Global trends and hotspots of Charcot–Marie–Tooth disease

Abstract

Background:

Charcot–Marie–Tooth disease (CMT) is a hereditary neuropathy characterized by progressive sensory and motor deficits. Over the past decade, the volume of CMT-related publications has increased steadily. This study aimed to examine global publication trends, key contributors, collaborative networks, and emerging research hotspots in CMT through bibliometric analysis.

Methods:

Publications related to CMT from 2014 to 2024 were retrieved from the Web of Science Core Collection. VOSviewer 1.6.20, CiteSpace 6.3.R1, and the R package “bibliometrix” were used to analyze annual publication output, country and institutional contributions, journal distribution, author collaboration, citation networks, keyword co-occurrence, and research trends.

Results:

A total of 1531 CMT-related articles were identified, with an annual growth rate of 1.75%. The average number of citations per document was 15.56. The United States ranked first in research output, contributing 324 publications and accounting for 21.2% of the total, followed by China and Italy. The University of London and University College London were the leading institutions, and the Journal of the Peripheral Nervous System and Neuromuscular Disorders were the main publication venues. Keyword analysis showed that “mutations,” “neuropathy,” and “gene” were the most frequent terms, reflecting sustained attention to genetic mechanisms, disease pathogenesis, and molecular diagnosis. Citation burst analysis identified “variants” and “deoxyribonucleic acid (DNA) methylation” as recent burst keywords beginning in 2022, suggesting increasing interest in variant interpretation, epigenetic regulation, and advanced sequencing-based approaches, including whole-genome sequencing.

Conclusion:

This bibliometric analysis shows that global CMT research maintained steady growth from 2014 to 2024, with the United States, China, Italy, the United Kingdom, and several leading institutions making major contributions. Current research remains centered on genetic mechanisms and molecular diagnosis, while recent hotspots indicate a shift toward variant interpretation, epigenetic regulation, whole-genome sequencing, and personalized medicine. Future bibliometric studies should include additional databases and underrepresented regions to provide a more comprehensive understanding of the evolving CMT research landscape.

1. Introduction

Charcot–Marie–Tooth disease (CMT), also known as hereditary motor and sensory neuropathy, is a genetically inherited peripheral neuropathy characterized by progressive muscle weakness and atrophy in the distal limbs, coupled with sensory impairments.^[1] CMT is classified based on nerve conduction velocity into demyelinating forms (CMT1), axonal forms (CMT2), and intermediate forms (CMT3).^[2] It affects at least 1 in every 2500 individuals and is predominantly inherited in an autosomal dominant manner; however, autosomal recessive and X-linked inheritance patterns are also observed.^[3] Symptoms typically commence before the age of 20 and exhibit a gradual progression; nevertheless, severe early-onset forms, such as Dejerine–Sottas syndrome, may present during infancy.^[4] The pathogenesis of CMT is multifaceted, involving gene mutations, abnormal myelination, axonal damage and regeneration, metabolic disturbances, and environmental factors.^[3] More than 100 genes have now been implicated in CMT, including PMP22, MPZ, and GJB1, and recent studies continue to expand the genetic spectrum of the disease through the identification of novel pathogenic mechanisms and variant types. Mutations in these genes affect proteins required for Schwann-cell function, myelin formation, axonal integrity, and peripheral nerve maintenance, leading to impaired nerve conduction, distal muscle weakness, sensory loss, and atrophy.^[5,6] Recent experimental and genetic studies have further highlighted axonal degeneration pathways and mosaic pathogenic variants as important areas of investigation in CMT, including SARM1-related axon-degeneration mechanisms and de novo somatic mosaicism involving CMT-associated genes such as INF2 and TRPV4.^[7,8] Thus, the pathogenesis of CMT involves overlapping mechanisms, including demyelination, axonal degeneration, metabolic dysfunction, and genetic or epigenetic modifiers.^[5,6,9] Currently, there are no curative treatments for CMT; management remains largely symptomatic and supportive, including rehabilitation exercise, physical therapy, orthotic support, pharmacological treatment for complications, and selected surgical procedures to improve lower-limb function, prevent contractures, and maintain ambulation in individuals with severe impairment.^[10,11]

While many patients with CMT can maintain daily activities with varying degrees of functional limitation, severe cases may develop disabling or life-threatening complications, including respiratory muscle weakness, cardiac abnormalities, lung infections, and pressure sores related to prolonged immobility.^[12] CMT also affects psychological well-being and quality of life, with some patients experiencing low self-esteem, anxiety, and limitations in education or professional development.^[13] Because CMT is genetically heterogeneous and clinically variable, research in this field has expanded across multiple areas, including molecular genetics, disease mechanisms, diagnostic technologies, rehabilitation, quality of life, and therapeutic development. However, this growing literature is dispersed across different disciplines, countries, institutions, and journals, making it difficult to obtain a comprehensive understanding of global research activity, collaboration patterns, and evolving hotspots. The genetic nature of CMT also makes genetic counseling and prenatal diagnosis important for families with reproductive plans. In this context, bibliometric analysis provides a quantitative approach to mapping publication trends, influential contributors, co-authorship networks, citation patterns, keyword evolution, and emerging research themes over time.^[14–16] Therefore, a bibliometric study of CMT can clarify the current research landscape, identify underexplored areas, and provide evidence to guide future scientific collaboration and research planning.

This bibliometric approach not only elucidates the current status and advancements in CMT research but also identifies emerging hotspots and informs future scientific inquiries and interdisciplinary collaborations. By examining patterns of co-authorship and citation, this study reflects on the historical trajectory of research, highlights evolving paradigms, and suggests potential future directions. Furthermore, it offers a detailed perspective on the interactions among various research domains, underscoring the multidisciplinary nature of contemporary studies and indicating areas where further interdisciplinary cooperation could significantly enhance our understanding of CMT.

2. Materials and methods

2.1. Search strategies and data source

A bibliometric analysis was conducted using the Web of Science Core Collection (WoSCC). WoSCC was selected as the primary data source because it provides standardized bibliographic records, complete cited-reference information, author affiliations, institutional addresses, journal data, and citation metrics that are compatible with VOSviewer, CiteSpace, and the R package “bibliometrix.” In addition, WoSCC has been widely used in bibliometric studies because of its strict journal selection criteria and relatively consistent citation indexing across scientific disciplines. Although Scopus and PubMed include additional biomedical records, WoSCC was used in this study to ensure data consistency, reduce heterogeneity in citation indexing, and enable reproducible citation and co-citation analyses.

The search was conducted on December 14, 2024, to minimize the influence of subsequent database updates. The study period was set from January 1, 2014, to December 14, 2024. The search strategy was as follows: TS = (“Charcot-Marie-Tooth disease” OR “CMT disease” OR “Charcot-Marie-Tooth disease type 1” OR “Charcot-Marie-Tooth disease type 2” OR “intermediate Charcot-Marie-Tooth disease” OR CMT1 OR CMT2 OR “Charcot-Marie-Tooth Syndrome” OR “Charcot-Marie-Tooth Hereditary Neuropathy”).^[17] Only English-language articles were included. Records were exported from WoSCC in “Full Record and Cited References” and “Plain Text” formats. The extracted information included publication year, title, author names, author affiliations, institutional affiliations, countries or regions, journal title, keywords, citation counts, and cited references.

Because the abbreviations CMT1 and CMT2 may also refer to concepts unrelated to CMT disease, all retrieved records were manually reviewed by title, abstract, author keywords, and Keywords Plus to exclude irrelevant studies. Search validation was conducted by checking whether key CMT publications known to the authors and highly cited CMT studies included in recent reviews were captured by the search strategy. In addition, a random sample of retrieved records was manually examined to assess the relevance of the search results. The validation process confirmed that the search strategy captured the major CMT-related literature included in the final bibliometric analysis.

The inclusion criteria were as follows: publications focused on CMT disease or closely related inherited peripheral neuropathies; articles published between 2014 and 2024; English-language publications; and document type restricted to articles. The exclusion criteria were as follows: duplicate records; non-English publications; reviews, editorials, letters, meeting abstracts, corrections, book chapters, and other non-original research documents; and records in which CMT1 or CMT2 referred to unrelated terms, such as plant chromomethylase or other non-neurological topics.

Two reviewers independently screened all retrieved records according to the predefined eligibility criteria. Disagreements were resolved through discussion, and, when necessary, a third reviewer was consulted. Interreviewer agreement was assessed using Cohen’s kappa coefficient. After screening, the final dataset included 1531 articles.

Before analysis, author names, institutional names, and country or region names were standardized to reduce duplication caused by spelling variants, abbreviations, and address-format inconsistencies. For institutional analysis, merged records were reviewed manually. Institutions were treated as separate entities when they appeared as distinct organizations in WoSCC affiliation data. Therefore, the University of London and University College London were retained as separate institutions because WoSCC indexed them as separate affiliation entities and because University College London functions as an independent research institution within the federal University of London system. This decision was considered when interpreting institutional rankings.

2.2. Statistical analysis and visualization

Bibliometric and visualization analyses were conducted using VOSviewer, version 1.6.20 (Centre for Science and Technology Studies, Leiden University, Leiden, the Netherlands), CiteSpace, version 6.3.R1 (Drexel University, Philadelphia, Pennsylvania, USA),^[18] R (R Foundation for Statistical Computing, Vienna, Austria), and the R package bibliometrix.^[19] These tools were used to analyze publication output, citation performance, country and institutional contributions, journal distribution, author collaboration, co-citation patterns, keyword co-occurrence, and citation bursts.^[18–20] Publication trends were summarized by annual publication counts and annual growth rate.

VOSviewer was used to construct collaboration and co-occurrence networks. The following analyses were performed: country collaboration, institutional collaboration, journal co-occurrence or bibliographic coupling, author collaboration, and keyword co-occurrence. The counting method was full counting. The normalization method was association-strength normalization. In the network maps, node size represented publication count, citation count, or keyword frequency, node color indicated cluster membership, and link thickness represented total link strength. CiteSpace was used to identify temporal trends and burst keywords. The time slicing was set from 2014 to 2024, with 1 year per slice. The node type was set as keyword for burst detection. The selection criteria were Top N = 50 per slice. The pruning methods included Pathfinder and pruning the merged network. Burst detection was performed using Kleinberg algorithm as implemented in CiteSpace. The strongest burst keywords and their active periods were used to identify emerging research hotspots.

The bibliometrix package in R was used for descriptive bibliometric analysis and performance evaluation. Main functions included convert2df(), biblioAnalysis(), summary(), biblioshiny(), and networkPlot(). These analyses were used to calculate annual scientific production, country productivity, author productivity, source productivity, citation indicators, and collaboration indicators. Journal performance was further described using Journal Citation Reports quartiles and Impact Factors, based on the most recent 2024 data available at the time of analysis.

No formal hypothesis testing was performed because this was a descriptive bibliometric study. Quantitative results were presented as frequencies, percentages, citation counts, average citations, and network indicators. Sensitivity analyses were limited by the use of a single database and predefined search strategy. To reduce potential retrieval bias, manual screening and search validation were conducted as described above.

Citation impact was assessed using total citations, average citations per document, and citation-based indicators, including the H-index, G-index, and M-index.^[21,22] Because older publications had more time to accumulate citations than recent publications, citation-age bias was considered during interpretation. Where applicable, citations per year were calculated as a supplementary indicator using the following formula: citations per year = total citations/(2024 − publication year + 1). However, the main bibliometric rankings were based on total citation counts to remain consistent with commonly used bibliometric methods. Journal performance was evaluated using Journal Citation Reports quartiles and Impact Factors, with the most recent 2024 data ensuring timely assessments. These metrics provided valuable insights into the academic influence and prestige of journals within the field. By integrating these analytical approaches, this study offers a systematic overview of the CMT research landscape, highlighting key contributors, prominent institutions, and evolving research directions. The results furnish valuable guidance for future studies and promote collaborative opportunities to enhance the understanding and treatment of CMT.

3. Results

3.1. Overview of publications

The comprehensive selection process is illustrated (Fig. 1). Between 2014 and 2024, a total of 1531 documents were published, indicating sustained interest in this research domain, with an annual growth rate of 1.75%. These publications involved 9405 authors, with 32.85% exhibiting international co-authorship, which signifies considerable global collaboration. Each document had an average of 9.92 coauthors, and the total number of references cited reached 29,139, supported by 3016 distinct keywords, thereby showcasing the depth and diversity of the research. As demonstrated in Figure 2, the publication trend reflects a gradual growth trajectory, with annual outputs fluctuating over the years. The number of publications reached its zenith at 170 in 2020, followed by a slight decline, with 140 publications in 2023 and 138 in 2024. The average number of citations per document stands at 15.56, illustrating the enduring impact of these works. Furthermore, the average document age of 4.75 years emphasizes the ongoing advancement and relevance of the field within the research community.

Figure 1.

Flowchart of the literature screening process.

Figure 2.

Annual number of publications.

3.2. Analysis of countries

The global distribution of publications showed that research output was concentrated in several countries. In this study, publication volume refers to the number of articles published, whereas scientific impact was evaluated using citation-based indicators, including total citations, average citations per article, and, where applicable, H-index-related metrics. The USA had the highest publication volume, with 324 articles, accounting for 21.2% of all publications. It also had the highest total citation count, with 7285 citations, and an average citation rate of 22.5 citations per article, indicating that the USA ranked highly in both productivity and citation impact. The multiple-country publication (MCP) ratio (the proportion of publications involving international collaboration) was 0.349, suggesting a moderate level of international collaboration. China ranked second in publication volume, with 201 articles, accounting for 13.1% of all publications. However, most publications from China were single-country publications (publications involving authors from only 1 country), and its MCP ratio was relatively low. China ranked fifth in total citations, with 1599 citations, indicating that its publication volume was high, whereas its citation impact was comparatively lower than that of the USA, the UK, and Germany.

The UK published fewer articles than the USA and China but showed a higher MCP ratio of 0.632 and achieved the second-highest total citation count, with 2293 citations and an average citation rate of 24.1 citations per article. These findings suggest that the UK had a strong citation impact and a high level of international collaboration despite a smaller publication volume. Germany also demonstrated notable scientific impact, with an MCP ratio of 0.536, 1745 total citations, and an average citation rate of 20.8 citations per article. These results indicate that Germany contributed substantially to the field in terms of citation impact and international collaboration rather than publication volume alone (Table S1, Supplemental Digital Content 1). The USA and China were the leading countries in publication volume, followed by Italy, Japan, and the UK, whereas the UK and Germany showed relatively high citation impact and international collaboration indicators (Fig. 3A). The international collaboration network showed that the USA occupied a central position, with frequent collaborative links with China, the UK, and Germany (Fig. 3B). China also showed collaborative links with the USA and several European countries. Countries such as Belgium and Cyprus had smaller publication outputs but relatively high MCP ratios, suggesting that their contributions were more collaboration-oriented.

Figure 3.

Analysis of countries. (A) Distribution of corresponding author’s publications by country. (B) Visualization map depicting the collaboration among different countries.

3.3. Analysis of institutions

Institutional publication volume and scientific impact were analyzed separately. Institutional publication volume was measured by the number of articles, whereas institutional impact was interpreted using citation-related metrics and network indicators when available. The University of London had the highest publication volume, with 299 articles, followed by University College London and Sungkyunkwan University (Fig. 4A). These institutions were also positioned centrally in the collaboration network, indicating strong collaborative activity with national and international partners (Fig. 4B). However, publication volume alone does not necessarily indicate higher scientific impact; therefore, institutional rankings should be interpreted together with citation metrics and network centrality indicators.

Figure 4.

Analysis of institutions. (A) Top 10 institutions by article count and rank. (B) Visualization map depicting the collaboration among different institutions.

3.4. Analysis of journals

The 1531 articles included in this study were published in 423 journals. For journals, publication volume was defined as the number of CMT-related articles published, whereas scientific impact was assessed using citation indicators, total link strength, journal quartile, and impact factor where available. The Journal of the Peripheral Nervous System had the highest publication volume, with 91 articles, followed by Neuromuscular Disorders with 55 articles and Brain (Table S2, Supplemental Digital Content 2). These journals represented major publication venues for CMT research.

The journal network analysis showed that Brain and Neurology were central nodes, indicating strong connections with other journals in the citation or coupling network (Fig. 5A). In the coupling network, 69 journals met the threshold of at least 5 links. The journals with the highest total link strength included the Journal of the Peripheral Nervous System with a total link strength of 27,008, followed by Neuromuscular Disorders and Brain (Fig. 5B). These findings indicate that some journals contributed primarily through publication volume, whereas others demonstrated influence through citation relationships and network connectivity.

Figure 5.

Analysis of journals. (A) The co-occurrence networks of journals. (B) The coupling networks of journals.

3.5. Analysis of authors

A total of 9405 authors contributed to CMT research during the study period. Author productivity was evaluated by publication count, whereas author impact was assessed using citation counts, H-index, and total link strength in the co-authorship network. Reilly Mary R. had the highest publication count, with 84 articles, and the highest citation count, with 2055 citations. Shy Michael E. ranked second in publication volume, with 75 articles and 1661 citations. Both authors had high H-index values of 25, suggesting strong scientific impact in addition to high productivity.

Pareyson Davide and Laura Matilde were also important contributors, with 61 and 44 publications, respectively, and substantial citation counts (Table S3, Supplemental Digital Content 3). The co-authorship network showed several major collaborative clusters, reflecting different research focuses. The green cluster included Reilly Mary R. and Herrmann David N.; the blue cluster included Pareyson Davide; and the orange cluster included Choi Byung-Ok and Chung Ki Wha (Fig. 6). Total link strength further indicated that Reilly Mary R. and Pareyson Davide had strong collaborative connections. These results suggest that leading authors differed in their roles, with some contributing mainly through high publication productivity, others through citation impact, and others through collaborative network centrality.

Figure 6.

Analysis of authors.

3.6. Analysis of keywords

The analysis of keywords elucidates the research hotspots and trends within the field. “Mutations” emerges as the most frequently occurring keyword, with a total of 274 instances and the highest total link strength of 956. This is followed by “neuropathy” (238) and “gene” (186) (Table S4, Supplemental Digital Content 4). These keywords signify a substantial interest in genetic factors, disease mechanisms, and clinical outcomes. Areas of emphasis include genetic research, as evidenced by terms such as “gene,” “mutations,” and “protein,” alongside clinical topics such as “Marie-Tooth disease” and “sensory neuropathy.” Additionally, methodological themes, including “reliability” and “natural history,” warrant attention. The keyword co-occurrence network illustrates the interconnected nature of these research topics, with key nodes such as “mutations” and “neuropathy” functioning as central hubs. Earlier research congregates around broader topics like “gene” and “protein,” whereas more recent studies focus on specific subfields, including “phenotype” and “diagnosis.” The timeline analysis reveals a chronological shift in focus, with darker hues representing earlier studies and lighter shades denoting newer topics (Fig. 7A and B). Citation burst analysis underscores the evolving trends within the field. Recent bursts, including “variants” and “DNA methylation,” indicate emerging research areas in 2022, while earlier bursts, such as “sensory neuropathy” and “mitofusin 2 mutations,” highlight foundational topics that have significantly influenced ongoing investigations (Fig. 7C). This progression reflects an increasing specificity and refinement in research focus, propelled by advancements in molecular and clinical methodologies.

Figure 7.

Keyword co-occurrence and burst keyword. (A) Visual analysis of keyword co-occurrence network. (B) Visual analysis of keyword co-occurrence network and timeline change. (C) Top 20 keywords with the strongest citation bursts.

4. Discussion

This study systematically analyzed global research trends and emerging hotspots in CMT using a bibliometric approach. The findings indicate a steadily growing body of literature, with research activity concentrated in several key countries, institutions, authors, and journals. The leading position of the USA may be explained by its high publication volume, citation impact, and central role in international collaboration networks, reflecting the presence of established neuromuscular research centers, rare disease programs, and long-standing multicenter collaborations. The UK, although producing fewer articles than the USA, showed high average citations and a high MCP ratio, suggesting that internationally collaborative studies may enhance scientific visibility and impact. China and Italy underscore global interest in CMT research, with China contributing substantially to publication volume and Italy maintaining an important role in clinically oriented and collaborative studies.

At the institutional and author levels, the prominence of the University of London, University College London, Sungkyunkwan University, Reilly Mary R., Shy Michael E., Pareyson Davide, and Choi Byung-Ok may reflect sustained access to well-characterized patient cohorts, expertise in inherited neuropathies, and contributions to genotype–phenotype correlation, clinical classification, and translational research. These patterns are consistent with previous studies emphasizing the importance of genetic heterogeneity, clinical characterization, and collaborative research in CMT. Similarly, the concentration of publications in specialized journals such as the Journal of the Peripheral Nervous System and Neuromuscular Disorders indicates that disease-specific journals remain central platforms for CMT research, whereas publication in broader journals reflects the relevance of CMT to neurogenetics and peripheral nerve biology. Overall, leadership in this field appears to be shaped not only by publication volume but also by citation impact, international collaboration, specialized clinical resources, and methodological expertise.

4.1. Highly cited literature

The analysis of highly cited literature is important because it helps identify publications that have shaped the knowledge base and research direction of the CMT field. However, this analysis also revealed a methodological issue related to abbreviation ambiguity. Two highly cited studies by Stroud et al^[23] and Dubin et al^[24] were retrieved because they used the abbreviation CMT2 to refer to chromomethylase 2 in Arabidopsis thaliana, rather than CMT disease type 2. Although these articles had high citation counts, their biological content is unrelated to inherited peripheral neuropathy and should not be interpreted as influential CMT disease literature. Therefore, we did not use these studies to infer biological mechanisms, therapeutic targets, or epigenetic contributions to CMT disease. Their presence instead highlights the importance of manual screening when broad abbreviations such as “CMT1” and “CMT2” are included in bibliometric search strategies.

After excluding abbreviation-related noise, the highly cited CMT-relevant studies mainly reflected several core directions in the field. In 2015, Fridman et al^[25] analyzed clinical and genetic data from 1652 patients within the Inherited Neuropathy Consortium and characterized major CMT subtypes, including CMT1A and CMT2A. This study is highly influential because it provided large-scale evidence on the genetic distribution and clinical burden of CMT, supporting future natural history studies and trial design. In 2016, Morrow et al^[26] evaluated magnetic resonance imaging biomarkers for tracking disease progression in neuromuscular disorders, including CMT type 1A. This work contributed to the development of objective outcome measures, which are essential for monitoring disease progression and evaluating therapeutic efficacy in clinical trials. In 2018, Nicolas et al^[27] identified KIF5A as a gene associated with amyotrophic lateral sclerosis and further linked KIF5A mutations to CMT type 2 and hereditary spastic paraplegia. This finding expanded the phenotypic spectrum of KIF5A-associated disorders and emphasized the shared mechanisms between inherited neuropathies and related neurodegenerative diseases, particularly axonal transport dysfunction. The highly cited literature analysis suggests that influential CMT research is concentrated around large international cohorts, genotype–phenotype correlation, disease progression biomarkers, and genes connecting CMT with overlapping neurological disorders. At the same time, the inclusion of non-CMT records caused by abbreviation ambiguity underscores the need for careful manual validation in bibliometric studies.

4.2. Emerging keywords and research hotspots

Keyword co-occurrence, burst detection, and overlay visualization jointly suggested that CMT research has gradually shifted from broad disease description toward genetic diagnosis, genotype–phenotype correlation, disease monitoring, and trial-oriented outcome assessment. In the present study, traditional keywords such as “mutation,” “neuropathy,” and “Charcot-Marie-Tooth disease” showed high occurrence frequencies and occupied central positions in the co-occurrence network, indicating that genetic etiology and peripheral neuropathy remain the conceptual foundation of this field. In contrast, more recent burst keywords, including “variant,” “whole-exome sequencing,” “whole-genome sequencing,” “natural history,” “outcome measure,” “Schwann cell,” and “myelin sheath,” appeared mainly in the later period of the overlay map and showed burst activity, suggesting emerging research priorities.

Among these hotspots, “variant” showed a high occurrence frequency and was located close to keywords related to genotype–phenotype correlation and genetic diagnosis . This pattern indicates that the field is moving beyond gene discovery toward more refined variant interpretation and clinical stratification. This interpretation is supported by recent CMT studies on GJB1-related CMTX1.^[28,29] Record et al^[29] analyzed an international cohort of patients with GJB1 variants and improved variant classification, genotype–phenotype correlation, and longitudinal assessment. These findings are directly relevant to the bibliometric signal because accurate variant interpretation is essential for diagnosis, prognosis, patient stratification, and future trial design.^[30–32]

Epigenetics-related terms, including “DNA methylation,” emerged as recent keywords in the analysis. This suggests that epigenetic regulation may be an emerging but still relatively limited topic in CMT research. However, the current evidence should be interpreted cautiously. Some cited studies discuss epigenetic regulation, DNA methylation, histone modification, or ribonucleic acid-mediated mechanisms in broader neurological disease models rather than directly in CMT. Therefore, these findings should be described as potential or emerging mechanisms rather than established drivers of CMT pathogenesis. Where model organisms such as Drosophila melanogaster are used, their relevance to CMT should be considered indirect unless the study specifically models CMT-related genes or phenotypes.^[33–37] The appearance of these terms may indicate growing interest in mechanisms that could contribute to phenotypic variability beyond coding mutations, but further CMT-specific validation is required.

Keywords related to cellular and subcellular pathology, such as “Schwann cell,” “myelin sheath,” and “axon,” reflected sustained interest in the biological structures most directly affected in demyelinating and axonal forms of CMT. Rather than providing a broad summary of clinical guidelines, these bibliometric findings suggest that mechanistic studies of Schwann-cell dysfunction, myelin maintenance, and axonal degeneration remain central because they connect genetic defects with clinical phenotypes and potential therapeutic targets.^[38,39] In addition, the co-occurrence of these terms with “next-generation sequencing” and “diagnosis” indicates that mechanistic classification is increasingly integrated with molecular diagnosis, particularly after PMP22 duplication testing and broader genetic sequencing approaches.^[8,40]

Outcome-related keywords also emerged as an important hotspot. “Quality of life,” “outcome measure,” and “natural history” were mainly located near clinically oriented keywords. This indicates increasing attention to patient-centered evaluation and clinical trial readiness.^[41,42] The development and validation of the pediatric Charcot–Marie–Tooth disease quality of life measure by Ramchandren et al^[41] directly supports this trend, as it provided a disease-specific patient-reported outcome for children with CMT across international study sites. Together with longitudinal clinical assessment tools, these studies suggest that the field is moving toward standardized outcome measurement, which is necessary for future interventional trials.^[43,44]

Sequencing-related keywords represented another major recent hotspot. “Whole-exome sequencing” showed frequent occurrences, whereas “whole-genome sequencing” appeared more recently in the overlay map. These findings indicate a transition from targeted gene testing and whole-exome sequencing toward more comprehensive genomic approaches. CMT studies using whole-genome sequencing have shown additional diagnostic value in patients who remain unsolved after PMP22 duplication screening and whole-exome sequencing, particularly for structural variants, noncoding variants, and regions with insufficient exome coverage.^[45–47] This hotspot therefore reflects a methodological shift in the field: improving diagnostic yield, resolving atypical cases, and expanding the known genetic architecture of CMT. The keyword evolution suggests that recent CMT research hotspots are concentrated in 4 main directions: refined genetic variant interpretation, cautious exploration of epigenetic or regulatory mechanisms, integration of cellular pathology with molecular diagnosis, and development of sequencing strategies and outcome measures for trial readiness. These trends indicate that the field is moving from descriptive classification toward precision diagnosis, longitudinal monitoring, and translational research.^[7]

4.3. Research significance and limitations

This study maps the global CMT research landscape from 2014 to 2024, identifying major contributors, collaboration patterns, influential journals, highly cited literature, and emerging hotspots. These findings may help new researchers understand key themes and assist established researchers in identifying collaboration opportunities and developing topics. However, several limitations should be noted. First, the use of only WoSCC and English-language records may have introduced database and language bias, potentially underestimating contributions from non-English studies or literature indexed only in other databases. Second, raw citation counts may be affected by citation-age bias, meaning older publications may appear more influential simply because they have had more time to accumulate citations. Third, abbreviation ambiguity involving terms such as “CMT1” and “CMT2” may have introduced false-positive records, including plant CMT2 or cross-disease literature, which could affect citation and keyword analyses. Fourth, author and institutional name disambiguation may influence productivity and collaboration rankings. Fifth, results may vary according to threshold settings, keyword merging, and pruning parameters. Finally, because findings were not cross-validated with Scopus or PubMed, their generalizability may be limited. Future studies should integrate multiple databases, apply citation normalization, and test the robustness of different bibliometric parameters.

5. Conclusion

This bibliometric analysis mapped global CMT research from 2014 to 2024 by evaluating publication trends, collaboration networks, influential contributors, journals, citation patterns, and keyword evolution. Beyond confirming the steady growth of CMT-related publications, this study provides several bibliometric insights: CMT research output and influence remain concentrated in a limited number of countries, institutions, and authors, suggesting that international collaboration networks and specialized neuromuscular centers play an important role in shaping the field; highly cited literature and keyword evolution indicate a gradual shift from broad genetic characterization toward more applied themes, including variant interpretation, sequencing-based diagnosis, disease monitoring, and outcome measurement; and abbreviation-related retrieval noise, particularly involving “CMT2,” highlights the need for careful manual validation in bibliometric studies of diseases with commonly used abbreviations. These findings may help researchers identify influential collaborations, core journals, and emerging topics with translational potential. Future studies should integrate multiple databases, refine search strategies to reduce abbreviation bias, and pay greater attention to underrepresented regions and less-studied CMT subtypes.

Author Contributions

Conceptualization: Hongshi Fan.

Writing – original draft: Hongshi Fan, Yan Wang.

Data curation: Dongliang Xiang.

Formal analysis: Dongliang Xiang.

Writing – review & editing: Fei Pei, Yan Wang.