Department of Biomedical Engineering, University of Virginia School of Engineering & Applied Science, Charlottesville, Virginia, USA.
Department of Pediatrics, Neurology & Public Health, University of Virginia School of Medicine, Charlottesville, Virginia, USA.
Muscle Nerve. 2024 Dec;70(6):1205-1214. doi: 10.1002/mus.28263. Epub 2024 Sep 30.
INTRODUCTION/AIMS: Spinal muscular atrophy (SMA) and Duchenne muscular dystrophy (DMD) are progressive neuromuscular disorders characterized by severe muscle weakness and functional decline (Pillen et al., Muscle Nerve 2008; 37(6):679-693). With new therapeutics, objective methods with increased sensitivity are needed to assess muscle function. Ultrasound imaging is a promising approach for assessing muscle fat and fibrosis in neuromuscular disorders. This study builds on prior work by combining ultrasound-based measurements of muscle size, shape, and quality, relating these measures to muscle strength, and proposing a multivariable image-based estimate of muscle function.
Maximum voluntary elbow flexion torque of 36 participants (SMA, DMD, and healthy controls) was measured by hand-held dynamometry and elbow flexor muscles were imaged using ultrasound. Muscle size (cross-sectional area, maximum Feret diameter or width, and thickness), quality (echogenicity, texture anisotropy index), and cross-sectional shape (diameter ratio) were measured. Multivariable regression was used to select ultrasound measurements that predict elbow flexion torque.
Significant differences were observed in muscle size (decreased), shape (thinned), and quality (decreased) with increased disease severity and compared to healthy participants. CSA (brachioradialis R = 0.51), maximum Feret diameter (biceps R = 0.49, brachioradialis R = 0.58) and echogenicity (brachioradialis R = 0.61) were most correlated with torque production. Multivariable regression models identified that muscle size (CSA, maximum Feret diameter) and quality (echogenicity) were both essential to predict elbow flexion torque (R = 0.65).
A multivariable approach combining muscle size and quality improves strength predictions over single variable approaches. These methods present a promising avenue for the development of sensitive and functionally relevant biomarkers of neuromuscular disease.
简介/目的：脊髓性肌萎缩症（SMA）和杜氏肌营养不良症（DMD）是两种进行性神经肌肉疾病，其特征为严重的肌肉无力和功能下降（Pillen 等人，《肌肉神经》，2008 年；37（6）：679-693）。随着新的治疗方法的出现，需要使用更敏感的客观方法来评估肌肉功能。超声成像在评估神经肌肉疾病中的肌肉脂肪和纤维化方面具有广阔的应用前景。本研究在先前工作的基础上，结合了肌肉大小、形状和质量的基于超声的测量方法，将这些测量方法与肌肉力量联系起来，并提出了一种基于多变量图像的肌肉功能估计方法。
使用手持式测力计测量 36 名参与者（SMA、DMD 和健康对照组）的最大自主肘部弯曲扭矩，并使用超声对肘部屈肌进行成像。测量肌肉大小（横截面积、最大 Feret 直径或宽度和厚度）、质量（回声性、纹理各向异性指数）和横截面积形状（直径比）。使用多元回归选择可预测肘部弯曲扭矩的超声测量方法。
随着疾病严重程度的增加以及与健康参与者的比较，观察到肌肉大小（减小）、形状（变薄）和质量（降低）存在显著差异。CSA（肱桡肌 R=0.51）、最大 Feret 直径（肱二头肌 R=0.49，肱桡肌 R=0.58）和回声性（肱桡肌 R=0.61）与扭矩产生的相关性最强。多元回归模型表明，肌肉大小（CSA、最大 Feret 直径）和质量（回声性）对预测肘部弯曲扭矩都很重要（R=0.65）。
结合肌肉大小和质量的多变量方法可提高单一变量方法的预测强度。这些方法为开发神经肌肉疾病的敏感且与功能相关的生物标志物提供了有前途的途径。
