD and IMAT results that were associated with age and physical performance measures, and were of good predictive value for the progression of myosteatosis or sarcopenia. The protocol was also validated on animal skeletal muscle samples that showed a good representation of histological lipid content with positive correlations, further supporting the clinical application for the rapid evaluation of muscle quality and objective quantification of skeletal muscle at the peripheral for sarcopenia assessment.

Keywords: HR-pQCT; aged skeletal muscle; animal model; intramuscular infiltration; sarcopenia.

Figure 1

Representative images showing representation of the IMAT analysis protocol and the resulting 3D images. (A) In the original image, the ROI was selected using the muscle contours determined by the STA protocol, which is (B) depicted in green. (C) The intended segmentation threshold with an upper threshold of 100 HU did not produce a realistic result and was, therefore, rejected. (D) The upper threshold of −20 HU was defined manually. (E) The STA protocol produces a 3D segmentation, with the bones depicted in grey, the muscles in red, and fat in transparent green. (F) Application of the IMAT segmentation produces a 3D image, as shown with the IMAT depicted in green and the muscle depicted in transparent red.

Figure 2

(A) IMAT% was significantly higher in OVX rats than in SHAM rats (* p < 0.05). (B) No significance was detected for MD between the two groups.

Figure 3

Morphological differences and fat infiltration of skeletal muscles in SHAM and OVX rats. (A) The OVX group presented more intramuscular fat tissue than the SHAM group by H&E taken at 20× magnification, where fat tissues are indicated by black arrows. (B) The OVX group showed a higher ORO signal (red area) than the SHAM group. (C) Quantitative analysis revealed that the ORO area of the OVX group was significantly higher (* p < 0.05, Student’s T-test).

Figure 4

MicroCT fat-related parameter of IMAT% was highly correlated with ORO-positively stained area (r = 0.898, r² = 0.806; p < 0.05).