Trends and hotspots in the studies of spinal metastasis: A bibliometric analysis from 2000 to 2023

Abstract

Background:

Spinal metastasis (SM) is described as a metastatic malignant bone tumor with a high mortality rate and often leads to symptoms related to spinal cord or nerve compression, such as pain and debilitating neurological dysfunction. In this study, we explored the research hotspots and trends in SM using bibliometric analysis, which provided reliable novel hints and pathways for future exploration.

Methods:

We reviewed articles and reviews on SM published in the Web of Science Core Collection between 2000 and 2023. VOSviewer and CiteSpace were used to conduct the bibliometric and knowledge map analyses.

Results:

A total of 2325 original articles and reviews published in 489 academic journals by 10,460 authors from 2231 affiliations in 61 countries/regions were retrieved. The United States was the largest contributor. The University of Toronto was the leader in relevant research. Arjun Sahgal was the most published author and Peter C Gerszten had the most co-citations. The Journal of Neurosurgery Spine published the most SM related articles, and Spine was the most commonly cited journal. The preeminent areas of scholarly inquiry concerning SM were centered on minimally invasive treatments, radiation therapy, and the prognosis and management of patients with SM.

Conclusion:

SM is a common type of metastatic bone tumor, and its treatment is increasingly shifting towards minimally invasive surgery and radiation therapy, with a growing emphasis on the prognosis and management of SM as a key area for future research.

1. Introduction

Spinal metastasis (SM) represents a prevalent form of metastatic malignancy, affecting a considerable proportion of cancer patients, ranging from 20% to 40%, over the course of disease progression. Among these individuals, up to 20% develop symptomatic complications that are attributable to spinal cord compression.^[1] The incidence of symptomatic SM is expected to increase concomitantly with the increasing elderly population and advancements in cancer management. Progression in surgical methodologies and refinements in diverse treatment modalities have typically led to enhanced overall survival outcomes for patients diagnosed with SM.^[2] Minimally invasive surgery and stereotactic radiation therapy have resulted in substantial advancements in the treatment of SM.^[3] Implementation of a multidisciplinary approach aimed at optimizing patient outcomes in the management of spinal metastases is crucial.^[4] The diagnosis and treatment of SM are advancing toward greater maturity, underscoring the critical need for researchers to enhance the efficiency and directionality of their investigations. Therefore, a bibliometric analysis is essential to fully assess the current state of research, identify emerging areas of interest, and outline future prospects in the SM field.

The term “bibliometrics” is commonly credited to Pritchard, who introduced it in 1969.^[5] After more than half a century, bibliometrics has evolved into a more sophisticated discipline. Bibliometrics involves measuring “document properties and document-related processes.”^[6] Its research scope includes analyzing word frequency, citations, co-occurrences, and various document counts, such as publications by authors, research groups, or countries, among other metrics.^[7] CiteSpace is highly effective for visualizing information and providing quantitative and visual insights into specific domains. VOSviewer, developed by Eck and Waltman, is a freely available program known for its robust capabilities in co-occurrence and co-citation analyses.^[8] Currently, few bibliometric studies have focused on SM. Thus, the purpose of this study is to utilize VOSviewer and CiteSpace to analyze SM research, aiming to illuminate the present status and research hotspots within this field.

2. Materials and methods

2.1. Data sources and search strategies

We obtained bibliographic data using the Web of Science Core Collection (WOSCC) database, specifically, the Science Citation Index Expanded. WOSCC is one of the most important and reliable databases in bibliometric analysis, covering the vast majority of the other databases, and the vast majority of current bibliometric analyses are conducted on WOSCC. To minimize the potential impact of rapid database updates, all publications from 2000 to 2023 were systematically retrieved and downloaded from the Science Citation Index Expanded of the WOSCC database on February 28, 2024. The search terms were as follows: (((TS= (“spinal metastases”)) OR TS= (“vertebral metastases”)) OR TS= (“spinal metastasis”)) OR TS= (“vertebral metastasis”). The research encompassed original articles and reviews specifically authored in English; early access, meeting abstracts, editorial materials, letters, proceedings papers, corrections, news items, and retracted publications were excluded. Eventually, 2325 studies were retrieved, and the data were preserved and archived in a textual format.

2.2. Data collection and cleaning

Original data were extracted from the WOSCC. The data encompassed various parameters, including the number of publications (NP), citations, H-index, publication year, authors, affiliations, countries/regions, journals, references, and keywords. VOSviewer v.1.6.20 (Centre for Science and Technology Studies, Leiden University, Leiden, The Netherlands) and CiteSpace 6.2.R4 (Chaomei Chen, supported by Drexel University, College of Information Science and Technology, Philadelphia) were used to analyze the collected data.

2.3. Bibliometric analysis

The NP was used to evaluate production and the number of citations, excluding self-citations (NC), served as a measure of impact. The H-index was used to assess the contribution to the field of research and forecast forthcoming scientific advancements. The H-index, proposed by Hirsch, offers a straightforward and reliable method for comparing the scientific productivity and visibility of institutions. This is defined as the number of papers with a citation count equal to or greater than H.^[9] The co-occurrence of keywords identifies the most prevalent keywords in the same literature and co-citation occurs when 1 item references 2 items.^[10] The analysis of co-occurring keywords alongside the co-cited literature provides insights into the research hotspots pertaining to SM. Burst Detection, an algorithm devised by Kleinberg in 2003,^[11] serves as an efficient analytical tool for detecting sudden surges in the popularity of a quote or keyword within a defined timeframe. This functionality enables the swift identification of concepts or topics that have experienced heightened discussion activity during the specified period. In the graph, the blue line represents the overall timeframe, and the red line highlights the specific period during which the reference burst has occurred. Bibliometric maps were generated using VOSviewer and CiteSpace to obtain a more comprehensive analysis based on co-occurrence and co-citations.

3. Results

3.1. Publication overview

In this study, we retrieved a total of 2325 publications published from 2000 to 2023 through the retrieval method, including 1955 original articles and 370 reviews. The total NC was 21,242, the H-index of all these publications was 97, and the average NC per item was 24.34.

3.2. Annual trend of publication quantity

Changes in NP over time can objectively indicate the overarching developmental trajectory of an investigation. The number of studies published on SM between 2000 and 2023 correlated with the year of publication. A correlation coefficient (R²) of 0.8921 denoted a robust positive association between the publication year and number of studies published on SM, as illustrated in Figure 1A. The annual NP exhibited a generally increasing trend and peaked in 2020 (Fig. 1B). The NP growth rates remained relatively stable over the study period, indicating consistent development in SM research. The annual publication quantity trend is illustrated by a polynomial fitting curve (Fig. 1A), which demonstrates that SM research has progressed over the years.

Figure 1.

(A) Curve fitting of the total annual growth trend of publications. (B) Number of publications by year from 2000 to 2023.

3.3. Contributions of countries/regions

The publications were from 61 countries/regions. Table 1 delineates the 10 most influential countries/regions according to publication count, citations, and H-index. The United States had the highest NP (772), followed by China (378) and Germany (210). The United States, Canada, and Japan ranked the highest in terms of citation counts, with figures of 23,245, 8493, and 6373, respectively. Similarly, the United States, Canada, and Germany held the top 3 positions for the H-index, with values of 80, 50, and 41, respectively. Figure 2A illustrates the country’s co-authorship network, with the United States emerging as the leader among all countries/regions, followed by China, Germany, Canada, and Japan. Figure 2B illustrates the overlay visualization map of country/region co-authorship; among the countries/regions leading the research, China was relatively late in the development of SM research. Figure 2C shows the density visualization for the countries/regions in the co-authorship network map, where yellow indicates a higher frequency than blue. Figure 2D and E illustrates the annual NP of the 10 countries with the highest influence. In Figure 2D, the size and color of the circles are representative of the NP values, and the magnitude of the area reflects the number of yearly publications in Figure 2E. Figure 2F shows the regional distribution of the annual publications in different countries/regions.

Table 1.

Top 10 productive country/region in the field of SM studies.

Rank	Country	NP	NC	H-index	Average per item
1	USA	772	23,245	80	34.79
2	China	378	3502	31	10.14
3	Germany	210	5819	41	28.77
4	Canada	209	8493	50	43.97
5	Japan	202	6373	36	33.35
6	Italy	152	3857	32	26.42
7	England	125	3723	36	30.78
8	South Korea	116	2179	29	19.62
9	France	115	3065	31	27.77
10	Netherlands	96	3394	35	37.02

NC = count of citations, excluding self-citations, NP = number of publications, SM = spinal metastasis.

Figure 2.

(A) Co-authorship network visualization map of countries/regions related to SM. (B) Co-authorship overlay visualization map of countries/regions related to SM. (C) Co-authorship density visualization map of countries/regions related to SM. (D) Annual output trend of the top 10 productive countries/regions (the colors and size of the circles represent NP). (E) Stacked area chart of the top 10 productive countries/regions. (F) Geographic distribution of studies on SM from 2000 to 2023. NP = number of publications, SM = spinal metastasis.

3.4. Analysis of affiliations

A total of 2231 affiliations have contributed to research in this field, and 151 of these institutions published 7 or more articles. Table 2 illustrates the 10 most prominent affiliations in terms of published articles: The University of Toronto had the highest NP (159), followed by Sunnybrook Health Science Center (113), and the University of Texas System (112). The University of Toronto had the highest NC (6758) and H-index (46). All 10 affiliations were from the United States and Canada. Figure 3A illustrates the affiliations’ co-authorship network visualization map. Figure 3B shows the overlay visualization map of co-authorship affiliations; affiliations represented in blue conducted SM research earlier than those represented in yellow. Figure 3C presents a density visualization map for the affiliation network. Green indicates an infrequent appearance, whereas yellow indicates a higher frequency. Figure 3D shows a timeline view of the affiliation co-authorship cluster, which was divided into 13 clusters. Smaller numbers indicate that more affiliations are included, and #0 photodynamic therapy (PDT) ranks 1st. Figure 3E shows the top 20 affiliations with the highest occurrence of bursts.

Table 2.

Top 10 productive affiliations in the field of SM studies.

Rank	Affiliation	NP	NC	Country	H-index	Average per item
1	University Of Toronto	159	6758	CANADA	46	46.08
2	Sunnybrook Health Science Center	113	4114	CANADA	36	40.25
3	University Of Texas System	112	5815	USA	39	54.07
4	Johns Hopkins University	101	4433	USA	36	45.62
5	Memorial Sloan Kettering Cancer Center	96	5630	USA	37	62.23
6	Utmd Anderson Cancer Center	95	5455	USA	38	59.66
7	Harvard University	91	3300	USA	28	37.9
8	University Health Network Toronto	88	4251	CANADA	33	50.93
9	University OF California System	74	3042	USA	28	41.86
10	Harvard Medical School	64	1548	USA	22	25.83

NC = count of citations, excluding self-citations, NP = number of publications, SM = spinal metastasis.

Figure 3.

(A) Co-authorship network visualization map of affiliations related to SM. (B) Co-authorship overlay visualization map of affiliations related to SM. (C) Co-authorship density visualization map of affiliations related to SM. (D) Timeline view of affiliation co-authorship cluster analysis in the SM related studies. (E) Top 20 affiliations with the strongest citation bursts. SM = spinal metastasis.

3.5. Performance of authors and co-cited authors

A total of 10,460 authors have contributed to this field, with more than 150 authors each contributing to over 8 articles. Table 3 lists the 10 most productive authors and their affiliations and countries. All authors were from the United States and Canada, except for Verlaan (Netherlands). Sahgal (University of Toronto) had the highest number of published articles (NP = 73) and the highest H-index (32). Bilsky (Memorial Sloan Kettering Cancer Center) was the most cited author (NC = 4752). Figure 4A–C illustrates the collaboration patterns among these authors, the duration of each author’s contribution to the field (authors represented in yellow published articles at later time points than those depicted in blue), and the frequency of occurrence for each author (yellow indicates more frequent than blue). From these figures, we can see that Sahgal, Sciubba, and Rhines were the top 3 authors who cooperated the most closely with other authors. Co-cited authors refer to 2 or more authors who are cited together in 1 or more articles, and are related by their co-occurrence. Among the 25,796 co-cited authors, 154 authors had at least 49 citations. The visualization maps of the network and density of the co-cited authors are shown in Figure 4D and E. Gerszten, Tokuhashi, and Sahgal are the 3 most frequently cited authors. The co-cited author cluster timeline view is shown in Figure 4F; #0 stereotactic body radiotherapy, #1 survival, and #2 vertebroplasty were the most popular themes, and the use of carbon fiber in the treatment of SM has become more popular in recent years.

Table 3.

Top 10 active authors in the field of SM studies.

Rank	Author	Affiliation	Country	NP	NC	H-index	Average per item
1	Sahgal, Arjun.	University of Toronto	Canada	73	3300	32	49.29
2	Sciubba, Daniel M.	Johns Hopkins University	USA	62	2317	24	38.71
3	Bilsky, M.	Memorial Sloan Kettering Cancer Center	USA	59	4752	31	84.12
4	Rhines, Laurence D.	University of Texas System	USA	53	3617	29	70.45
5	Gokaslan, Ziya L.	Johns Hopkins University	USA	50	2916	26	59.98
6	Laufer, Ilya.	Memorial Sloan Kettering Cancer Center	USA	49	2498	24	53.63
7	Yamada, Yoshiya	Memorial Sloan Kettering Cancer Center	USA	41	3573	24	90.29
8	Fehlings, Michael G.	University of Toronto	Canada	36	2490	26	70.92
9	Verlaan, Jorrit-Jan.	Utrecht University	Netherlands	35	673	16	20.31
10	Chang, Eric L.	University of Texas System	USA	33	2445	25	76.64

NC = count of citations, excluding self-citations, NP = number of publications, SM = spinal metastasis.

Figure 4.

(A) Co-authorship network visualization map of authors related to SM. (B) Co-authorship overlay visualization map of authors related to SM. (C) Co-authorship density visualization map of authors related to SM. (D) Co-cited network visualization map of authors related to SM. (E) Co-cited density visualization map of authors related to SM. (F) Timeline view of authors co-cited cluster analysis in the SM related studies. SM = spinal metastasis.

3.6. Analysis of journals and co-cited journals

Articles related to SM have been published in more than 489 journals, with more than 120 journals each contributing more than 5 articles. Table 4 shows the 10 most productive journals and displays their impact factor (IF) in addition to other basic items. The most prolific journal was the Journal of Neurosurgery Spine (NP = 117), and the most cited journal was Spine (NC = 5599), which had the highest H-index (34). Neurosurgery had the highest IF (4.8), followed by Frontiers in Oncology (4.7) and Spine Journal (4.5). The journal network visualization map is shown in Figure 5A; the Journal of Neurosurgery Spine, Spine, World Neurosurgery play a pivotal role in this domain of research. Figure 5B shows the journal overlay visualization map related to SM, and the yellow keywords denote a later appearance than the blue ones. The density visualization of the journals is shown in Figure 5C, where each journal depicted in the density visualization is color-coded to represent its density. Blue denotes a lower frequency and yellow signifies a higher frequency. Co-cited journals are 2 or more journals that are cited concurrently by researchers. Figure 5D and E illustrates the co-cited network and journal density visualization map related to SM (the colors indicate the same as in Figure 5D and E). Among the 5363 co-cited journals, more than 150 journals were cited over 76 times. Spine, International Journal of Radiation Oncology Biology Physics, and Journal of Neurosurgery Spine are the 3 most frequently cited journals. These 3 magazines were also the ones most closely associated with the others.

Table 4.

Top 10 productive journals in the field of SM studies.

Rank	Journal	NP	NC	H-index	Average per item	IF (2022)
1	Journal of Neurosurgery Spine	117	4193	34	37.25	2.8
2	Spine	107	5599	34	54.21	3
3	World Neurosurgery	86	884	16	10.48	2
4	European Spine Journal	69	2515	25	37.19	2.8
5	International Journal of Radiation Oncology Biology Physics	54	3643	31	68.89	7
6	Journal of Clinical Neuroscience	52	688	16	13.37	2
7	Spine Journal	50	1645	24	33.86	4.5
8	Neurosurgery	41	1117	19	27.56	4.8
9	Frontiers in Oncology	40	275	8	6.95	4.7
10	Journal of Neuro Oncology	35	919	18	26.4	3.9

NC = count of citations, excluding self-citations, NP = number of publications, SM = spinal metastasis.

Figure 5.

(A) Network visualization map of journals related to SM. (B) Overlay visualization map of journals related to SM. (C) Density visualization map of journals related to SM. (D) Co-cited network visualization map of journals related to SM. (E) Co-cited density visualization map of journals related to SM. SM = spinal metastasis.

3.7. Citation analysis and top 10 most cited articles

A total of 141 documents were cited more than 73 times among the 2325 documents. Figure 6A, B, and C presents the citation network, overlay, and density visualization maps of these documents, respectively. As shown in Figure 6A, all articles were categorized into 8 clusters, with 6 of these clusters containing more than 10 articles each. Cluster 1 (red) predominantly comprised articles focusing on stereotactic body radiation therapy (SBRT) in treating SM, discussing aspects such as treatment efficacy, prognostic assessment, and safety. Cluster 2 (green) included articles primarily addressing the treatment and prognostic evaluation of SM, with topics ranging from surgical interventions to prognostic factors and predictive modeling. Cluster 3 (blue) included articles discussing the applications, innovations, and advancements of SBRT in the treatment of SM. Cluster 4 (yellow) included articles focusing on innovations and improvements in the diagnosis and treatment of SM, with an emphasis on early intervention and treatment strategies. Cluster 5 (purple) includes articles that primarily focused on the treatment and management of SM, including discussions on therapeutic approaches, prognostic evaluations, and multidisciplinary treatment strategies across various domains. Cluster 6 (light blue) included articles that primarily discuss interventional methods and techniques for treating SM and vertebral compression fractures. Figure 6B presents a visualization map of the cited documents. Each circle color signifies the average citation year, with yellow indicating years later than blue. Figure 6C presents a citation network density visualization map. Green indicates lower occurrence, and yellow indicates higher occurrence.

Figure 6.

(A) Citation network visualization map of articles related to SM. (B) Citation overlay visualization map of articles related to SM. (C) Citation density visualization map of articles related to SM. (D) Top 10 most cited articles in the field of SM studies. (Each article is shown by the colors and size of the circle). SM = spinal metastasis.

Table 5 and Figure 6D list the top 10 most cited publications, 9 of which are original articles and 1 a review. The article written by Tomita et al in 2001 and published in Spine ranked 1st with a total NC of 902. This article designed a scoring system for SM based on data consisting of retrospectively evaluated prognostic factors. These factors include malignancy grade, visceral metastases, and bone metastases.^[12] Tokuhashi et al evaluated the prognostic accuracy of this scoring system by comparing predicted survival with actual survival in 118 patients who used the system after 1998 and in 246 patients who were evaluated retrospectively.^[13] Fisher et al developed a spinal instability tumor scoring system using evidence from an expert consensus and systematic reviews. This system predicts spinal stability in neoplastic lesions by considering symptoms, radiographic criteria, and other key factors. It guides surgical consultation decisions and evaluates spinal instability caused by tumors.^[14] Gerszten PC et al demonstrated the potential for improving long-term palliative care in patients with SM treated with radiosurgery by evaluating the clinical improvement in 500 SM patients treated with radiosurgery.^[15] Fourney et al reviewed a group of patients with SM who underwent vertebroplasty (VP) and kyphoplasty (KP), and demonstrated that percutaneous VP and KP can largely relieve pain in these patients, as evidenced by an evaluation of postoperative pain improvement.^[16] For example, Jarvik et al evaluated the sensitivity and accuracy of plain radiography, magnetic resonance imaging, computed tomography, and bone scanning for diagnosing low back pain due to SM. Of these, magnetic resonance imaging and bone scanning are considered more sensitive and accurate, whereas plain radiographs also have relatively high accuracy and sensitivity in certain situations.^[17] Laufer et al built a NOMS decision framework, including neurological, oncological, mechanical, and systemic considerations, and determined the optimal combination of radiotherapy and surgery.^[18] A retrospective study published by Yamada et al demonstrated that high-dose, single-fraction image-guided intensity-modulated radiotherapy could be a safe and effective noninvasive treatment for SM and provided tumor control with minimal impact on the quality of life.^[19] A review published by Tree et al showed that SBRT offers durable local control and the potential for improved progression-free survival in oligometastatic disease outside the liver and lungs. However, randomized trials are still needed to confirm its effectiveness.^[20] Cox et al developed consensus guidelines for stereotactic radiosurgery to treat SM. Radiation oncologists and neurological surgeons independently contoured the target structures in the 10 metastatic spine cases. Their contours were analyzed using STAPLE and kappa statistics to assess agreement.^[21]

Table 5.

Top 10 most cited articles in the field of SM studies.

Rank	Year	Article	Total citation	IF (2022)	Type of study
1	2001	Tomita K, Kawahara N, Kobayashi T, Yoshida A, Murakami H, Akamaru T. Surgical strategy for spinal metastases. Spine (Phila Pa 1976). 2001 Feb 1;26(3):298–306.	902	3	Article
2	2005	Tokuhashi Y, Matsuzaki H, Oda H, Oshima M, Ryu J. A revised scoring system for preoperative evaluation of metastatic spine tumor prognosis. Spine (Phila Pa 1976). 2005 Oct 1;30(19):2186–91.	753	3	Article
3	2010	Fisher CG, Dipaola CP, Ryken TC, et al A novel classification system for spinal instability in neoplastic disease: an evidence-based approach and expert consensus from the Spine Oncology Study Group.Spine (Phila Pa 1976). 2010 Oct 15;35(22):E1221–9.	668	3	Article
4	2007	Gerszten PC, Burton SA, Ozhasoglu C, Welch WC. Radiosurgery for spinal metastases: clinical experience in 500 cases from a single institution. Spine (Phila Pa 1976). 2007 Jan 15;32(2):193–9.	519	3	Article
5	2003	Fourney DR, Schomer DF, Nader R, et al Percutaneous vertebroplasty and kyphoplasty for painful vertebral body fractures in cancer patients. J Neurosurg. 2003 Jan;98(1 Suppl):21–30.	488	4.1	Article
6	2002	Jarvik JG, Deyo RA. Diagnostic evaluation of low back pain with emphasis on imaging. Ann Intern Med. 2002 Oct 1;137(7):586–97.	415	39.2	Article
7	2013	Laufer I, Rubin DG, Lis E, et al The NOMS framework: approach to the treatment of spinal metastatic tumors. Oncologist. 2013 Jun;18(6):744–51.	410	5.8	Article
8	2008	Yamada Y, Bilsky MH, Lovelock DM, et al High-dose, single-fraction image-guided intensity-modulated radiotherapy for metastatic spinal lesions. Int J Radiat Oncol Biol Phys. 2008 Jun 1;71(2):484–90.	386	7	Article
9	2021	Tree Alison C, Khoo, Vincent S, Eeles Rosalind A, et al Stereotactic body radiotherapy for oligometastases. Lancet oncology. 2013 Jan 1;14(1):28–37.	378	51.1	Review
10	2012	Cox BW, Spratt DE, Lovelock M, et al International Spine Radiosurgery Consortium consensus guidelines for target volume definition in spinal stereotactic radiosurgery. Int J Radiat Oncol Biol Phys. 2012 Aug 1;83(5): e597–605.	362	7	Article

IF = impact factor, SM = spinal metastasis.

3.8. Co-cited references and reference burst

A comprehensive retrieval yielded 35,856 articles, of which 420 garnered more than 20 citations. Figure 7A shows a timeline view of the co-cited reference clusters, which are divided into 10 clusters. Cluster #0, which was characterized by the highest number of references, was primarily centered on SBRT. This suggests a significant connection between this topic and SM studies, warranting further comprehensive investigation. Cluster #1 contained the most research on these topics in the last 3 years, suggesting that studies on microwave ablation (MWA) have become popular in recent years. Figure 7B shows the top 20 co-cited references with strong occurrence bursts in SM research. For example, and article by Spratt et al, published in Lancet Oncology in 2019, and by Barzilai et al, published in Neurosurgery in 2018, have been the most popular articles in recent years. The 1st study established a practical algorithm that can help clinicians manage patients with SM by examining the factors pivotal to patient management and evaluating the contributions of multidisciplinary approaches (including image-guided stereotactic radiosurgery, separation surgery, VP, and minimally invasive local ablative approaches) and their respective management options.^[22] The 2nd article synthesized the current evidence-based findings concerning SM treatment and incorporated them into the NOMS decision framework. This framework encompasses 4 crucial decision points pertinent to metastatic spinal tumors: neurology, oncology, mechanical stability, systemic disease, and medical comorbidities.^[1] These 2 articles can assist clinicians in developing treatment strategies and managing patients with SM. The article by Patchell et al published in Lancet in 2005 had the strongest burst strength, and showcased the superiority of direct decompression surgery combined with postoperative radiotherapy over radiotherapy alone for patients experiencing spinal cord compression attributed to metastatic tumors. This conclusion stems from a randomized, multi-institutional trial without blinding.^[23]

Figure 7.

(A) Timeline view of co-cited references cluster analysis in SM related research. (B) Top 20 co-cited references with strong occurrence bursts in SM related research. SM = spinal metastasis.

3.9. Analysis of research hot spots

The keywords of an article are a distillation of the subject matter, and a summary of the keywords of articles in a particular field summarizes the content of the area of study. The keywords serve as indicators of research hotspots in this field. The analysis of keywords can also indicate the time of occurrence of keywords whose frequency of occurrence has changed, thereby revealing changes in research hotspots within the field.

A total of 10 clusters of research hotspots have been formed (Fig. 8D), and #0 spinal metastasis ranks 1st. Each cluster comprises numerous closely related keywords. The smaller the cluster number, the more keywords the cluster contains, indicating a higher popularity of the research area represented by the cluster.

Figure 8.

(A) Co-occurrence network visualization map of keywords in SM related articles. (B) Co-occurrence overlay visualization map of keywords in SM related articles. (C) Co-occurrence density visualization map of keywords in SM related articles. (D) Timeline view of keywords cluster analysis in SM related research. (E) Top 25 keywords with strong occurrence burst in SM related research. SM = spinal metastasis.

Figure 8A illustrates the keyword co-occurrence network visualization map. After conducting keyword co-occurrence analysis using VOSviewer, we found a total of 5842 keywords, of which 154 appeared more than 25 times. All keywords were categorized into 4 clusters. Cluster 1 (red) primarily encompassed studies and trials relevant to SM, disease assessment, different treatments, and their effects. Cluster 2 (green) primarily focused on various malignant tumors, prognoses, and surgery-related issues. Cluster 3 (blue) mainly addressed various cancers, metastases, and diagnostic modalities. Cluster 4 (yellow) included various surgical procedures. The top 10 most frequently occurring keywords were: “spinal metastases,” “surgery,” “cancer,” “management,” “cord compression,” “disease,” “spinal metastasis,” “survival,” “radiotherapy,” and “spine.” Among them, “spinal metastases,” “surgery,” and “cord compression” were the most strongly associated, representing core elements in the field of SM research. These keywords represent fundamental aspects of SM research field. The keywords represented in blue in Figure 8B indicate an earlier occurrence than those in yellow. Figure 8C shows the frequency of keyword appearance, with yellow appearing more frequently than green. Figure 8E shows the top 25 representative burst keywords.

3.10. Analysis of surgical treatment

Because a significant portion of the articles included in this study focused on surgical treatments, we selected 350 articles related to surgery that were published between 2021 and 2023 for keyword timeline clustering analysis. As shown in Figure 9, the articles are categorized into 11 clusters based on keywords: #0 intraoperative blood loss, #1 sarcopenia, #2 quality of life, #3 spinal metastasis, #4 metastatic spinal cord compression, #5 radiofrequency ablation, #6 separation surgery, #7 wound infection, #8 surgical treatment, #9 spinal metastatic disease, and #10 nanoparticle. The lower the cluster number, the more articles it contains. These clusters represent prominent topics in the surgical treatment of SM and illustrate their distribution over time.

Figure 9.

Timeline view of keywords cluster analysis in surgically relevant research.

4. Discussion

4.1. General information

The spine is the most common site of bone metastasis, and the number of patients with SM is likely to increase as the elderly population continues to grow. Approximately 5% to 10% of individuals diagnosed with cancer are projected to develop metastatic spinal tumors,^[24] a condition that often leads to neurological dysfunction and pain. SM is receiving increasing attention from researchers. This study examines the hotspots and trends in SM studies by analyzing the state of SM research between 2000 and 2023. For the 1st bibliometric analysis of SM, the work covered a broader range of topics, was more systematic in its approach, and provided a comprehensive bibliometric analysis of the entire SM research ecology. This perspective provides a more holistic overview of the field and fills a notable gap in the current literature.

Drawing on the WoSCC database, this study employed VOSviewer and CiteSpace for bibliometric analysis to delineate hotspots and trends in SM research. A comprehensive dataset comprising 2325 reviews and original articles from 2000 to 2023 was amassed. An incremental trend in annual paper publications was observed, peaking in 2020 (Fig. 1A). This surge signifies the burgeoning interest and scholarly depth in SM, underscoring its escalating prominence in academia.

Among the top-ranked countries/regions, the United States has emerged as the foremost contributor, reflecting its strong productivity in this field, and China has witnessed rapid development in recent years. Seven institutions and 7 authors from the U.S. secured positions within the top 10 in SM research, attesting to its premier institutions and specialized research cohorts. Moreover, the U.S. demonstrated the highest NC and H-index, indicative of its tremendous research progress over the past 23 years. The analysis of collaboration networks shows a wide range internationally, with the United States standing out in terms of its participation in collaborative efforts, indicating that research in this area in the United States has grown rapidly over the past 23 years. In contrast, despite the sizable number of papers published in China, there is a clear disconnect between quantity and quality. This dissonance, illustrated in Figure 2D and E, may stem from China’s late start in SM research despite its rapid development in recent years. Chinese scholars and institutions should prioritize improving their research quality to narrow the gap between the quantity and quality of their outputs. Figure 2F illustrates the uneven regional distribution of SM research. Developed countries are the leading contributors to SM research. However, as the population ages and society evolves, SM has emerged as a crucial global research topic. Therefore, developing countries must intensify their focus on SM to alleviate patient suffering and extend their lives. It is imperative to remain informed about the advances and hot topics in SM research.

The University of Toronto has published the most research on SM, as it has the highest NP, NC, and H-index (Fig. 3A). From Figure 3E, we can see that Harvard University and Harvard Medical School have conducted SM research with the strongest burst values. Brigham and Women’s Hospital and Massachusetts General Hospital were also early in their research, which have been well developed in recent years. Of all the research institutions, PDT was the most popular research area. PDT is a promising, minimally invasive, nonthermal alternative for managing SM, showing potential for enhancing vertebral bone structural integrity when employed either independently or in conjunction with bisphosphonates, as evidenced in animal studies.^[25] Phase I trials have shown the potential of PDT as a supplementary modality alongside vertebral cement augmentation to enhance mechanical stability and relieve pain symptoms in patients,^[26] yet further comprehensive investigations are warranted.

In Figure 6A, the publications are categorized according to the topics covered, including treatment outcomes, safety considerations, surgical management, and various other relevant aspects. This categorization offers a structured overview of the diverse topics covered in the literature on the treatment and management of SM, highlighting the breadth and depth of research in this field. The article by Tomita et al in 2001 was found to be most closely associated with other articles, followed by those written by Gerszten et al in 2007, and articles written by Sutcliffe et al in 2013 (Fig. 6A). Of the top 10 most-cited publications, 9 were original articles and 1 was a review. The main focus of these articles was the diagnosis of SM, choice of treatment options, and efficacy of radiation and surgical treatments. These publications have far-reaching significance and provide clinical guidance for the diagnosis and treatment of SM. Among these 10 articles, 3 were on SM related scoring systems and 4 were on radiation therapy, suggesting that the assessment of SM patients and radiation-related therapies are more popular in the field. An article published in Spine by Tomita et al in 2001 ranked 1st. From a clustering perspective, studies on SBRT have been the most numerous, and studies on MWA are the most topical in recent years.

Six of the top 10 prolific journals boasted an IF exceeding 3, with the majority being specialty journals within the realms of spine and neurosurgery. This trend implies that articles concerning SM are predominantly published in such specialized journals. However, it also suggests that publishing in highly rated journals can be challenging. Sahgal is the most published author and Bilsky the most cited; both are from Canada. SBRT was ranked 1st in the authors’ co-cited cluster analysis. SBRT for SM is a novel therapeutic approach designed to administer elevated biologically effective doses to metastatic lesions while minimizing radiation exposure to surrounding healthy tissues, and is increasingly used in the treatment of SM.^[27] An open-label, multicenter, randomized, controlled phase 2/3 trial demonstrated significant treatment efficacy and biomechanical safety advantages of SBRT over conventional external beam radiotherapy (EBRT).^[28]

4.2. Hotspots and frontiers

Hotspot analysis can reflect trends and cutting-edge topics in a research field. From Figure 8E, “outcm,” “vertebral metastases,” and “polymethylmethacrylate” were the top 3 keywords with the strongest burst strength, representing the most important research in this field, and “microwave ablation” and “thermal ablation” were the most novel. This suggests that ablation for the treatment of SM is a potential trend in the future. As depicted in Figure 8B, the research emphasis in the field of SM is progressively gravitating towards minimally invasive treatments and radiation therapy, and the prognosis and management of patients with SM has also become a hot topic. Minimally invasive spinal interventions have emerged as the logical preference for managing SM, which mitigate soft tissue damage and minimize blood loss, and are becoming increasingly popular for patients who cannot tolerate open surgery or have a limited life expectancy. VAC and thermal ablation are the 2 most commonly used treatments. VAC consists mainly of VP and KP, which primarily reduce the symptoms associated with SM. Evidence indicates that VAC is an outstanding alternative to medication and traditional surgical interventions in patients with SM. This procedure offers rapid pain relief with minimal disruption of the adjuvant therapy and minimal complications.^[29] The main application of thermal ablation in SM is radiofrequency ablation (RFA); however, MWA has become popular in recent years (Fig. 8E). RFA for SM yields noteworthy pain alleviation, decreased disability, and an enhanced quality of life.^[30] Short- to medium-term analgesia in patients with painful spinal metastases is both effective and safe.^[31] MWA has certain advantages over RFA; it demonstrates superior effectiveness in high-impedance tissues, such as bone, thereby enabling deeper penetration and more efficient heating. However, in the context of spinal lesions, its capacity to swiftly administer substantial energy to a broad ablation area may pose a drawback, potentially resulting in peripheral overheating and subsequent serious complications.^[32] However, further studies and clinical data are required. Furthermore, cryoablation has demonstrated safety and efficacy in alleviating pain and controlling SM progression.^[33] VAC and RFA are often combined in clinical practice. The combination of RFA and VAC may theoretically have a synergistic effect on pain management and local tumor control, although further research is still needed.^[34] In the realm of radiation therapy, owing to the inadequate outcomes observed with low-dose palliative conventional EBRT in addressing spinal metastases, numerous cancer centers have commenced exploration into the potential of SBRT as an alternative treatment avenue for these patients. These techniques have demonstrated persistent pain relief and effective local tumor management.^[15] However, another study found no significant disparity between SBRT and EBRT across multiple parameters. Hence, further investigations are warranted to establish SBRT as a superior clinical choice to EBRT.^[35] PDT has demonstrated potential as an alternative to thermal ablation in phase I clinical trials. Its application as an adjunctive therapy for vertebral cement augmentation has also been proven safe; however, further in-depth clinical studies are still needed.^[26] Researchers have developed assessment systems to manage and predict the prognosis of patients with SM effectively. Commonly utilized frameworks include the NOMS framework, Spinal Instability Neoplastic Score, and Tokuhashi and Tomita scoring systems.^[36] It is important to acknowledge that these decision-making systems occasionally lack the precision required to provide direct guidance. Therefore, researchers should further study multidisciplinary guidelines to enhance treatment.

Upon further analysis of surgically relevant articles, we identified that the themes of “intraoperative blood loss” and “sarcopenia” warrant additional attention. These topics have gained significant focus in the context of surgical treatment for SM in recent years but have not been highlighted in previous analyses. Preoperative embolization is crucial in the treatment of SM using traditional surgical procedures such as en bloc spondylectomy and separation surgery. This is widely regarded as an effective method for preventing significant intraoperative bleeding.^[37] Notably, the current study indicates that preoperative embolization may not be beneficial for the surgical treatment of SM originating from non-hypervascular primary tumors.^[38] In addition to the pathological classification, angiographic evaluation is also crucial.^[39] Kumar et al developed a scoring system for assessing the surgical duration and blood loss, which has become an important clinical guide for the surgical treatment of SM.^[40] Additionally, minimally invasive surgery is regarded as a safer and more effective method for treating SM in the thoracolumbar spine owing to its associated reduction in blood loss.^[41] The cluster “sarcopenia” includes studies on the prediction of outcomes in SM patients treated with surgery. Sarcopenia is a reliable indicator of mortality in patients with spinal metastases, offering valuable insights into survival probabilities and aiding in the decision-making process regarding treatment options.^[42] Changes in the psoas muscle area are considered a reliable predictor of patient survival.^[43] Given the debilitated state of many patients with cancer, it is imperative to conduct a preoperative evaluation of their nutritional status, manage sarcopenia, and enhance their postoperative nutritional status, all of which are crucial components of systemic therapy following the surgical treatment of SM.^[44,45] An important avenue for future research is to utilize machine learning algorithms for the preoperative evaluation of SM patients.^[46]

In conclusion, studies on SM have demonstrated a shift toward minimally invasive treatment and radiation therapy, emphasizing the importance of management and prognostic prediction and concluding that evaluation systems and multidisciplinary guidelines need to be further explored and refined.

4.3. Limitation

Bibliometric analysis provides a valuable perspective for understanding the research priorities within a discipline and predicting the evolution of research areas. However, our study has certain limitations. It incorporates only English-language articles and reviews from SCI extended indexed journals, potentially excluding valuable contributions from non-English sources. This study did not include content from databases such as PubMed, Embase, Scopus, and Google Scholar, which would have affected the reliability of the analyzed results to some extent. It is important to recognize that VOSviewer and CiteSpace are primarily designed for keyword and citation-based analysis rather than full-text content examination, which may affect the depth of insights provided by this approach. It was possible that noteworthy newly published articles were excluded owing to the inherent lag in data collection. It is worth noting that traditional quality assessment tools are not applicable to bibliometric analyses and therefore have some implications for scientific rigor. Limitations of this study also include citation bias in Western research institutions, neglect of low-impact journals, and the possibility that multidisciplinary interdisciplinary research (e.g., AI, molecular mechanisms) may not have been fully included. We advocate that future studies should encompass a broader range of databases and employ advanced analytical techniques to provide a more thorough comprehension of the current state of SM research.

5. Conclusion

Bibliometric analysis indicated a significant and rapid expansion of research in the field of SM. Notably, the United States emerged as the foremost contributor, holding the top position among high-output countries. Furthermore, the University of Toronto has emerged as a leader in this field, with the highest number of published studies. Current and future research efforts are predominantly focused on investigating minimally invasive treatments and radiation therapy for SM. Moreover, the prognostic considerations and management of patients with SM remain the focal points of scholarly discourse and investigation.

Acknowledgments

Thanks to the authors of all the original studies included in this study.

Author contributions

Conceptualization: Luming Kong, Huapeng Guan.

Data curation: Jinbao Liu, Wenzhe Bai, Zelin Yue, Pengpeng Qi.

Investigation: Jinbao Liu, Wenzhe Bai, Zelin Yue, Pengpeng Qi.

Methodology: Luming Kong, Huapeng Guan.

Software: Luming Kong, Huapeng Guan.

Visualization: Jinbao Liu, Wenzhe Bai, Zelin Yue, Pengpeng Qi.

Writing – original draft: Luming Kong, Huapeng Guan.

Writing – review & editing: Nianhu Li.