Basic research for ultrasound-guided injection into skeletal muscle lesions in an experimental animal model

Abstract

Aims: Ultrasound-guided injection techniques are expected to enhance therapeutic efficacy for skeletal muscle injuries and disorders, but basic knowledge is lacking. The purpose of this study was to examine the diagnostic accuracy of ultrasound for abnormal skeletal muscle lesions, and to examine the distribution patterns of solution and cells injected into abnormal muscle lesions under ultrasound guidance.

Methods: A cardiotoxin (CTX)-induced muscle injury model was used. Briefly, CTX was injected into tibialis anterior muscle in rats under ultrasound observation. First, the diagnostic accuracy of abnormal muscle lesions on ultrasound was examined by comparing ultrasound findings and histology. Next, Fast Green solution and green fluorescent protein (GFP)-labelled cells were simultaneously injected into the abnormal muscle lesions under ultrasound guidance, and their distribution was evaluated.

Results: Evaluation of short-axis ultrasound images and cross-sectional histological staining showed a strong correlation (r = 0.927; p < 0.001) between the maximum muscle damage area in ultrasound and haematoxylin and eosin (H&E) staining evaluations. Histological analysis showed that ultrasound-guided injection could successfully deliver Fast Green solution around the myofibres at the site of injury. In contrast, the distribution of injected cells was very localized compared to the area stained with Fast Green.

Conclusion: This experimental animal study demonstrated the potential of ultrasound to quantitatively visualize abnormalities of skeletal muscle. It also showed that ultrasound-guided injections allowed for highly accurate distribution of solution and cells in abnormal muscle tissue, but the patterns of solution and cell distribution were markedly different. Although future studies using a more clinically relevant model are necessary, these results are important findings when considering biological therapies for skeletal muscle injuries and disorders.

Fig. 1

Ultrasound-guided cardiotoxin injection into rat tibialis anterior muscle. In the macroscopic photographs, probe positions during long-axis/short-axis imaging are shown. Blue arrows indicate the direction of needle puncture. In the ultrasound images, positions of the needle in the muscle tissue are indicated with white arrows. Scale bars = 2 mm.

Fig. 2

Ultrasound images and histological sections of tibialis anterior 24 hours after cardiotoxin (CTX) injection. a) Long-axis and b) short-axis ultrasound images. c) In a short-axis ultrasound image, lesion containing more hypoechoic area than normal muscle tissue is outlined with a dotted line. d) to i) Histological analysis performed using the cross-sectional sections. Staining with haematoxylin and eosin (HE), phalloidin, and anti-F4/80 antibody is shown. g) to l) Stained images of the area bordering CTX-induced muscle injury and surrounding tissue. Myofibres, endomysium, and perimysium at the site of muscle injury were stained with anti-collagen 1 antibody, Picrosirius red (PSR) without/with polarized microscopy, anti-laminin antibody, phalloidin, and 4′,6-diamidino-2-phenylindole (DAPI). In the merged image, myofibres were labelled with phalloidin (red), nuclei with DAPI (blue), and endomysium with laminin (green). Scale bars = 1 mm in a) to f), 500 μm in g) to i), and 100 μm in j) to l).

Fig. 3

Quantitative comparison between ultrasound and histological measurements of muscle injury area. a) Typical ultrasound images and cross-sectional haematoxylin and eosin (HE) staining of tibialis anterior 24 hours after cardiotoxin (CTX) injection. b) The comparison of skeletal muscle injury areas measured in ultrasound and HE-stained images. c) Bland-Altman plot of the differences (y-axis) and means (x-axis) of the muscle injury areas measured by HE staining and ultrasound (n = 12). Scale bars = 1 mm.

Fig. 4

The distribution of the injected solution and cells. a) to j) Images of cryosections immediately created after injection. a) Brightfield image. b) and c) Images of staining with Picrosirius red (PSR). d) to g) Images of area indicated by solid line in panel a). d) Brightfield image. e) Image of staining with anti-laminin antibody. f) Image of staining with PSR. g) Image of staining with anti-collagen 1 antibody. h) to j) Images of area indicated by dotted line in panel a). h) Brightfield image. i) GFP-positive cells. j) Image of phase contrast. k) Graph showing a comparison between Fast Green-stained and GFP-positive areas. Scale bars = 1 mm for upper panels, 200 μm for middle panels, and 100 μm for lower panels.

Fig. 5

The relationship between muscle membranes and distribution of injected cells. a) and b) Images of staining with anti-collagen 1 antibody. Solid line in panel a) indicates area shown in panel b). The cells were distributed in a pattern along the perimysium. c) to h) Images of injected cells in injured muscle tissue. c) and d) Images of staining with Picrosirius red. Fragmented red signals were observed at the margins of the cell aggregates. e) to h) The merged images labelled with laminin (red) and nuclei with 4′,6-diamidino-2-phenylindole (DAPI) (blue). No clear contact between the laminin staining and cell aggregates was observed. Scale bars = 100 μm.