Adolescent idiopathic scoliosis (AIS) is an important group of disorders presenting with rotational asymmetry of the spine, often diagnosed at the onset of puberty. As its name “idiopathic” suggests, the scoliosis has a variety of distinctive factors: connective tissue (Marfan syndrome, etc.); neuromuscular (cerebral palsy, spinal muscular atrophy, etc.); and structural (hemi-spine, etc.). A great deal of work has been done to better understand the clinical manifestations of adolescent idiopathic scoliosis and to find more appropriate treatments. The comparison of the efficacy of bracing, surgical correction and fusion has been the subject of clinical research. The majority of scholars have concluded that bracing is effective in slowing the progression of small angles (30-40°) in skeletally immature patients, while posterior spinal fusion is usually more effective in rapidly developing patients with large angles (40-50°) and in older adolescents with severe angles. The above conclusions are based on an understanding of the natural evolution of AIS, with large angles and degree of skeletal immaturity being important risk factors for disease progression. Another group of scholars has attempted to avoid the obvious risks of bracing and surgical treatment by studying AIS and further implementing treatments that target the intrinsic abnormalities of AIS. These studies include the search for the underlying etiology, the understanding of pathogenesis, and the analysis of associations with fatal events. It is difficult to draw definitive conclusions from these studies because the findings may be either etiologic or pathogenic, or they may be secondary to the disease itself. The extensive research on the etiology of AIS encompasses a wide range of aspects: from different animal models to studies in the fields of histology, immunofluorescence, genetics, and molecular biology of blood and biopsy specimens from patients with AIS, to imaging, hormonal, and neurological studies of patients with AIS, and so on. In this article, we will present the current status of the various studies: structural (intervertebral disc collagen, paravertebral muscle fibers, bone), neurological (including central and peripheral nervous system), and growth and developmental. Our aim is to present the multiple arguments for and against the various etiologic hypotheses and to describe current advances in the etiology of AIS. Epidemiology Although epidemiologic investigation in large populations has always been a relatively tractable subject, the important differences in reporting suggest that it is not. Epidemiologic statistics of AIS in large populations have been accomplished by many authors: Shand and Eisberg’s small-film examination of 5,000 tuberculosis investigations in the state of Delaware resulted in a statistical prevalence of 0.5% in patients with scoliosis greater than 10° and age greater than 14 years. Others have reported prevalence rates of 0.133% (Minnesota) and 13.6% (Las Vegas). To understand these differences, we should note the context of these investigations. Moe and Kennedy surveyed orthopedic records of scoliosis patients seen at several major medical centers in Minnesota, using the entire Minnesota school-age population as the subject of the survey, and concluded that the prevalence rate was 0.133%. Their investigation was limited in part by the fact that the selection of subjects included only patients seen by orthopedic surgeons and did not include patients with AIS who had been missed. kane and Moe acknowledged these shortcomings and labeled the 0.133% prevalence rate the lower limit of the prevalence rate in their paper. Kane and his colleagues applied a loose diagnostic criterion: a Cobb angle greater than or equal to 5°. They counted a total of 3,492 patients and came up with a result of 13.6%. Undoubtedly, epidemiologic results depend not only on sample size but also on the criteria adopted for the study. Current studies require the criterion of a Cobb angle of at least 10°. Genetics Large-scale population surveys have found that affected families have a higher prevalence compared to the general population. This difference has led scholars to accept the importance of genetic factors in the development of the disease.Riseborough and Wynne investigated 207 patients and 2662 family relatives and found a prevalence of 11% in first-degree relatives; 2.4% in second-degree relatives; and 1.4% in third-degree relatives. Some scholars have conducted studies on the prevalence of AIS in twins of monozygotic and dizygotic twins. Concordance of prevalence in monozygotic twins (i.e., both twins have AIS) and nonconcordance in dizygotic twins (only one has AIS) can increase the reliability of the genetic etiology of AIS.DeGeorge and Fisher reported concordance in monozygotic twins to be 6/6, and nonconcordance in dizygotic twins to be 2/8.However, in a critique of this work, Cowell et al. used a 10° cobb angle as a diagnostic criterion and found a nonuniformity of 5/8 in dizygotic twins; Carr reported a concordance of 3/3 in monozygotic twins and 3/3 in dizygotic twins; and Kesling and Reinker reviewed all the data on twins with scoliosis and reported a concordance of 73% in monozygotic twins and 36% in dizygotic twins. Because all of these rates are higher than the prevalence of first-degree relatives in large-scale population-based statistics, these studies support a genetic etiology for AIS. Despite the consensus of information on the familial distribution of AIS, the mode of inheritance of the disease has been a subject of discussion among scholars.Cowell et al. chose 17 families (192 individuals) with the disease for their study, and no cases of male-to-male inheritance were detected, which led to the inference that AIS is a concomitant X-chromosome genetic disease.Miller et al. studied 14 affected families (136 individuals), however, their results did not support the X chromosome correlation of AIS. Other small sample researchers have found cases of male-to-male inheritance, and they argue against the X-chromosome correlation argument in favor of an autosomal dominant mode of inheritance. In a large sample study, Wynne Davis also supported a dominant mode of inheritance, but most of the samples in his study group lacked X-ray imaging evidence. Subsequently, Riseborough and Wynne Davis applied more stringent diagnostic criteria to 2,869 patients with AIS, and the results of this large-scale study were more supportive of a multifactorial mode of inheritance. If AIS is genetically transmissible, then its cause can be found by an either/or approach: genetic analysis of the affected genealogy.Carr et al. found four genealogies with an autosomal mode of inheritance. Carr et al. identified 4 families with autosomal inheritance and used genetic markers to rule out collagen type I and II structural genes as factors in the pathogenesis of AIS. Miller et al. used a similar approach in 11 families (52 individuals) and ruled out collagen type I, FBN1 (microfibrillar protein 15), and elastin as factors in the pathogenesis of the disease. If this approach can be applied to other areas (neurological abnormalities, vestibular abnormalities, hormonal regulation, growth and development), it may contribute to the discovery of the pathogenesis of AIS. Growth and Development and Hormones There are clinical observations that the major progression of scoliosis often occurs during the rapid growth period. For these reasons, growth and development factors have been the focus of numerous AIS etiologic studies. In a representative Scandinavian study, Nordwall and Willner found that in the age-appropriate group, patients with AIS were taller than the comparison group, and that this difference existed regardless of whether or not the reduction in height due to scoliosis had been corrected in patients with AIS. In Norway, Skogland and Miller conducted a prospective study of 62 adolescent girls whose height was 2.5 standard deviations above the mean and found that the prevalence of AIS in the 62 girls was 21%, which was significantly higher than the reported mean distribution. This finding provides further evidence for the association between height and scoliosis. In Yugoslavia, a representative study by Buric and Momcilovic showed that the association between height and AIS is not limited to Scandinavia. Their study confirmed the observation that in the age-appropriate population, the height of AIS patients was significantly higher than that of the comparison group. They also measured the ratio of standing height to sitting height in an attempt to detect overgrowth of the trunk relative to the limbs. However, they did not find a significant difference in leg/torso ratios between the AIS patients and the comparison group, although such a difference had been previously demonstrated by Willer. In Sweden, Willer conducted a longitudinal study of the growth and development of patients who progressed to AIS and a comparison group, and found that patients with AIS were significantly taller than the comparison group, even when they were undiagnosed. Furthermore, the growth rate of AIS patients increased at the age of 8 and 9 years; however, the difference in growth rate between AIS patients older than 10 years of age and the comparison group was not significant.Haggllund et al. applied the Infant-Child-Puberty Development Model in an attempt to identify the period of peak growth in children with AIS. They noted that during puberty, AIS patients were above average height; however, early menarche and late pubertal development resulted in minimal height growth. Nordwall and Willner applied the “Greulich” and “Pyle” methods to the growth of AIS patients. Nordwall and Willner used the “Greulich” and “Pyle” methods to study the bone age of a representative group of AIS patients and a comparison group. They found that the bones of the 11-12 year old female patients with AIS were more mature than those of the age-appropriate comparison group, while the bone age of the 15-17 year old female patients with AIS was less mature than that of the comparison group. They also proposed the hypothesis that the accelerated period of childhood development and skeletal maturation depends on an internal environment of elevated growth hormone secretion. A large number of studies on growth hormone have been conducted, with some results supporting the above hypothesis and some opposing it.Yamada et al. applied L-arginine stimulation to detect changes in GH levels in a total sample of 10 patients, with no specific age restriction, and they did not find any significant elevation of GH levels in these AIS patients.Misol et al. applied arginine, insulin, and glucose tolerance assays in 15 AIS patients. assays and other methods in 15 patients with AIS (mean age 13.8 years) and again did not find significant elevations in GH levels. In Misol’s experiment, they chose several discrete time points to measure GH levels and compare them to a control group at the same time. A limitation of this approach, however, is that there is a natural cycle of GH secretion, as reported in the literature, and this cycle affects the values of GH levels at particular points in time. Another important research topic is that, as clinical observations have shown, in patients with AIS, the younger the age, the faster the growth rate, so it is necessary for one to evaluate GH levels in both younger and older patients with AIS.Skogland and Miller applied naphthacin and levodopa stimulation experiments to detect GH levels. They calculated the area of the time-response curve to reflect the total amount of GH secretion (ICGH). They found that within the same group, the GH levels of AIS patients were not significantly higher than those of the comparison group; however, when they compared data from different age groups, they found that there was a significant increase in total GH secretion in patients aged 7-12 years, whereas there was no such trend in the group of patients older than 12 years.Ahl et al. adopted an experimental approach in which the GH secretion level was measured every 20 minutes over a 24-hour period, and a similar curve was obtained. secretion levels, also yielded a similar area under the curve, thus confirming Akogland’s study. There seems to be no longer any doubt about the association between AIS and abnormal growth and development. The findings suggest that higher developmental rates in younger patients may be associated with the modulation of elevated GH levels. If this hypothesis is correct, is there a more fundamental abnormality that contributes to the elevated GH secretion or to the impaired feedback of normal GH secretion regulation? There is also the obvious fact that most individuals with rapid growth do not develop AIS; therefore, there should be a stimulus in AIS that causes this asymmetric growth. the hypothesis proposed by Goldberg et al. states that there should be genes in the genome that are responsible for the control and temporal ordering of growth and development to ensure a normal symmetric pattern. Genes are involved in the growth and development of the individual, but growth and development itself may only be a predisposing factor for AIS rather than the cause of AIS. It is the intrinsic abnormality that causes asymmetric growth that is the underlying cause of AIS malformations. Connective Tissue Abnormalities It has long been hypothesized that defects in spinal structures (e.g., ligaments, discs, or paravertebral muscles) may contribute to the etiology of AIS. The fact that patients with a variety of primary connective tissue anomalies often have secondary scoliosis adds credence to this hypothesis. However, an important issue is that the connective tissue abnormalities exhibited by these patients may also be secondary to AIS. This possibility adds to the complexity of the issue. Several studies have attempted to find a link between AIS and connective tissue abnormalities, but have not been successful. 90% of disc collagen consists of type I and II collagen, which has been the subject of investigation. the study by Beard et al. did not find that the distribution of type I and II collagen in scoliotic discs differed significantly from that of normal discs, and his study lacked a quantitative analysis of the collagen. As already mentioned, Carr et al. did not identify a link between collagen types I and II markers, and they found that the genetic markers for AIS were independently dominantly inherited. et al. found a significant decrease in glucosaminoglycans (GAGs) and a corresponding increase in collagen in the nucleus pulposus of the intervertebral discs of patients with AIS, and they also noted that the GAGs consisted of short-chain compounds. They concluded that these changes may be one of the etiologic factors of AIS, but they still could not rule out the possibility that it was a secondary manifestation of AIS. They also suggested that there was an abnormal degradation of GAGs in patients with AIS, but they were unable to prove it at the time, and Zaleske et al. confirmed that there was indeed a decrease in the amount of GAGs in patients with AIS through similar experiments. Furthermore, they detected an increase in acid phosphatase (one of the markers of lysosomal activity) in the discs of the patients, leading them to hypothesize that the increase in acid phosphatase and the decrease in GAGs confirmed Pedrini’s conjecture that the decrease in GAGs stemmed from an increase in its degradation. et al. also compared the intervertebral discs of AIS patients and patients. The results showed that the connective tissue changes in the discs of patients with AIS were more severe than in discs with spinal herniation, and that these changes did not appear to be the first cause of AIS. The discs at the apex of the lateral curvature had the lowest levels of GAGs compared to the discs at the base of the lateral curvature. This fact raises the question of whether the alterations in GAGs are likely to be a secondary manifestation of increased degradation induced by lateral bending pressures, rather than the primary etiologic factor of AIS. Ghosh et al. confirmed Pedrini’s findings of reduced collagen in the nucleus pulposus of discs with AIS and noted that the degree of collagen reduction was related to the location and angle of the disc in lateral bending.Ghosh et al. found that the ratio of keratanin to chondroitin phosphate in the annulus fibrosus of convex discs versus concave discs in lateral bending was highest at the apex of the bend.The studies of Taylor and Ghosh further support the argument that connective tissue changes are secondary to the stresses of lateral bending. provide evidence for the argument that they are secondary alterations. Other scholars have also focused on the possible place of the elastic fiber system in the etiology of AIS. Morphologic studies of skin biopsies have shown elastic fiber abnormalities in the middle and deep dermis in 28 of 34 patients with AIS. In Hadley’s study, specimens of the ligamentum flavum from 23 patients with AIS were retained intraoperatively, and abnormalities in the arrangement of elastic fibers were found in 18 of these specimens. In four of these specimens, exuberant growth of fibroprogenitor cells in in vitro experiments suggested impaired binding of secreted microfibrillar proteins to the extracellular matrix. Most studies have failed to provide conclusive evidence for the hypothesis of connective tissue abnormalities as the cause of AIS. Localized abnormalities at the apex of scoliosis make it unlikely that connective tissue abnormalities are the cause of AIS, but more likely that they are merely a secondary alteration. Muscle Abnormalities Many scholars have noted paraspinal muscle tissue abnormalities as a possible etiology of AIS. Secondary scoliosis caused by myopathies such as Duchenne muscular dystrophy has led to the study of skeletal muscle aspects in patients with AIS. But how to explain the above phenomenon, whether it is a primary etiology or just a secondary alteration in the disease process, has been the main question that has troubled people. The multifidus muscle is a paraspinal muscle group that plays a major role in spinal mobility and has therefore been extensively studied histochemically and structurally.Khosla et al. studied biopsy specimens of the multifidus muscle in patients with AIS using light and electron microscopy and found that the defects of the myofilm at the tendon junctions were more prominent in the concave surface of the lateral curvature.Khosla et al. proposed a hypothesis that an increase of the membrane permeability may lead to an excessive uptake of calcium ions, thus leading to an overload of the skeletal muscle and a decrease of the skeletal muscle. and thus lead to localized hypercontraction and hyperextension of the skeletal muscle, further leading to altered cell morphology. Many authors have applied histological and histochemical techniques to confirm that type I slow contracting muscle fibers predominate in the convex side of the lateral bend. These fibers contract with greater force, thus creating greater tension on the convex side. Ford et al. counted neuromuscular spindles in biopsy specimens and found that in the paraspinal muscles of patients with AIS, there was a smaller than normal distribution of neuromuscular spindles, which may indicate a defect in the muscle afferent system. fidler and Jowett found that the multifidus muscle was shorter on the convex side of scoliosis than on the concave side, and they attributed this to contraction of the multifidus muscle, which was the cause of the deformity. The question now is: are these changes the direct cause of the deformity or are they secondary to the deformity?Slager et al. performed different neurological and myological studies on muscle fibers from 19 patients with non-specific scoliosis. They found that these patients, like the AIS patients, had a predominance of type I and type 2A in their muscle fibers. From this they deduced that the changes in the muscle may be related to persistent stress and contraction.The same histochemical studies were carried out by Zetterberg et al. They measured capillary counts and enzyme activities. They found that capillary density, prolyl phosphate dehydrogenase, and lactate dehydrogenase were higher on the convex side than on the concave side; and they similarly found and further confirmed a dominant distribution of type I and type 2A fibers. The consistency of these findings suggests that muscle alterations should be the result of sustained stress and secondary to AIS. Muscles play an important role in maintaining spinal morphology at rest and in motion. The hypothesis that paraspinal muscle imbalance leads to AIS deformity seems plausible, especially supported by the results of early histologic studies. However, the similarity of the alterations occurring in AIS and non-idiopathic scoliosis, as well as the metabolic alterations, are more suggestive of the conclusion that the muscle alterations should be the result of persistent stress, and therefore should be secondary to the deformity, rather than the primary etiology. NEUROLOGICAL STUDIES The central control of spinal posture depends on the acquisition of three types of information: proprioceptive information from the foot; visual afferents; and vestibular signals. Impairment of the above factors in the neural system of postural control became one of the hypotheses of the etiology of AIS.Barrios et al. developed a model of unilateral spinal injury in rabbits, where the occurrence of scoliosis in the study group and the control in the comparison group were the focus of the study. Scholars believe that loss of proprioception causes asymmetric muscle atrophy, which ultimately leads to deformity. The sensitivity of vibratory sensation was also measured and it was found that patients with AIS were more sensitive to said sensation compared to controls.McInnes et al. conducted a similar study, however the results were the opposite of the above. Not only did they find that scoliosis patients were more insensitive to vibratory stimuli, but they also pointed out that the application of the Biological Vibratory Threshold Measure (as used by Wyatt) to identify AIS was unreliable.Keessen et al. used the Spatial Orientation Scale to examine the proprioceptive accuracy of the upper extremity in patients with AIS, and found that the spatial accuracy of patients with AIS was significantly poorer. In addition, balance determinants, mechanical platforms, and tape recordings have been applied to study patients’ proprioceptive abilities. People measured the swing amplitude of patients in stable and unstable states with eyes open and closed, respectively. The results revealed that patients with AIS exhibited greater oscillations under gravity compared to controls. The vestibular system has long been a focus of research interest because of its important role in maintaining body posture. One important vestibular and neurological disorder that has been investigated is nystagmus. Yamada et al. applied the Frensel ophthalmoscope to detect nystagmus, but did not find significant differences in patients with AIS. However, it is believed that the Frensel spectacles are not a sensitive method. Sahlstrand and Lidstrom and Petruson used a nystagmus current tracer to measure the heat and position of nystagmus. Their results suggested the presence of vestibular imbalance (high frequency of nystagmus) as well as a possible brainstem disorder (dysfunctional nystagmus rhythm) in patients with AIS. However, they considered the above conclusions to be inconclusive because the vestibular system defect, a central nervous system abnormality, could be a secondary alteration of the scoliosis deformity. The alterations in the brain itself are considered to be one of the etiologic factors of AIS through the study of young bipedal rats. In their experiments, disruption of stereotactic function in the posterior hypothalamus caused scoliosis in 16 of 103 rats. The researchers concluded that the central damage caused dysfunction of the brainstem reticular formation, which interfered with the central system controlling posture.Sahlstrand and Lidstrom applied electroencephalography and nerve conduction studies to further confirm that central dysfunction was indeed present in patients with AIS compared to controls. Their results showed that AIS patients did have a higher chance of abnormal EEG performance, especially during paroxysmal seizures; however, peroneal nerve conduction experiments in the patients did not show significant differences.Geissele et al. performed MRI scans on many AIS patients and controls, and the results suggested that there was significant involvement of brain tissue. Of these, asymmetry of the ventral aspect of the corticospinal tract was the most common abnormality. Rosenbach’s experiment was applied to study the visual afferents. They found that in patients with left eye dominance was associated with right sided convexity; whereas in patients with right eye dominance was associated with left sided convexity. The researchers hypothesized that patients with left eye dominance might tilt to the right to refine sensory perception, and that this might act as a trigger for axial asymmetry. Evidence suggests that patients with AIS have proprioceptive and vestibular abnormalities that may be caused by first-ever vestibular deficits and central deficits.Sahlstrand et al. investigated first-ever vestibular dysfunction in patients with AIS. Of 24 patients with abnormal labyrinthine function, only 3 had a history that might suggest first vestibular impairment. Thus, so-called first vestibular abnormalities may not exist. On the other hand, different forms of central disorders may also contribute to the etiology. The brainstem, including the reticular system and the vestibular nuclei, which are responsible for controlling symmetrical axial movements, could also explain the various vestibular abnormalities described above. MRI studies by Geissele et al. seem to support this hypothesis, but further confirmation requires studies to accurately characterize and confirm the abnormalities in neural pathways. Melatonin Melatonin, because of its central metabolites and indirect effects on growth, has always attracted the interest of scholars.Machid et al. constructed a model of AIS in chicks and found the following: chicks in which the pineal gland was removed invariably developed scoliosis. When the pineal gland was implanted sub-muscularly in these chicks, the chance of scoliosis decreased to 10%. In the next experiments, 5-HT and melatonin were applied to the pineal gland-removed chicks, and melatonin was found to greatly reduce the incidence of scoliosis. More animal experiments showed that melatonin has an inhibitory effect on growth and gonadal maturation. Some scholars have concluded that melatonin may inhibit 5-HT-related stimulation of growth hormone. Melatonin is also thought to control growth and development by regulating the release of growth hormone, which is reflected by the circadian rhythm of the hormone: GH levels are high during the day and low during the night, whereas melatonin levels are low during the day and high during the night. In human subjects, melatonin values in blood and urine samples measured every 12 hours in patients with AIS did not show significant differences from comparison controls. However, in studies with blood samples measured every 3 hours, one finds significantly reduced melatonin levels compared to patients with non-progressive scoliosis and comparison groups. People have also applied pinealectomy to mammals (rats, etc.) in an attempt to better correlate data from chicks with data from humans. Pinealectomy failed to induce scoliosis in quadrupedal rats; it did so after removal of the forelegs and tail. The use of melatonin, among other things, inhibited the development of scoliosis in these rats. Studies in chicks and bipedal rats suggest that melatonin has a role in the etiology of AIS. The link between melatonin, the central nervous system, and growth and development is unique, which makes it even more attractive as a possible etiology of AIS. However, further exploration of melatonin’s pathogenic mechanisms is needed to confirm melatonin’s place in the etiology of AIS. It is possible that a reduction in melatonin could cause asymmetric growth or cause asymmetric central nervous system behavior, but these need to be further substantiated. Platelets Platelets have also been a topic of interest for scholars exploring the etiology of AIS. Platelets have a similar contractile protein system to skeletal muscle. Moreover, because platelets have little to do with the mechanistic aberrations of AIS, platelet abnormalities are more reflective of a premiere cytologic disorder. In summary, platelet abnormalities deserve attention as a functional cellular disorder that may be associated with AIS. Platelet abnormalities include ultrastructural alterations; reduced aggregation stimulated by epinephrine and adenosine 5N-diphosphate; reduction of β-adrenergic receptors; and lack of membrane myosin heavy chains of condensin. These abnormalities may reflect defects in the contractile protein system in patients with AIS. Moreover, some scholars have found elevated intraplatelet calcium levels. Because platelet myosin-regulated contractile movements require the involvement of calcium-dependent enzymes, abnormal calcium levels may reflect alterations in the protein contractile system in patients with AIS. Studies of calmodulin levels in AIS patients further support the above hypothesis. Calmodulin is a calcium-binding receptor protein that regulates the contractile protein system in skeletal muscle and platelets.Cohen et al. used enzymatic analysis to find elevated levels of calmodulin in platelets of patients with AIS.Kindsfater et al. repeated these experiments with a more accurate immunological assay, and came to the same conclusions, finding that levels of calmodulin were higher in patients with progressive AIS compared to those with stable AIS and controls. The levels of calmodulin were elevated compared to stable AIS and controls. Conclusion The study of the etiology of AIS is complicated by the large number of hypotheses that have been proposed. The scope of research includes various fields such as genetics, molecular, histochemistry, and dermatoglyphics. The results of early studies showed muscle and connective tissue differences in patients with AIS compared to controls. And one also found differences in the organization of the convex and concave surfaces of the scoliosis. However, although early studies suggested that these changes were the cause of the AIS deformity, the current view is that they are secondary to AIS. Patients have an early accelerated developmental process. Although this early accelerated development may not be the etiology of AIS, it may provide the basis for the development of scoliosis. During accelerated development, the muscle may lose its ability to maintain symmetry. Alternatively, some etiology may have led to accelerated growth and development while causing AIS. The control of axial posture is closely linked to the vestibular system. There is no doubt about the existence of vestibular abnormalities in AIS. Although the findings suggest abnormalities in brainstem regulation, much more refined work is needed to further search for abnormalities in the neural pathways that cause asymmetric malformations. Studies of platelets and melatonin have shown their association with AIS, but a great deal of work is still needed to establish their place in the process leading to the final deformity. The very nomenclature of idiopathic scoliosis points to its lack of a clear etiology. Consequently, studies attempting to explain AIS in terms of a single etiology have always ended in uncertainty and failure. It is impossible for a single gene product (collagen, muscle fibers) to cause AIS. In studies, the differences between patients with AIS and comparison groups of patients, other than malformations, are important but subtle. Several studies have shown that AIS patients compared to controls do not show differences in initial experiments (GH levels, melatonin levels), but by applying more sensitive test methods, significant differences appear. In addition, others have found equally significant differences in progressive AIS relative to non-progressive AIS. One thus concludes that there may be a subtle hormonal or cellular level physiologic environment that becomes the causative agent of asymmetric growth, and that an abnormal homeostatic system may exacerbate the deformity. This theory suggests that the etiology is the result of a number of concurrent factors. On the other hand, it may be the case that some malformations are caused by melatonin disorders, while others are caused by brainstem abnormalities, and still others by growth abnormalities. Because of the uncertainty of the current study, these malformations are lumped together under the generalized diagnosis of AIS. This group of disorders also suggests a multifactorial etiology, but the various factors should have acted independently and together to cause the final asymmetrical malformation. Moreover, different factors may occur simultaneously and be linked by genetics, or some may be caused by others. As we have seen, although correlates of AIS have been identified, the search for the ultimate pathologic mechanism remains a complex project. Some scholars have attempted to use the study of the etiology of AIS to guide the determination of its prognosis. They have tried platelet data, equilibrium, and rotational measurements to identify patients at potential risk for progression. However, since the true etiology and pathogenesis of AIS are unknown, their attempts seem somewhat haphazard, and it is not surprising that they ultimately failed. However, if the etiology of AIS can be recognized, then tools can be produced that can be used by clinicians to identify patients at risk for large-angle deformities from those with small-angle static. Obviously, this will require further research, but a great deal of important information and data has now been summarized. Joining this information together and ultimately achieving a perfect explanation of the pathogenesis of AIS will be the next step in the progression of AIS research.