Predictors of spine deformity progression in adolescent idiopathic scoliosis: A systematic review with meta-analysis

Table 1 Ovid medline search strategy: Ovid MEDLINE^®, Ovid MEDLINE^® In-Process and Other Non-Indexed Citations, Ovid MEDLINE^®, Daily and Ovid OLDMEDLINE^®

No.	Search syntax
1	“Adolescent idiopathic scoliosis”.ab,ti.
2	(AIS and scoliosis).ab,ti.
3	Scoliosis/ and (exp adolescent/ or exp child/)
4	Or/1-3
5	“Curve progression”.ab,ti.
6	“Disease susceptibility”.ab,ti.
7	Prediction.ab,ti.
8	“Disease progression”.ab,ti.
9	Exp disease progression/
10	Disease susceptibility/
11	“Predictive value of tests”/
12	Exp decision support techniques/
13	Or/5-12
14	Scoliosis/ra
15	(Ogilvie JW or Ward K*).au. and scoliosis.ab,ti.
16	“Scoliscore”.mp.
17	“Axial biotech”.mp.
18	Moreau A*.au. and scoliosis.ab,ti.
19	4 and 13
20	13 and 14
21	Or/15-20
22	(genetic adj2 test*).ab,ti.
23	“Genetic predisposition”.ab,ti.
24	“Single nucleotide polymorphism”.ab,ti. Or (SNP and polymorphism).ab,ti.
25	Genetic Testing/
26	Exp genetic predisposition to disease/
27	Polymorphism, single nucleotide/
28	Or/22-27
29	4 and 28
30	30. 21 or 29

Table 2 Excluded publications

No.	Excluded publications
1	Buchan JG, Alvarado DM, Haller GE, Cruchaga C, Harms MB, Zhang T, Willing MC, Grange DK, Braverman AC, Miller NH, Morcuende JA, Tang NL, Lam TP, Ng BK, Cheng JC, Dobbs MB, Gurnett CA. Rare variants in FBN1 and FBN2 are associated with severe adolescent idiopathic scoliosis. Hum Mol Genet 2014; 23: 5271-5282
2	Danielsson AJ, Nachemson AL. Radiologic findings and curve progression 22 years after treatment for adolescent idiopathic scoliosis: comparison of brace and surgical treatment with matching control group of straight individuals. Spine (Phila Pa 1976) 2001; 26: 516-525
3	Grauers A, Danielsson A, Karlsson M, Gerdhem P: Familial heredity of idiopathic scoliosis unrelated to age at diagnosis and prognosis. Eur Spine J 2012; 21: S314
4	Inoue M, Minami S, Nakata Y, Kitahara H, Otsuka Y, Isobe K, Takaso M, Tokunaga M, Nishikawa S, Maruta T, Moriya H. Association between estrogen receptor gene polymorphisms and curve severity of idiopathic scoliosis. Spine (Phila Pa 1976) 2002; 27: 2357-2362
5	Lonstein JE, Carlson JM. The prediction of curve progression in untreated idiopathic scoliosis during growth. J Bone Joint Surg Am 1984; 66: 1061-1071
6	Lowe TG, Burwell RG, Dangerfield PH. Platelet calmodulin levels in adolescent idiopathic scoliosis (AIS): can they predict curve progression and severity? Summary of an electronic focus group debate of the IBSE. Eur Spine J 2004; 13: 257-265
7	Machida M, Dubousset J, Yamada T, Kimura J. Serum melatonin levels in adolescent idiopathic scoliosis prediction and prevention for curve progression--a prospective study. J Pineal Res 2009; 46: 344-348
8	Miyake A, Kou I, Takahashi Y, Johnson TA, Ogura Y, Dai J, Qiu X, Takahashi A, Jiang H, Yan H, Kono K, Kawakami N, Uno K, Ito M, Minami S, Yanagida H, Taneichi H, Hosono N, Tsuji T, Suzuki T, Sudo H, Kotani T, Yonezawa I, Kubo M, Tsunoda T, Watanabe K, Chiba K, Toyama Y, Qiu Y, Matsumoto M, Ikegawa S. Identification of a susceptibility locus for severe adolescent idiopathic scoliosis on chromosome 17q24.3. PLoS One 2013; 8: e72802
9	Nault ML, Mac-Thiong JM, Roy-Beaudry M, deGuise J, Labelle H, Parent S. Three-dimensional spine parameters can differentiate between progressive and nonprogressive patients with AIS at the initial visit: a retrospective analysis. J Pediatr Orthop 2013; 33: 618-623
10	Nault ML, Mac-Thiong JM, Roy-Beaudry M, Turgeon I, de Guise J, Labelle H, Parent S: Three-Dimensional Spinal Morphology can Differentiate Between Progressive and Non-Progressive Patients With Adolescent Idiopathic Scoliosis at the Initial Presentation. Spine (Phila Pa 1976) 2014
11	Ogilvie JW. Update on prognostic genetic testing in adolescent idiopathic scoliosis (AIS). J Pediatr Orthop 2011; 31: S46-S48
12	Ogura Y, Takahashi Y, Kou I, Nakajima M, Kono K, Kawakami N, Uno K, Ito M, Minami S, Yanagida H, Taneichi H, Yonezawa I, Tsuji T, Suzuki T, Sudo H, Kotani T, Watanabe K, Chiba K, Toyama Y, Matsumoto M, Ikegawa S. A replication study for association of 5 single nucleotide polymorphisms with curve progression of adolescent idiopathic scoliosis in Japanese patients. Spine (Phila Pa 1976) 2013; 38: 571-575
13	Ogura Y, Takahashi Y, Kou I, Nakajima M, Kono K, Kawakami N, Uno K, Ito M, Minami S, Yanagida H, Taneichi H, Yonezawa I, Tsuji T, Suzuki T, Sudo H, Kotani T, Watanabe K, Chiba K, Toyama Y, Matsumoto M, Ikegawa S. A replication study for association of 53 single nucleotide polymorphisms in a scoliosis prognostic test with progression of adolescent idiopathic scoliosis in Japanese. Spine (Phila Pa 1976) 2013; 38: 1375-1379
14	Patten SA, Moldovan F. Could genetic determinants of inner ear anomalies be a factor for the development of idiopathic scoliosis? Med Hypotheses 2011; 76: 438-440
15	Peng Y, Liang G, Pei Y, Ye W, Liang A, Su P. Genomic polymorphisms of G-protein estrogen receptor 1 are associated with severity of adolescent idiopathic scoliosis. Int Orthop 2012; 36: 671-677
16	Qiu XS, Tang NL, Yeung HY, Lee KM, Hung VW, Ng BK, Ma SL, Kwok RH, Qin L, Qiu Y, Cheng JC. Melatonin receptor 1B (MTNR1B) gene polymorphism is associated with the occurrence of adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2007; 32: 1748-1753
17	Qiu XS, Tang NL, Yeung HY, Qiu Y, Cheng JC. Genetic association study of growth hormone receptor and idiopathic scoliosis. Clin Orthop Relat Res 2007; 462: 53-58
18	Roye BD, Wright ML, Williams BA, Matsumoto H, Corona J, Hyman JE, Roye DP, Vitale MG. Does ScoliScore provide more information than traditional clinical estimates of curve progression? Spine (Phila Pa 1976) 2012; 37: 2099-2103
19	Sanders JO, Khoury JG, Kishan S, Browne RH, Mooney JF, Arnold KD, McConnell SJ, Bauman JA, Finegold DN. Predicting scoliosis progression from skeletal maturity: a simplified classification during adolescence. J Bone Joint Surg Am 2008; 90: 540-553
20	Soucacos PN, Zacharis K, Gelalis J, Soultanis K, Kalos N, Beris A, Xenakis T, Johnson EO. Assessment of curve progression in idiopathic scoliosis. Eur Spine J 1998; 7: 270-277
21	Soucacos PN, Zacharis K, Soultanis K, Gelalis J, Xenakis T, Beris AE. Risk factors for idiopathic scoliosis: review of a 6-year prospective study. Orthopedics 2000; 23: 833-838
22	Stokes IA, Aronsson DD. Disc and vertebral wedging in patients with progressive scoliosis. J Spinal Disord 2001; 14: 317-322
23	Sun X, Qiu Y, Zhu Z, Zhu F, Wang B, Yu Y, Qian B. Variations of the position of the cerebellar tonsil in idiopathic scoliotic adolescents with a cobb angle & gt; 40 degrees: a magnetic resonance imaging study. Spine (Phila Pa 1976) 2007; 32: 1680-1686
24	Sun X, Zhu ZZ, Qiu Y, Wang B, Li WG, Zhu F, Yu Y, Qian BP, Ma WW. [The role of initial bone mineral status in predicting the early outcome of brace treatment in girls with adolescent idiopathic scoliosis]. Zhonghua Waike Zazhi 2008; 46: 1066-1069
25	Takahashi Y, Kou I, Takahashi A, Johnson TA, Kono K, Kawakami N, Uno K, Ito M, Minami S, Yanagida H, Taneichi H, Tsuji T, Suzuki T, Sudo H, Kotani T, Watanabe K, Chiba K, Hosono N, Kamatani N, Tsunoda T, Toyama Y, Kubo M, Matsumoto M, Ikegawa S. A genome-wide association study identifies common variants near LBX1 associated with adolescent idiopathic scoliosis. Nat Genet 2011; 43: 1237-1240
26	Takahashi Y, Matsumoto M, Karasugi T, Watanabe K, Chiba K, Kawakami N, Tsuji T, Uno K, Suzuki T, Ito M, Sudo H, Minami S, Kotani T, Kono K, Yanagida H, Taneichi H, Takahashi A, Toyama Y, Ikegawa S. Lack of association between adolescent idiopathic scoliosis and previously reported single nucleotide polymorphisms in MATN1, MTNR1B, TPH1, and IGF1 in a Japanese population. J Orthop Res 2011; 29: 1055-1058
27	Takahashi Y, Matsumoto M, Karasugi T, Watanabe K, Chiba K, Kawakami N, Tsuji T, Uno K, Suzuki T, Ito M, Sudo H, Minami S, Kotani T, Kono K, Yanagida H, Taneichi H, Takahashi A, Toyama Y, Ikegawa S. Replication study of the association between adolescent idiopathic scoliosis and two estrogen receptor genes. J Orthop Res 2011; 29: 834-837
28	Tang NL, Yeung HY, Hung VW, Di Liao C, Lam TP, Yeung HM, Lee KM, Ng BK, Cheng JC. Genetic epidemiology and heritability of AIS: A study of 415 Chinese female patients. J Orthop Res 2012; 30: 1464-1469
29	Vijvermans V, Fabry G, Nijs J. Factors determining the final outcome of treatment of idiopathic scoliosis with the Boston brace: a longitudinal study. J Pediatr Orthop B 2004; 13: 143-149
30	Wiley JW, Thomson JD, Mitchell TM, Smith BG, Banta JV. Effectiveness of the boston brace in treatment of large curves in adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2000; 25: 2326-2332
31	Wise CA, Gao X, Shoemaker S, Gordon D, Herring JA. Understanding genetic factors in idiopathic scoliosis, a complex disease of childhood. Curr Genomics 2008; 9: 51-59
32	Wu H, Ronsky JL, Cheriet F, Harder J, Küpper JC, Zernicke RF. Time series spinal radiographs as prognostic factors for scoliosis and progression of spinal deformities. Eur Spine J 2011; 20: 112-117
33	Yamauchi Y, Yamaguchi T, Asaka Y. Prediction of curve progression in idiopathic scoliosis based on initial roentgenograms. A proposal of an equation. Spine (Phila Pa 1976) 1988; 13: 1258-1261
34	Yang T, Jia Q, Guo H, Xu J, Bai Y, Yang K, Luo F, Zhang Z, Hou T. Epidemiological survey of idiopathic scoliosis and sequence alignment analysis of multiple candidate genes. Int Orthop 2012; 36: 1307-1314
35	Yang T, Xu JZ, Jia QZ, Guo H, Luo F, Ye Q, Bai Y. [Comparative analysis of sequence alignment of SH3GL1 gene as a disease candidate gene of adolescent idiopathic scoliosis]. Zhonghua Waike Zazhi 2010; 48: 435-438
36	Yang Y, Wu Z, Zhao T, Wang H, Zhao D, Zhang J, Wang Y, Ding Y, Qiu G. Adolescent idiopathic scoliosis and the single-nucleotide polymorphism of the growth hormone receptor and IGF-1 genes. Orthopedics 2009; 32: 411
37	Ylikoski M. Spinal growth and progression of adolescent idiopathic scoliosis. Eur Spine J 1993; 1: 236-239
38	Ylikoski M. Growth and progression of adolescent idiopathic scoliosis in girls. J Pediatr Orthop B 2005; 14: 320-324
39	Yu Ws, Chan Ky, Yu FWP, Yeung Hy, Ng BKW, Lee Km, Lam Tp, Cheng JCY: Abnormal bone quality versus low bone mineral density in adolescent idiopathic scoliosis: a case-control study with in vivo high-resolution peripheral quantitative computed tomography. Spine Journal 2013
40	Zhang HQ, Lu SJ, Tang MX, Chen LQ, Liu SH, Guo CF, Wang XY, Chen J, Xie L. Association of estrogen receptor beta gene polymorphisms with susceptibility to adolescent idiopathic scoliosis. Spine (Phila Pa 1976) 2009; 34: 760-764
41	Zhao D, Qiu GX, Wang YP, Zhang JG, Shen JX, Wu ZH, Wang H. Association of calmodulin1 gene polymorphisms with susceptibility to adolescent idiopathic scoliosis. Orthop Surg 2009; 1: 58-65

Table 3 Selected studies

Ref.	Study design	Publication	Spine deformity	Age (mean/range)	Gender	n	Treatment	Initial Cobb angle(degree)	Follow-up	Drop out	Progression of deformitycriteria	Analysis/Method ofprediction	Indicesused	Prediction validity(progressive vs stable spine deformities)
Upadhyay et al[73]	RCS	Art.	Thoracic, thoracolumbar, lumbar	< 18 Risser sign ≤ 2	NS	85	Brace	20-45	Until skeletal maturity or surgical tretment	NS	Cobb increasing ≥ 5o, and/or vertebral rotation ≥ 5o	Comparative analysis of progress vs stable cases	Predictor: Increase of Cobb angle and/or vertebral rotation ≥ 5o at 1-2 mo follow-up during brace treatment	OR = 33.23 (95%Cl: 4.0-270.4) P < 0.001 (1) Sensitivity: 39%; (2) Specificity: 98%; (3) +PV: 93%; and (4) -PV: 72%
Peterson et al[62]	PChS	Art.	Thoracic, thoracolumbar	10-15	F (100%)	159	Observation (120) Electrical stimulation (39)	25-30	Until skeletal maturity	NS	Cobb increasing ≥ 6o	Multiple logistic regression modeling	Predictors: (1) Risser sign (0-1); (2) Apical level (uperTh12); (3) Imbalance (10 mm); and (4) Age	(1) Sensitivity: 81%2; (2) Specificity: 81%; (3) +PV: 82%; and (4) -PV: 80%
Ajemba et al[60]	RChS	Art.	NS	12.3 (10-15)	F (87%) M (14%)	44	Observation (30) Brace (14)	18-49	1 yr - skeletal maturity	NS	Cobb increasing ≥ 5o	6 multiple support vector classifier models	Predictors: (1)16 Lenke Rad. Indices; (2)Wrist X-ray; (3) Age; (4) Sex; and (5) Growing index	(1) Sensitivity: 67%-91%2; (2) Specificity: 22%-67%; (3) +PV: 73%-86%; and (4) -PV: 43%-67%
1Cheung et al[64]	PCS	Art.	Right thoracic	10-16	F (87%) M (13%)	30	NS	10-60	4-5 mo	NS	Cobb increasing >10o	Multiple regression modeling, nomogram	Predictor: (1) Spinal grows velocity (≥ 11 mm/yr); (2) Paraspinal EMG activity concave/convex ≥ 1.3	(1) Sensitivity: 69%-79%2; (2) Specificity: 69%-79%; and (3) +PV: 60%-89%
1Danielsson et al[63]	PChS	Art.	Thoracic, Thoracolumbar	10-15 (skeletal)	F (100%)	92	Observation Brace and electrical stimulation	25-35	16 yr	14%	Cobb increasing ≥ 6o	Rate comparison	Predictor: Premenarche at inclusion vs menarche at inclusion	OR = 2.523 (95%Cl: 1.0-6.11) P = 0.05 (1) Sensitivity: 60%; (2) Specificity: 63%; (3) +PV: 53%; and (4) -PV: 70%
Kindsfater et al[46]	RCS	Art.	Thoracic, thoracolumbar	11-20	F (71%) M (29%)	17	Observation (7) Brace (10)	34 (15-90)	< 1 yr	NS	Cobb > 30o; Increasing > 10o /yr	Comparative analysis of progress vs stable cases	Predictor: Level of platelet calmodulin (ng/μg of protein): progressive 1.4-10.7; stable < 1.4 (P = 0.001)	OR = 275.03 (95%Cl: 4.8-15724.2) P = 0.007 (1) Sensitivity: 100%; (2) Specificity: 100%; (3) +PV: 100%; and (4) -PV: 100%
1Lowe et al[47]	PChS	Art.	King I-V	Adolescents	F (93%) M (7%)	55	Observation (28) Brace (17) Fusion(10)	≤ 25	1-3 yr	9.80%	Cobb increasing > 10o /yr	Comparative analysis of progress vs stable cases	Predictor: Increasing of platelet calmodulin level during first year of observation	OR = 11.03 (95%Cl: 1.7-69.9) P = 0.02 (1) Sensitivity: 69%; (2) Specificity: 83%; (3) +PV: 85%; and (4) -PV: 67%
Sun et al[67]	RCS	Art.	Thoracic, thoracolumbar, lumbar	10-16	F (100%)	142	Brace	20-40	0.6-5.9 yr	NS	Cobb exceeding 45o, surgical tretment	Multiple logistic regression modeling	Predictors: (1) Premenarche; (2) Curve > 30o; and (3) Risser sign: 0-1	OR: 5.1-11.52P ≤ 0.002 (1) Sensitivity 72%-89%3; (2) Specificity 48%-77%; (3) +PV: 20%-33%; and (4) -PV: 94%- 97%
1Sun et al[66]	RCS	Art.	Thoracic, thoracolumbar	10-15	F (100%)	68	Brace	20-40	3-6 mo	NS	Cobb increasing > 6o, or exceeding 45o	Comparative analysis of progress vs stable cases	Predictors: (1) Premenarche; (2) Curve > 30o; (3) L2-L4 BMD < 0.76 g/cm2; and (4) Thoracic curve	OR: 6.6-11.22 (0.001 > P < 0.072) (1) Sensitivity: 74.5%; (2) Specificity: 64.7%
Hung et al[65]	PCS	Art.	Thoracic, thoracolumbar, lumbar	11-16	F (100%)	324	Observation	20-30	0.5-3.5 yr	NS	Cobb increasing > 6o,	Multiple logistic regression modeling	Predictors: (1) Age at diagnosis < 13 yr; (2) Premenarche; (3) Risser sign: 0-1; (4) Curve pattern: thoracic or thoracolumbar; and (5) Initial Cobb angle > 30o; Osteopenia: decreased hip neck BMD at concave side	OR: 2.1-4.62 (0.001 > P < 0.044) (1) Sensitivity: 76% (95%Cl: 69-83); (2) Specificity: 70% (95%Cl: 62-77)
Lam et al[74]	PChS	Art.	NS	11-16	F (100%) Chinese population	294	Observation (192), Brace (102)	> 10; Mean: 26 (St. D, 8.2o)	Mean, 3.4 yr (St. D, 1.57o)	NS	Cobb increasing > 6o	Multiple logistic regression modeling	Predictors: (1)Age at diagnosis 11-13 yr, (2) Premenarche; (3) Initial Cobb angle > 25o; and (4) Ultrasound bone stiffness index (calcaneus) Z-score ≤ 0	OR: 2.0-8.62 (0.0001 > P < 0.2) (1) Sensitivity: 84.7%; (2) Specificity: 66.5%
Lee et al[68]	RCS	Art.	NS	10-17	F (82.3%) M (17.7%)	1858 450	Brace (331)	10-30	NS	NS	Cobb > 30o	Risk assessment	Predictor: Initial Cobb angle ≥ 26ovs 8o-10o	Hazard ratio, 8.82 (95%Cl: 6.85-11.31)
Tan et al[36]	PCS	Art.	NS	7-14	F (84.9%) M (15.1%)	186	Observation Brace	> 10	1-8 yr	18%	Cobb ≥ 30o	Risk assessment	Predictor: Initial Cobb angle ≥ 25ovs < 25o	OR = 24.62 (95%Cl: 9.9-60.6) P < 0.001 (1) Sensitivity: 68%3; (2) Specificity: 92%; (3) +PV: 68%; and (4) -PV: 92%
Modi et al[77]	RCS	Art.	Thoracic, thoracolumbar	10-15	F (84%) M (16%)	113	Brace	40-56	Until skeletal maturity (Risser sign ≥ 4); average: 34 ± 13 mo	NS	Cobb increasing ≥ 5o	Comparative analysis of progress vs stable cases	Predictor: Rib-vertebral angle at convex side of the curve apex after brace treatment (< 65ovs≥ 65o )	OR = 5.63 (95%Cl: 2.2-13.9) P < 0.001 (1) Sensitivity: 45%; (2) Specificity: 87%; (3) +PV: 69%; and (4) -PV: 71%
Qiu et al[69]	RCS	Art.	Thoracic, thoracolumbar	10-20	Chinese population	120	Brace	25-40	2.5 ± 0.35 yr	NS	Cobb increasing ≥ 5o	Comparative analysis of progress vs stable cases	Predictor: NTF3 gene: rs11063714, genotype GG vs AA	OR = 3.33 (95%Cl: 1.0-10.9) P = 0.08 (1) Sensitivity: 43%; (2) Specificity: 82%; (3) +PV: 56%; and (4) -PV: 72%
Xu et al[70]	RCS	Art.	Thoracic, thoracolumbar, lumbar	10-15	F (87%) M (13%)	312	Brace	20-40	0.6-2.2 yr	NS	Cobb increasing ≥ 5o and/or surgical correction	Logistic regression modeling	Predictors: (1) ERα gene: rs9340799, allele G; (2) TPH1 gene: rs10488682, allele A; (3) Risser sign O-1; and (4) Curve ≥ 30o	OR: 1.2-3.62 0.0001 > P < 0.1 (1) Sensitivity: 51%; (2) Specificity: 82%; and (3) Correct predictions: 75%
Yeung et al[75]	RCS	Art.	NS	12-16	F (100%) Chinese population	340	Observation	> 20	Until skeletal maturity, 16 years old or surgical intervention	NS	NS	Comparative analysis of Cobb angle in following genotypes of IGF1 SNP rs5742612: TT; TC; and CC	Predictor: TT (mean Cobb, 38 ± 12.1, n = 169) vs CC (mean Cobb, 33o± 9.0, n = 33), P = 0.01 Cut-point: Cobb, 35.7o	OR = 2.13 (95%Cl:1.0-4.4) P = 0.1 (1) Sensitivity: 88%; (2) Specificity: 22%; (3) +PV: 57%; and (4) -PV: 61%
1Ward et al[58]	RChS	Art.	Severe: 8% Moderate/ mild: 92%	9-13 at diagnosis	F (100%) F (100%) M (100%)	277 257 163	NS	> 10	Till skeletal maturity or severe deformity	NS	Severe: Cobb > 40o Moderate: Cobb 25o-40o	Multiple logistic regression modeling	Predictor: Scale (1-200 ) based on 53 SNP markers; cut point, 40: 1-40 ( ≤ 1% risk of progression)	OR=16.83 (95%Cl: 6.6-42.7) P < 0.001 (1) Sensitivity: 91%; (2) Specificity: 63%; (3) +PV: 17%; and (4) -PV: 99%
1Bohl et al[59]	RCS	Art	NS	≥ 10	F (81%) M (19%)	16	Brace	20-40	1 yr after brace discontinuation or skeletal maturity	36%	Cobb > 45o	Comparative analysis: patients with Cobb > 45ovs Cobb < 45o logistic regression modeling	Predictor: Scale (1-200 ) based on 53 SNP markers and initial Cobb angle: cut-point, 160: 160-200 (high risk of curve progression with Cobb > 45o) vs < 160 (low risk of curve progression with Cobb > 45o)	OR = 21.03 (95%Cl:1.5-293.3) P = 0.05 (1) Sensitivity: 78%; (2) Specificity: 86%; (3) +PV: 88%; and (4) -PV: 75%
Zhao et al[71]	RChS	Art	Double curves: thoracic, thoracolumbar or lumbar	10-20	Cases (AIS): F (90%) M (10%) Controls: F (75%) M (25%) Chinese population	67 100	Surgical correction	30-90	NS	NS	Cobb ≥ 30o	Comparative analysis of cases vs healthy controls	Predictors: (1) ER1 gene: rs2234693, allele T; (2) CALM 1 gene: rs12885713, allele T	OR: 1.7-1.83 0.01 > P < 0.05 (1) Sensitivity: 28%-69%; (2) Specificity: 44%-82%; (3) +PV: 45%-51%; and (4) -PV: 63%-68%
Zhou et al[72]	RCS	Art.	NS	11-18	F (100%) Chinese population	241	NS	20-100	Until skeletal maturity	54%	NS	Comparative analysis of severe cases (mean Cobb, 36o± 13o) vs moderate cases (mean Cobb, 29o± 7.4o)	Predictor: Il-17RC gene: rs708567, genotype GG, Cut-point: Cobb angle, 32.5o	OR = 3.43 (95%Cl: 1.4-8.3) P = 0.007 (1) Sensitivity: 94%; (2) Specificity: 17%; (3) +PV: 60%; and (4) -PV: 69%
Moreau et al[44]	RChS	Art.	Thoracic, thoracolumbar, lumbar	13-20	Cases (AIS): F (83%) M (17%) Controls: F (65%) M (35%)	41 17	Surgical correction	30-90	NS	NS	NS	Comparative analysis of AIS cases (mean Cobb, 54o± 14o) vs controls (non- idiopathic deformities)	Predictor: low inhibition of forskolin stimulated cAMP by melatonin in osteoblasts vs significant inhibition of forskolin stimulated cAMP by melatonin in osteoblasts	OR=3.93 (95%Cl: 0.45-33.7) P = 0.3 (1) Sensitivity: 20%; (2) Specificity: 94%; (3) +PV: 89%; and (4) -PV: 33%
Akoume et al[16]	PCS	Art	Asymptomatic subjects at-risk of AIS	5-15	F (65%) M (35%)	31	Observation	≤ 10	2 yr	NS	Cobb > 10o	Comparative analysis of cases with developed AIS spine deformity (mean Cobb, > 10o) vs subjects at risk, but without deformity	Predictor: peripheral blood mononuclear cells electrical impedance after melatonin or iodomelatonin administration: < 120 ohms vs≥ 120o homs	OR = 18.53 (95%Cl: 8.7-392.5) P = 0.03 (1) Sensitivity: 33%; (2) Specificity: 100%; (3) +PV: 100%; and (4) -PV: 70%
Akoume et al[61]	RChS	Art	NS	NS	NS	162 794	NS	NS	NS	NS	Cobb angle ≥ 45o Cobb angle 10o-44o	Comparative analysis of the G proteins functional status	Predictor: type of peripheral blood mononuclear cells G protein response to electrical stimulation: FG2 vs FG1 or FG3	OR = 2.63 (95%Cl: 1.9-3.7) P < 0.001 (1) Sensitivity: 26%; (2) Specificity: 88%; (3) +PV: 56%; and (4) -PV: 67%
Yamamoto et al[76]	RCS	Art.	NS	9-15	F (100%)	28	Analysis of curve history	5-59	05-2 yr	NS	Cobb increasing > 4o	Comparative analysis of progressive cases vs stable	Predictor: Brain stem function, abnormal vestibular-eye test vs normal	OR = 24.03 (95%Cl: 2.4-240.6) P = 0.007 (1) Sensitivity: 91%; (2) Specificity: 71%; (3) +PV: 67%; and (4) -PV: 92%

¹Study used industrial or other sponsorship;

²Predictive characteristics reported in the collected publication;

³Predictive characteristics calculated by authors of current review using extracted results. PChS: Prospective cohort study; RChS: Retrospective cohort study; PCS: Prospective case series; RCS: Retrospective case series; +PV: Positive predictive value; -PV: Negative predictive value; NS: Not specified; EMG: Electromyography; SNP: Single nucleotide polymorphism; NTF3: Neurotropin-3; IGF1: Insulin-like growth factor 1; ERα: Estrogen receptor alpha; Il-17RC: Interleukin 17 receptor C; CALM1: Calmodulin1.

Table 4 Risk of bias assessment

Ref.	Questions for evaluation														Score
Upadhyay et al[73]	Yes	Yes	Unsure	No	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	No	Yes	8
Peterson et al[62]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unsure	No	No	Unsure	Yes	Yes	Yes	10
Ajemba et al[60]	Yes	Yes	No	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	Yes	Yes	10
Cheung et al[64]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	No	Yes	10
Danielsson et al[63]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unsure	Yes	Yes	Yes	Yes	Yes	Yes	13
Kindsfater et al[46]	Yes	Yes	No	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	Yes	Yes	10
Lowe et al[47]	No	No	Yes	Yes	Yes	No	No	Unsure	Yes	Yes	Yes	Yes	No	No	7
Sun et al[67]	Yes	Yes	No	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	Yes	Yes	10
Sun et al[66]	Yes	Yes	No	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	Yes	Yes	10
Hung et al[65]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	Yes	No	10
Lam et al[74]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	Yes	Yes	11
Lee et al[68]	Yes	Yes	No	Yes	Yes	Yes	No	Unsure	No	No	Unsure	Yes	No	Yes	7
Tan et al[36]	Yes	Yes	Yes	Yes	Yes	Yes	No	Unsure	Yes	Yes	Yes	Yes	Yes	Yes	12
Modi et al[77]	Yes	No	No	Yes	Yes	Yes	Yes	Unsure	No	No	Unsure	Yes	No	No	6
Qiu et al[69]	Yes	Yes	No	No	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	No	No	7
Xu et al[70]	Yes	Yes	No	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	No	Yes	9
Ward et al[58]	Yes	No	No	Yes	Yes	Yes	Yes	Unsure	No	No	Unsure	No	Yes	Yes	7
Bohl et al[59]	Yes	Yes	No	Yes	Yes	No	Yes	Unsure	No	Yes	No	Yes	Yes	Yes	9
Zhao et al[71]	Yes	No	Unsure	Yes	Yes	Yes	Yes	Unsure	No	No	Unsure	Yes	No	Yes	7
Zhou et al[72]	Yes	Yes	No	Yes	No	Yes	Yes	Unsure	Yes	Yes	No	Yes	Yes	Yes	10
Moreau et al[44]	Yes	Yes	No	Yes	No	Yes	Yes	Unsure	No	No	Unsure	Yes	No	Yes	7
Akoume et al[16]	Yes	Yes	No	Yes	Yes	Yes	Yes	Unsure	Yes	No	Unsure	Yes	No	Yes	9
Akoume et al[61]	No	No	No	No	Yes	No	Unsure	Unsure	No	No	Unsure	Yes	No	No	2
Yamamoto et al[76]	Yes	No	No	Yes	Yes	No	No	Unsure	Yes	No	Unsure	Yes	Yes	No	6
Yeung et al[75]	Yes	No	No	Yes	No	Yes	Yes	Unsure	No	No	Unsure	Yes	Yes	Yes	7

Questions for evaluation: (1) Were inclusion/exclusion criteria clearly described? (2) Were confounding factors identified, and taken into consideration as selection criteria? (3) Was study prospective? (4) Were number/rate of male and female enrolled into the study reported? (5) Were criteria of curve progression clearly identified? (6) Was measurement of curve progression clearly described? (7) Were enrolled patients analyzed in the same treatment group (bracing, observation, physiotherapy) to which they were allocated? (8) Was statistician blind to the status of subjects enrolled into the study? (9) Was follow-up period clearly identified? (10) Was drop out rate reported? (11) Was drop out rate acceptable? (< 25%); (12) Are reports of the study free of suggestion of selective outcome reporting? (13) Was impact of gender taken into consideration during analysis? and (14) Was impact of confounding factors taken into consideration during analysis?

Table 5 Summary table of meta-analysis of association between studied characteristics and progressive adolescent idiopathic scoliosis

Studiedcharacteristics	Study	Participants	Heterogeneity		Summary statistics			P value	Level of evidence (GRADE)
	(n)	(n)	I2(%)	Level	Pooled odds ratio	95% confident limits
						Lower	Upper
Age (< 13 yr)	3	760	59	Moderate	2.7	1.8	4.6	0.001	Low
Osteopenia	3	686	51	Moderate	2.8	1.4	5.6	0.005	Low
Brain stem dysfunction	1	28	NA	NA	24.0	2.4	240.3	0.007	Very low
Multiple indices1	7	1057	35	Moderate	9.6	6.1	15.2	< 0.001	Low
Curve pattern	4	607	59	Moderate	2.3	1.2	4.6	0.017	Low
Curve progression during bracing	1	85	NA	NA	33.2	4.0	272.9	0.001	Very low
Initial Cobb angle	8	3719	90	High	7.6	4.2	13.6	< 0.001	Low
Melatonin signaling	2	89	0	Low	6.5	1.1	38.2	0.037	Low
Platelet calmodulin	2	72	39.9	Moderate	39.9	2.2	735.9	0.013	Low
Premenarche	6	980	64	High	4.0	2.0	7.9	< 0.001	Low
Rib-vertebral angle	1	113	NA	NA	5.6	2.2	13.9	< 0.001	Very low
Skeletal immaturity	4	891	50	Moderate	2.8	1.6	4.8	< 0.001	Low
SNP CALM1	1	67	NA	NA	1.7	1.0	2.9	0.036	Very low
SNP ER1	2	379	63	High	2.4	1.3	4.7	0.009	Low
SNP IGF1	1	340	NA	NA	2.1	0.9	4.5	0.054	Very low
SNP IL17RC	1	312	NA	NA	1.5	0.9	2.4	0.074	Very low
SNP NTF3	1	120	NA	NA	3.3	1.0	10.9	0.050	Very low
SNP TPH1	1	312	NA	NA	2.1	1.0	4.4	0.052	Very low
SNPs(53), ScoliScore test	2	713	0	Low	17.2	7.1	41.5	< 0.001	Low
Gi proteins functional status	1	956	NA	NA	2.6	1.9	3.7	< 0.001	Very low

¹Multiple indices included combinatorial radiographic, demographic, and physiologic characteristics. NA: Not available; SNP: Single nucleotide polymorphism; CALM1: Calmodulin 1; ER1: Estrogen receptor 1; IGF1: Insulin-like growth factor 1; IL17RC: Interleukin-17 receptor C; NTF3: Neurotrophin 3; TPH1: Tryptophan hydroxylase 1.