Abstract
Objective
IS (idiopathic scoliosis) is a common spinal condition occurring in otherwise completely healthy adolescents. The root cause of IS remains unclear. This systematic review will focus on an update of genetic factors and IS etiology. Though it is generally accepted that the condition is not due to a single gene effect, etiology studies continue looking for a root cause including genetic variants. Though susceptibility from multiple genetic components is plausible based on known family history data, the literature remains unclear regarding multifactorial genetic influences. The objective of this study was to critically evaluate the evidence behind genetic causes (not single gene) of IS through a systematic review and strength-of-study analysis of existing genetic and genome-wide association studies (GWAS). We used the protocol of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).
Methods
PubMed was searched for the terms IS, scoliotic, spinal curve, genetic, gene, etiology, polymorphisms. Articles were assessed for risk-of-bias. Level-of-evidence grading was completed via Oxford Centre for Evidence-Based Medicine criteria. The assessment scores factor strength of a study in determining a positive or negative association to a gene etiology.
Results
After screening of 36 eligible papers, 8 relevant studies met inclusion criteria at this time, 3 were in favor of a genetic factor for IS, whereas 5 studies were against it.
Conclusion
Based on the literature analyzed, there is moderate evidence with a low risk-of-bias that does not clarify a genetic cause of IS. The 2 studies in favor of a genetic etiology were completed in homogeneous populations, limiting their generalizability. Relying on a genetic etiology alone for IS may over simplify its multifactorial nature and limit appreciation of other influences.
Keywords: Idiopathic scoliosis, Spine, Pediatrics, Genetics, Etiology, Multifactorial
1. Introduction
Idiopathic Scoliosis (IS) is a common spinal condition affecting 2–4% of the population worldwide,1,2 occurring in otherwise completely healthy adolescents. IS is characterized by complex three-dimensional changes in the normal spinal anatomy, including deformities in the axial, coronal and sagittal planes.3 The underlying cause of the disease is not fully known. Multiple disciplines have studied the etiology including clinical medicine specialties, molecular biology/genetics, anthropology and biomechanical research and to date have sought to elucidate the etiology of IS without conclusion or consensus. Most studies accept the concept that the etiology of the condition is not due to or does not originate from a single gene variant. While a genetic or familial component is likely based on known common family history data,4, 5, 6 studies continue the search for a multifactorial genetic root cause. Wynne-Davis and Riseborough showed that prevalence of scoliosis among first-degree relatives was significantly higher than in the general population in British and American cohorts.7,8 For example, Tang et al. reported a sibling reoccurrence risk of 18% in a female Chinese cohort9 and Grauers et al. had 56% of treated patients report one or more relatives with scoliosis, indicating a slightly higher risk of treatment when there is a family history of scoliosis.10 It is important to note etiological studies should also synthesize a better understanding of the multiple factors that may be involved, including inherent susceptibility, growth, biomechanics of IS, bone health and other related issues— and not only a multifactorial genetic variants. In the current ongoing discussion of etiology of IS, this review will focus on published genetics studies.
The objective of this review is to critically evaluate the evidence behind a multifactorial genetic cause of IS through a systematic review and strength-of-study analysis of genetic and genome-wide association studies (GWAS) using the modern meta-analyses protocol of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).
2. Methods
Our review followed the PRISMA guidelines for conducting systematic reviews. PRISMA is an evidence-based minimum set of items for quality reporting in systematic reviews and meta-analyses developed in 2009.11 Specifically, this framework incorporates the concept of systematic review as an iterative process, the distinction between conducting and reporting research, and differences between study-level and outcome-level assessment of bias risk.11 According to these guidelines, a structured search, study selection, risk-of-bias assessment of individual studies, and level of evidence grading was completed.
2.1. Search and selection
The PRISMA search and selection process consists of four phases: identification, screening, eligibility, and inclusion.11 First, during identification, PubMed was searched extensively for articles published between 1950 and 2017. The following terms were queried: idiopathic scoliosis, scoliotic, spinal curve, genetic, gene, etiology, polymorphisms. The titles, abstracts, and full-text were screened for relevance and any uncertainties were discussed among the study team. Next, during, screening, any duplicate records were removed and articles that were meta-analyses, review articles, editorials, or those lacking online access were excluded (Fig. 1) (See Appendix A).
Fig. 1.
PRISMA flow diagram.
2.2. Risk-of-bias assessment
The remaining articles were assessed for risk of bias using a scoring system for scoliosis literature review, modified from Schlosser et al.12 This was completed via a six-item questionnaire in which each answer ‘yes’ equates to 1 point and each answer ‘no’ equates to 0 points (Table 1). A score of 6 indicated low risk-of-bias and a score of 0 was a high risk-of-bias. Schlosser et al. used a pre-defined cut-off value of ≤3, which is half of the total score.12 Therefore, articles with a ≥4 score were included in the qualitative analysis and those with a score of ≤3 were excluded (Appendix B). Any uncertainties or disagreements reached consensus through discussion within the study team.
Table 1.
Risk-of-bias assessment.7
| Item | Scoring |
|---|---|
| Selection | |
| 1. Is the control group representative for normal adolescents? | 1 = community control; 0 = hospital control, 0 = no description of source |
| 2. Was other pathology excluded that possibly influences the outcome? | 1 = yes; 0 = no or no description |
| Comparability | |
| 3. Were the same inclusion and exclusion criteria (except for spinal deformity) used for AIS and healthy adolescents? | 1 = yes; 0 = no or no description |
| Exposure/outcome | |
| 4. Were observers blinded to AIS/healthy adolescent status? | 1 = yes; 0 = no or no description |
| 5. Was the data collection performed in the same standardized way for AIS cases and healthy adolescents? | 1 = yes; 0 = no or no description |
| 6. Was the primary outcome parameter for AIS cases and healthy adolescents available? | 1 = available for >90% of AIS and healthy adolescents; 0 = available for <90% of AIS and healthy adolescents |
2.3. Level of evidence
After this risk-of-bias assessment was completed, the remaining articles also underwent level of evidence grading. The criteria used was from the Oxford Centre for Evidence-Based Medicine (Table 2).13 The scale ranging from 1a to 5, with 1a being the highest level of evidence and 5 being the lowest. Furthermore, a grade of recommendation is assigned from the Oxford criteria based on the level of evidence. If the level of evidence is 2a, it will then receive a B grade of recommendation, meaning the study has a moderate level of evidence.
Table 2.
Oxford centre for evidence-based medicine level of evidence.8
| Grade of Recommendation | Levels of Evidence | Type of Study |
|---|---|---|
| A | 1a | SR (with homogeneity) of RCTs and of prospective cohort studies |
| 1b | Individual RCT with narrow confidence interval, prospective cohort study with good follow-up | |
| 1c | All or none studies, all or none case series | |
| B | 2a | SR (with homogeneity) of cohort studies |
| 2b | Individual cohort study | |
| 2c | Outcomes research, ecological studies | |
| 3a | SR of case control studies, SR of 3b and better studies | |
| 3b | Individual case control study, nonconsecutive cohort study | |
| C | 4 | Case series/case report, poor quality cohort studies |
| D | 5 | Expert opinion, bench research |
SR, systematic reviews; RCTs, randomized control trials.
2.4. Qualitative analysis
The included studies were assessed in a qualitative analysis. The grade of risk-of-bias factored towards the strength of a study in providing a positive or negative single-gene etiology of IS. For example, if a study received a score of 6 on risk-of-bias assessment, it has a low risk-of-bias.
3. Results
The PRISMA flow diagram is seen in Fig. 1.11 In the identification phase, 36 relevant articles were identified. Next, in the screening phase, 19 articles were excluded for being duplicate records and articles that were meta-analyses, review articles, editorials, lacking online access or citations used for the paper. In the eligibility phase, 9 more articles were excluded due to an assessed high risk of bias (score ≤ 3). This left 8 articles to be included in the qualitative analysis. Of these 8, two had a risk-of-bias score of 4, four had a risk-of-bias score of 5 and two had a risk-of-bias score of 6 (very good scores).
Level of evidence was then assessed using the criteria from the Oxford Centre for Evidence-Based Medicine.13 All 8 articles received a grade of recommendation of B based on the Oxford criteria. Of the 8 studies analyzed, 3 were in favor of a single genetic link for IS, whereas the other 5 studies were against a single-gene etiology (Table 3) . Of note, 4 of the 8 included studies were GWAS (Table 3).
Table 3.
Summary of the 8 articles included in the quantitative analysis.
| Citation | Study Design | Grade of Recommendation | Level of Evidence | Risk-of-bias Score | In favor of a Gene Etiology? | Gene/Region | Population Studied |
|---|---|---|---|---|---|---|---|
| Sharma et al. (2015)9 | GWAS | B | 3b | 4 | Yes (only in females) | PAX1 | Caucasian and Japanese |
| Zhu et al. (2017)10 | GWAS | B | 3b | 4 | No | MEIS1, MAGI1, AND TNIK | Han Chinese |
| Shyy et al. (2010)11 | Cohort | B | 2b | 5 | No | GPR30, hMel-1B, and RORalpha | Caucasian |
| Gao et al. (2014)12 | Cohort | B | 2b | 5 | Yes | 15q24 microdeletion region | Non-Hispanic White |
| Chettier et al. (2015)1 | Case-control | B | 3b | 5 | No | LBX1 | Caucasian |
| Mcgregor et al. (2011)13 | Case-control | B | 3b | 5 | No | LOX, LOX1, LOX2, LOX3 and LOX4 | Caucasian |
| Zhu et al. (2015)14 | GWAS | B | 3b | 6 | No | AJAP1, PAX3, and BCL-2 | Han Chinese |
| Gao et al. (2013)15 | GWAS | B | 3b | 6 | Yes | LBX1 | Han Chinese |
*GWAS = genome wide association study.
Together, these assessments would indicate, that based on current literature, there is moderate evidence with a low risk-of-bias against a single genetic etiology for IS.
4. Discussion
The objective of this study was to conduct a systematic review and strength-of-study analysis using PRISMA guidelines in order to assess the current literature evidence supporting a genetic etiology of IS. After extensive screening, those articles were included with a moderate level of evidence and low risk-of-bias. The current evidence does not strongly clarify a gene theory for the etiology of IS.
In our systematic review, there were 3 studies that were in favor of a possible gene theory. One study in favor of a gene theory was conducted by Gao et al. It was a case-control study in which three SNPs near the LBX1 gene (rs11190870, rs11598564, and rs625039) were found to be associated with IS predisposition.20 The authors noted the mechanism is unexplained and they did not find any association between these variants and curve severity. This study was conducted in a Han Chinese population in southern China, so the generalizability of the results to other populations is unclear. Chettier et al. performed a GWAS in a Caucasian population to further investigate the association of the LBX1 locus with IS.1 Instead of using the more common method of unphased SNPs as clinical predictors, they used haplotypes to see if they could uncover risk without compromising the effect size. The authors identified two clinically relevant haplotypes in the LBX1 region: a recessive risk haplotype and a co-dominant protective haplotype. Individuals carrying both haplotypes have neither increased nor decreased risk, suggesting they may neutralize each other; a mechanism is unknown. Another study was a GWAS of American and Japanese subjects, sought to explain the sexual dimorphism observed in IS. The authors identified associations with 20p11.22 SNPs (single nucleotide polymorphisms).14 The associated segment is a PAX1 enhancer locus, which encodes for transcription factors involved in spine development. The precise mechanism of how this relates to IS is unexplained. While this found an association for females, the lack of male association works against a gene theory for IS overall.
Our review found 5 studies that do not support a genetic theory. Zhu et al. conducted a GWAS, in which they identified three novel susceptible loci: 3p14.1, 2p14, and 3q26.2.15 The authors also confirmed a previously located region of 20p11.22, but in females only. Since four genes were identified or confirmed, this study would not support a single gene theory unless all 4 happened to be regulators of an unidentified single gene. All patients in this study were from a Chinese Han population. Zhu et al. also completed a GWAS in a Han Chinese population and identified three new loci that may be associated with IS at 1p36.32 near AJAP1, 2q36.1 near PAX3, and 18q21.33 near BCL-2.19 A limitation of this study is the lack of experiments supporting the functional role of these newly defined susceptibility genes, and lack of consensus among papers on these loci. Regardless, the study would argue against a gene theory if the multiple different loci have a relationship with the development of IS.
Shyy et al. evaluated the coding, splice-site, and promoter regions of three melatonin-related receptors by DNA sequencing for variants associated with development of scoliosis.16 They concluded that though there was a relationship observed from previous animal studies with melatonin and scoliosis, they did not find an association in humans with melatonin receptors and IS. While the study did not screen the noncoding, or intronic, regions of the three genes, the authors explain this as a relatively accepted practice.16 Gao et al. completed a 24-patient cohort study of mostly non-Hispanic white subjects with early onset scoliosis using high-density microarray genotyping compared to 482 IS patients.17 The authors analyzed copy number variation and absence of heterozygosity and found that the novel mutations that were identified in 3 early onset scoliosis patients were not associated with their IS cohort. Finally, McGregor et al. tested a hypothesis from animal models, which implicated the lysyl oxidase enzyme in the development of scoliosis.18 The authors conducted a case-control genetic association study to determine if LOX- LOX4 were associated with the IS phenotype. They did not find any evidence of IS association with common variants and identified SNPs of these five human lysyl oxidase genes. Though a large cohort of IS patients were studied, the study was not powered to detect small risks. The power of the study was also weakened by use of an unscreened control population, which can potentially lead to misclassification bias.
The etiology of IS is likely multifactorial with complex interplay between various factors. For example, non-genetic risk factors, such as bipedal verticality,21, 22, 23 inherent torsion,21,22,24 spine immaturity,21,22,24 biomechanics,21,22,25,26 and bone health (osteopenia)27 have been reported. In the studies included in this systematic review, other important environmental and physiologic factors were not considered, thereby limiting the researchers’ ability to evaluate gene-environment interactions for a multifactorial disease. Genetic association studies are a means of identifying risk variants in complex traits. Of the total reported gene associations, less than 5% are estimated to explain the overall genetic contribution to IS disease risk.14 Though certain genes were found to have small associations in limited populations, they were far from describing the full phenotype of IS. It is also important to note that the studies included in this analysis were conducted primarily in Asian or Caucasian populations only, further limiting the generalizability of the results. For all these reasons, the current literature does not clarify a gene etiology for IS.
Our current review is not without limitations. The screened sample size is small, with 36 initial articles from a single though from a very large database. Five articles were excluded due to limited availability of the full-text rather than the merit of the article. Attempts were made to contact other institutional libraries as well as corresponding authors for access, but these were unsuccessful. Of these five studies, based on the abstracts alone, three were in favor and two were against of a gene theory. We also combined three independent methods of assessments-the PRISMA method of qualitative analysis, a risk-of-bias assessment adapted from Schlosser et al. and a level of evidence assessment from the Oxford Centre for Evidence-Based Medicine.
Gorman et al. performed an extensive review of 34 candidate gene studies and 16 genome-wide and linkage studies published from 1992 to 2011.28 In contrast to our study, they did not exclude studies based on quality metrics such at the PRISMA guidelines, risk-of-bias or level of evidence. Despite their larger inclusion set, they found nearly all studies were significantly underpowered and/or failed to be replicated by larger subsequent studies. Their comprehensive analysis covered 35 different target genes or loci corresponding to a wide variety of developmental and physiologic systems/pathways including connective tissue, bone formation/metabolism, melatonin signaling, and puberty/growth. While a “good” genetic association study likely needs thousands of individuals to detect common variant risk alleles,29 they found only 3 studies had over 1000 patients. These findings support a thesis of synthesizing multiple diverse disciplines in 2 ways; first by demonstrating that no study to date has provided strong evidence for clarifying a gene cause of large percentage of IS, and second, that the genetic loci that have been associated originate in a diverse number of body systems and areas of development. This is certainly not to say that genetics do not play a role in IS susceptibility, as the evidence for a degree of heritability is well-appreciated. However, the current evidence reviewed here and by Gorman et al. do not clearly explain a genetic common pathway for the etiology of IS.
Additionally, in Grauers et al. detailed but non-systematic review on studies of IS etiology6 they suggest that better and newer genetic methodologies may help us identify rare variants, but acknowledge that current evidence does not clarify a genetic root cause.
As Gorman et al. proposes, if future genetic association studies can be better designed by increasing their power and controlling for gender, age, ethnic and phenotypic variety (such as curve type, magnitude or progression risk, etc.) we may find there exist associations of genetic markers that help understand a subset of patient's heritable risk for IS. A comprehensive approach that integrates other potential factors, including association studies of genetic variants, non-genetic factors including anthropology,21 the complex interaction with pathoanatomy,3 a predisposition to deformation of the immature spine during growth21 and biomechanics of the human spine,3 will assist the critical thinking necessary in a discussion of elucidation of etiology of IS. A synthesis of these genetic, biomechanical and physiologic pathways, will aid in creating a complete picture that increases our understanding of the nature of IS. An understanding of the etiology can aid in earlier diagnosis and developing innovative and better-personalized treatments plans for IS patients.
5. Conclusion
Based on the systematic review in this paper under PRISMA guidelines, there is moderate evidence with a low risk-of-bias that does not clarify an understanding of a genetic cause of IS.
Source of funding
This funding was supported by an internal grant from Lurie Children's Hospital.
Declaration of competing interest
None of the authors have any conflict of interests to disclose.
Appendix A.
Table 4.
Summary of the articles excluded in the screening phase.
| Citation | Reason for Exclusion |
|---|---|
| Liang et al. (2014) | Meta-Analysis |
| Cao et al. (2016) | Meta-Analysis |
| Londono et al. (2014) | Meta-Analysis |
| Latalski et al. (2017) | Review Paper |
| Burwell et al. (2016) | Review Paper |
| Wajchenberg et al. (2016) | Review Paper |
| Wise et al. (2008) | Review Paper |
| Dayer et al. (2013) | Review Paper |
| Giampietro et al. (2015) | Editorial Paper |
| Ogura et al. (2016) | No Online Access |
| Zhonghua et al. (2008) | No Online Access |
| Zhonghua et al. (2009) | No Online Access |
| Xia et al. (2007) | No Online Access |
| Zhou et al. (2012) | No Online Access |
Appendix B.
Table 5.
Summary of the articles excluded in eligibility phase.
| Citation | Question 1 | Question 2 | Question 3 | Question 4 | Question 5 | Question 6 | Risk-of-bias Score |
|---|---|---|---|---|---|---|---|
| Nada et al. (2017) | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
| Al-Othman et al. (2017) | 0 | 0 | 0 | 0 | 1 | 1 | 2 |
| Haller et al. (2016) | 0 | 0 | 0 | 0 | 1 | 1 | 2 |
| Kou et al. (2013) | 1 | 0 | 0 | 0 | 1 | 1 | 3 |
| Miyake et al. (2013) | 1 | 0 | 0 | 0 | 1 | 1 | 3 |
| Xu et al. (2017) | 1 | 0 | 0 | 0 | 1 | 1 | 3 |
| Qin et al. (2017) | 0 | 0 | 1 | 0 | 1 | 1 | 3 |
| Sharma et al. (2011) | 1 | 0 | 0 | 0 | 1 | 1 | 3 |
| Shyy et al. (2010) | 0 | 1 | 0 | 0 | 1 | 1 | 3 |
Questions 1–6 are from the risk-of-bias assessment in Table 1.
References
- 1.Chettier R., Nelson L., Ogilvie J.W., Albertsen H.M., Ward K. Haplotypes at LBX1 have distinct inheritance patterns with opposite effects in adolescent idiopathic scoliosis. PLoS One. 2015;10(2) doi: 10.1371/journal.pone.0117708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Grauers A., Einarsdottir E., Gerdhem P. Genetics and pathogenesis of idiopathic scoliosis. Scoliosis Spinal Disord. 2016;11:45. doi: 10.1186/s13013-016-0105-8. 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Van Blarcum G.S., Wagner S.C., Meyer R.M., Kang D.G., Lehmen R.A. The biomechanics of spinal deformity in adolescent idiopathic scoliosis. J Spine. 2014;4(197) doi: 10.4172/2165-7923.1000197. [DOI] [Google Scholar]
- 4.Riseborough E.J., Wynne-Davies R. A genetic survey of idiopathic scoliosis in Boston, Massachusettes. J Bone Joint Surg Am. 1973;55(5):974–982. [PubMed] [Google Scholar]
- 5.Tand N.L., Yeung H.Y., Hung V.W. Genetic epidemiology and heritability of AIS: a study of 415 Chinese female patietns. J Orthop Res. 2013;30(9):1464–1469. doi: 10.1002/jor.22090. [DOI] [PubMed] [Google Scholar]
- 6.Grauers A., Danielsson A., Karlsson M., Ohlin A., Gerdhem P. Family hisotry and its assocaition to curve sive and treatment in 1,463 patients with idiopathic scoliosis. Eur Spine J. 2013;22(110):2421–2426. doi: 10.1007/s00586-013-2860-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wynne-Davies R. Familial (idiopathic) scoliosis. A family survey. J Bone Joint Surg Br. 1968;50(1):24–30. [PubMed] [Google Scholar]
- 8.Riseborough E.J., Wynne-Davies R. A genetic survey of idiopathic scoliosis in Boston, Massachusetts. J Bone Joint Surg Am. 1973;55(5):974–982. [PubMed] [Google Scholar]
- 9.Tang N.L., Yeung H.Y., Hung V.W. Genetic epidemiology and heritability of AIS: a study of 415 Chinese female patients. J Orthop Res. 2012;30(9):1464–1469. doi: 10.1002/jor.22090. [DOI] [PubMed] [Google Scholar]
- 10.Grauers A., Danielsson A., Karlsson M., Ohlin A., Gerdhem P. Family history and its association to curve size and treatment in 1,463 patients with idiopathic scoliosis. Eur Spine J. 2013;22(11):2421–2426. doi: 10.1007/s00586-013-2860-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Moher D., Liberati A., Tetzlaff J., Altman D.G., The PRISMA Group Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7) doi: 10.1371/journal.pmed1000097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Schlosser T.P.C., van der Heijden G.J.M.G., Versteeg A.L., Castelein R.M. How idiopathic is adolescent idiopathic scoliosis? a systematic review of associated abnormalities. PLoS One. 2014;9(5) doi: 10.1371/journal.pone.0097461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Opeyemi D.O., Rhee J.S. Rating evidence in medical literature. AMA J Ethics. 2011;13(1):46–51. doi: 10.1001/virtualmentor.2011.13.1.pfor2-1101. [DOI] [PubMed] [Google Scholar]
- 14.Sharma S., Londono D., Eckalbar W.L. A PAX1 enhancer locus is associated with susceptibility to idiopathic scoliosis in females. Nat Commun. 2015;18(6):6452. doi: 10.1038/ncomms7452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zhu Z., Xu L., Leung-Sang Tang N. Genome-wide association study identifies novel susceptible loci and highlights Wnt/beta-catenin pathway in the development of adolescent idiopathic scoliosis. Hum Mol Genet. 2017;26(8):1577–1583. doi: 10.1093/hmg/ddx045. [DOI] [PubMed] [Google Scholar]
- 16.Shyy W., Wang K., Gurnett C.A. Evaluation of GPR50, hMel-1B and ROR-alpha melatonin-related receptors and the etiology of adolescent idiopathic scoliosis. J Pediatr Orthop. 2010;30(6):539–543. doi: 10.1097/BPO.0b013e3181e7902c. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gao X., Gotway G., Rathjen K., Johnston C., Sparagana S., Wise C.A. Genomic analysis of patients with unexplained early-onset scoliosis. Spine Deform. 2014;2(5):324–332. doi: 10.1016/j.jspd.2014.04.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mcgregor T.L., Gurnett C.A., Dobbs M.B. Common polymorphisms in human lysyl oxidase genes are not associated with the adolescent idiopathic scoliosis phenotype. MBC Med Genet. 2011;12(8):92. doi: 10.1186/1471-2350-12-92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhu Z., Tang N.L., Xu L. Genome-wide association study identifies new susceptibility loci for adolescent idiopathic scoliosis in Chinese girls. Nat Commun. 2015;6:8355. doi: 10.1038/ncomms9355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Gao W., Peng Y., Liang G. Association between common variants near LBX1 and adolescent idiopathic scoliosis replicated in the Chinese Han population. PLoS One. 2013;8(1) doi: 10.1371/journal.pone.0053234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kouwenhoven J.W.M., Smit T.H., van der Veen A.J., Kingma I., van Dieen J.H., Castelein R.M. Effects of dorsal versus ventral shear loads on the rotational stability of the thoracic spine: a biomechanical porcine and human cadaveric study. Spine. 2007;7(32):2545–2550. doi: 10.1097/BRS.0b013e318158cd86. [DOI] [PubMed] [Google Scholar]
- 22.Sarwark J.F., Castelein R.M., Maqsood A., Aubin C.E. The biomechanics of induction in adolescent idiopathic scoliosis. J Bone Joint Surg Am. 2019;101(6):e22. doi: 10.2106/JBJS.18.00846. [DOI] [PubMed] [Google Scholar]
- 23.Schlösser T.P.C., Janssen M.M.A., Hogervorst T. The odyssey of sagittal pelvic morphology during human evolution: a perspective on different Hominidae. Eur Spine J. 2014;23(7):1433–1441. doi: 10.1016/j.spinee.2017.03.016. [DOI] [PubMed] [Google Scholar]
- 24.Kouwenhoven J.W.M., Vincken K.L., Bartels L.W., Castelein R.M. Analysis of preexistent vertebral rotation in the normal spine. Spine. 2006;31(13):1467–1472. doi: 10.1097/01.brs.0000219938.14686.b3. [DOI] [PubMed] [Google Scholar]
- 25.Villemure I., Aubin C.E., Dansereau J., Labelle H. Simulation of progressive deformities in adolescent idiopathic scoliosis using a biomechanical model integrating vertebral growth modulation. J Biomech Eng. 2002;124(6):784–790. doi: 10.1115/1.1516198. [DOI] [PubMed] [Google Scholar]
- 26.Shi L., Wang D., Driscoll N. Biomechanical analysis and modelling of different vertebral growth patterns in adolescent idiopathic scoliosis and healthy subject. Scoliosis. 2011;23(11):6. doi: 10.1186/1748-7161-6-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lam T.P., Yip B.H.K., Yu F.W.P. Improving low bone mass in girls with AIS using calcium and vitamin D supplementation- a randomized controlled study. Bone Abstr. 2015 doi: 10.1530/boneabs4.P93. [DOI] [Google Scholar]
- 28.Gorman K.F., Julien C., Moreau A. The genetic epidemiology of idiopathic scoliosis. Eur Spine J. 2012;21:1905–1919. doi: 10.1007/s00586-012-2389-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Timpson N.J., Lawlor D.A., Harbord R.M. C-reactive protein and its role in metabolic syndrome: mendelian randomisation study. Lancet. 2006;366(9501):1954–1959. doi: 10.1016/S0140-6736(05)67786-0. [DOI] [PubMed] [Google Scholar]