1 Introduction
Scoliosis is a three-dimensional (3D) spinal deformity characterized by lateral curvature in the coronal plane and abnormal physiological curvature in the sagittal plane. Clinical guidelines issued by the Society on Scoliosis Orthopaedic and Rehabilitation Treatment (SOSORT) propose graded intervention strategies based on deformity severity: patients with a Cobb angle < 10° only require regular follow-up monitoring; those with 10° to 20° can receive targeted exercise training and rehabilitation interventions; brace therapy combined with rehabilitation is preferred for 20° to 45°; and surgical correction is recommended when the Cobb angle exceeds 45°.[
1] Rehabilitation approaches, including bracing, exercise training, traction, manual therapy, and electrical stimulation, can effectively improve spinal morphology and functional status in patients.[
2] Surgical interventions for scoliosis are dominated by posterior or anterior spinal fusion, which features complex procedures, large surgical trauma, long operation duration, high blood loss, and considerable perioperative risks.[
3] Therefore, perioperative rehabilitation management and enhanced recovery strategies are crucial for improving surgical outcomes and accelerating postoperative recovery.[
4-
6]
In recent years, the management concept of scoliosis has undergone rapid iteration. Traditional single corrective rehabilitation and conservative treatment models are gradually being updated by perioperative systematic functional rehabilitation, patient-reported outcome (PRO)-based evaluation systems, multidisciplinary collaborative management, and individualized precision rehabilitation strategies. However, the research in this field presents scattered, interdisciplinary, and rapidly updating features. At present, there remains a lack of systematic and macroscopic overview studies that sort out the evolutionary trajectory of global scoliosis rehabilitation research, clarify the internal knowledge structure, and identify emerging research frontiers.
Bibliometric analysis and visual research are effective methods to grasp disciplinary frontiers and clarify research contexts systematically. The SCI-E database of the Web of Science Core Collection is an authoritative platform for scientific statistics and academic evaluation.[
7–
10] At present, no systematic bibliometric study has been reported in the field of scoliosis rehabilitation. Based on citation data and visualization tools, this study systematically sorts out the knowledge structure and development context of this field to provide a reference for researchers to grasp research directions and formulate scientific research decisions.
Different from general bibliometric research, this work specifically targets the pain point of unclear research context and unsynchronized theoretical and clinical development in scoliosis rehabilitation. It quantitatively and visually sorts out the research hotspots, evolution laws, and research gaps in this field, which can not only help researchers grasp the developing frontier but also provide targeted evidence for optimizing clinical rehabilitation schemes and formulating standardized rehabilitation management procedures for scoliosis patients.
2 Materials and methods
2.1 Data retrieval
Data were retrieved from the Web of Science Core Collection (SCI-E) database via the Library of Capital Medical University. The search strategy was formulated as follows: TS = (rehabilitation OR habilitation OR recovery) AND (scoliosis OR spinal curvatures). Language was restricted to English, and document types were limited to articles and reviews. The time span covered the period from 1985 to 2023. To avoid deviations caused by real-time database updates, the entire retrieval was completed on a single day, February 20, 2023. Full records and cited reference data were exported in both plain text and tab-delimited formats.
2.2 Bibliometric analysis
This study analyzed the annual publication and citation trends of the included literature and plotted the proportion map of publications by country. The publication volume, total citations, and H-index of the top 10 countries were extracted. For the top 20 highly cited literature, the author, title, journal, publication year, total citations, and average citations per article were recorded. The publication output, H-index, total citations, and average citations per article of the top five authors, institutions, and countries were sorted out, and the top 10 research directions, funding projects, and journals were statistically analyzed.
A bibliometric platform was used to explore the annual publication output of the top 10 countries, international cooperation patterns, and the annual distribution of keywords. VOSviewer 1.6.11 was adopted to construct collaboration networks of authors and institutions, as well as the mapping of literature co-citation, journal cocitation, author co-citation, and keyword co-occurrence. CiteSpace 6.1.R6 was applied, with the time span set from January 1992 to February 2023 and a single time slicing under default parameters. On this basis, literature co-citation timeline analysis, keyword clustering analysis, as well as highly cited burst literature and burst keyword analysis were performed.
3 Results
3.1 Publications and citations
A total of 308 articles and reviews were included from January 1992 to February 2023, with 4,333 total citations, an average citation of 14.07 per paper, and an H-index of 35. Figure 1 showed the annual publication output showed a fluctuating upward trend over the past 30 years, reaching the highest level of 42 publications in 2022. Citation records began in 1996 with three citations, and the annual citation peak appeared in 2021 at 556 times.
3.2 Publications, citations, and H-index of different countries in the world
A total of 30 countries have published literature in this field, among which the United States, China, and Italy occupy a leading position. The United States published its first paper in 1992, while China began its first publication in 1999. No relevant publications were produced in 1994, 1996, and 1998. The top five countries by publication output were the United States (76 publications), China (56 publications, accounting for 18.18%), Italy (41 publications), Poland (30 publications), and Canada (19 publications). In terms of total citations, the United States ranked first with 1,264 citations. Regarding the H-index, the United States and Italy tied for first place, both with an H-index of 19 (Fig. 2).
3.3 Top 20 most cited articles on rehabilitation in scoliosis
Top 20 most cited articles on rehabilitation in scoliosis are shown in Table 1. Of the 308 included publications, the top 20 highly cited papers received 45 to 244 citations, with the average citation per paper ranging from 2.17 to 17.43 times. The top article had 244 citations and was published in 2010 by Isaacs et al.,[
11] followed by Monticone et al.[
12] with 108 citations in 2014 and Lenssinck et al.[
13] with 105 citations in 2005. The oldest article was published by Noordeen et al.[
18] in 1994, and the most recent article in the top 20 was published in 2017 by Adogwa et al.[
25]
3.4 Contribution of authors, organizations, and countries
A total of 1,392 authors, 540 institutions, and 30 countries have participated in research in this field. The top three productive authors were Negrini S (17 publications, Italian Spine Science Institute), Zaina F (16 publications, the same institution), and Romano M (nine publications, the same institution). The top three publishing institutions included the Italian Spine Science Institute (13 publications), the University of California System (10 publications), and the University of Brescia (nine publications). What's more, the corresponding records, H-index, total citations, and average citations of the top five authors, organizations, and countries were meanwhile shown in Table 2.
3.5 Contribution of research directions, journals, and funds
A total of 37 research directions, 129 funding programs, and 138 journals were involved in this field. Among the top 10 research directions, Orthopedics ranked first with 122 publications, an H-index of 24, total citations of 1,996, and an average citation of 16.36 per article. Rehabilitation achieved the highest average citation of 19.98 per paper (Table 3).
In terms of journals, Spine led the list with 37 publications, an H-index of 18, total citations of 980, and an average citation of 26.49 per article. Disability and Rehabilitation had the highest average citation rate at 33.00 times per paper. BMC Musculoskeletal Disorders published the largest number of highly cited papers (three articles). The Journal of Bone and Joint Surgery had the highest impact factor of 6.56 (Table 4).
Regarding funding support, the National Natural Science Foundation of China ranked first with six supported publications, followed by the United States Department of Health and Human Services (five publications) and the National Institutes of Health (four publications) (Fig. 3).
3.6 Co-authorship analysis of publications
Negrini S had the most co-authorship strength (total link strength = 25), with 14 documents and 463 citations, followed by Zaina F (total link strength = 24) with 11 documents and 377 citations, and Donzelli S (total link strength = 12) with six documents and 88 citations (Fig. 4A). Besides, the strongest collaborative country was Italy (total link strength = 35) with 41 documents and 1,027 citations, followed by Poland (total link strength = 29) with 30 documents and 472 citations, and the United States (total link strength = 27) with 75 documents and 1,263 citations (Fig. 4B). Moreover, the closest collaborative organization was the Italian Scientific Spine Institute (total link strength = 18) with 13 documents and 501 citations; the second organization was the University of Medical Sciences (total link strength = 15) with eight documents and 266 citations, and the third was Tzaneio General Hospital of Piraeus (total link strength = 14) with five documents and 229 citations (Fig. 4C).
3.7 Co-citation analysis of publications
The most co-cited reference (
n = 30) was “Effects of bracing in adolescents with idiopathic scoliosis,” published by Weinstein SL on
The New England Journal of Medicine in 2013,[
31] The second reference (
n = 22) was “2011 SOSORT guidelines: Orthopedic and Rehabilitation treatment of idiopathic scoliosis during growth,” published by Negrini S on
Scoliosis in 2012 (one) (Fig. 5A). On the other hand, the most co-citation journal was
Spine (
n = 1,819), followed by
European Spine Journal (
n = 484), and
Scoliosis Spinal Disorder (
n = 442) and (Fig. 5B). Furthermore, the most cocitation author was Negrini S (
n = 289), the second was Weiss HR (
n = 127) and the third was Weinstein SL (
n = 112) (Fig. 5C).
3.8 Co-citation timeline of references and burst analysis
Co-citation of references for a timeline diagram was drawn using Citespace software (Fig. 6). References in the same cluster are arranged on the timeline in chronological order of publication. “Adolescent idiopathic scoliosis,” “SOSORT guidelines,” and “Asian adolescent idiopathic scoliosis” were the clusters with most published references. According to the year of publications, “treatment” and “idiopathic scoliosis” were the clusters with the earliest references. “Adolescent idiopathic scoliosis” and “Asian adolescent idiopathic scoliosis” were the clusters with the latest references. The top 25 references with the highest burst value were shown in Figure 7.[
19,
32,
33]
3.9 Keywords visualization analysis
The top 20 keywords with the highest burst value were demonstrated in Figure 8. The time period occupied by red on the right was the duration of the keywords. According to the burst strength and duration of the keywords, the transformation of domain research direction can be roughly divided into three stages. The first stage was from 1998 to 2011, the keywords were “anesthesia,” “anterior instrumentation,” and “pulmonary function.” The second stage was from 2012 to 2018, the keywords were “quality of life,” “complication,” and “reliability.” The third stage was from 2019 to 2023, the keywords were “Cobb angle,” “posterior spinal fusion,” and “risk factor.”
4 Discussion
Scoliosis refers to one or more parts of the spine presenting a certain curve in the coronal plane. There will also be a rotation of the vertebral body, or sagittal convexity, or kyphosis. It is a deformity produced in three dimensions.[
34] Scoliosis can be divided into functional and structural types.[
35] Functional scoliosis is temporary, usually caused by lumbar disc herniation, poor posture, hip contracture, and other reasons; it can be totally cured by eliminating the cause and through some correction treatments. Structural scoliosis is caused by lesions of the spine and associated tissues, such as congenital scoliosis, idiopathic scoliosis, and so on. The probability of adolescent idiopathic scoliosis is as high as 75% to 80%, which seriously affects the health of patients. It will have negative physiological as well as psychological effects on children and adolescent.[
36,
37] The research focus is gradually shifting from traditional structural correction and passive conservative rehabilitation to perioperative whole-cycle functional recovery, patient-centered subjective rehabilitation evaluation, multidisciplinary integrated management, and personalized intelligent rehabilitation intervention.
At present, there are two main methods of treatment for scoliosis: rehabilitation treatment and surgical treatment.[
38] The commonly used rehabilitation methods include functional rehabilitation training, posture training, core strength training, stability training, and so on, which can improve spinal circumference. Meanwhile, the development of scoliosis can be controlled by the asymmetric stress of both sides of the muscle group, and it has a good effect on adolescent scoliosis.[
39] Posterior spinal fusion surgery is the most effective treatment for scoliosis when the Cobb angle is > 50° or when nonoperative treatment has failed, which is defined as the Cobb angle being > 45° and progressing more than 5° within 6 months. As we all know, posterior spinal fusion surgery for scoliosis has some disadvantages, such as long operation time, excessive bleeding, large trauma, and slow recovery. A perioperative rehabilitation program based on the International Classification of Functioning, Disability, and Health concept was used in the treatment of scoliosis.[
40] Enhanced recovery after surgery has also been widely used for scoliosis surgery perioperative management in recent years.[
4–
6]
Bibliometrics mainly collects bibliographic databases and bibliometric features, using the literature database to retrieve literature on related topics, and employing statistical methods to analyze and study multiple pieces of information in the literature in an all-round way, which helps researchers to timely understand the research direction and development trend of related fields.[
41,
42] Scholars can also find current research hotspots and hotspot directions in a specific field by using visual analysis software to further analyze the literature, as visual analysis has been widely applied in the field of orthopedics.[
43] However, there is insufficient research on the long-term rehabilitation prognosis of scoliosis, uneven development of personalized rehabilitation technology, a lack of unified clinical rehabilitation evaluation standards, and relatively lagging research on rehabilitation intervention for special populations.
From the results of bibliometric analysis and VOSviewer, we can see that a total of 308 articles and reviews were retrieved, which have a sum of 4,333 citations, an average of 14.07 citations per item, and an H-index of 35. The number of publications and citations showed a fluctuating increase year by year. The United States, China, and Italy contributed the most publications. The highest-yield author, organization, country, research directions, funding sources, and journals were Negrini S, Italian Scientific Spine Institute, USA, Orthopedics, National Natural Science Foundation of China, and Spine, respectively. However, there was not much collaboration among authors, organizations, and countries.
From the results of bibliometric analysis and Citespace, we can see that “adolescent idiopathic scoliosis,” “SOSORT guidelines,” and “Asian adolescent idiopathic scoliosis” are the largest cluster terms. In addition, “Cobb angle,” “posterior spinal fusion,” and “risk factor” are the latest two burst terms in recent years. As all we know, the proportion of adolescent idiopathic scoliosis is as high as 75% to 80% in all kinds of scoliosis. The SOSORT produced its first guidelines in 2005 and renewed them in 2016, which provides the most important principles to guide scoliosis surgery and rehabilitation.[
44,
45] More importantly, the safety of posterior spinal fusion for scoliosis is increasingly being emphasized by spinal surgeons in recent years.[
46]
Our study aims to outline the intellectual connections within the dynamic changes of scientific knowledge in the field of rehabilitation and recovery in scoliosis by making good use of the citation database and software tools. These results can benefit scholars by providing a better understanding of future research directions and offering sufficient evidence for policy-making. We emphasize that the emerging research frontiers identified in this study can provide targeted directions for subsequent basic and clinical research. Meanwhile, the research trend characteristics summarized in this paper can provide references for clinicians to optimize rehabilitation strategies, formulate individualized rehabilitation plans, and promote the standardized and intelligent development of scoliosis rehabilitation.
5 Limitations
It should be noted that this study only used a single database, the Web of Science Core Collection, and did not include InCites, MEDLINE, and so on, resulting in limited data coverage. Only English literature was included, while literature in other languages was not, leading to language bias. In addition, citation frequencies are affected by publication time, and retrieval results may vary slightly at different times. Last but not least, there were no potential biases in citation metrics, such as the “Matthew effect” and the time lag in citation accumulation. Additionally, the impact of interdisciplinary research on rehabilitation would be further explored, such as engineering and biomechanics. In the future, the scope of databases and languages can be expanded to improve the comprehensiveness of the analysis.
6 Conclusions
In summary, research on rehabilitation in scoliosis is in a rapid development stage. Annual publications in this field have been continually increasing since 1992, especially in the last 2 years. The United States, China, and Italy have made the most contributions to this field. The most high-yield author, organization, country, research directions, funds, and journals were Negrini S, Italian Scientific Spine Institute, USA, Orthopedics, National Natural Science Foundation of China, and Spine, respectively. “Adolescent idiopathic scoliosis,” “SOSORT guidelines,” and “Asian adolescent idiopathic scoliosis” are the largest cluster terms, which indicates the future hotspots on rehabilitation in scoliosis, while “Cobb angle,” “posterior spinal fusion,” and “risk factor” are the latest two burst terms in recent years, which indicate that surgery and its safety are becoming the research frontiers and trends.
The Author(s) 2026. Published by Wolters Kluwer Health, LLC. on behalf of Higher Education Press. This is an open access article distributed under the Creative Commons Attribution License 4.0 (CCBY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.