➢ Most of the muscular and connective-tissue abnormalities in patients with adolescent idiopathic scoliosis have been widely recognized as being secondary to the development of the spine deformity.
➢ Several recent genome-wide association studies (GWAS) have indicated that gene polymorphisms are associated with adolescent idiopathic scoliosis. Additional association studies of these candidate genes in other populations are needed to clarify the role of these genes in the etiology of adolescent idiopathic scoliosis.
➢ The study of the pathogenesis of adolescent idiopathic scoliosis has progressed considerably in terms of both genetic analyses and analyses based on the unique phenomena of the scoliotic spine. Further research based on recent findings may improve our understanding of the pathogenesis of adolescent idiopathic scoliosis.
Understanding the cause of adolescent idiopathic scoliosis is important for elucidating its pathogenesis. Epidemiological investigations have demonstrated that the incidence of idiopathic scoliosis is approximately 2% in patients with a Cobb angle of <11° and is approximately 0.3% to 0.5% among those with a Cobb angle of >20°. The proportion of patients with adolescent idiopathic scoliosis who are need of treatment is only 0.1% to 0.3%. The wide range of curve magnitudes in patients with idiopathic scoliosis suggests the presence of multiple genetic and environmental factors1-3. Various theories have been proposed regarding the cause of idiopathic scoliosis, but none have been widely accepted3. Based on historical and recent findings regarding the cause of idiopathic scoliosis, the present report reviews the muscular theory, neurological theory, connective tissue theory, bone growth mismatch theory, genetic theory, and experimental animal models of this disorder.
Muscular and Neurological Theory
The most representative and important report on the muscular theory of adolescent idiopathic scoliosis was based on the electromyographic findings of Le Febvre et al.4. That study revealed potential asymmetry of the erector spinae muscles in patients with idiopathic scoliosis. Following that report, many reports on the abnormality of the erector spinae muscles in patients with idiopathic scoliosis were published5,6. The differences in the fine structure of the left and right sides of the erector spinae muscles in patients with idiopathic scoliosis also have been reported5,6. However, because these differences between the left and right sides of the erector spinae muscles in patients with idiopathic scoliosis are similar to the behavior observed in the muscle fibers of healthy individuals after endurance training, Zetterberg et al. concluded that the difference is secondary to the asymmetrical pressure on the erector spinae muscles due to the curvature of spine7. Saatok et al. biopsied muscle tissue at the time of operative treatment of idiopathic scoliosis, and, although the histological differences between these muscle tissues supported the findings of the previous reports, the authors also concluded that the difference is secondary to the asymmetrical pressure on the erector spinae muscles due to the curvature of spine8. Recently, most of these muscular abnormalities have been widely recognized as being secondary to the development of the spine deformity7,8.
Several studies on platelet abnormalities in patients with idiopathic scoliosis were performed in the 1980s on the basis of the similarity of the cytoskeleton of skeletal muscle cells and platelet cells9-11. Although recent studies have not confirmed the presence of morphological abnormalities10,11, the abnormal platelet aggregation ability in patients with idiopathic scoliosis with a large Cobb angle has been confirmed. Abnormal blood levels of calmodulin, which promotes muscle contraction and platelet aggregation in a calcium concentration-dependent manner, have been reported11,12. In addition, the blood concentration of melatonin, which is a calmodulin antagonist, is reduced in patients with idiopathic scoliosis12,13.
Numerous studies have suggested that nervous system and peripheral nerve disorders potentially cause scoliosis14. Mochida reported that the retrograde axonal transport of horseradish peroxidase in the spinal muscle, characterized by the tonic muscle, was distinctly reduced in patients15. Reduction of the retrograde axonal transport should be regarded as a primary neurogenic change related to the causation of idiopathic scoliosis. Studies involving the electromyographic test for nystagmus, proprioception testing, vibration perception testing, inspection, and sense-of-balance testing for patients with idiopathic scoliosis have been reported15-17. Somatosensory abnormalities in patients with idiopathic scoliosis recently were reported by Cheng et al.16. Because somatosensory abnormalities are correlated with the magnitude of the Cobb angle, these abnormalities also have been suggested to be secondary changes according to the curvature of the spine15. In patients with scoliosis, it is difficult to draw a distinction between the causes and consequences of the abnormalities in proprioceptive and postural control.
On the basis of radiographic findings, Roth proposed that shortening of the spinal cord can lead to idiopathic scoliosis18. On the basis of magnetic resonance imaging (MRI) findings, Porter recently reported that the mismatch of bone growth and spinal cord growth causes the spinal cord to be too short relative to the growth of the spine and tethering of the spine, leading to scoliosis19. However, Deacon et al., in an analysis of intraspinal abnormalities in patients with idiopathic scoliosis with use of whole-spine MRI, reported that patients with true idiopathic scoliosis have no abnormal findings such as Chiari malformation or tethered spinal cord20. A recent report involving the use of MRI indicated that the frequency of abnormalities of the spinal cord, such as Chiari malformation, is approximately 10% in patients with idiopathic scoliosis21, a value that was substantially higher than had been reported previously. Most neuromuscular theories are based on the observation that patients with neuromuscular disease often develop secondary scoliosis, but the different characteristics of the spinal curvature seen in association with neuromuscular scoliosis as compared with typical adolescent idiopathic scoliosis suggest the existence of different mechanisms in the development of idiopathic scoliosis.
Connective Tissue Theory
Because connective tissue diseases such as Marfan syndrome and Ehlers-Danlos syndrome potentially cause spinal deformity, a large number of histological studies were performed in the 1970s to investigate abnormalities in the intervertebral disc and skin in patients with idiopathic scoliosis22. Bushell et al. reported on the abnormal localization of collagen fibers in the anulus fibrosus on the convex side of the spine in a patient with idiopathic scoliosis23. The content of collagen fibers was low on the concave side of the scoliotic spine23. Despite these reports, this collagen degeneration is considered to be secondary to the physical changes associated with the progression of scoliosis23,24.
Bone Growth Mismatch Theory
Abnormal spinal growth is another etiologic theory because scoliosis develops and progresses during skeletal growth25,26. An imbalance of growth that appears to exist between the anterior and posterior structures of the spine has been hypothesized as a contributing factor in the etiology of adolescent idiopathic scoliosis26-28. Other reports have postulated that the etiology of scoliosis is related to the development of relative thoracic lordosis28,29. Willner and Johnson reported that rapid anterior spinal growth compared with posterior spinal growth results in thoracic hypokyphosis with subsequent buckling of the vertebral column, leading to the rotational deformity of scoliosis29. However, the cause for this mismatch of anterior and posterior spinal column growth has not been reported.
The involvement of genetic factors in the pathogenesis of idiopathic scoliosis has been widely recognized. The hereditary nature of idiopathic scoliosis was reported in studies from the United States and Scotland during the 1960s and 1970s30,31. Following those reports, many familial studies on the incidence of idiopathic scoliosis have been conducted. Wynne-Davies reported that idiopathic scoliosis is found in approximately 11% of the offspring of first-degree relatives and 2.4% of offspring of second-degree relatives31. They also reported that idiopathic scoliosis is caused by a variety of genes and has a dominant mode of inheritance on the basis of the findings of a study of 114 patients with idiopathic scoliosis and their first, second, and third-degree relatives31. After that report, incomplete penetrance, various phenotypes, and a variety of modes of inheritance of idiopathic scoliosis (mainly autosomal dominant and X-linked dominant) were reported32-34. Many studies on idiopathic scoliosis have involved monoamniotic twins. Esteve reported that the incidence of scoliosis is approximately 73% in one twin when scoliosis is present in the other twin35. To clarify the involvement of specific genes in the development of idiopathic scoliosis, many genetic linkage analyses have been conducted on families with a high prevalence of scoliosis. Miller et al. reported candidate idiopathic scoliosis loci on chromosomes 6, 9, 16, and 17 on the basis of the results of a linkage analysis of patients with scoliosis and their families (total, 202 people) in 200536. A general limitation of the linkage analysis at that time was the difficulty of narrowing down a chromosome region to a specific gene associated with a specific disease. The fact that many different loci have been reported as candidate loci of idiopathic scoliosis suggests that scoliosis is not a single-gene disease. Miller proposed that the diversity of genes between patients is one reason for the difficulty of identifying specific candidate genes associated with idiopathic scoliosis37. Even if a single gene causes a disease, all patients with an abnormality in this gene do not necessarily have an identical phenotype. Moreover, there are a number of genetic abnormalities that confer the same phenotype. Thanks to recent advances in genetic and bioinformatic technologies, which enable researchers to determine the gene or genes responsible for a disease, genetic polymorphisms associated with idiopathic scoliosis have been identified in genes encoding estrogen receptors, melatonin receptors, elastic fibers, chromodomain helicase DNA-binding protein 7 (CHD7), interleukin (IL)-6, matrix metalloproteinase 3, growth hormone receptor, and various collagens38-41. Several recent genome-wide association studies (GWAS) also have indicated gene polymorphisms associated with adolescent idiopathic scoliosis. Sharma et al. performed GWAS of 327,000 single-nucleotide polymorphisms in 419 families affected by adolescent idiopathic scoliosis and found the strongest evidence of an association with chromosome 3p26.3 single-nucleotide polymorphisms in the proximity of the CHL1, DSCAM, and CNTNAP242.
However, to date, many of these genes have been excluded from a causative role in scoliosis, whereas others are under further investigation43. Zhou et al. conducted a case-control study with use of data from 648 patients with adolescent idiopathic scoliosis and 573 healthy adolescents from the Chinese Han population and confirmed that IL-17RC gene polymorphisms were associated with adolescent idiopathic scoliosis in this Chinese Han population, whereas CHL1, CNTNAP2 and DSCAM genes were not44.
Takahashi et al. conducted a genome-wide association and replication study of 1376 Japanese females with adolescent idiopathic scoliosis and identified a susceptibility locus at chromosome 10q24.3145. They reported that patients with genetic polymorphisms in the vicinity of LBX1 (the gene encoding ladybird homeobox-1) have a 1.56-fold higher risk of adolescent idiopathic scoliosis. A polymorphism in the vicinity of LBX1 also has been associated with adolescent idiopathic scoliosis in a similar study involving a Chinese Han population46. Jiang et al. reported that the LBX1 gene polymorphism is involved in both the progression of the curve and the development of scoliosis with use of a polymerase chain reaction-based method that targeted 946 Chinese patients with idiopathic scoliosis47. LBX1 is a homeobox protein. During muscle development, LBX1 regulates when the precursor cells of muscle cells of the limb migrate to the desired position, although its transcriptional target genes are not yet known48. It would be important to investigate the function of LBX1 to understand the pathogenesis of idiopathic scoliosis.
In recent years, concurrent with genetic research on the pathology of idiopathic scoliosis, other researchers have investigated whether the progression of the spinal curve could be predicted. Ward et al., in a study of fifty-seven patients with progressive idiopathic scoliosis who had had a failure of brace treatment, reported that the expression levels of multiple genes can predict the progression of scoliosis49. On the basis of their results, tests to predict the progression of scoliosis with use of the saliva of patients have been commercialized in the United States50. Additional association studies of these candidate genes in other populations are needed to clarify the role of these genes in the etiology of adolescent idiopathic scoliosis.
Analysis of the Results of the Unique Physical Characteristics of Scoliotic Patients and Laboratory Animals
Adam, in 1865, reported that the apical vertebrae in patients with scoliosis lose their normal thoracic kyphosis in the sagittal plane and that lordotic changes in the thoracic spine were observed during the autopsies of patients with scoliosis51. Adam also reported significant differences (p < 0.05) in terms of rotation and curvature of the spine in patients with idiopathic scoliosis when the forward bending test (Adam forward bending test) was compared with the standing position, findings that were consistent with the results of the autopsy. However, this important fact has not been duly recognized in recent decades. Early reports on scoliosis research indicated earlier growth of the trunk in patients with idiopathic scoliosis as compared with healthy patients51; however, it was later suggested that the characteristics of the control group in this report were inaccurate52. Thus, no difference in growth was apparent in comparison with that in less kyphotic healthy individuals51. However, in a prospective clinical study of 130 patients with idiopathic scoliosis, Archer and Dickson reported that patients with idiopathic scoliosis who had a large Cobb angle (>15°) were taller than patients who had a small Cobb angle (<10°) but that growth was equal in both groups52. It is not clear whether this difference stems from the decrease in thoracic kyphosis or from genetic differences. The abnormal growth rates and growth patterns in patients with idiopathic scoliosis were questioned in a recent study because the absolute size and growth patterns of the body rely on the genetic background53. Nearly 100 years after the report by Adam, Somerville created an experimental rabbit model in which a change in the spine similar to idiopathic scoliosis developed following the induction of lordosis by wiring the posterior laminae to stop the growth of the posterior elements of the spine and detaching the soft tissue on one side28. Those results supported those reported by Adam51. In 1984, Dickson et al. investigated the importance of the sagittal profile of the spine in the development of scoliosis and found that the progression of scoliosis was stopped when the lordotic spine was surgically restored into kyphosis in the same experimental rabbit model54. Experimental animal models have shown that spinal cord injury causes scoliosis immediately after injury55. Langenskiold and Michelsson reported that the experimental costotransversectomy (excision of a part of a rib along with the transverse process of a vertebra) of rabbits caused neuromuscular-like scoliosis55. De Salis et al. reported that the coagulation of the segmental artery below the rib and transverse processes immediately caused neuromuscular-like scoliosis in an experimental rabbit model56. These experimental scoliosis models currently are not considered to be accurate models of idiopathic scoliosis but are more representative of neuromuscular scoliosis because of the lack of changes that are characteristic of idiopathic scoliosis, such as rotational deformity and lordotic change in the spine.
In contrast, Thillard reported that chicks with experimental pinealectomy exhibited idiopathic scoliosis-like changes57. It is believed that the pathology of scoliosis in the chick is similar to the pathology in humans because the chick is a bipedal animal58. Subsequently, idiopathic scoliosis-like changes induced by experimental pinealectomy in the rat were confirmed in several studies by Machida and colleagues58,59. Regarding the mechanism of the generation of idiopathic scoliosis-like changes in the pinealectomy model, Machida et al. reported the importance of melatonin, which is a hormone secreted from the pineal gland as a precursor to tryptophan59. Machida et al. reported that the deficiency of melatonin had a role in the development of scoliosis in the chicken pinealectomy model59,60. They reported that blood melatonin was lower both in the chicken pinealectomy model and in patients with idiopathic scoliosis as compared with control patients during sleep and that the administration of melatonin prevented the progression of scoliosis both in the chicken pinealectomy model and in patients with adolescent idiopathic scoliosis60-62. They found that melatonin administration stopped the progression of scoliosis in the pinealectomy rat model62. Because thoracic lordosis may occur at the beginning of scoliosis in these experimental models, the authors reported the importance of lordosis of the thoracic spine in the mechanism of scoliosis development62. In addition, even after pinealectomy, scoliosis did not develop in all quadripedal gait animals, and idiopathic scoliosis-like spinal deformity occurred only if bipedal gait was experimentally developed in rats62. This finding strongly suggests the importance of posture in the development of scoliosis. Kono et al. reported a decreased number of osteoblasts in pinealectomized chickens as compared with chickens that underwent a sham operation and suggested that melatonin deficiency may lead to reduced bone density and idiopathic scoliosis63. Recently, Moreau et al. reported a significantly lower expression of MT2 (melatonin receptor) in patients with adolescent idiopathic scoliosis (p = 0.02) and found that decreased MT2 expression was correlated with abnormal arm span as part of abnormal systemic skeletal growth64,65. Melatonin is a hormone that is relatively safe to use, such as for insomnia treatment, and clinical studies on the treatment of idiopathic scoliosis with melatonin will provide more insight into the preventive activity and mechanism of this hormone.
The early detection and treatment of progressive idiopathic scoliosis are important, as is the elucidation of its pathogenesis. The study of the pathogenesis of adolescent idiopathic scoliosis has progressed considerably in terms of both genetic analyses and analyses based on the unique phenomena of the scoliotic spine (Table I)66. Additional research based on recent findings may improve our understanding of the pathogenesis of adolescent idiopathic scoliosis.
Source of Funding: No external funding was utilized for this investigation.
Investigation performed at the Department of Orthopedic Surgery, National Center for Musculoskeletal Disorders, Murayama Medical Center, Tokyo, Japan
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2014 by The Journal of Bone and Joint Surgery, Incorporated