Efficacy of computer-aided design and computer-aided manufacturing-based braces for adolescent idiopathic scoliosis: a systematic evaluation and meta-analysis

Xiaogang Shen

1,2

Dailiang Zhang

Fei Yuan

Qian Peng

Chunjiang Lv

Hongzhe Qi

Mengyan Zhao

Gongzhen Chen

Hao Ye

Gan Gao

Tao Guo

4✉

Shen 1

1 People’s Hospital of Anyue County Ziyang China

2 Guizhou University of Traditional Chinese Medicine Guiyang China

3 The First Affiliated Hospital of Chengdu Medical College Chengdu China

4 Guizhou Provincial People’s Hospital Guiyang China

5 Orthopedics Hospital of Sichuan Province Chengdu China

6 Beijing Jishuitan Hospital, Guizhou Hospital Guiyang China

Xiaogang Shen,MM^1,2#, Dailiang Zhang,MM^3#, Fei Yuan,MM^4#, Qian Peng,MM⁵, Chunjiang Lv,MM⁶,Hongzhe Qi,MM², Mengyan Zhao,MM², Gongzhen Chen,MM², Hao Ye,MM², Gan Gao,MD⁴, Tao Guo,MD^4*

1. People's Hospital of Anyue County, Ziyang, China;

2.Guizhou University of Traditional Chinese Medicine, Guiyang, China;

3.The First Affiliated Hospital of Chengdu Medical College, Chengdu, China;

4.Guizhou Provincial People's Hospital, Guiyang, China;

5.Orthopedics Hospital of Sichuan Province, Chengdu, China;

6. Beijing Jishuitan Hospital, Guizhou Hospital,Guiyang,China;

#Xiaogang Shen, Dailiang Zhang, and Fei Yuan are co-first authors

*Corresponding author: Tao Guo, MD, Chief Physician, Professor, Master and doctoral supervisor, Department of Orthopedics of Guizhou Provincial People's Hospital, Direction: Basic and Clinical Research on Spinal Related Diseases.

Abstract

Background

Adolescent idiopathic scoliosis (AIS) is commonly treated conservatively, with computer-aided design and manufacturing (CAD/CAM) emerging as a modern approach. The superiority of CAD/CAM braces over traditional hand-made orthotics remains unestablished, prompting an investigation into their comparative efficacy.

Objective

To systematically review and meta-analyze the available evidence on the effectiveness of CAD/CAM braces for AIS, incorporating biomechanical analysis, finite element analysis, or 3D printing.

Methods

A systematic search was conducted in PubMed, Web of Science (WOS), Embase, MEDLINE, Cochrane Library, China National Knowledge Infrastructure (CNKI), WanFang, VIP, and China Biology Medicine disc (CBM) databases through April 1, 2024. The search targeted randomized controlled trials (RCTs) and randomized controlled crossover trials (RCT-CTs). Eligible studies were assessed for quality, and data were extracted for meta-analysis.

Results

The initial search identified 13 trials involving 539 participants (1629 individual data points). After applying inclusion criteria, 11 trials with 469 participants (1424 individual data points) were included in the meta-analysis. The results indicated that CAD/CAM braces provided only marginally better or equivalent outcomes compared to traditional braces. The therapeutic advantages of CAD/CAM methods were not conclusively demonstrated by the current evidence.

Conclusion

The evidence base for the superiority of CAD/CAM braces in AIS treatment is insufficient. Future research with larger sample sizes and higher methodological quality is essential to enhance the evidence-based approach to medical practice in this field.

Keywords:

Adolescent idiopathic scoliosis

computer-aided design

computer-aided manufacturing

orthoses/supports

meta-analysis

Introduction

Adolescent idiopathic scoliosis (AIS), a three-dimensional spinal deformity of unknown etiology, exhibits a global prevalence ranging from 0.93–12%(Negrini et al. 2018) and is more common in girls(Hengwei et al. 2016; Zheng et al. 2017). The primary therapeutic approaches, selected based on the degree of spinal curvature, are progressive and include observation, exercise therapy, the use of braces, and surgical intervention.(Negrini et al. 2018) Cobb's angle is a critical metric for evaluating the severity of spinal conditions such as scoliosis(Zhang et al. 2022). It quantifies the degree of spinal curvature and axial rotation, as well as the asymmetry of the trunk and the presence of kyphosis. This angle can be precisely determined through imaging modalities like spinal X-rays(Ghaneei et al. 2018; Speirs et al. 2024). Imaging techniques, particularly spinal X-rays, are instrumental in assessing the spine. Adolescent idiopathic scoliosis, if not managed, can lead to a progression of the spinal curve and associated back pain, as well as more severe implications such as cardiovascular issues and psychosocial challenges(Kan et al. 2023). In extreme cases, uncontrolled scoliosis may result in pulmonary complications, potentially culminating in fatal outcomes(Weiss et al. 2016).

Orthotic therapy is the only proven conservative treatment for scoliosis.(Negrini et al. 2015; Stokes & Luk 2013; Zhang & Li 2019) TOrthotic devices are available in a diverse array of types, and their design and fabrication techniques are in a state of continuous advancement. The predominant methods currently in use encompass both traditional craftsmanship and cutting-edge technologies such as computer-aided design (CAD) and computer-aided manufacturing (CAM). These may be augmented with biomechanical analysis or finite element analysis to ensure precision and efficacy. Additionally, the incorporation of 3D printing has revolutionized the field, offering customized and intricate solutions tailored to individual patient needs.(Karavidas 2019; Weinstein et al. 2013; Weiss et al. 2017) Moreover, the integration of modern CAD/CAM technology with biomechanical and finite element analyses is reaching a level of sophistication, enhancing the precision and functionality of orthotic devices. These advanced methods allow for the creation of CAD/CAM-supported orthoses that can be meticulously crafted through Computer Numerical Control (CNC) machining or additive manufacturing techniques like 3D printing. Despite the technological advancements, the debate on the comparative effectiveness of CAD/CAM-produced braces continues among researchers, with numerous studies examining their clinical outcomes and performance.(Karavidas 2019; Zheng et al. 2023) Ultimately, the design and production process of such advanced orthoses demands substantial investment in acquiring sophisticated software and hardware. Additionally, it necessitates the expertise of professionals with backgrounds in engineering or computer science. Consequently, bothmedical institution and patients are compelled to weigh the financial implications, carefully evaluating the cost-benefit ratio to ensure a justifiable return on their investment.

The objective of this research is to conduct a comparative analysis of the effectiveness between CAD/CAM orthoses and those crafted through traditional manual methods. The study will also examine the benefits of integrating CAD/CAM with biomechanical or finite element analysis in the context of traditional manual braces. Furthermore, it aims to assess the performance of braces that utilize CAD/CAM in conjunction with biomechanical or finite element analysis against those produced solely by CAD/CAM. Additionally, the study will explore the advantages of 3D printing over traditional manual fabrication. The ultimate goal is to contribute to the evidence base for evidence-based medicine, offering healthcare providers and patients a more informed basis for decision-making when selecting orthotic devices. This research seeks to provide a comprehensive set of data to guide the choice of braces, ensuring that selections are aligned with the best available scientific and clinical evidence.

Materials & methods

Study design and registration

This was a systematic evaluation and meta-analysis following the PRISMA guidelines(Page et al. 2021) and was registered with the PROSPERO system (CRD42024535347).

Search methodology

A thorough literature search was conducted across multiple databases, including PubMed, EMBASE, MEDLINE, Web of Science (WOS), and the Cochrane Library, as well as Chinese databases such as China Biomedical Database (CBM), China National Knowledge Infrastructure (CNKI), Wanfang Data, and VIP Medical Information System, up to April 1, 2024. The search strategy involved a combination of titles, keywords, subject headings, and free-text terms, tailored to the specific indexing and retrieval mechanisms of each database (as detailed in Table 1). Furthermore, supplementary sources were identified through manual searches, and the search scope was not restricted by language, ensuring a comprehensive and inclusive review of the available literature.

Inclusion criteria

The inclusion criteria for this study were established in accordance with the PICOS framework(Amir-Behghadami & Janati 2020) :

P (Population)

Individuals with a confirmed diagnosis of Adolescent Idiopathic Scoliosis (AIS) who were treated with a brace.

I (Intervention)

The experimental group received braces fabricated using CAD/CAM technology, which may encompass biomechanical analysis, finite element analysis, or 3D printing techniques.

C (Comparison)

The control group was treated with either conventional handmade braces or braces produced solely by CAD/CAM methods without additional advanced analyses.

O (Outcome)

The primary outcomes of interest included the Cobb angle measurements for lordosis or thoracic curvature, thoracolumbar curvature, kyphosis and lordosis, as well as the parietal rotation angle.

S (Study Design)

Eligible studies were limited to clinical randomized controlled trials or randomized crossover controlled trials.

This structured approach ensured that the study focused on relevant comparisons and outcomes, contributing to a rigorous evaluation of the efficacy of different brace technologies in the treatment of AIS.

Exclusion criteria

The exclusion criteria for this study were delineated as follows:

Duplicate Publications

Articles or studies that have been published more than once.

Non-Original Research

Clinical case reports, animal experiments, review articles, correspondence, laboratory studies, meta-analyses, dissertations, or conference abstracts.

Vague Criteria

Studies with ambiguous diagnostic or efficacy criteria that could compromise the reliability of the findings.

Mixed Populations

Studies involving heterogeneous populations, such as those that include other musculoskeletal disorders alongside scoliosis, leading to potential confounding of the results.

Data Integrity

Literature with incomplete or erroneous data, where the necessary statistical parameters, specifically the mean and standard deviation, could not be retrieved or calculated.

These criteria were applied to ensure that the study encompassed only the most relevant and methodologically sound research, thereby upholding the integrity and validity of the analysis.

Research screening

The literature search was conducted autonomously by three investigators—Zhang Dailiang, Yuan Fei, and Zhao Mengyan—with the collected articles subsequently compiled into EndNote databases to facilitate the elimination of duplicate entries. Articles that failed to align with the predefined inclusion and exclusion criteria were initially culled based on a review of their titles and abstracts. The full texts of the articles that passed the initial screening were then scrutinized in detail to determine their suitability for inclusion in the study. Discrepancies among the three researchers were addressed through collaborative deliberation. In instances where a consensus could not be achieved, a fourth reviewer, Tao Guo, was engaged to arbitrate and render a final verdict. This rigorous and transparent process ensured that the study's selection criteria were applied consistently and objectively.

Research assessment

Two researchers, Hongzhe Qi and Qian Peng, independently utilized the Cochrane Risk of Bias tool (Sterne et al. 2019) to evaluate the methodological rigor and potential biases within the selected studies. This comprehensive instrument assesses seven critical domains: sequence generation (randomization method), allocation concealment, blinding of participants and personnel, blinding of outcome assessment, completeness of outcome data, reporting bias, and other relevant aspects. The risk of bias for each study was classified into one of three categories: 'low risk', 'unclear risk', or 'high risk'. The researchers then cross-verified their assessments and engaged in discussions to resolve any inconsistencies, ensuring a thorough and accurate evaluation of the included studies' quality.

Data extraction

Three researchers (Shen Xiaogang, Chen Gongzhen, and Lv Chunjiang) independently extracted data from the included studies. The extracted data included authors, year of publication, type of study, number of subjects, age, Risser's sign, intervention, control, pre- and post-intervention Cobb's angle, duration of follow-up, and other outcome indicators. One of the studies(Cottalorda et al. 2005) gave Cobb angles for all outcomes but no mean ± standard deviation, we calculated the required mean ± standard deviation using SPSS 26.0. One study(Cobetto et al. 2017) reported improvement rates and between-group P values for thoracic (T) and lumbar (L) Cobb, and the outcome mean was obtained directly from the baseline mean - degrees of decline or baseline mean * (1 - improvement rate).SD was obtained from the estimation of the P value according to the method suggested in the Cochrane Handbook, section 6.5.2.3, by first inputting the value "= tinv(P, P, P, P)" into any cell of the Microsoft Excel software. "= tinv(P, ν)" and press enter to estimate the t-value, v is the total degree of freedom, obtained from n1 + n2 -2, and then according to the formula

$\:\text{S}\text{E}=\frac{{\text{M}\text{D}}_{\text{E},\:\text{t}\text{w}\text{o}\:\text{g}\text{r}\text{o}\text{u}\text{p}\text{s}}}{\text{t}}$

The standard error is obtained, and then Eq.

$\:\text{S}\text{D}=\frac{\text{S}\text{E}}{\sqrt{\frac{1}{{\text{n}}_{1}}+\frac{1}{{\text{n}}_{2}}}}$

The standard deviation was calculated. The Kyphosis and Lordosis of this study reported differences and between-group p-values but not the standard deviation of the outcome, and the same p-value calculation was used to obtain the SD. A study(Cobetto et al. 2016) of T and L Cobb Corner reported the difference between baseline and outcome, and according to the comments in the Cochrane Handbook section 6.5.2.8, the formula was used to

$\:{\text{C}\text{o}\text{r}\text{r}}_{\text{E}}=\frac{{\text{S}\text{D}}_{\text{E},\text{b}\text{a}\text{s}\text{e}\text{l}\text{i}\text{n}\text{e}}^{2}+{\text{S}\text{D}}_{\text{E},\text{f}\text{i}\text{n}\text{a}\text{l}}^{2}-{\text{S}\text{D}}_{\text{E},\text{c}\text{ℎ}\text{a}\text{n}\text{g}\text{e}}^{2}}{2\times\:{\text{S}\text{D}}_{\text{E},\text{b}\text{a}\text{s}\text{e}\text{l}\text{i}\text{n}\text{e}}^{}\times\:{\text{S}\text{D}}_{\text{E},\text{f}\text{i}\text{n}\text{a}\text{l}}^{}}$

The correlation coefficient was estimated, assuming that the intervention did not affect the variability of the outcome metrics, so that the SD of the outcome was replaced by the SD of the baseline, and the resulting CorrE values for both groups were ≈ 0.8, and then, according to the pre-and post-intervention differences formula

$\:{\text{S}\text{D}}_{\text{E},\text{c}\text{ℎ}\text{a}\text{n}\text{g}}=\sqrt{{\text{S}\text{D}}_{\text{E},\text{b}\text{a}\text{s}\text{e}\text{l}\text{i}\text{n}\text{e}}^{2}+{\text{S}\text{D}}_{\text{E},\text{f}\text{i}\text{n}\text{a}\text{l}}^{2}-\left(2\times\:\text{C}\text{o}\text{r}\text{r}\times\:{\text{S}\text{D}}_{\text{E},\text{b}\text{a}\text{s}\text{e}\text{l}\text{i}\text{n}\text{e}}\times\:{\text{S}\text{D}}_{\text{E},\text{f}\text{i}\text{n}\text{a}\text{l}}\right)}$

which was transformed to obtain the outcome standard deviation formula

$\:{\text{S}\text{D}}_{\text{E},\text{f}\text{i}\text{n}\text{a}\text{l}}=\frac{2\times\:\text{C}\text{o}\text{r}\text{r}\times\:{\text{S}\text{D}}_{\text{E},\text{b}\text{a}\text{s}\text{e}\text{l}\text{i}\text{n}\text{e}}\pm\:\sqrt{{(-2\times\:\text{C}\text{o}\text{r}\text{r}\times\:{\text{S}\text{D}}_{\text{E},\text{b}\text{a}\text{s}\text{e}\text{l}\text{i}\text{n}\text{e}})}^{2}-4\times\:({\text{S}\text{D}}_{\text{E},\text{b}\text{a}\text{s}\text{e}\text{l}\text{i}\text{n}\text{e}}^{2}-{\text{S}\text{D}}_{\text{E},\text{c}\text{ℎ}\text{a}\text{n}\text{g}\text{e}}^{2})}}{2}$

. Alternatively, the Cochrane Handbook section 6.5.2.7 suggests that the method of standard deviation can be borrowed directly from similar studies, and one of the studies we included(Cobetto et al. 2014) used this method, borrowing from a similar study by the same team(Desbiens-Blais et al. 2012). The three authors engaged in collaborative discussions and performed calculations to reconcile their differences. Subsequently, they consulted with Guo Tao to finalize the decision-making process. In parallel, efforts were made to contact the original corresponding author of the studies in quest of more precise data. However, these attempts were unsuccessful as no response was received. Despite this, the authors remained committed to ensuring the accuracy and integrity of the data included in their analysis.

Comprehensive Quantitative Analysis (Meta-Analysis)

The primary outcome metrics of this study were primary curvature Cobb angle, thoracic curvature Cobb angle, lumbar curvature Cobb angle, kyphosis, kyphosis, and parietal rotation in the experimental and control groups included in the study. Using Review Manager 5.4.1, a meta-analysis of outcome indicators after the included study interventions was performed. Heterogeneity tests were performed using the I² statistic(Wang et al. 2024) I² greater than 50% indicated significant heterogeneity and data were analyzed using a random effects model. Otherwise, a fixed effects model was used. Mean difference (MD) and 95% confidence interval (95% CI) were also calculated.

Results

Results of the literature search

As depicted in Fig. 1, the initial systematic search yielded a total of 432 potential studies. Of these, 419 were subsequently excluded for various reasons, including 175 duplicates, 222 that were conference papers, laboratory or theoretical investigations, dissertations, case reports, and so on. Following this, an additional 22 studies were deemed ineligible after a thorough examination of the full text. This rigorous screening process resulted in 13 studies that met the inclusion criteria for review.

Among these 13 studies(Chunxin et al. 2019; Cobetto et al. 2014; Cobetto et al. 2017; Cobetto et al. 2016; Cottalorda et al. 2005; Desbiens-Blais et al. 2012; Guy et al. 2021; Hongsheng et al. 2023; Labelle et al. 2007; Lin et al. 2022; Raux et al. 2014; Weiwei et al. 2022; Wong et al. 2005), one study(Raux et al. 2014) could not provide the necessary mean data, which was a requirement for our analysis. Furthermore, it was noted that two of the studies(Cobetto et al. 2017; Cobetto et al. 2016) were actually the same study reported at different points in time, with one having a larger sample size. To ensure the robustness of our meta-analysis, we opted for the version with the more extensive data(Cobetto et al. 2017).

Consequently, a total of 11 studies were ultimately incorporated into the quantitative meta-analysis, ensuring a high standard of quality and relevance in our review.

Study Characteristics

As presented in Table 2, the 13 studies that were included in our analysis comprised a mix of randomized controlled trials and randomized controlled crossover trials, with the majority being randomized controlled trials, totaling 9 studies. This selection ensures a robust evidential base for the meta-analysis, adhering to the gold standard of clinical research methodology.(Chunxin et al. 2019; Cobetto et al. 2017; Cobetto et al. 2016; Guy et al. 2021; Hongsheng et al. 2023; Labelle et al. 2007; Lin et al. 2022; Weiwei et al. 2022; Wong et al. 2005) and 4 studies were randomized controlled crossover trials(Cobetto et al. 2014; Cottalorda et al. 2005; Desbiens-Blais et al. 2012; Raux et al. 2014). A total of 1424 participants were enrolled across the studies under review, with 120 individuals specifically allocated to the comparison between CAD/CAM braces and conventional braces. This participant pool provides a substantial sample size for assessing the efficacy of the different bracing interventions(Chunxin et al. 2019; Cottalorda et al. 2005; Raux et al. 2014; Wong et al. 2005). In the comparative analysis of the CAD/CAM versus conventional brace group, the primary outcome measure focused on the primary flexion Cobb angle. As for the group examining CAD/CAM combined with biomechanical analysis or finite element analysis against the conventional brace group, a total of 410 comparative analyses were conducted. This comprehensive evaluation ensures a thorough examination of the impact of advanced bracing technologies in comparison to traditional methods(Cobetto et al. 2014; Desbiens-Blais et al. 2012; Labelle et al. 2007; Weiwei et al. 2022). A total of 410 individual assessments were conducted to compare thoracic kyphosis, thoracolumbar kyphosis, and overall kyphosis measures between groups. Specifically, these evaluations contrasted the outcomes of CAD/CAM braces enhanced with biomechanical analysis or finite element analysis against those of braces fabricated solely by CAD/CAM methods. This meticulous approach allowed for a detailed examination of the effects of incorporating advanced analytical techniques into the brace design process(Cobetto et al. 2017; Cobetto et al. 2016; Guy et al. 2021). Involved 922 visits and studied thoracic kyphosis, thoracolumbar kyphosis, kyphosis and kyphosis anterior; 3D printing versus traditional bracing group(Hongsheng et al. 2023; Lin et al. 2022) involved 72 visits and studied only the main curvature Cobb angle. While the primary focus of the studies was on key outcome measures such as Cobb angles and kyphosis assessments, several additional parameters were also reported, offering a more holistic view of the impact of bracing interventions. These included spinal length, height, sitting height, plumb line offset distance, quality of life metrics like the SF-36, spinal deformity-specific measures such as the C7 vertebral body center to mid-sacral plumb line (C7-CSVL), apical vertebral deviation (AVT), comfort levels, SRS-22, and the Trunk Appearance Perception Scale (TAPS). However, due to insufficient data and a lack of uniformity across studies, these variables could not be quantitatively synthesized in a meta-analysis, limiting our ability to draw systematic conclusions from them.

Table 2

, Study characteristics
NO.	Studies (Year)	Country (region)	Type of STUDY	Sample size (male: Female)	Age	Risser	Cobb (°)	Experimental Group			Control group			Follow-up	Main outcome measures
NO.	Studies (Year)	Country (region)	Type of STUDY	Sample size (male: Female)	Age	Risser	Cobb (°)	Sample size	Age	Interventions	Sample size	Age	Interventions	Follow-up	Main outcome measures
1	Xu CX, et al,2019^[25]	China	RCT	20 (5:15)	10–16	NA	25–40	10 (3:7)	12.8 ± 2.15	Rodin 4D + CNC milling machine engraving	10 (2:8)	13 ± 1.87	Ordinary plaster bracing	2 years	Cobb Angle of the main curve and length of the spine
2	Zhao WW, et al, 2022^[26]	China	RCT	62	10–15	0–3	25–50	31 (2:29)	12.52 ± 1.56	CAD/CAM + CNC milling machine engraving	30 (5:26, 1 case detached)	12.87 ± 1.48	Ordinary plaster bracing	1 year and 2 years	Height, sitting height, vertical deviation, SF-36, Cobb, C7-CSVL, AVT
3	Zhang HS, et al, 2023^[27]	China	RCT	50	10–17	NA	< 40	25 (6:19)	14.07 ± 2.09	CAD + 3D printing	25 (7:18)	14.01 ± 2.14	Ordinary plaster bracing	1 year	Cobb Angle and effective rate before and at 3, 6, 9, and 12 months
4	Cottalorda J, et al, 2005^[19]	France	RCOD	30 (4:26)	11–16	NA	15–30	30	NA	CAD/CAM + CNC milling machine engraving	30	NA	Ordinary plaster bracing	instant	Cobb Angle and comfort
5	Wong MS, et al, 2005^[31]	Hong Kong, China	RCT	40 (F)	10–14	0–2	25–45	20	12.7 ± 1.0	CAD/CAM + CNC milling machine engraving	20	12.5 ± 1.0	Ordinary plaster bracing	instant	Cobb Augle, Apical vertebral rotation
6	Labelle H, et al, 2007^[32]	Canada	RCT	48 (2:46)	10–16	0–3	20–45	24 (1:23)	12.9 ± 1.4	CAD/CAM	24 (1:23)	12.9 ± 1.4	Ordinary plaster bracing	instant	Thoracic Cobb, Lumbar Cobb, Kyphosis, Lordosis
7	Blais FD, et al 2012^[23]	Canada	RCOD	6 (F)	11–13	0–3	Chest: 29 Waist: 24	6	11–13	CAD/CAM	6	11–13	Ordinary plaster bracing	instant	Thoracic Cobb, Lumbar Cobb, Kyphosis, Lordosis
8	Cobetto N,et al,2014^[22]	Canada	RCOD	15 (F)	11–14	0–1	20–45	15	11–14	CAD/CAM + FEM	15	11–14	Ordinary plaster bracing	instant	Thoracic Cobb, Lumbar Cobb, Kyphosis, Lordosis, Comfort
9	Raux S,et al,2014^[28]	France	RCOD	30 (4:26)	13.3	NA	25	30	13.1	CAD/CAM + CNC milling machine engraving	30	13.5	Ordinary plaster bracing	instant	Initial Angle, Thoracic Cobb, Lumbar Cobb, Comfort
10	Cobetto N,et al,2016^[21]	Canada	RCT	40	10–16	0–2	20–45	21	13.2 ± 1.4	CAD/CAM + FEM	19	13.0 ± 1.3	CAD/CAM	instant	Initial Angle, Thoracic Cobb, Lumbar Cobb Kyphosis, Lordosis
11	Cobetto N,et al,2017^[20]	Canada	RCT	48	NA	0–2	20–40	25	NA	CAD/CAM + FEM	23	NA	CAD/CAM	instant	Thoracic Cobb, Lumbar Cobb, Kyphosis, Lordosis, PMC
12	Guy A, et al, 2021^[30]	Canada	RCT	120	≥ 10	0–2	20–45	59	NA	CAD/CAM + FEM	61	NA	CAD/CAM	2 years	Initial Angle, 2-year Thoracic, Lumbar, Kyphosis, Lordosis, Apical vertebral rotation, SRS-22, and compliance
13	Lin YM,et al,2022^[29]	Hong Kong, China	RCT	30	10–14	0–2	20–40	15	12.4 ± 0.8	CAD/CAM + 3Dprinting	15	12.0 ± 1.0	(4-Axis Carving Machine	2 years	Initial Angle and Cobb, SRS-22, TAPS at 3 and 24 months
	Total			539				311			308
CAD/CAM: computer-aided design and computer-aided manufacturing, FEM: finite element analysis, 3D: 3D-printed manufactured support, C7-CSVL: distance from the center of the C7 vertebral body to the mid-sacral plumb line, AVT: apex vertebral deviation, Apical vertebral rotation: apex vertebral rotation, RCOD: Randomised Cross-over Control Test. Thoracic Cobb: thoracic curvature, Lumbar Cobb: lumbar curvature, Kyphosis: kyphosis, Lordosis: lordosis, PMC: planes of maximum curvature, SRS-22: Scoliosis Research Society 22-item questionnaire, TAPS: Trunk Appearance Perception Scale, NA: not availa

Across all studies, the demographic profile of the patients was notably consistent, with ages predominantly ranging from 10 to 17 years. Risser's sign, an indicator of skeletal maturity, was minimally at grade 0 and maximally at grade 3. The initial Cobb angle measurements were generally clustered between 20–45°, with one study also including angles of 15–20°. The majority of participants were female, reflecting the typical prevalence of idiopathic scoliosis.

The inclusion criteria in these studies were presumably aligned with the guidelines recommended by the Scoliosis Research Society (SRS) and the Society on Scoliosis Orthopedic and Rehabilitation Treatment (SOSORT), ensuring a standardized approach to patient selection and treatment.

Research risk of bias assessment

The randomization sequence methodology was clearly defined in 12 of the studies, as illustrated in Fig. 2A. The majority of the randomized controlled trials had an unclear or high risk regarding allocation concealment and blinding practices. Although blinding for outcome assessment was not explicitly stated in all trials, our experience suggests that it would be challenging to maintain complete objectivity among imaging assessment staff due to potential outcome assessment bias. However, the completeness of the outcome data was generally good, with only three trials reporting occasional data dislodgement and one study mentioning adherence assessment, which was not reflected in the reported outcomes. It is important to note that the studies focused solely on the immediate Cobb angle correction achieved with bracing, thus we did not identify any studies as being at risk for selective reporting. Figures 2A and 2B present an overview of the methodological quality assessment of the included literature and the corresponding percentage breakdown of the methodological quality ratings.

GRADE system evaluation results

We used the Cochrane Collaboration Network GRADEprofiler 3.6.1 system(Alonso-Coello et al. 2016) software to assess the quality of the included evidence. In our evaluation of four subgroups across a total of 15 metrics, it was observed that the evidence quality for the CAD/CAM combined with Finite Element Method (FEM) versus conventional brace control was low for all parameters, including Thoracic (T), Thoracolumbar/Lumbar (TL/L), Kyphosis, and Lordosis. Within the subgroup comparing CAD/CAM enhanced with FEM to CAD/CAM-only controls, while the evidence quality for the T Cobb, Lordosis, Apical rotation-MT, and Apical rotation-TL/L was moderate, the evidence quality for TL/L and Kyphosis remained low. Similarly, the evidence quality for CAD/CAM braces versus conventional braces was low, and for 3D printed braces versus conventional braces, it was moderate. The downgrade in evidence quality could be attributed to several factors: (1) uncertainties in the methodological rigor of included studies, particularly regarding allocation concealment and blinding; (2) broader confidence intervals or instances where they crossed the line of equivalence; (3) higher heterogeneity among the studies; and (4) small sample sizes, which can limit statistical power. Figure 3 provides a visual representation of these assessments, illustrating the variability in evidence quality across different metrics and study groups.

meta-analysis results

Since three of the included trials(Cobetto et al. 2014; Cobetto et al. 2017; Cobetto et al. 2016) did not directly report standard deviations, we finally obtained them according to the method described in 2.7. Since one study(Raux et al. 2014) did not have a mean was excluded, and another(Cobetto et al. 2016) study was an earlier publication of the same trial by the same research team, we chose the latest published report of one(Cobetto et al. 2017) which was the most recently published report. Eleven studies were finally included in the meta-analysis, and the Cochrane Handbook for the Systematic Evaluation of Interventions(Higgins JPT 2023) suggests that a meta-analysis can be a statistical combination of the results of two or more independent studies, so we divided four subgroups according to the different modes of control, namely: the CAD/CAM with plain support group(Chunxin et al. 2019; Cottalorda et al. 2005; Wong et al. 2005) CAD/CAM + FEM with plain support group(Cobetto et al. 2014; Desbiens-Blais et al. 2012; Labelle et al. 2007; Weiwei et al. 2022) CAD/CAM + FEM vs. CAD/CAM alone group(Cobetto et al. 2017; Guy et al. 2021) CAD/CAM + 3D printing with normal support group.(Hongsheng et al. 2023; Lin et al. 2022). Based on the different comparisons of specific flexion types, the two subgroups of CAD/CAM + FEM with the plain brace and CAD/CAM + FEM with CAD/CAM alone were further subdivided into subgroups: T Cobb, TL/L Cobb, Kyphosis, and Lordosis, respectively, and CAD/CAM + FEM with CAD/CAM alone was further subdivided into Apical rotation-MT and Apical rotation-TL/L subgroups. The specific meta-analysis of each group is as follows.

A total of three trials(Chunxin et al. 2019; Cottalorda et al. 2005; Wong et al. 2005) (120 in total) comparing the effectiveness of CAD/ CAM-made braces with hand-made braces [30,31] (Fig. 2). The level of heterogeneity I² = 42% and the MD (fixed effects model) was − 0.66 (95% CI:-2.79-1.46. The combined difference value P = 0.54. (Fig. 4)

A total of four trials(Cobetto et al. 2014; Desbiens-Blais et al. 2012; Labelle et al. 2007; Weiwei et al. 2022) (410 in total) compared the effectiveness of CAD/ CAM + FEM fabricated braces versus conventional braces on the Cobb angle of the primary thoracic curvature (Fig. 5), with a level of heterogeneity of I2 = 19% and an MD (fixed-effects model) of -0.76 (95% CI:-3.71-2.19), and a difference in the combined difference of P = 0.61. In terms of TL/L, Kyphosis, and Lordosis only three trials were compared(Cobetto et al. 2014; Desbiens-Blais et al. 2012; Labelle et al. 2007) (Fig. 5), with a heterogeneity of 0% for TL/L Cobb, an MD (fixed-effects model) of -3.18. (95% CI: -6.81-0.46), and a combined difference of P = 0.09. Heterogeneity was 0% for Kyphosis Cobb, an MD (fixed-effects model) of -1.16 (95% CI: -4.95-2.19), and a difference of P = 0.09 after combination. -4.95-2.63), post-merger difference P = 0.55. Heterogeneity for Lordosis Cobb was 4%, with an MD (fixed effects model) of -0.76 (95% CI: -3.50-1.99), post-merger difference P = 0.59. Total heterogeneity for the merger was 0% with an MD (fixed effects model) was − 1.3 (95% CI:-2.89-0.3), P = 0.64 for heterogeneity between subgroups, and P = 0.73 for test difference.

A total of 2 trials(Cobetto et al. 2017; Guy et al. 2021) (553 in total) compared the effectiveness of CAD/ CAM + FEM fabricated braces with CAD/CAM fabricated braces alone (Fig. 6), where the level of heterogeneity for T Cobb I² = 78%, MD (random effects model) was − 1.43 (95% CI:-7.54-4.67), and the difference between the combinations was P = 0.65. Heterogeneity for TL/L Cobb was 82% with an MD (random effects model) of -0.60 (95% CI: -7.39-6.20) and a combined difference of P = 0.86. Heterogeneity for Kyphosis Cobb was 90% with an MD (random effects model) of -6.06 (95% CI. -20.74-8.61), post-merger difference P = 0.42. Heterogeneity in Lordosis Cobb was 86%, with an MD (random effects model) of 2.24 (95% CI: -4.60-9.09), post-merger difference P = 0.52. The total heterogeneity of the groups combined was 77%, with an MD (random effects model) was − 0.86 (95% CI:-3.87-2.14), with P < 0.001 for heterogeneity between subgroups and P = 0.57 for the difference after merging. For parietal rotation (Fig. 7), there were a total of 269 participants in the comparisons, with a level of heterogeneity I² = 0% for the Apical rotation-MT Cobb, and a level of P = 0.0001 for the MD(random effects model) -0.04 (95% CI:-0.06-0.02), with a difference of P = 0.001 after merging. The heterogeneity of Apical rotation-TL/L Cobb was 71%, with an MD (random effects model) of -0.49 (95% CI:-2.82-1.84), P = 0.68 for the combined difference. combined total heterogeneity was 71% with an MD (random effects model) of -0.37 (95% CI:-1.37 to 0.64), P = 0.48 for between-subgroup heterogeneity, and P = 0.70 for the combined difference.

A total of 2 trials(Hongsheng et al. 2023; Lin et al. 2022) (66 in total) compared the effectiveness of 3D printed braces with conventional braces (Fig. 8). The level of heterogeneity I² = 72%, MD (random effects model) was − 0.58 (95% CI:-6.41 to 5.26), and the combined post-test difference P = 0.85.

Sensitivity and heterogeneity analyses (see supplementary material for details)

We reviewed all steps and analyzed subgroups with large heterogeneity (I²>75%) using both the fixed and random effects model(Wang et al. 2024). As well as changing the effect indicator to a standardized mean difference to ensure the robustness of the results and sensitivity to outliers. For providing both p-value and difference endings while using p-value estimation to obtain standard deviation(Cobetto et al. 2017), at the beginning we used P-value estimation, and to verify sensitivity, we also estimated the standard deviation by difference, which did not significantly change the final endings of the meta-analysis, even though different standard deviations were obtained(Cobetto et al. 2016). Thus, we can have greater confidence in the robustness of our findings. However, when the model and outcome measures were altered, divergent results emerged between the 3D printing and traditional handmade brace groups. Our analysis suggests that despite both trials taking place in China, there were several factors that may have contributed to the variability: the smaller sample sizes, the greater divergence of Lin's trial from the norm, and the differing follow-up durations—two years for Lin's study and one year for Zhang's. Additionally, the control braces in each study were produced using different methods—Lin's controls were crafted by a carver, while Zhang's were handmade. These discrepancies could have affected the robustness of the meta-analysis outcomes. Consequently, there is a need for further high-quality, homogeneous studies to bolster the reliability and robustness of the results.

Publication Bias Analysis

Given the limited number of trials per group(Higgins JPT 2023), which is fewer than ten participants, we refrained from creating funnel plots. However, to evaluate the potential bias, we delved into the original research studies and inferred that the modest sample sizes, absence of blinding, and other factors could have skewed our findings. These considerations are detailed in the assessments found in sections 3.3 and 3.4.

Discussion

Qualitative systematic evaluation

This systematic review and meta-analysis conducted a comparative analysis of the effectiveness between CAD/CAM-fabricated braces and traditional braces, as well as the combination of CAD/CAM with FEM (Finite Element Method) and 3D printing technologies against conventional braces for the treatment of Adolescent Idiopathic Scoliosis (AIS). The study incorporated high-quality randomized controlled trials and randomized controlled crossover trials that adhered to the standards set by the Scoliosis Research Society (SRS) or the Society on Scoliosis Orthopedic and Rehabilitation Treatment (SOSORT)(Negrini et al. 2018). The methodological rigor was generally high, yet the evidence quality was categorized as moderate to low. The majority of the trials indicated that braces manufactured using CAD/CAM technology yielded outcomes that were comparable or slightly superior to those of traditional braces. Nonetheless, the constraints imposed by the small sample sizes and the evidence quality may restrict the broader applicability of the study's conclusions.

Research into the use of CAD/CAM technology, either in isolation or in conjunction with biomechanics, finite element analysis, and 3D printing for brace fabrication, has primarily engaged software systems such as Rodin 4D, CATIA5R21, ORTEN, OrtenfileR, Biosculptor, Canfit, and similar tools. Despite this, no efficacy studies have been conducted to specifically compare the design and manufacturing outcomes of braces using various software systems in a controlled manner. Notably, Rodin 4D software has emerged as the most prevalent choice, being utilized in eight of the studies we have included in our analysis(Chunxin et al. 2019; Cobetto et al. 2014; Cobetto et al. 2017; Cobetto et al. 2016; Desbiens-Blais et al. 2012; Guy et al. 2021; Hongsheng et al. 2023; Weiwei et al. 2022). Furthermore, participant-specific group factors, such as curve type and spinal flexibility, which could potentially impact the efficacy of the brace, have been taken into account in our analysis(Strube et al. 2021). Currently, a limited number of studies, specifically four (Cottalorda et al. 2005; Guy et al. 2021; Lin et al. 2022; Raux et al. 2014), have explored patient-specific factors such as curve type and spinal flexibility, which may introduce subjectivity into the evaluation of brace effectiveness across different manufacturing techniques. Moreover, in randomized controlled crossover trials, it is crucial to minimize confounding variables, including the sequence of different manufacturing methods and the duration between brace usage periods.

Our review identified seven studies that solely evaluated the braces' effectiveness based on the angle of correction achieved within the brace(Cobetto et al. 2014; Cobetto et al. 2017; Cobetto et al. 2016; Cottalorda et al. 2005; Desbiens-Blais et al. 2012; Labelle et al. 2007; Raux et al. 2014; Wong et al. 2005), while five trials examined the long-term efficacy of the braces(Chunxin et al. 2019; Guy et al. 2021; Hongsheng et al. 2023; Lin et al. 2022; Weiwei et al. 2022). While the in-brace correction angle is a significant metric for assessing brace quality, the ultimate aim is to enhance long-term quality of life through sustained treatment. Patient adherence and quality of life are intrinsically linked to long-term treatment success, and thus, these factors should be assessed. This comprehensive evaluation approach is also endorsed by the Scoliosis Research Society (SRS) and the Society on Scoliosis Orthopedic and Rehabilitation Treatment (SOSORT)(Negrini et al. 2018). Regrettably, among the five trials with long-term follow-ups exceeding one year (Chunxin et al. 2019; Guy et al. 2021; Hongsheng et al. 2023; Lin et al. 2022; Weiwei et al. 2022), only one study took adherence into account(Guy et al. 2021), and three trials evaluated the quality of life(Guy et al. 2021; Lin et al. 2022; Weiwei et al. 2022). In general, traditional brace interventions have shown limited effectiveness. These conventional support devices heavily rely on the expertise of orthopedic practitioners, consume substantial materials, and are less precise compared to CAD/CAM or FEM techniques. They are also susceptible to issues such as friction, pressure sores, and restricted movement.

On the other hand, braces manufactured using CAD/CAM technology can achieve therapeutic outcomes that are comparable to or slightly better than those of traditional braces. The primary benefits of CAD/CAM lie in the automation of the production process, which can reduce manufacturing time to one-third of that required by traditional methods, enhance hygiene standards, and decrease the consumption of raw materials, among other advantages. The benefits of integrating CAD/CAM with FEM are particularly evident at the design stage of three-dimensional brace creation. This approach saves approximately half the time compared to traditional methods, resulting in the production of braces that are thinner and lighter. It ensures precise pressure application, enhanced comfort, and achieves clinical outcomes for Cobb angle correction that are either comparable or marginally superior(Bidari et al. 2021). The clinical results for the correction of the Cobb angle are either on par with or slightly improved over traditional approaches.

Quantitative meta-analysis evaluation

Our findings indicate that braces manufactured using CAD/CAM technology yield results that are comparable to those of traditional braces. Although there is a slight edge for CAD/CAM braces, with a mean difference of 0.66° in favor of the CAD/CAM method, this difference is not statistically significant. This is particularly true when considering that the value is less than the margin of measurement error, which is estimated to be between 3–5°(Lechner et al. 2017). The 95% confidence interval for this difference spans from − 2.79° to 1.46°, encompassing zero, which suggests that the observed difference is not clinically meaningful. While a previous review suggested that CAD/CAM technology could potentially offer superior in-brace correction (IBC) compared to traditional manual braces(Karavidas 2019), our meta-analysis results did not conclusively demonstrate a clear advantage for CAD/CAM. This aligns with the perspective of Qian Zheng(Zheng et al. 2023). The 95% CIs for the comparison between CAD/CAM + FEM and conventional braces included zero, and the mean difference fell within the margin of error. This indicates that braces fabricated using CAD/CAM + FEM or biomechanical principles did not exhibit a significant advantage, a conclusion that is also in accordance with Qian Zheng's findings(Zheng et al. 2023). This conclusion aligns with that of Qian Zheng et al. When it comes to the comparison between CAD/CAM + FEM and CAD/CAM-only braces, only the mean difference in Kyphosis Cobb correction, which was − 6.06, exceeded the measurement error, hinting at a minor advantage. However, with a P-value of 0.42, this difference was not statistically significant. The Apical rotation-MT Cobb dimension, despite having a P-value of 0.001 suggesting a significant difference, had a mean difference (MD) of -0.04, which is below the measurement error threshold of 3–5°. Thus, we still deem this advantage to be negligible. The same applies to all other bending planes, including the parietal rotation aspect, which do not provide sufficient evidence to suggest an advantage.

Although two individual trials comparing 3D printing with conventional braces showed promising outcomes, the evidence was not robust. We anticipate that future studies, particularly larger and more homogeneous ones, will enhance the robustness of the findings.

Limitations

The quantity and sample size of high-quality clinical studies, specifically randomized controlled trials or randomized controlled crossover trials, eligible for inclusion in this systematic review and meta-analysis were constrained. While the Cochrane Handbook posits that meta-analyses can be conducted by statistically amalgamating the findings of two or more independent studies(Higgins JPT 2023), an expanded pool of high-quality, pertinent studies would be instrumental in drawing more dependable conclusions. This could potentially alter our current evidence-based conclusions in the future.

Moreover, the majority of the trials we reviewed are deficient in assessing patient flexibility, adherence, and long-term quality of life indicators. This shortfall may significantly contribute to the Scoliosis Research Society (SRS) and the Society on Scoliosis Orthopedic and Rehabilitation Treatment (SOSORT) having insufficient confidence in the efficacy analyses of brace therapy(Negrini et al. 2018). Consequently, there is a pressing need for additional high-quality, long-term follow-up studies to provide conclusive evidence and enhance the reliability of clinical practice recommendations.

Conclusion

Physicians and patients may opt for CAD/CAM-based brace manufacturing when circumstances allow. It is essential, however, to be mindful of the costs and equipment investment challenges associated with this technology. This awareness can inform appropriate recommendations to medical professionals, patients, orthopedic surgeons, and healthcare organizations operating under varying conditions and resource availability. At the very least, braces manufactured using CAD/CAM technology offer therapeutic benefits that are at least equivalent to those of traditional braces. Moreover, they come with the added benefits of being lighter, more visually appealing, and more resource-efficient in terms of material usage.(Bidari et al. 2021; Cobetto et al. 2017; Cobetto et al. 2016). It is important to acknowledge the financial implications of adopting CAD/CAM manufacturing for brace production. This process necessitates the expertise of skilled CAD designers and tends to be more costly than traditional methods. Additionally, while biomechanical simulation or finite element analysis techniques offer advanced design capabilities, they also present certain limitations. These techniques demand more intricate measurements and the involvement of highly specialized personnel compared to CAD/CAM alone. Furthermore, current finite element analysis models often fail to accurately represent muscles and soft tissues, which are critical components in the biomechanics of the human body. The theoretical assumptions and simplifications inherent in such modeling may not fully capture the complex mechanical properties of patients with Adolescent Idiopathic Scoliosis (AIS). As a result, while these advanced methods hold promise, they may not yet provide a wholly realistic depiction of the biomechanical environment within which braces function(Clin et al. 2010; Desbiens-Blais et al. 2012; Fortin et al. 2007). In recent years, advancements have been made in the simulation of soft tissue representation, with new methods being developed and investigated to more accurately reflect the actual physical characteristics of patients. These improvements aim to enhance the realism of biomechanical models, thereby providing a more precise understanding of the interactions between braces and the patient's body(Guan et al. 2020; Zhang et al. 2021). In addition to FEM or biomechanical simulation, CAD/CAM combined with other advanced techniques such as 3D ultrasound-guided assessment(Lou et al. 2015) can provide a radiation-free method to determine the optimal pressure level and position for optimal in-brace correction. Also, recent studies have attempted to incorporate CAD/CAM-manufactured asymmetric brace design(Sy et al. 2016) wearable technology(Lin et al. 2020) adjustable braces(Lin et al. 2020) 4D printing(Javaid & Haleem 2020), and other new concepts into the manufacturing of AIS supports, which also deserves further research. It is suggested that large-sample, high-quality randomized controlled studies combining CAD/CAM with more new concepts and technologies could be explored in the future to guide future clinical practice. Furthermore, and of critical importance, we advocate for future studies to meticulously record and report the mean and standard deviation of outcomes associated with brace therapy. It is also recommended that additional indicators, such as adherence to treatment, aesthetic considerations, disability levels, pain, and quality of life, be incorporated into the scope of patient assessments. By expanding the range of measured outcomes, researchers can gain a more comprehensive understanding of the impact of brace therapy on patients' overall well-being and daily functioning(Negrini et al. 2018; Negrini et al. 2015; Negrini et al. 2016).

Ultimately, the SOSORT guidelines also prioritize quality of life as a primary outcome measure for assessing Adolescent Idiopathic Scoliosis (AIS)(Negrini et al. 2018). Incorporating this into research can lead to a more holistic and accurate efficacy analysis, capturing the nuanced feedback from AIS patients regarding braces fabricated by various methods. It also paves the way for scholars to conduct more comprehensive systematic reviews and meta-analyses in the future. This, in turn, can contribute to a higher standard of evidence-based medicine, offering healthcare providers and patients with AIS more reliable data to inform their treatment decisions.

Author Contribution

Xiaogang Shen, Dailiang Zhang, and Tao Guo had full access to all of the data in the study and took responsibility for the integrity of the data and the accuracy of the data analysis.Concept and design: Xiaogang Shen, Dailiang Zhang, and Tao Guo.Acquisition, analysis, or interpretation of data: Xiaogang Shen, Dailiang Zhang, Fei Yuan, Hongzhe Qi, Qian Peng, and Mengyan Zhao.Drafting the manuscript: Xiaogang Shen, Dailiang Zhang, Gongzhen Chen, and Chunjiang LvCritical review of the manuscript for important intellectual content: Gan Gao, Tao Guo.Statistical analysis: Xiaogang Shen, Dailiang Zhang, Hao Ye.Administrative, technical, or material support: Gan Gao, Tao Guo.Supervision: Xiaogang Shen, Gan Gao, Tao Guo#Xiaogang Shen, Dailiang Zhang, and Fei Yuan are co-first authors.

Concept and design: Xiaogang Shen, Dailiang Zhang, and Tao Guo.

Acquisition, analysis, or interpretation of data: Xiaogang Shen, Dailiang Zhang, Fei Yuan, Hongzhe Qi, Qian Peng, and Mengyan Zhao.

Drafting of the manuscript: Xiaogang Shen, Dailiang Zhang, Gongzhen Chen, and Chunjiang Lv

Critical review of the manuscript for important intellectual content: Gan Gao, Tao Guo.

Statistical analysis: Xiaogang Shen, Dailiang Zhang, Hao Ye.

Administrative, technical, or material support: Gan Gao, Tao Guo.

Supervision:Xiaogang Shen,Gan Gao,Tao Guo

#Xiaogang Shen, Dailiang Zhang, and Fei Yuan are co-first authors.

Funding

This research is supported by the following projects: National Natural Science Foundation of China (82260431), Guizhou Provincial Department of Science and Technology (Guizhou Science and Technology Cooperation support [2023] General 196), Guizhou Science and Technology Department Project (Guizhou Science and Technology he Foundation-ZK [2022] General 247), Guizhou Provincial people's Hospital Doctor Foundation (GZSYBS202105).

Institutional Review Board Statement

Not applicable.

Informed Consent

Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Competing Interests

The authors declare there are no competing interests.

Electronic Supplementary Material

Below is the link to the electronic supplementary material

Supplementary Material 1

References

Alonso-Coello P, Schunemann HJ, Moberg J, Brignardello-Petersen R, Akl EA, Davoli M, Treweek S, Mustafa RA, Rada G, Rosenbaum S, Morelli A, Guyatt GH, Oxman AD, and Group GW. 2016. GRADE Evidence to Decision (EtD) frameworks: a systematic and transparent approach to making well informed healthcare choices. 1: Introduction. BMJ 353:i2016. 10.1136/bmj.i2016

Amir-Behghadami M, and Janati A. 2020. Population, Intervention, Comparison, Outcomes and Study (PICOS) design as a framework to formulate eligibility criteria in systematic reviews. Emerg Med J 37:387. 10.1136/emermed-2020-209567

Bidari S, Kamyab M, Ghandhari H, and Komeili A. 2021. Efficacy of Computer-Aided Design and Manufacturing Versus Computer-Aided Design and Finite Element Modeling Technologies in Brace Management of Idiopathic Scoliosis: A Narrative Review. Asian Spine J 15:271–282. 10.31616/asj.2019.0263

Chunxin X, Jing Z, Xiangxi Y, Xiaohua C, Hua W, Shijun Y, and Min. S. 2019. Clinical efficacy of Rodin 4D technique versus traditional plaster technique for fabrication of adolescent idiopathic scoliosis orthosis. Chinese Journal of Rehabilitation Medicine 34:941–944.

Clin J, Aubin CE, Parent S, and Labelle H. 2010. A biomechanical study of the Charleston brace for the treatment of scoliosis. Spine (Phila Pa 1976) 35:E940-947. 10.1097/BRS.0b013e3181c5b5fa

Cobetto N, Aubin CE, Clin J, Le May S, Desbiens-Blais F, Labelle H, and Parent S. 2014. Braces Optimized With Computer-Assisted Design and Simulations Are Lighter, More Comfortable, and More Efficient Than Plaster-Cast Braces for the Treatment of Adolescent Idiopathic Scoliosis. Spine Deform 2:276–284. 10.1016/j.jspd.2014.03.005

Cobetto N, Aubin CE, Parent S, Barchi S, Turgeon I, and Labelle H. 2017. 3D correction of AIS in braces designed using CAD/CAM and FEM: a randomized controlled trial. Scoliosis Spinal Disord 12:24. 10.1186/s13013-017-0128-9

Cobetto N, Aubin CE, Parent S, Clin J, Barchi S, Turgeon I, and Labelle H. 2016. Effectiveness of braces designed using computer-aided design and manufacturing (CAD/CAM) and finite element simulation compared to CAD/CAM only for the conservative treatment of adolescent idiopathic scoliosis: a prospective randomized controlled trial. Eur Spine J 25:3056–3064. 10.1007/s00586-016-4434-3

Cottalorda J, Kohler R, Garin C, Genevois P, Lecante C, and Berge B. 2005. Orthoses for mild scoliosis: a prospective study comparing traditional plaster mold manufacturing with fast, noncontact, 3-dimensional acquisition. Spine (Phila Pa 1976) 30:399–405. 10.1097/01.brs.0000153346.40391.3b

Desbiens-Blais F, Clin J, Parent S, Labelle H, and Aubin CE. 2012. New brace design combining CAD/CAM and biomechanical simulation for the treatment of adolescent idiopathic scoliosis. Clin Biomech (Bristol, Avon) 27:999–1005. 10.1016/j.clinbiomech.2012.08.006

Fortin D, Cheriet F, Beausejour M, Debanne P, Joncas J, and Labelle H. 2007. A 3D visualization tool for the design and customization of spinal braces. Comput Med Imaging Graph 31:614–624. 10.1016/j.compmedimag.2007.07.006

Ghaneei M, Komeili A, Li Y, Parent EC, and Adeeb S. 2018. 3D Markerless asymmetry analysis in the management of adolescent idiopathic scoliosis. BMC Musculoskelet Disord 19:385. 10.1186/s12891-018-2303-4

Guan T, Zhang Y, Anwar A, Zhang Y, and Wang L. 2020. Determination of Three-Dimensional Corrective Force in Adolescent Idiopathic Scoliosis and Biomechanical Finite Element Analysis. Front Bioeng Biotechnol 8:963. 10.3389/fbioe.2020.00963

Guy A, Labelle H, Barchi S, Audet-Duchesne E, Cobetto N, Parent S, Raison M, and Aubin CE. 2021. Braces Designed Using CAD/CAM Combined or Not With Finite Element Modeling Lead to Effective Treatment and Quality of Life After 2 Years: A Randomized Controlled Trial. Spine (Phila Pa 1976) 46:9–16. 10.1097/BRS.0000000000003705

Hengwei F, Zifang H, Qifei W, Weiqing T, Nali D, Ping Y, and Junlin Y. 2016. Prevalence of Idiopathic Scoliosis in Chinese Schoolchildren: A Large, Population-Based Study. Spine (Phila Pa 1976) 41:259–264. 10.1097/BRS.0000000000001197

Higgins JPT TJ, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). 2023. Cochrane Handbook for Systematic Reviews of Interventions version 6.4 (updated August 2023). Available from www.training.cochrane.org/handbook.: Cochrane.

Hongsheng Z, Hanwei H, Jinwu W, Junwei Z, Rixin L, Zijie L, Peng W, Yuanjing X, Zanbo W, and Keming W. 2023. Evaluation of the effectiveness of precision medicine 3D-printed scoliosis orthosis for the treatment of scoliosis. World Complex Medicine 9:17–19 + 32.

Javaid M, and Haleem A. 2020. Significant advancements of 4D printing in the field of orthopaedics. J Clin Orthop Trauma 11:S485-S490. 10.1016/j.jcot.2020.04.021

Kan MMP, Negrini S, Di Felice F, Cheung JPY, Donzelli S, Zaina F, Samartzis D, Cheung ETC, and Wong AYL. 2023. Is impaired lung function related to spinal deformities in patients with adolescent idiopathic scoliosis? A systematic review and meta-analysis-SOSORT 2019 award paper. Eur Spine J 32:118–139. 10.1007/s00586-022-07371-z

Karavidas N. 2019. Bracing In The Treatment Of Adolescent Idiopathic Scoliosis: Evidence To Date. Adolesc Health Med Ther 10:153–172. 10.2147/AHMT.S190565

Labelle H, Bellefleur C, Joncas J, Aubin CE, and Cheriet F. 2007. Preliminary evaluation of a computer-assisted tool for the design and adjustment of braces in idiopathic scoliosis: a prospective and randomized study. Spine (Phila Pa 1976) 32:835–843. 10.1097/01.brs.0000259811.58372.87

Lechner R, Putzer D, Dammerer D, Liebensteiner M, Bach C, and Thaler M. 2017. Comparison of two- and three-dimensional measurement of the Cobb angle in scoliosis. Int Orthop 41:957–962. 10.1007/s00264-016-3359-0

Lin Y, Cheung JPY, Chan CK, Wong SWF, Cheung KMC, Wong M, Wong WC, Cheung PWH, and Wong MS. 2022. A Randomized Controlled Trial to Evaluate the Clinical Effectiveness of 3D-Printed Orthosis in the Management of Adolescent Idiopathic Scoliosis. Spine (Phila Pa 1976) 47:13–20. 10.1097/BRS.0000000000004202

Lin Y, Lou E, Lam TP, Cheng JC, Sin SW, Kwok WK, and Wong MS. 2020. The Intelligent Automated Pressure-Adjustable Orthosis for Patients With Adolescent Idiopathic Scoliosis: A Bi-Center Randomized Controlled Trial. Spine (Phila Pa 1976) 45:1395–1402. 10.1097/BRS.0000000000003559

Lou EH, Chan AC, Donauer A, Tilburn M, and Hill DL. 2015. Ultrasound-assisted brace casting for adolescent idiopathic scoliosis, IRSSD Best research paper 2014. Scoliosis 10:13. 10.1186/s13013-015-0037-8

Negrini S, Donzelli S, Aulisa AG, Czaprowski D, Schreiber S, de Mauroy JC, Diers H, Grivas TB, Knott P, Kotwicki T, Lebel A, Marti C, Maruyama T, O'Brien J, Price N, Parent E, Rigo M, Romano M, Stikeleather L, Wynne J, and Zaina F. 2018. 2016 SOSORT guidelines: orthopaedic and rehabilitation treatment of idiopathic scoliosis during growth. Scoliosis Spinal Disord 13:3. 10.1186/s13013-017-0145-8

Negrini S, Minozzi S, Bettany-Saltikov J, Chockalingam N, Grivas TB, Kotwicki T, Maruyama T, Romano M, and Zaina F. 2015. Braces for idiopathic scoliosis in adolescents. Cochrane Database Syst Rev 2015:CD006850. 10.1002/14651858.CD006850.pub3

Negrini S, Minozzi S, Bettany-Saltikov J, Chockalingam N, Grivas TB, Kotwicki T, Maruyama T, Romano M, and Zaina F. 2016. Braces for Idiopathic Scoliosis in Adolescents. Spine (Phila Pa 1976) 41:1813–1825. 10.1097/BRS.0000000000001887

Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, Chou R, Glanville J, Grimshaw JM, Hrobjartsson A, Lalu MM, Li T, Loder EW, Mayo-Wilson E, McDonald S, McGuinness LA, Stewart LA, Thomas J, Tricco AC, Welch VA, Whiting P, and Moher D. 2021. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Int J Surg 88:105906. 10.1016/j.ijsu.2021.105906

Raux S, Kohler R, Garin C, Cunin V, and Abelin-Genevois K. 2014. Tridimensional trunk surface acquisition for brace manufacturing in idiopathic scoliosis. Eur Spine J 23 Suppl 4:S419-423. 10.1007/s00586-014-3337-4

Speirs JN, Parent S, Kelly M, Upasani VV, Petcharaporn M, Bryan TP, and Newton PO. 2024. Three-dimensional vertebral shape changes confirm growth modulation after anterior vertebral body tethering for idiopathic scoliosis. Spine Deform. 10.1007/s43390-024-00856-5

Sterne JAC, Savovic J, Page MJ, Elbers RG, Blencowe NS, Boutron I, Cates CJ, Cheng HY, Corbett MS, Eldridge SM, Emberson JR, Hernan MA, Hopewell S, Hrobjartsson A, Junqueira DR, Juni P, Kirkham JJ, Lasserson T, Li T, McAleenan A, Reeves BC, Shepperd S, Shrier I, Stewart LA, Tilling K, White IR, Whiting PF, and Higgins JPT. 2019. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 366:l4898. 10.1136/bmj.l4898

Stokes OM, and Luk KD. 2013. The current status of bracing for patients with adolescent idiopathic scoliosis. Bone Joint J 95-B:1308–1316. 10.1302/0301-620X.95B10.31474

Strube P, Gunold M, Muller T, Leimert M, Sachse A, Pumberger M, Putzier M, and Zippelius T. 2021. Influence of curve morphology and location on the efficacy of rigid conservative treatment in patients with adolescent idiopathic scoliosis. Bone Joint J 103-B:373–381. 10.1302/0301-620X.103B2.BJJ-2020-1113.R2

Sy N, Borysov M, Moramarco M, Nan XF, and Weiss HR. 2016. Bracing Scoliosis - State of the Art (Mini-Review). Curr Pediatr Rev 12:36–42.

Wang Y, DelRocco N, and Lin L. 2024. Comparisons of various estimates of the I2 statistic for quantifying between-study heterogeneity in meta-analysis. Stat Methods Med Res:9622802241231496. 10.1177/09622802241231496

Weinstein SL, Dolan LA, Wright JG, and Dobbs MB. 2013. Effects of bracing in adolescents with idiopathic scoliosis. N Engl J Med 369:1512–1521. 10.1056/NEJMoa1307337

Weiss HR, Karavidas N, Moramarco M, and Moramarco K. 2016. Long-Term Effects of Untreated Adolescent Idiopathic Scoliosis: A Review of the Literature. Asian Spine J 10:1163–1169. 10.4184/asj.2016.10.6.1163

Weiss HR, Tournavitis N, Nan X, Borysov M, and Paul L. 2017. Workflow of CAD / CAM Scoliosis Brace Adjustment in Preparation Using 3D Printing. Open Med Inform J 11:44–51. 10.2174/1874431101711010044

Weiwei Z, Dun W, and Huabai L. 2022. Computer-assisted fabrication of orthoses for correction of adolescent idiopathic scoliosis. Chinese Journal of Orthopaedic Surgery 30:983–988.

Wong MS, Cheng JC, and Lo KH. 2005. A comparison of treatment effectiveness between the CAD/CAM method and the manual method for managing adolescent idiopathic scoliosis. Prosthet Orthot Int 29:105–111. 10.1080/17461550500069547

Zhang K, Xu N, Guo C, and Wu J. 2022. MPF-net: An effective framework for automated cobb angle estimation. Med Image Anal 75:102277. 10.1016/j.media.2021.102277

Zhang Q, Chon T, Zhang Y, Baker JS, and Gu Y. 2021. Finite element analysis of the lumbar spine in adolescent idiopathic scoliosis subjected to different loads. Comput Biol Med 136:104745. 10.1016/j.compbiomed.2021.104745

Zhang Y, and Li X. 2019. Treatment of bracing for adolescent idiopathic scoliosis patients: a meta-analysis. Eur Spine J 28:2012–2019. 10.1007/s00586-019-06075-1

Zheng Q, He C, Huang Y, Xu T, Jie Y, and Ma CZ. 2023. Can Computer-Aided Design and Computer-Aided Manufacturing Integrating with/without Biomechanical Simulation Improve the Effectiveness of Spinal Braces on Adolescent Idiopathic Scoliosis? Children (Basel) 10. 10.3390/children10060927

Zheng Y, Dang Y, Wu X, Yang Y, Reinhardt JD, He C, and Wong M. 2017. Epidemiological study of adolescent idiopathic scoliosis in Eastern China. J Rehabil Med 49:512–519. 10.2340/16501977-2240

Table 1

Yes