Introduction
Adolescent idiopathic scoliosis (AIS) is a disease characterized by vertebral rotation and lateral curvature of the spine that impairs physical growth during adolescence [1]. It affects 2%–4% of children aged <16 years globally, frequently beginning mildly in early adolescence and worsening as skeletal maturity approaches, with 0.04%–0.3% of adolescents developing severe AIS with curves of >40° [2,3]. The etiology of AIS has been investigated through genetic, epigenetic, endocrine, and developmental factors, but it remains largely unknown [3–5].
Several researchers evaluated the proportion and quantity of types I and II skeletal muscle fibers in AIS in the mid-1970s and revealed that paraspinal muscular imbalance was one of the main hallmarks of AIS [6,7]. Ultrastructural assessment of the multifidus muscle in progressive idiopathic scoliosis in the 1980s indicated sarcolemma and myotendinous junction changes [8]. AIS is predominantly associated with paraspinal nerve or muscle pathology changes according to previous studies [9]. In 2015, Luciano et al. [10] presented cases of patients with AIS who did not demonstrate the typical symptoms of reduced limb muscle strength or restricted respiratory muscle function. However, paraspinal muscle biopsies from two of these patients with AIS indicated a high likelihood of core myopathy. Hypothetically, this is because AIS may be a particular kind of neuromuscular disease [10,11]. Moreover, neuromuscular illnesses, such as Duchenne muscular dystrophy (DMD), which has dystrophin protein deficit in the muscles, impair the paraspinal muscles and induce scoliosis in late stage [12]. Additionally, abnormal expression of plasma proteins, such as recombinant vitamin D binding protein, has altered AIS pathogenesis [13]. Both of these observations indicate that the proteins in the paraspinal muscles affect spinal morphology. Therefore, further etiological studies are warranted to identify whether specific neuromuscular diseases are related to AIS. Paraspinal myopathy may be a key factor in the initial pathogenesis of idiopathic scoliosis [9].
We conducted a prospective case-control study based on the hypothesis that AIS onset and progression may be related to neuromuscular diseases and protein abnormalities using histological methods and immunohistochemical markers to assess the pathological changes in the paraspinal muscles and their correlation with the severity of the spinal deformity.
Materials and Methods
Patients
This prospective case-control study began in November 2018 at Peking University Third Hospital (registration number: ChiCTR2100048484). This study was approved by Ethics Committee of Peking University Third Hospital (no., M2019488). Informed consent forms for surgery were obtained from all the patients or their legal surrogates. The authors affirm that this manuscript did not contain any individual person’s data or images which need to provide informed consent for publication. The case series included 40 patients with AIS, all of whom underwent posterior scoliosis correction surgery. Their average age was 16.51±5.20 years (Table 1). The inclusion criteria were as follows: being an adolescent aged 8–19 years, previously diagnosed with AIS by a physician, having a normal body mass index (18.5–23.9 kg/m2), and having no related complications. All participants had main thoracic/lumbar curves with a Cobb angle of >40° by the standing posteroanterior radiographs, Risser sign of ≥3°, and met the surgical indications. This study excluded neurological, muscular, neuromuscular, and rheumatologic illnesses, malignancies, and surgical correction histories. None of the patients in this series underwent preoperative physiotherapy or wore braces. Additionally, the patients were separated into mild (Nash–Moe 0 and I) and severe (Nash–Moe II and III) Nash–Moe groups to identify if dystrophin protein deficiency increased with vertebral rotation. Controls were 20 patients with CS and 20 with SDD.
Pathological sampling and specimen preparation
The experienced surgeon performed a paraspinal muscle biopsy of the multifidus muscle in both the concave and convex regions. Performing the biopsy slightly away from the tendon tissue is important to prevent nonspecific pathological changes. The biopsy specimen was wrapped in semihumid saline gauze and immediately transported to the laboratory. In the laboratory, the tissue was drained with blotting paper, embedded in tragacanth and OCT compound (Tissue-Tek), and then rapidly frozen in isopentane precooled in liquid nitrogen. This meticulous procedure prevents autolytic changes or irreversible artifacts in the muscle tissue within 30 minutes. Cryostat sections were cut to a 7–10 μm thickness. Conventional hematoxylin and eosin staining, histochemical staining (NADH-TR), and EnVision two-step immunohistochemical staining were conducted under standard techniques that utilized the following primary antibodies: dystrophin-1 (anti-dystrophin rod domain; Leica Biosystems, Deer Park, IL, USA), -2 (anti-dystrophin C-terminal; Leica Biosystems), -3 (anti-dystrophin N-terminal; Leica Biosystems), and myosin (Zsbio, Beijing, China) [14]. We utilized polyclonal antibodies to stain dystrophin and used normal tissue from a cadaver as a control to confirm the antibody specificity. Our results are consistent across multiple measurements, minimizing the potential for sampling errors.
Interpretation of the pathological results
The following items were observed and assessed: (1) muscle atrophy, compensatory hypertrophy, and muscle degeneration degrees; (2) whorled fibers or internal nuclei presence; (3) muscle fiber type proportion and distribution; (4) muscle tissue interstitial condition, and (5) dystrophin expression. No generally recognized standard exists for grading these pathological features. Therefore, we used the predominant quartering method in pathology, which is also a customary approach, to conduct statistical processing. We refer to Sang-Jun’s method [15] to classify dystrophin staining into four levels, namely positive, uneven dyeing or dizzy lineation, light-colored lineation or discontinuous lineation, and negative. Additionally, other features refer to “Robbins basic pathology [16]” and “Muscle biopsy: a practical approach [14],” Two pathologists independently analyzed each subject, with no inconsistent judgments. The pathological results were compared between different patient groups and Nash–Moe classification in AIS. Correlation analysis was conducted between the pathological results and the Cobb angle in patients with AIS.
Statistical analysis
The IBM SPSS ver. 22.0 (IBM Corp., Armonk, NY, USA) was used for clinical and pathological data analyses. Descriptive statistics were expressed as mean with standard deviation and count data were presented in numbers. All p-values of <0.05 were considered statistically significant. The independent t-test and Mann-Whitney U test were conducted to analyze the differences in continuous variables between the groups. Fisher’s exact test and χ2 analyses were utilized to assess the differences among the categorical variables. The Cobb angle and dystrophin protein depletion were investigated based on Spearman correlation analysis with IBM SPSS.
Results
Clinical data of the enrolled patients
The AIS cohort consisted of 13 male and 27 female patients, with average ages of 16.5±5.20 years and Cobb angle of 54.84°±12.67°. Based on the Nash–Moe type classification, 7, 23, and 10 patients were classified as types I, II, and III, respectively. Additionally, we included 20 patients with SDD (including lumbar spinal stenosis and degenerative thoracolumbar kyphosis) and 20 patients with CS (Table 1).
The histological changes of the paraspinal muscles and comparisons between the AIS and CS groups
Patients with AIS demonstrated less diffuse and small-group atrophy and a greater degree of large-group atrophy than those with CS (p=0.008) (Table 2), and the percentage of muscle atrophy was more severe (p=0.019) (Table 2). The degree of degeneration (p=0.067) (Table 2) in patients with AIS and the degree of edema was more severe (p=0.028) (Table 2). Most importantly, dystrophin-1, -2, and -3 were more decreased or absent in this group (p<0.001) (Table 2, Fig. 1).
The histological changes of the paraspinal muscles and comparisons between the AIS and SDD groups
Patients with AIS had significantly less muscle atrophy than those with SDD (p=0.001) (Table 3), but significantly more degeneration (p=0.002) (Table 3) and edema (p<0.001) (Table 3). Additionally, the myofiber atrophic pattern of patients with AIS was mainly type II (p=0.020) (Table 3). Moreover, dystrophin-1, -2, and -3 were more decreased or absent in this group (p<0.001) (Table 3, Fig. 1).
Discussion
AIS is a prevalent three-dimensional spine deformity characterized by a lateral spinal curvature of >10° [2]. The exact causes of scoliosis are unclear, but genetic, structural, and environmental factors may contribute to its development. The curvature of scoliosis frequently worsens during puberty, a time when the paraspinal muscles are rapidly growing [17]. Rehabilitation exercises for the paraspinal muscles, such as Schroth gymnastics, are effective in treating scoliosis [18]. Additionally, scoliosis frequently appears in the late stages of neuromuscular diseases and worsens with sagittal deformities [19]. Research on the association between paraspinal muscle lesions and scoliosis could provide new insights into the disease’s pathogenesis, with muscle biopsy being the most effective method for such studies [20].
Paraspinal muscle atrophy and degeneration
The patients with AIS in this study demonstrated higher paraspinal atrophy and atrophy incidence than those previously reported by Wajchenberg et al. [9]. Further, we indicate that the sampling location may affect atrophy, in addition to the disease course and the Cobb angle. This study sampled the paraspinal muscle closer to the tendon than before investigations, which may explain the increased atrophy rate [9].
Compared to patients with CS, those with AIS had a large group of atrophy, indicating that AIS may be a unique neuromuscular disease since neuromuscular disease muscle atrophy is mostly spread along the diseased nerve. Patients with AIS had a larger percentage of muscular atrophy than those with CS, possibly because their muscle atrophy was mostly in the big bundles. Moreover, muscular atrophy may play a role in AIS development.
Patients with AIS were significantly more severe in terms of the degree of edema, indicating that patients with AIS had severe lesions.
We studied 20 patients with SDD to understand muscle’s role in AIS’s pathogenesis. Patients with SDD had more muscle atrophy due to their longer clinical course, and those with AIS demonstrated more severe paraspinal muscle degeneration, indicating more serious myogenic lesions. This indicates that the muscles of patients with SDD have age-related changes, whereas patients with AIS demonstrated pathological degeneration.
Patients with AIS are generally young; hence, they have more type II muscle fibers, also known as “rapid muscle fibers” responsible for movement [21], than those with SDD. Spinal movement is restricted due to deformities, causing more atrophy of type II muscle fibers in AIS.
Abnormal expression of dystrophin in the paraspinal muscles of AIS
We are particularly interested in the dystrophin. The C-terminal of dystrophin-2 is key for connecting to the extracellular structure by relating two cytoplasmic proteins [22] which significantly affects muscle morphology and function in dystrophin-1, -2, and -3.
Dystrophin is a 427-kD structural protein and a cytoplasmic protein associated with sarcolemma. It binds the actin cytoskeleton to the transmembrane dystrophin–glycoprotein complex [23,24] and maintains membrane integrity and cellular homeostasis [25]. In our study, three antibodies targeting the rod-domain, N-terminal, and C-terminal dystrophins were utilized to stain the paraspinal musculoskeletal membranes in patients with AIS.
Our results indicated a significant loss of dystrophin in the paraspinal muscles of patients with AIS compared with those with SDD, confirming that patients with AIS have lower dystrophin levels. Additionally, we investigated patients with CS to identify if dystrophin insufficiency is a secondary effect of scoliosis. Patients with CS typically develop scoliosis before any paraspinal muscle pathology and do not show changes in dystrophin expression; thus, the difference in dystrophin levels between patients with AIS and those with CS indicates that dystrophin reduction in AIS is not incidental and may be important for the disease’s development. Longitudinal studies are required to identify if paraspinal muscle pathology precedes scoliosis in patients with AIS. This result of decreased dystrophin in the paravertebral muscles of patients with AIS is novel and has not been previously reported.
DMD gene mutation disables dystrophin expression, which causes DMD [26]. DMD, a fatal X-related recessive illness, affects the skeletal, respiratory, and cardiac muscles. Most patients are boys because of their genetic character [27]. Mutations in the gene of DMD and Becker muscular dystrophy (BMD) reduce or produce abnormal dystrophin, which destroys the sarcolemma, causes cell necrosis and apoptosis, muscle atrophy in the limbs, and severe respiratory muscle and diaphragm muscle weakness, and some patients may even have scoliosis caused by paraspinal muscle involvement in the later stages [12].
This trial excluded participants with DMD or BMD symptoms, such as limb or chest muscle weakness. However, the patients with AIS in this study demonstrated a unique pattern of dystrophin protein deficiency, indicating a specific dystrophinopathy type. This study revealed pathogenic changes only in the paraspinal muscles, causing asymmetrical spine traction and scoliosis, unlike BMD or DMD.
Dystrophin is lacking in DMD except in revertant fibers. Truncated dystrophin reduces the functional protein intensity at different epitopes, such as the N-terminal and C-terminal, according to the “read frame” hypothesis in BMD [28]. This study revealed reduced immunostaining of AIS dystrophin at the rod-domain and C-terminal epitopes and almost no immunostaining at the bundle of the atrophic fibers’ N-terminal epitope. This immunological expression pattern resembles that of myodystrophy, but the aberrant expression of antimyodystrophy in AIS muscle biopsy tissues with atrophic muscle fibers was more pronounced compared with that in DMD or BMD. This does not imply that AIS should be classified as a neuromuscular disease, and it indicates a hypothesis that muscular lesions may play a role in AIS occurrence and development, providing a basis for future research.
Further confirm the role of the dystrophin
Patients CS and SDD exhibited the role of dystrophin in AIS etiology. We classified dystrophin into Nash–Moe and Cobb angle subgroups for further study.
Patients with higher Nash–Moe classification demonstrated increased dystrophin-2 depletion in the convex paraspinal muscles. This may be because the muscles on the convex side bear more tension, and they are prone to rotating to the convex side as they atrophy, similar to the research by Labrom [29]. Spinal rotation increased with dystrophin deficiency severity on the convex side, supporting its role in AIS. A significant association was observed between the Cobb angle and dystrophin-2 staining abnormalities on the concave side of the multifidus muscle. These results indicate that dystrophin-2 changes are associated with both pathological muscle changes and spinal scoliosis, indicating a potential role in AIS etiology.
This study had some limitations. First, the SDD and CS groups had confounding factors, including age and spinal muscle degeneration. However, patients with SDD demonstrated no significant dystrophin deficiency; thus, confounding factors, such as degeneration, likely did not affect its expression. Likewise, patients with CS, who have scoliosis but no dystrophin loss, indicate that scoliosis itself does not affect dystrophin expression. Therefore, the inclusion of patients with SDD and CS as controls helps achieve our purpose even with some confounding factors. Secondly, due to space limitations, we did not discuss some features that do not impact dystrophin. Finally, ethical concerns make it difficult to collect paraspinal muscle biopsies before or during the early scoliosis stages.
In summary, we conducted an ethically approved experiment indicating that decreased dystrophin levels may be related to AIS. However, more conclusive evidence requires research on muscles other than paraspinal muscles and further molecular biological studies or animal experiments, particularly focusing on dystrophin-2. In particular, we further knock out dystrophin through animal experiments to observe the lateral bending of the animals. Additionally, long-term patient follow-up is required to investigate if further reductions in dystrophin are associated with the curve progression in AIS.
Conclusions
Dysfunction of the paraspinal muscles plays a crucial role in AIS formation and progression. Dystrophin protein levels may significantly decrease or even disappear in AIS, and this loss is associated with scoliosis severity. More research on the paraspinal muscle function and dystrophin protein in AIS is warranted to explore its etiology.
Key Points
The etiology of adolescent idiopathic scoliosis (AIS) remains unclear, involving genetics, epigenetics, endocrinology, developmental imbalance, and increasing focus on paraspinal muscles.
AIS patients exhibit significantly reduced dystrophin expression in paraspinal muscles, which may contribute to scoliosis, with reductions correlating to higher Cobb angles and rotation degrees.
Reduced dystrophin levels in paraspinal muscles may significantly impact muscle morphology and function, as it plays a key role in linking two cytoplasmic proteins to the extracellular matrix, especially dystrophin C-terminal.
Longitudinal studies and animal models are need-ed to explore the relationship between dystrophin deficiency and AIS progression further.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
