In this study, we searched the Web of Science Core Collection (WOSCC) database, which contains a large number of documents in the biomedical field. The search used the following formats: TS = (microRNA*) AND TS = (Spinal cord injury). The time span was January 1, 2008 to January 31, 2024. After carefully searching the existing literature, we screened out 1,087 potential articles. We limited the types of articles to Article and limited the language to English, finally including 754 articles (Figure 1).

Search results are downloaded in a “plain text file” format, with complete records and references in the WOSCC database. Data are extracted after careful screening and importing into Microsoft Excel 2016 for analysis: annual publications, citation numbers, countries, institutions, authors, journals, highly cited articles, co-cited articles, keywords, and H-index. The summarized data are imported into the online bibliometrics analysis platform,² and a fitting curve is used to predict the number of published documents.

To analyze the bibliometrics of SCI and miRNA, we visually analyzed selected data. These data clarified research trends, influence, and distribution. This paper used VOSviewer software for visual analysis of authors, keywords, keyword co-occurrence timeline, density maps, journals, highly cited literature, and co-cited literature. CiteSpace software was used for detailed visualization of countries/regions cooperation network map, institutions cooperation network map, strongest citation burst, keyword clustering, journal dual graph overlay, etc. The node size represents the number of nodes; the lines between nodes represent cooperation between institutions, with thicker lines indicating closer relationships. From these results, inferences about relationships between research hotspots and frontiers were made.

Arrowsmith (see text footnote 1) was utilized to identify relationships between miRNAs and SCI and evaluate potential miRNA mechanisms in SCI. Keywords filtered through Arrowsmith served as prediction groups, while those filtered through VOSviewer were confirmation groups. Venn diagrams were drawn for both to determine potential keywords for miRNA and SCI research. R software was used for gene ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of potential genes with correlation capacity ≥0.95. The STRING online database³ was used to create a potential gene protein–protein interaction (PPI) network, visualized with Cytoscape software. Cytoscape software also aided in screening genes within the PPI network.

Analyzing publication trends enables understanding of the developmental speed and changing attention to miRNAs in SCI treatment (Figure 2). From 2008 to 2024, 754 papers were published. Annual publication numbers rapidly increased from 2013 to 2019, peaking in 2019 (n = 102). Publications in 2019–2022 were relatively flat, but decreased significantly in 2023 (n = 55). With only January 2024 included, it is currently the year with the fewest publications (n = 6). The function formula of the fitting curve is “y = 1.2421x^2.265, R² = 0.9874,” indicating the research growth curve in this field fits well. These results show studying miRNAs is an important direction in SCI.

We found that a total of 50 countries/regions have published literature on the role of miRNAs in SCI, with China being the most prominent country, accounting for the largest number of publications (n = 527; 59.48%) and citations (n = 10,155; 40.32%) (Figures 3A,B; Table 1). With the globalization of knowledge and technology, international cooperation is increasingly close, and the United States plays an important role in the cooperation network. European countries have the closest cooperation (Figure 3C). Overall, although China has the largest number of publications, it still lacks international cooperation and needs to strengthen exchanges in the future.

Papers in this field have been published by 802 institutions. Most of the top 10 universities in publication count and citation count are from China. China Medical University published the most articles (n = 40; 2.67%) and received the most citations (n = 891; 2.26%), ranking first in both categories (Figures 4A,B; Table 2). Johns Hopkins University ranks first in centrality (coefficient = 0.05), while Chinese research institutions rank relatively high, showing that Chinese universities have made remarkable achievements and great influence in the field of miRNAs related to SCI, but there is still room for improvement (Figure 4C). Nevertheless, analyzing the scale and degree of inter-institutional collaboration reveals that institutions with more publications did not extensively collaborate with each other (Figure 4D).

Related literature was published in 278 journals, with the most published article in European Review for Medical and Pharmacological Sciences (n = 25; 6.4%), the most cited journal in Experimental Neurology (n = 715; 6.9%) (Figures 5A,B; Table 3), and the highest H-index in European Review for Medical and Pharmacological Sciences (Figure 5C). Considering journal classification and influence factors, research quality needs further improvement. Analysis of superimposed journal graphs shows yellow bars indicate articles in molecular/biology/immunology journals often cite articles in molecular/biology/genetics journals (Figure 5D). Per cooperative network analysis, journals fell into four clusters, with the large node/density of European Review for Medical and Pharmacological Sciences, Frontiers in Molecular Neuroscience, and Neural Regeneration Research indicating these journals’ important roles in the field (Figures 5E,F).

A total of 3,867 authors have published relevant papers, with Ma, H publishing the most papers (n = 11, 3.63%) and also being the most cited author (n = 357, 4.68%), indicating extensive research in this field (Figures 6A,B; Table 4). Wang, Y, with the highest H-index, is the most influential author (Figure 6C). Many academic groups are active in this field, such as Ma, H, Feng, SQ, He, XJ, Lu, HB, etc. Authors within groups collaborate closely, but inter-group collaboration is insufficient (Figure 6D).

Table 5 shows the top 10 cited publications, led by Xin HQ’s article in Stem Cells, with 521 citations. It reported Mir-133b promotes neural plasticity and recovery post-stroke treatment via transfer of exosome-enriched particles. Liu NK’s article in Experimental Neurology, “Altered microRNA expression post-traumatic spinal cord injury,” was cited 153 times (Table 6). VOSviewer software is used to draw the visual map and visual density view of the cited documents and co-cited documents in order to obtain better visual effect. Only references with citation frequency ≥ 40 are shown in the cited chart. Xin (2013) was cited by more literatures (Figures 7A,B). Only references with citation frequency ≥ 30 are displayed in the co-cited reference map. Liu NK (2009) and Bartel Dp (2004) have attracted wide attention (Figures 7C,D).

Keywords are words that have a high degree of generalization of the overall research content of an article. Through co-occurrence analysis of 3,151 keywords in all literatures, it was found that research mainly focuses on SCI, expression, apoptosis, microRNAs, etc. (Figure 8A). The LLR algorithm was used to perform cluster analysis on all keywords, drawing a cluster map with a total of nine clusters. The smaller the classification number, the more keywords that classification contains. #0 neuropathic pain, #1 central nervous system; each tag is interrelated and developed, rather than existing independently (Figure 8B). According to keyword co-occurrence timeline chart of the references, this development and change can also be confirmed from 2018 to 2019, SCI related research content is the richest (Figure 8C).

Emergent words refer to keywords that appear within a short time or are used frequently reflecting a change in research direction within a field. Based on keyword intensity and duration research direction can be roughly divided into three stages: (1) 2008–2015 focusing on changes in RNA and related gene expression post spinal cord injury and stem cell therapy; (2) 2015–2018 focusing on peripheral nerve injury treatment and exploring in vivo SCI treatment in mice; and (3) 2018–2024 primarily studying extracellular vesicles for SCI repair and neuroinflammation alleviation. This development and change is also confirmed by the co-citation time chart of references. Keyword co-occurrence timeline reflects the richest SCI research content from 2018 to 2019 (Figure 8D).

We selected the keywords from the Arrowsmith project as the prediction group (n = 2,779) and the keywords extracted by VOSviewer as the confirmation group (n = 728). A Venn diagram showed 204 common keywords, which can serve as potential research directions for studying miRNAs in SCI (Figure 9A). We extracted genes corresponding to these potential keywords for Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. The GO analysis showed biological processes (BP) focused on cell responses to hormones, oxidative stress, phosphorus metabolism, growth factors, and oxidative stress. Cellular components (CC) were mainly concentrated in receptor complexes, transcriptional regulatory complexes, and vesicle cavities. Molecular functions (MF) were remarkably rich in transcription factor binding and kinase binding (Figure 9B). KEGG pathway enrichment analysis showed the PD-1/PD-L1 signaling pathway played a certain role (Figure 9C). A Protein–Protein Interaction (PPI) network was constructed and key genes were screened. It was found that AKT, MAPK, FOXO, and SMAD were potential targets related to miRNA treatment of SCI, indicating these genes played important roles in regulating cell pathophysiological processes. They are key players in SCI treatment (Figures 9D,E).

Spinal cord injury is a nervous system disease causing physical and psychological harm to patients and a heavy economic burden on society (17). It can be divided into primary and secondary injury. Primary injury is typically mechanical, while secondary injury can lead to insufficient perfusion of gray matter post-injury and spread to surrounding white matter (18, 19). Injury-induced oxidative stress, inflammation, glial or fibrotic scar formation, demyelination, and nerve injury can result in partial paralysis (20). Current research focuses on neuroprotection, neurorepair, and scar inhibition therapy. Neuroprotective therapies focus on reducing/preventing secondary damage (6). The goal of nerve repair therapy is to reshape neurovascular recovery of the nervous system (21). Scar inhibition is a potential strategy to control scar growth and promote axonal regeneration by inhibiting astrocyte proliferation and hypertrophy (22, 23). Current clinical treatments mainly involve surgical decompression and drug therapy. Surgical decompression can improve spinal cord compression, increase blood flow at the injury site, and reduce expansion. However, surgical treatment can lead to greater trauma and can only provide symptomatic relief rather than a fundamental cure (24, 25). Injectable drug therapy can also cause immune rejection, as well as excessive drug release and rapid degradation (26). In the context of medical treatment, the central objective is to mitigate hindrances to axon growth, ensuring a microenvironment that is conducive to nerve development. As part of this, there is an ongoing exploration of strategies that could potentially rebuild damaged neural circuits. Simultaneously, there is a concerted effort to promote functional recovery, enhancing the overall quality of life for patients. Lastly, there is a continuous search for suitable grafts that could support and guide the process of axon regeneration, ensuring optimal outcomes, miRNA-based therapy, blood vessel intervention and combination of multiple treatments will play a key role in the treatment of SCI in the future (27).

MicroRNAs are considered important biomarkers and therapeutic targets in SCI development (28). They exhibit tissue specificity and stability; miRNA content changes post-SCI are prerequisites for biological effects (29). MiRNAs serve as pathological biomarkers for Alzheimer’s disease, epilepsy, brain injury, and other conditions (30, 31). Bioinformatics analyses show abnormal miRNAs’ potential target genes post-SCI involve SCI pathogenesis, including inflammation, oxidative stress, apoptosis, and neuronal death (32). They also impact astrocytes and scar formation (33). Research on miRNAs and SCI can inform SCI diagnosis and treatment, but miRNAs’ pathophysiological significance remains incompletely determined. We use various bibliometrics tools for quantitative analysis, citation analysis, and visualization to reveal research hotspots and prospects. Compared to traditional review and meta-analysis, our approach explores research trends across multiple dimensions, providing richer, deeper, and more diversified information. It offers new reference and direction for SCI treatment and development.

Bibliometrics can systematically classify and analyze literatures on different topics, enabling better understanding of correlations between hot spots, trends, and topics within a research field (34). This study applies bibliometrics methods to examine literature on miRNAs’ role in SCI restoration. From 2008 to 2022, related literature rose yearly, totaling 754 articles. Publications remained high and entered rapid growth. Involving 50 countries/regions, 802 institutions, 278 journals, and 3,867 authors, indicating sustained researcher interest and a hot topic in SCI restoration.

China currently has clear advantages in terms of national paper count, citation count, institutional ranking, and author ranking. However, cooperative network analysis shows the United States occupies an important position in the cooperative network, with high H-index for research institutions, indicating the United States also plays an important role in promoting related work. China’s overall link strength in the cooperation network remains insufficient; international exchanges and cooperation should be strengthened. Ma, H and Feng, SQ from China are field leaders and may continue to hold leadership positions, but their research teams are primarily domestically based. Journal research shows European Review for Medical and Pharmacsciences is the most published journal, also with the highest H-index, while Experimental Neurology is the most cited. Considering journal division and impact factors, research quality in this field needs further improvement. Chinese research institutions and scholars can advance miRNA research in SCI recovery by strengthening international cooperation, introducing advanced methods and technologies, critically evaluating and improving research papers, and actively participating in academic conferences. Such efforts can enhance academic level and reputation, and contribute important scientific achievements to the field.

Cited and co-cited documents refer to research in which one document cites another, and multiple documents cite the same one, respectively (35). Research references help scholars understand hotspots and frontiers in a field, identify key documents with important value, reveal academic networks, and strengthen knowledge dissemination. Combining literature contents in Tables 5, 6, we found inflammation, neuron protection, axon regeneration, scar formation, and specific mechanisms post-SCI remain research foci.

Keyword analysis extracts keywords or key phrases through processing and statistically analyzing large document, article, or data sets, enabling understanding of research topics, hotspots, and basic content. This study’s keywords are spinal cord injury, expression, apoptosis, and miRNAs. Changes in miRNA gene expression lead to increased apoptosis and improved SCI. This strategy remains a hot research topic. miR-7a treats SCI by inhibiting neuronal apoptosis and oxidative stress (36), while miR-487b inhibits inflammation and neuronal apoptosis in SCI by targeting Ifitm3 (37). Based on keyword cluster analysis, nerve injury closely relates to inflammation. Currently, combined with keyword co-occurrence timeline and Reference citation bursts, keywords gradually transition to extracellular vesicles and neuroinflammation. Long et al. reported astrocyte-derived exosomes rich in miR-873a-5p inhibit neuroinflammation post-traumatic brain injury through microglial phenotypic regulation (38). Future directions include exosomes carrying miRNAs to inhibit inflammation for SCI treatment.

According to GO analysis and KEGG pathway, miRNAs actively participate in pathophysiological activities post-SCI. The PD-1/PD-L1 signaling pathway is involved in inhibiting neuroinflammation and improving traumatic brain injury (39). PPI network predicts AKT, MAPK, and FOXO as important pivotal genes, potentially key in apoptosis, autophagy, and inflammation. Currently, these genes are widely studied in SCI research. miR-21-5p, miR-92b-3p, and miR-34a all inhibit neuronal apoptosis and promote nerve regeneration via AKT (40–42). Excessive miR-340-5p reduces spinal cord injury-induced neuroinflammation and apoptosis by modulating P38-MAPK signaling (43). FOXO is widely reported in anti-inflammatory and anti-cancer studies (44, 45). SMAD plays an important role in SCI remyelination (46). Elucidating miRNA-target gene regulatory mechanisms may achieve reliable therapeutic effects and more comprehensive, effective rehabilitation programs for SCI patients.

At present, effective clinical treatment methods for SCI remain lacking, and emerging treatment methods such as cell transplantation, epigenetic regulation, artificial scaffold transplantation, and gene therapy are the main research directions for researchers (47, 48). Secondary injury usually leads to changes in the spinal cord microenvironment, resulting in changes in miRNA expression (49). This discovery provides potential opportunities for gene intervention (50). MiR-33-5p alleviates rat SCI and protects PC12 cells from lipopolysaccharide-induced apoptosis (51). MiR-433-5p overexpression protects against acute spinal cord injury by activating MAPK1 (52). MiR-7a improves spinal cord injury by inhibiting neuronal apoptosis and oxidative stress (36). Although the use of miRNAs as a therapeutic agent has attracted considerable attention due to its therapeutic effect on SCI, The limitations of using miRNAs as a therapeutic tool lie in their delivery and activation in vivo (29).

Exosomes, as intercellular communication carriers, play an irreplaceable role in maintaining multicellular organism functional integrity. Protected by a phospholipid layer, miRNAs can be accurately transported to target sites and avoid hydrolysis by extracellular enzymes, enhancing blood–brain barrier penetration and preventing misdelivery. Jiang et al. (53) confirmed neuron-derived exosome-delivered miR-124-3p protects spinal cord from traumatic injury by inhibiting neurotoxic microglia and astrocyte activation. Liu et al. (54) demonstrated hypoxic-preconditioned mesenchymal stem cell-derived exosome shuttle miR-216a-5p repaired traumatic spinal cord injury by altering microglia M1/M2 polarization. Recent miRNA delivery technology advances raise high expectations for miRNA therapeutics. While no clinical trials yet use exosome-attached miRNAs to treat human SCI, experimental data strongly support this approach’s feasibility. We must further analyze its beneficial effects and potential risks before applying it clinically.