Skip to main content
Springer logoLink to Springer
. 2022 Mar 8;46(6):1314–1324. doi: 10.1007/s00268-022-06492-2

A Bibliometric Analysis of Robotic Surgery From 2001 to 2021

A Musbahi 1, C B Rao 2,, A Immanuel 1
PMCID: PMC9054892  PMID: 35258666

Abstract

Introduction

Bibliometric analyses are a method of evaluating the quality of research output in a certain domain. Robotic surgery has made vast leaps during the past 20 years and this paper aimed to assess some of the main areas of research using this method.

Methods

A search was undertaken for documents published between 2001 and 2021 from the World of Science database, using the keywords ‘robotic surgery’, ‘robotic assisted surgery’ and ‘robotic-assisted surgery. Results were compared using numerous bibliometric methodologies, and stratified by source-specific metrics, author-specific metrics and country-specific metrics.

Results

The search yielded 3839 documents, from 879 different sources. Only 2% of sources were found to be within Bradford’s Zone 1 of research and the most relevant sources were from the field of urology. The Journal of Urology and Surgical Endoscopy and other Techniques ranked highly among metrics such as H, G, M index and total citations. The top-rated authors had a H index of 15 in the field of robotic surgery and the total citations reached a peak at 1342. The USA, Japan and Italy were the most productive nations and increased collaborative research is leading to a greater number of multiple-centre publications.

Conclusion

Research into robotic surgery is still in its infancy with further reviews of the literature and greater output through large randomised controlled trials in multiple centres through collaborative research needed.

Introduction

Robotic surgery has advanced significantly in the last 20 years. Whilst beginning as stereotaxic systems in the late 1980s, for example the PUMA 200 [1], surgical robots have adapted to not only enable surgery with fewer cuts, but with better precision, accuracy, degrees of freedom and even magnification. A steady movement towards a fifth generation of autonomous robot is being made [2].

Specialties such as urology and gynaecology have long been trailblazers in robotic application and research with some expansion into general surgery and cardiothoracic surgery. The future frontiers of robotic surgery and its full capability are yet to be realised.

Research in robotic surgery was led with the publication of seminal works, such as Nix et al.’s randomised clinical trial on radical cystectomy [3]; however, the research field is young with higher levels of evidence required to prove equivalence or benefit over standard laparoscopic or open techniques.

Bibliometric analyses are defined as efforts to evaluate the quality of research through the measurement of various parameters of scholarly output. This enables a reader to gauge not only the volume of an author’s output or the rate, but an objective demonstration of the number of citations and relationships between authors and articles, not limited to peer-reviewed manuscripts. This then allows analysis of the impact and popularity of publications, authors, institutions and collaborative links.

This analysis can be used with numerous applications, including grant allocations, and by policymakers to set standards for research and direct suitable funding. This study aims to use bibliometric techniques to identify the research trends and patterns of robotic research output from the last 20 years (2001–2021).

Materials and methods

A comprehensive search of the literature was completed. This was taken from the ‘Web of Science Collection’, a subset of the ‘Web of Science’ database, which includes the Science Citation Index Expanded (SCIE), the Social Sciences Citation Index (SSCI) and the Arts and Humanities Citation Index (A&HCI). This database was chosen due to its standing in the academic world as one of the premier citation search platforms [4] and has been proved to be more accurate than rivals such as Scopus in fields such as its journal classification system [5, 6].

A search using the keywords, ‘robotic surgery’, ‘robotic assisted surgery’ and ‘robotic-assisted surgery’ was performed between January 2001 and January 2021.

After completing this search source-specific metrics, author-specific metrics and country-specific metrics were found. A keyword analysis of all the sources was also performed.

Source-specific metrics

The relevance of a source was measured using the total number of documents drawn from a source and were then clustered, using Bradford’s Law into zones. Zones as shown by Bradford’s Law as are indicative of their utility in a certain field [7, 8]. In this case, journals or sources in Zone 1 would be those with the highest productivity within robotic surgery and would represent “the core” of the literature. Total citations (TC) and the number of documents drawn per year per source were also included.

Author-specific metrics

The impact and relevance of authors were considered by drawing their number of documents with an absolute and a fractionalised value, used to understand their contributions in the context of both individual and collaborative research. Fractionalised counting allocates the credit of publication to co-authors in a fractional way, thus by comparing the total and fractional number of articles, one can analyse both participation and contribution to the field of robotic surgery, respectively [9, 10]. Total citations per author were also considered, along with the h, g and m index values. The h-index is a value that combines both publication and citation count to form a sole value. If an author publishes five articles, each with five citations, his/her h index would be 5. This would only increase if he/she published a sixth article and their total articles managed a minimum of six citations each. This begins to quantify both the quality and quantity of an author’s productivity [11]. The g index of an author is another metric of output, which is defined as an author’s top g articles that have been cited an average of g times or at least g2 times. The M index is an author’s H index/the years since their first publication. This tries to compare the output levels of author over time and takes into account early researchers.

Country-specific metrics

The number of documents per country, the number of single country publications (SCP) and their involvement in multiple country publications (MCP) was considered. An MCP ratio was calculated, which indicates the level of international collaboration in an evidence base. The number of countries involved is calculated as a ratio of the total number of publications with the first author being from that country.

Statistics

Data were collected and collated on Excel (Microsoft, United States). Statistical analysis was completed using IBM SPSS Statistics (IBM, United States).

Results

Demographics of the literature

A total of 3839 documents were found and analysed, from 879 different sources. A summary of this data is shown in Table 1. 13,378 different authors were involved in this research, 99.1% (13,258) of which were part of multi-authored documents. On average, there were 3.48 and 5.05 authors and co-authors, respectively, per document, with a mean of 0.287 documents per author. The mean Collaboration Index was 3.58. Furthermore, 68.1% (n = 2613) of all studies were articles, with the rest of the documents spread between book reviews, editorial materials, letters, proceedings papers, abstracts, news items, reprints and reviews.

Table 1.

Collection of overarching information regarding the collection

Main information about the collection Description Results
Main information about data Timespan 2001–2021
Sources (Journals, Books, etc.) 879
Documents 3839
Average years from publication 5.9
Average citations per documents 13.75
Average citations per year per doc 1.735
References 1
Document Types Article 2613
Article; proceedings paper 144
Book review 1
Correction 4
Editorial material 126
Letter 30
Meeting abstract 171
News item 1
Proceedings paper 195
Reprint 1
Review 552
Review; book chapter 1
DocumenT Contents
Keywords Plus (ID) 4170
Author's keywords (DE) 5436
Authors Authors 13,378
Author appearances 19,401
Authors of single-authored documents 120
Authors of multi-authored documents 13,258
Authors collaboration Single-authored documents 131
Documents per author 0.287
Authors per document 3.48
Co-Authors per documents 5.05
Collaboration index 3.58

As shown in Fig. 1 and Table 2, productivity has greatly increased over the last 20 years. Over the past 5 years, 52.6% (n = 2018) of the literature over the course of the past 20 years has been produced. The greatest proportional increase was between 2001–2005 and 2006–2010, with a 5.4 × increase in the number of documents created.

Table 3.

Table showing the resources in Zone 1

Source Rank Frequency Cumulative frequency Zone
Journal of Robotic Surgery 1 241 241 Zone 1
Surgical Endoscopy and Other Interventional Techniques 2 127 368 Zone 1
Urology 3 91 459 Zone 1
International Journal of Medical Robotics and Computer Assisted Surgery 4 87 546 Zone 1
Journal of Minimally Invasive Gynecology 5 77 623 Zone 1
Journal of Urology 6 66 689 Zone 1
Journal of Thoracic Disease 7 55 744 Zone 1
JSLS-Journal Of The Society Of Laparoendoscopic Surgeons 8 54 798 Zone 1
Gynecologic Oncology 9 53 851 Zone 1
Current Opinion In Urology 10 44 895 Zone 1
Journal Of Laparoendoscopic & Advanced Surgical Techniques 11 43 938 Zone 1
Innovations-Technology And Techniques In Cardiothoracic And Vascular Surgery 12 41 979 Zone 1
BJU International 13 40 1019 Zone 1
International Journal Of Gynecological Cancer 14 39 1058 Zone 1
World Journal Of Urology 15 36 1094 Zone 1
Annals Of Thoracic Surgery 16 35 1129 Zone 1
Journal Of Pediatric Urology 17 32 1161 Zone 1
Obesity Surgery 18 30 1191 Zone 1
European Urology 19 29 1220 Zone 1
Female Pelvic Medicine And Reconstructive Surgery 20 27 1247 Zone 1
Canadian Journal Of Urology 21 26 1273 Zone 1

Fig. 1.

Fig. 1

Graph showing the annual scientific production

Table 2.

Annual scientific productivity on robotic surgery

Year Articles
2001–2005 88
2006–2010 476
2011–2015 1257
2016–2021 2018

Source analysis

Sources were ranked by relevance to the topic with the total number of documents per source. The Journal of Robotic Surgery housed the greatest number of documents (n = 241), followed by Surgical Endoscopy and Other Interventional Techniques (n = 127) and Urology (n = 91) as the next most relevant. The top 25 most relevant sources are charted in Fig. 2. Building on this, 40% of the 25 most relevant sources were related to the specialty of urology, with gynaecology, thoracic surgery followed by general surgery.

Fig. 3.

Fig. 3

Pie Chart showing Bradford's Law

Fig. 2.

Fig. 2

Graph showing the distribution of the Top 25 most relevant sources

When sources were clustered using Bradford’s Law, only 2.39% (n = 21) sources were in Zone 1 this being the core of the literature, 1.33% (n = 117) were in Zone 2 and 84.3% (n = 741) were in Zone 3. 33.3% (n = 7) of the sources in Zone 1 were related to the field of urology.

Subsequently, sources were stratified by their h index, g index, m index and total citations, as shown in Fig. 4. The Journal of Urology was well represented in all three measures, with the top-ranked h index, the second g and m index and total citations. This indicates that there was not only a large amount of data from this journal, but it was influential in the field. Surgical Endoscopy and other Interventional Techniques had the highest number of total citations and, g index and third greatest m index ranking. This had the greatest impact using this variable, followed by Surgical Endoscopy and other Interventional Techniques. The greatest number of total citations was in Surgical Endoscopy and Other Interventional Techniques, again followed by the Journal of Urology. 28% (n = 7) of the sources with the highest number of total citations were in the field of urology.

Fig. 4.

Fig. 4

Graph showing the h index, g index, m index and total citations of the top 25 most productive sources

Author-specific analysis

Table 4 shows author-specific results for the top 20 most productive authors. All the top 20 authors had a h index of at least 5 with a median (IQR) of 9.5 (398), and the number of total citations ranged from 71 to 1342, with a median (IQR) of 341.5(5.75). The median (IQR) g index was 15 (3.5) and scores ranged from 7 to 26. Furthermore, the median (IQR) m index was 0.61(0.27) with a range from 0.375 to 1.14.

Table 5.

Table showing statistics regarding the country of origin of the research

Country Articles Frequency SCP MCP MCP/SCP Ratio
USA 1745 0.473028 1579 166 0.0951
China 275 0.074546 245 30 0.1091
Italy 209 0.056655 170 39 0.1866
Germany 180 0.048794 140 40 0.2222
United Kingdom 177 0.047980 133 44 0.2486
France 167 0.045270 138 29 0.1737
Canada 123 0.033342 95 28 0.2276
Japan 86 0.023313 77 9 0.1047
Australia 74 0.020060 61 13 0.1757
Korea 72 0.019517 59 13 0.1806
Turkey 67 0.018162 61 6 0.0896
India 56 0.015180 53 3 0.0536
Spain 56 0.015180 33 23 0.4107
Brazil 34 0.009217 23 11 0.3235
Belgium 30 0.008132 20 10 0.3333
Singapore 28 0.007590 20 8 0.2857
Switzerland 28 0.007590 20 8 0.2857
Israel 24 0.006506 16 8 0.3333
Denmark 23 0.006235 20 3 0.1304
Sweden 23 0.006235 17 6 0.2609
Romania 22 0.005964 20 2 0.0909
Greece 20 0.005422 16 4 0.2000
Netherlands 19 0.005150 13 6 0.3158
Austria 16 0.004337 13 3 0.1875
Iran 14 0.003795 13 1 0.0714
Saudi Arabia 14 0.003795 9 5 0.3571
Finland 12 0.003253 10 2 0.1667
Portugal 12 0.003253 7 5 0.4167
Ireland 11 0.002982 7 4 0.3636
Norway 11 0.002982 8 3 0.2727
Mexico 7 0.001898 6 1 0.1429
Egypt 6 0.001626 5 1 0.1667
Argentina 5 0.001355 4 1 0.2000
Chile 5 0.001355 4 1 0.2000
New Zealand 5 0.001355 5 0 0.0000
Thailand 5 0.001355 4 1 0.2000
United Arab Emirates 4 0.001084 3 1 0.2500
Colombia 3 0.000813 2 1 0.3333
Malaysia 3 0.000813 3 0 0.0000
Poland 3 0.000813 3 0 0.0000
Qatar 3 0.000813 1 2 0.6667
Czech Republic 2 0.000542 1 1 0.5000
Kuwait 2 0.000542 1 1 0.5000
Venezuela 2 0.000542 1 1 0.5000
Vietnam 2 0.000542 0 2 1.0000
Bulgaria 1 0.000271 1 0 0.0000
Indonesia 1 0.000271 0 1 1.0000
Luxembourg 1 0.000271 1 0 0.0000
South Africa 1 0.000271 1 0 0.0000

Table 6.

Table showing number of occurrences of keywords

Words Occurrences
Surgery 823
Outcomes 552
Experience 414
Cancer 304
Resection 271
Complications 231
Management 211
Learning-curve 192
Meta-analysis 138
Impact 133
System 118
Hysterectomy 117
Feasibility 104
Radical prostatectomy 104
Survival 103
Carcinoma 101
Lymphadenectomy 95
Children 91
Endometrial cancer 88
Trial 87
Laparotomy 86
Quality-of-life 86
Total mesorectal excision 82
Perioperative outcomes 75
Risk 74
Women 74
Initial experience 73
Accuracy 70
Laparoscopic surgery 69
Follow-up 67
Lymph-node dissection 67
Morbidity 67
Mortality 62
Tumours 62
Laparoscopy 60
Multicentre 59
Prostatectomy 59
Short-term outcomes 58
Replacement 57
Retropubic prostatectomy 57
Safety 57
Cost 55
Laparoscopic partial nephrectomy 54
Minimally invasive surgery 54
Repair 54
Performance 53
Rectal-cancer 50
Thoracic-surgery 49
Classification 47
Risk-factors 47

Table 4.

Author-specific results

Author h Index g Index m Index Total citations (TC)
Yang GZ 15 24 0.833 594
Stoyanov D 15 26 0.833 748
Kiaii B 10 14 0.588 233
Li J 5 9 0.385 96
Patel RV 9 18 0.600 328
Ahmad S 13 18 0.929 746
Holloway RW 15 18 1.071 810
Dasgupta P 7 16 0.412 275
Hubert J 8 17 0.421 355
Pigazzi A 11 17 0.688 1342
Wang Y 8 14 1.143 210
Kandil E 4 8 0.364 76
Patel VR 10 16 0.625 389
Darzi A 11 15 0.550 453
Fader AN 10 15 0.769 592
Gundeti MS 9 14 0.692 203
Kaouk JH 11 15 0.550 447
Kim S 5 13 0.556 169
Poignet P 6 10 0.375 119
Toloza EM 5 7 0.625 71

Country-specific analysis

Figure 5 shows the spread of corresponding authors by country. As demonstrated, the USA had the highest number of articles (n = 1745), followed by China (n = 275) and Italy (n = 209). Table 4 then splits the data, including the SCP and MCP values.

Fig. 6.

Fig. 6

Diagram showing distribution of the top 50 keywords, based on their total number of occurrences

Fig. 5.

Fig. 5

Graph showing the corresponding authors by countries

The median (IQR) number of articles per nation was 14 (51.25). When the documents were stratified by SCP and MCP, the median (IQR), respectively, was as follows: 13(29.25) and 3(8). The median (IQR) ratio of the two was 0.2 (0.208). Indonesia and Vietnam had the greatest international collaboration with an MCP/SCP of 1, followed by Qatar with 0.67, and Kuwait and Venezuela with 0.5 (Fig. 6).

Keywords analysis

The most used word, as expected, was surgery with 823 occurrences. Outcomes (552) and experience (414) were the next two.

Twelve of the top 50 keywords were based on specific procedures: resection, hysterectomy, radical prostatectomy, lymphadenectomy, laparotomy, total mesorectal excision, laparoscopic surgery, lymph-node dissection, laparoscopy, prostatectomy, retropubic prostatectomy and laparoscopic partial nephrectomy. Other common themes were those relating to patient outcomes, such experience and quality-of-life.

Figure 7 shows how keyword frequency has changed from 2008 to 2021. Earlier publications focus on specific surgical techniques, such as conduit urinary diversion; however, overtime, keywords became more generalised with examples such as outcomes and accuracy in 2019–2020. Furthermore, as time moved towards the end of the decade, patient populations such as paediatrics came into consideration, along with different specialties, such as even Trauma and Orthopaedics, with the involvement of ‘spine’ as a keyword.

Fig. 7.

Fig. 7

Graph showing the evolution of keywords from 2008 to 2021

Discussion

Research into robotic surgery is in its infancy but also changing and increasing. Over the past 20 years, over 3800 documents on the topic of robotic surgery have been compiled, gathering steam, especially from 2018 to 2020.

Despite this recent expansion, very few sources have fallen in the core of the literature, as shown by Bradford’s Law. Much of this research was in journals such as Urology, the Journal of Urology and the Journal of Minimally Invasive Gynecology, is related to urology or gynaecology, which has dominated the early stages of robotic research [12]. However, there has been a movement towards other specialties and subspecialties, such as general surgery and spinal surgery as shown by the keyword analysis.

Keyword analysis allows the reader to understand future frontiers of robotic surgery. The majority of research mentions surgical outcomes, complications and the importance of experience. If trends over time are considered, thoracic surgery and upper GI surgery are being increasingly mentioned in recent years, with keywords such as fundoplication and lobectomy cropping up during the final years of analysis.

When considering journal impact, the Journal of Robotic Surgery is the top-rated source through numerous metrics, such as relevance, Bradford Index, however, has only been active since 2007 and does not have an official impact factor, nor is it PubMed indexed.

Hirsch [13] hypothesised that after 20 years of research, a H index of 20 indicated a successful scientist, 40 indicated an outstanding scientist and 60 indicated extraordinary research output. As robotic research is still in its infancy, with few reaching even 20 years of research experience, thus the lack of any authors with a H index over 20 indicates the room for future research and growth in this field. The m index is a similar figure of the rate of productivity and shows a similar trend [14].

The most productive nations were the USA, China and Italy. This is expected, with the origins of robotic surgery beginning with Kwoh et al. [1] in California and market being dominated by the Da Vinci robotic system (Intuitive Surgical Inc, Sunnyvale, USA), originating in the United States, since 2000. Furthermore, The Business Research Company published a report in early 2020, indicating the North America was the largest region in the robotic surgical devices market in 2019, with Asia–Pacific growing at the greatest rate [15] (The Business Research Company, 2020). However, international collaboration levels remain low in these nations. Higher collaboration is present in nations with lower productivity. International collaboration has recently been realised with research such as the ROLARR trial [16], a product of collaboration between the United Kingdom, USA, Finland, Denmark, Italy and New Zealand.

What does the future hold?

As mentioned earlier, collaboration may prove to be key. Using the vast potential of centres worldwide could lead to increased numbers of large-scale RCTs, allowing surgeons to explore different populations. Furthermore, there is a niche for further systematic reviews of the literature. As time passes, authors will be able to increase their productivity, with increased funding, leading to greater metrics, such as H index’s moving towards 20 + etc. Other areas of research, such as bariatric surgery, resectional surgery and spinal surgery also provide new frontiers.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Kwoh YS, Hou J, Jonckheere EA, Hayati S. A robot with improved absolute positioning accuracy for CT guided stereotactic brain surgery. IEEE Trans Biomed Eng. 1988;35(2):153–160. doi: 10.1109/10.1354. [DOI] [PubMed] [Google Scholar]
  • 2.Ashrafian H, Clancy O, Grover V, Darzi A. The evolution of robotic surgery: surgical and anaesthetic aspects. BJA Br J Anaesth [Internet]. 2017;119(suppl_1):i72–84. Available from: 10.1093/bja/aex383 [DOI] [PubMed]
  • 3.Nix J, Smith A, Kurpad R, Nielsen ME, Wallen EM, Pruthi RS. Prospective randomized controlled trial of robotic versus open radical cystectomy for bladder cancer: perioperative and pathologic results. Eur Urol [Internet]. 2010;57(2):196–201. Available from: https://www.sciencedirect.com/science/article/pii/S0302283809010288 [DOI] [PubMed]
  • 4.Li K, Rollins J, Yan E. Web of Science use in published research and review papers 1997–2017: a selective, dynamic, cross-domain, content-based analysis. Scientometrics [Internet]. 2017/12/18. 2018;115(1):1–20. Available from: https://pubmed.ncbi.nlm.nih.gov/29527070 [DOI] [PMC free article] [PubMed]
  • 5.Wang Q, Waltman L. Large-scale analysis of the accuracy of the journal classification systems of Web of Science and Scopus. J Informetr [Internet]. 2016;10(2):347–64. Available from: https://www.sciencedirect.com/science/article/pii/S1751157715301930
  • 6.Mongeon P, Paul-Hus A. The journal coverage of Web of Science and Scopus: a comparative analysis. Scientometrics [Internet]. 2016;106(1):213–28. Available from: 10.1007/s11192-015-1765-5
  • 7.Bradford SC. Sources of information on specific subjects. Engineering. 1934;137:85–86. [Google Scholar]
  • 8.Venable GT, Shepherd BA, Roberts ML, Taylor DR, Khan NR, Klimo P. An application of Bradford’s law: identification of the core journals of pediatric neurosurgery and a regional comparison of citation density. Child’s Nerv Syst [Internet]. 2014;30(10):1717–27. Available from: 10.1007/s00381-014-2481-9 [DOI] [PubMed]
  • 9.Moed HF. Citation analysis in research evaluation. Vol. 9. Springer Science & Business Media; 2006.
  • 10.Sivertsen G, Rousseau R, Zhang L. Measuring scientific contributions with modified fractional counting. J Informetr [Internet]. 2019;13(2):679–94. Available from: https://www.sciencedirect.com/science/article/pii/S175115771830350X
  • 11.Choudhri AF, Siddiqui A, Khan NR, Cohen HL. Understanding bibliometric parameters and analysis. RadioGraphics [Internet]. 2015;35(3):736–46. Available from: 10.1148/rg.2015140036 [DOI] [PubMed]
  • 12.Hussain A, Malik A, Halim MU, Ali AM. The use of robotics in surgery: a review. Int J Clin Pract [Internet]. 2014 Nov 1;68(11):1376–82. Available from: 10.1111/ijcp.12492 [DOI] [PubMed]
  • 13.Hirsch JE. An index to quantify an individual’s scientific research output. Proc Natl Acad Sci U S A [Internet]. 2005 Nov 15;102(46):16569 LP–16572. Available from: http://www.pnas.org/content/102/46/16569.abstract [DOI] [PMC free article] [PubMed]
  • 14.Egghe L. Theory and practise of the g-index. Scientometrics [Internet]. 2006;69(1):131–52. Available from: 10.1007/s11192-006-0144-7
  • 15.Company TBR. Robotic Surgery Devices Global Market Report 2020–30: COVID-19 Growth and Change. Global: The Business Research Company; 2020 p. 1–300.
  • 16.Jayne D, Pigazzi A, Marshall H, Croft J, Corrigan N, Copeland J, et al. Effect of Robotic-Assisted vs Conventional Laparoscopic Surgery on Risk of Conversion to Open Laparotomy Among Patients Undergoing Resection for Rectal Cancer: The ROLARR Randomized Clinical Trial. JAMA [Internet]. 2017 Oct 24;318(16):1569–80. Available from: 10.1001/jama.2017.7219 [DOI] [PMC free article] [PubMed]

Articles from World Journal of Surgery are provided here courtesy of Springer

RESOURCES