Introduction

To date, over 2 million robot-assisted surgeries have been performed in the United States [1]. The first was performed in 1985. The Programmable Universal Machine for Assembly (PUMA) was used to obtain neurosurgical biopsies and for resection [2]. This was followed by the first minimally invasive surgery, a laparoscopic cholecystectomy, in 1987 [3]. The robotic and laparoscopic fields were subsequently combined with the development of the voice command AESOP robotic system in 1994. This system was primarily used for holding an endoscope to eliminate tremor and unnecessary movements [2]. This was followed by the ZEUS system which was comprised of 3 arms and a control console. The earliest version of the most commonly used platform today, the daVinci system was first utitlised in the hospital setting in 1999. This platform provides a 3-dimensional view and articulated instruments. Its first published use in humans was for a gastric fundoplication [4].

Currently, proponents of robotic-assisted surgery highlight the precision and cosmesis afforded by its use. Opponents note increased cost and prolonged operative times when compared to open and laparoscopic approaches. Although, it is most commonly utilised for urological, colorectal and gynecological surgery, it is gaining acceptance in several other fields including neurosurgery and orthopedic, cardiothoracic and maxillofacial surgery. As a result, several studies, reviews and case series have been published in this relatively novel field. We aimed to determine the most influential of these works using bibliometric analysis.

Bibliometric analysis records and analyses the citation history of individual manuscripts on a topic of interest. The number of citations of a particular manuscript in subsequent publications reflects its contribution to the field of interest and is often considered a proxy for how influential a work is. Thus, a large number of citations suggest a direct influence on the understanding and development of the field of interest. Bibliometric citation analysis has been used to analyse the most influential scientific papers in plastic, orthopedic and general surgery [5,6,7]. Such analysis has had not yet been performed for robotic surgery.

Methods

The terms ‘robotic surgery,’ ‘robot assisted surgery’ and ‘robot-assisted surgery’ were used to search Thomson Reuters Web of Science database. Available from: http://thomsonreuters.com/thomson-reuters-web-of-science/). The search was performed on Sept 1st 2018 and limited to full length, English language manuscripts. No limitations on publication date were applied. The returned dataset was sorted by number of citations. The 100 most cited manuscripts were identified and further evaluated. Number of citations, title, first and senior author, publishing institution of first author, year of publication and the country of origin of each manuscript were recorded and analyzed. This method was developed by Paladugu [7] and replicated by Kelly, Joyce and Kavanagh [8,9,10]. The 2017 and 5 year impact factors of each journal publishing the manuscripts were determined using InCites Journal Citation reports (Clarivate Analytics, 2018) and recorded.

Results

The Web of Science search returned 14,980 full-length English language papers. The number of citations derived from each of the top 100 cited works ranges from 492 (Nelson’s ‘Microbots for Minimally Invasive Surgery’) to 122 (Kreindler’s ‘Computer-Assisted And Robot-Assisted Resection of Thalamic Astrocytomas in Children’) [11, 12]. Published in 1991, Kreindler’s manuscript was also the oldest manuscript featured on the list. The most recent manuscript was published in 2015. This manuscript by Wakabayashi et al. reported a consensus statement on liver resection from the Second International Consensus Conference Held in Morioka, Japan [13]. The year which yielded the highest number of influential papers was 2009 (n = 15, 2608 citations) followed by 2007 (n = 9, 1855 citations, Fig. 1). The top 10 manuscripts are provided in Table 1. The complete list of 100 manuscripts is provided in Table 2. The majority (n = 4) of the top 10 featured urology, followed by maxillofacial surgery (n = 2), collaborations between multiple surgical subspecialities (n = 2), general surgery (n = 1) and medical engineering (n = 1).

Fig. 1
figure 1

a Number of top 100 publications per year. b Number of top 100 citations per year

Full size image
Table 1 Top 10 cited manuscripts
Full size table
Table 2 The top 100 manuscripts in robotic surgery
Full size table

The top 100 manuscripts were published in 38 journals (Table 3). European Urology published the most papers in the top 100 and also generated the largest number of citations with 15 papers and a total of 2595 citations. The United States of America is the country with the greatest number of publications in the top 100 (n = 68) followed by Korea and Italy (n = 4 each, Fig. 2). The Johns Hopkins University was the institution that had the greatest number of manuscripts with eight papers generating 1605 citations combined (Table 4). The University of Pennsylvania, Vattikuti Urology Institute, Henry Ford Health System, Yonsei University College of Medicine and Cleveland Clinic followed with five top 100 publications each.

Table 3 Journals Publishing the Top 100 manuscripts
Full size table
Fig. 2
figure 2

Countries of origin of the most cited manuscripts in robotic surgery

Full size image
Table 4 Top Publishing Institutions
Full size table

Seven authors had three or more first and or senior authorship in the top 100 list (Table 5). M. Menon, the Rajendra and Padma Vattikuti Chair in Oncology at the Vattikuti Urology Institute, Henry Ford Health System in Michigan, USA, had the most authorships with 3 first and 3 senior authorships. He has made several advances in the field of robotic prostatectomy including developing the Vattikuti Institute Prostatectomy.

Table 5 Authors with the most significant contributions to the top 100 manuscripts in robotic surgery
Full size table

The majority of top 100 manuscripts featured urological surgery (n = 28), followed by collaborations between multiple specialities (n = 15), colorectal surgery (n = 13), other fields (n = 11), maxillofacial surgery (n = 8), hepatobiliary surgery (n = 7), cardiothoracic surgery and gynecology (n = 5 each), upper gastrointestinal surgery, thyroid/endocrine surgery, orthopedics (n = 2 each) and general surgery and ortolaryngology (n = 1 each) (Fig. 3).

Fig. 3
figure 3

Top 100 by medical specialty

Full size image

The majority of studies were case series/reports (n = 42), followed by comparative studies (n = 24). Two animal studies were included (An Image-Directed Robotic System for Precise Orthopedic-Surgery and Robotic laparoscopic surgery: A comparison of the daVinci and Zeus Systems) [14, 15]. Two consensus statements were also included (Best Practices in Robot-assisted Radical Prostatectomy: Recommendations of the Pasadena Consensus Panel and Recommendations for Laparoscopic Liver Resection: A Report from the Second International Consensus Conference held in Morioka) [13, 16]. Two systematic reviews/meta-analyses were included (Robotic Versus Laparoscopic Partial Nephrectomy: A Systematic Review and Meta-Analysis and Positive Surgical Margin and Perioperative Complication Rates of Primary Surgical Treatments for Prostate Cancer: A Systematic Review and Meta-Analysis Comparing Retropubic, Laparoscopic, and Robotic Prostatectomy) [17, 18]. Three subspecialities contributed four multicentre trials to the top 100 (2 colorectal, 1 cardiothoracic and 1 maxillofacial surgery). The titles of the trials were: Robotic mitral valve surgery: A United States multicenter trial; Multicentric Study on Robotic Tumor-Specific Mesorectal Excision for the Treatment of Rectal Cancer; An international, multicentre, prospective, randomised, controlled, unblinded, parallel-group trial of robotic-assisted versus standard laparoscopic surgery for the curative treatment of rectal cancer and Transoral robotic surgery: A multicenter study to assess feasibility, safety, and surgical margins (Fig. 4) [19,20,21,22].

Fig. 4
figure 4

Manuscript types featured in the top 100 manuscripts

Full size image

Discussion

The evolution of the robotic platform is illustrated by this bibliometric analysis. Early reports, including the first paper in the Top 100, feature the first platform, the PUMA. In 1994, Taylor describes the early systems used in orthopedics and details the second generation platform which was in clinical trial at that time [14]. In 1995 Kavoussi report on the improved camera control using the robotic platform in a small case series of 11 patients undergoing laparoscopic pelvic surgery [23]. These early feasibility studies are followed by studies in orthopedic, cardiothoracic, colorectal and gynecological surgery from 1991 to 2001. Between 2000 and 2004, the majority of studies were case series and reviews. The number of case studies peaked at 23 between 2005 and 2009 and a surge in studies comparing robot assisted surgery to laparoscopic surgery was seen. In the later years, the majority of multicentre trials published their results. The first systematic reviews/meta-analyses were also published. It is interesting that a large number of review papers were published in the earlier time period when limited evidence was available. This reflects the interest in the platform at the time. Evidence based reviews followed several years later, between 2010 and 2015.

After safety was established, the focus of the literature shifted to improvements in outcomes. Multiple manuscripts on the effect of a learning curve were published as it was noted that, with time, oncological outcomes improved and operating time decreased. Atug et al. described the learning curved and improved oncologic margins in prostatectomy over a 2-year time period [18]. Similarly, Leninhan and Bokhari describe improved operating times and hospital lengths of stay in gynecological and colorectal surgery, respectively [24, 25]. Bokhari writes of their cohort that ‘the learning phase was achieved after 15 to 25 cases [25].’

Although several subspecialities contribute to the Top 100, urological surgery is the main contributor. Both of the systematic reviews and consensus statements were from Urology groups. Additionally, the journal publishing the highest number of publications was a urological journal. This is due to the early adoption of robotics for prostatectomy and expansion of the platform to include nephrectomy and other urological procedures. This is followed by colorectal surgery. In both specialties, the robot assisted approach is particularly beneficial while operating deep in the pelvis, i.e. for a prostatectomy or total mesorectal excision for rectal carcinoma.

A large number of collaborations with single institutions present their data together in the Top 100. Several studies including all robotic assisted procedures performed to a specific date in a single institution (i.e. combined urological, colorectal and gynaecological results) are found. Although providing early data primarily on safety, it is often difficult to determine individual specialty results within these papers.

Limitations

Several types of bias may have potentially affected our results. These biases are inherent to all bibliometric analyses and are well documented. Institutional, language, self-citation and powerful person bias and deliberate citation omission may result in disproportionate or inappropriate citations [26]. Often in bibliometric analyses, the majority of manuscripts are found in the earlier time periods of the study inclusion dates. A longer duration since publication often leads to the accumulation of a higher number of citations in older manuscripts. However, this is not found in our analysis. The majority of publications were between 2005 and 2009. This is likely due to the large number of case series published during this time.

Several multicentre collaborations are currently underway. Reports from these trials will in future alter the top 100 manuscripts as they will likely be cited heavily. Another limitation is the inclusion of only the senior and first authors and the institutions of the first author for analysis. In fact, several authors in the Top 100 may have contributed toward multiple manuscripts. However, including first and senior authors only reflects the authors traditionally with the most significant contribution to the manuscript.

In this work, we acknowledge the early observations and research from which the current platform has developed and will continue to develop. The ongoing advancements in robotic-assisted surgery means that the list of 100 most cited papers will change with these advancements. Thus, regular 5–10 yearly reviews of the most cited papers to keep up with advancements are warranted.

Conclusion

The 100 most cited manuscripts highlighted describe the progression of the robotic surgical platform from a basic platform used to steadily hold instruments to the 3D platform with articulated instruments used today. These manuscripts highlight the evolution from early feasibility studies to effectiveness studies and finally multicentre trials and meta-analyses. These studies demonstrate how robotic assisted surgery has gained acceptance in the fields of urological, colorectal, general, cardiothoracic, orthopedic, maxillofacial and neurosurgery. These works were cited over 120 times each reflecting their impact on the field as it is today.