Robotic surgery in obstetrics and gynecology: a bibliometric study

We used Clarivate’s “Web of Science” (WOS) to identify all relevant publications on robotic surgery in OBGYN [14, 15]. WOS is one of the leading citation search platforms. It was thoroughly studied and found to be more accurate than other platforms [14, 16]. The term “Robotic” was used based on the American Board of Obstetrics and Gynecology (ABOG) 2022 certifying examination topics list [17]. We restricted the query to document types labeled as “Articles” and “Reviews” under the WOS Category “Obstetrics Gynecology” and “Oncology”. Only articles with the terms “Robotic” or “Robot” in their title were included [18]. Articles were further independently manually reviewed by two researchers (R.M and G.L) to ascertain that the manuscript’s topic included robotic surgery in obstetrics and gynecology. Bibliometric data were obtained from the iCite database, a National Institutes of Health (NIH) database. The following bibliometric data were collected: average citations per year (CPY, calculated by as total number of citations divided by the number of years since publication); relative citation ratio (RCR, based on weighting the number of citations a paper receives to a comparison group within the same field); field citation ratio (FCR, calculated by dividing the number of citations an article has received by the average number of citations received by articles published in the same year and in the same research fields). WOS and iCite databases were queried on December 15th, 2022.

Each of the identified articles was evaluated for specific characteristics including the continent and country of origin of the corresponding author, country’s level of income as defined by the world bank by annual gross national income per capita [19], publishing journal, publication year, CPY, RCR, FCR, journal impact factor, subject matter, and study design. Journals’ impact factor was defined according to the 2022 Clarivate’s Journal Citation Report. Study design was categorized into the following: retrospective, prospective, case report, review, case series, randomized controlled trial, video article, and meta-analysis. We defined high CPY as the 90th percentile of CPY and divided the cohort into two groups: high CPY (≥ 90th percentile) and low CPY (< 90th percentile). The 90th percentile of CPY was calculated to be 6.6. We further divided the cohort into two groups by the year of publication, below the median (early period) and above the median (2015)—late period.

As the data used for this study are publicly available and do not include protected health information, Institutional Review Board review and approval was not required.

Statistical analyses were performed using SPSS version 29. We performed a descriptive analysis using a chi square test and Fisher exact test for categorical variables and a Wilcoxon rank-sum test for nonparametric continuous variables. Multivariable regression analysis was conducted to identify independent parameters associated with a high CPY. The regression analysis model included all factors that were found to be statistically significant in the univariate analysis. A 2-sided p value < 0.05 was considered statistically significant.

This is an observational study. The Cedars-Sinai Research Ethics Committee has confirmed that no ethical approval is required.

Most publications were of North American origin (n = 485, 57.9%, Table 1) predominantly from the U.S (n = 461, 55.0%), followed by European origin (n = 218, 26.0%) leaded by France and Italy (n = 49, 5.8%, n = 47, 5.6%, respectively). Overall, 788 (n = 94.0%) publications were from countries categorized as high-income countries. There were no publications from low-income countries and there were 50 publications (6.0%) from LMIC.

In a comparison of publications from LMIC versus high income countries (Table 2), all medians of bibliometric scores were lower in LMIC publications. The median year of publication for LMIC publications was more recent (2017 vs. 2015, p = 0.002).

Subjects of research that had higher representation in LMIC were benign gynecology [20 (40.0%) vs. 156 (19.8%)], endometriosis [4 (8.0%) vs. 23 (2.9%)], and fertility [2 (4.0) vs. 6 (0.8%)]. The types of publications that had higher representation in LMIC were case reports [11 (22.0%) vs. 68 (8.6%)], and case series [8 (16.0%) vs. 36 (4.6%)].

Most publications were in the field of gynecologic oncology (n = 344, 41.1%), followed by benign gynecology (n = 176, 21.0%) and urogynecology (n = 156, 18.6%) (Table 1, Fig. 3). Most studies were retrospective (n = 394, 47.0%), followed by prospective (n = 177, 21.1%) and case series/reports (n = 123, 14.7%). RCT comprised 3.6% (n = 30) of all publications.

The leading publishing journal was the Journal of Minimally Invasive Gynecology, with 151 (18.0%) publications, followed by two gynecologic oncology journals (Gynecologic Oncology—n = 105, 12.5%, International Journal of Gynecological Cancer—n = 60, 7.2%), and two urogynecology journals (International Urogynecology Journal—n = 60, 7.2%, Female Pelvic Medical Research—n = 55, 6.6%, Table 1).

The median Impact factor was 4.3 [1.9–5.3], with a median of 10.0 citations [3.0–27.0] and a CPY of 1.9 [0.7–3.8]. In an analysis of high CPY vs. low CPY (Table 3), the following topics were associated with higher CPY: oncology 45 (53.6%) vs. 299 (39.7%), and fertility 4 (4.8%) vs. 4 (0.5%). Urogynecology was associated with low CPY: 7 (8.3%) vs. 149 (19.8%). Publications from North America were associated with higher CPY: 62 (73.8%) vs. 423 (56.1%), as well as randomized controlled trials: 7 (9.5%) vs. 22 (2.9%).

In a multivariable regression analysis (Table 4), including publication year, impact factor, subject of article, continent, and study design, the following factors were associated with a higher CPY: journal’s impact factor [adjusted odds ratio (aOR) 95% confidence interval (CI) 1.30 (1.16–1.41)], subject of study being oncology [aOR 95% CI 1.73 (1.06–2.81)] and randomized controlled trials [aOR 95% CI 3.67 (1.47–9.16)]. Publication year was negatively independently associated with a high CPY [aOR 95% CI 0.93 (0.88–0.98)].

In a comparison of publications prior to 2015 vs. 2015 and later (Table 5), all median bibliometric scores were higher in the earlier period. The following areas of research had higher representation in the late period compared to the earlier period: fertility [8 (1.6%) vs. 0 (0%), respectively] and urogynecology [109 (22.4%) vs. 47 (13.4%), respectively]. After 2015 there has been a higher representation of publications from Asia [96 (19.7%) vs. 27 (7.7%)] and from LMIC [41 (8.4%) vs. 9 (2.6%)], a lower proportion of review articles [41 (8.4%) vs. 52 (14.8%)] and higher proportion of video publications [17 (3.5%) vs. 3 (0.9%)].

Robotic surgery research emerged approximately two decades ago. Its implementation in gynecology has been a major source of robotic surgery academic literature [6]. The first use of robotic surgery in gynecology was for tubal anastomosis in 1998 [20]. In 2005 the FDA approved the use of robotic surgery in gynecology and since then it has become a common approach in gynecology [21], mainly oncology. In the past decade the use of robotic surgical systems for all kinds of gynecological and non‐gynecological surgery has increased [22] (Figure S2). It has been used for various gynecologic procedures including hysterectomy, myomectomy, sacrocolpopexy, endometriosis surgery, gynecologic cancer and more [21]. According to Intuitive Surgical, the manufacturer of the da Vinci robotic Surgical System, the most widely used robotic device worldwide, 6730 system has been installed worldwide as of December 2021 [23].

Gynecologic oncology was the most frequently studied topic in gynecologic robotic surgery in our bibliometric study. In a recent general robotic surgery scientific review, the top two gynecological journals publishing robotic surgery research were the Journal of Minimally Invasive Gynecology with 77 publications, a journal that publishes both benign gynecology and gynecologic oncology articles, and the journal Gynecologic Oncology with 53 publications [6]. Since the FDA’s approval of the da Vinci Surgical System for gynecologic procedures it has integrated rapidly into the surgical treatment of endometrial cancer. This process lead to substantial research on this topic associating its use with favorable outcomes [24]. Later, as the LACC trial raised controversy concerning the use of minimally invasive radical hysterectomy as the primary treatment for early stage cervical cancer [25], associated with a need for further ongoing large randomized trials regarding the role of robotics in this setting [26]. Furthermore, robotic surgery has recently been integrated into cytoreductive surgery in ovarian cancer [27, 28], further expanding the gynecologic oncology literature on robotic surgery. Importantly, the oncologic research importance and impact is reflected by our finding that it is independently associated with a high CPY score.

We found that most publications originated in North America. According to Intuitive Surgical, 61.5% of the da Vinci systems installed until 2021 were in the U.S. The remaining systems were installed in Europe (17.8%), Asia (15.6%) and the rest of the world (5.1%). This report parallels the proportion of publications from different continents in our study (Fig. 4). Robotic surgery was originally developed to allow for remote surgery in space and for military purposes [29, 30]. Similarly, robotic surgery devices can potentially allow access to advanced surgery in remote areas, including low resource countries. Intuitive Surgical were the only providers in the global market for many years, and their hegemony is slowly decreasing with an 80% share in 2020 [31, 32]. Thus, this company’s report on the number of robotic surgical systems in LMIC, along with the proportion of publications from these countries, may provide a close approximation of their access and use of robot-assisted surgery. These data reveal a major disparity between high income countries and LMIC, that can be explained by the high estimated cost of robot-assisted laparoscopy systems, ranging $1.8–2.3 million U.S. dollars, the high maintenance costs, and the additional costs for the instruments, resulting in approximately $2000 additional U.S. dollars per robot-assisted case compared to laparotomy [33].

We did find an increase in the representation of LMIC during the recent period, corresponding to the recent introduction of robotic systems, leading to publications based on lower numbers, case series and case reports, and mostly on benign gynecology rather than oncologic diseases. Considering the potential benefits of robot-assisted surgery to patients and surgeons, this disparity raises the opportunity to improve quality of healthcare in LMIC.

The previous increase in the number of publications found in our study is in line with the ongoing increased rate of robotic surgery in gynecology. The rate of hysterectomies performed robotically has increased parallel to a decrease in abdominal hysterectomies, as well as conventional laparoscopic and vaginal hysterectomies, during the last two decades [34‐36]. The rate of myomectomies performed robotically has also increased since 2012 [21]. According to Intuitive Surgical, gynecology is the second largest U.S. surgical specialty using robotic systems in 2021. The number of procedures performed grew from approximately 282,000 in 2019 to 316,000 in 2021 [23]. While the number of publications might not be an exact proxy of clinical use of robotic surgery, it may reflect its emergence as a dominant surgical approach. It is unclear why the number of publications per year reached a peak in 2014 and hasn’t grown since then. It is possible that studies report robotic procedures as part of a cohort of minimally invasive surgeries, together with conventional laparoscopy. Indeed, a recent scientific review of robotic surgery literature reported an ongoing increase in publications per year [6]. However, between 2014 and 2018 there was a smaller increase in the publication rate. Unfortunately, this trend is not discussed in that review. As robotic gynecologic surgery may be advantageous both to patients and surgeons, this literary “plateau” should be further investigated. Alternatively, the innovation introduced by robotics has plateaued and further new technologies associated with the computer interface, such as image analysis, machine learning, and artificial intelligence are anticipated in order to be researched and reported.

Our bibliometric study has some limitations inherent to this type of studies. First, we used one literature database and may have missed robotic surgery studies. Second, we reported impact factor per journal according to the latest available impact factor list, and not by publication year’s specific impact factor. This may introduce skewing of the results. Furthermore, a topic-specific limitation is that articles studying robotic surgery may not be titled as such and thus were not included in our analysis. However, the selection of studies that focus on robotic surgery makes our bibliometric representation of the literature more specific.

We did not find a prior study evaluating publications and bibliometrics of robotic surgery in OBGYN. Thus, this study may be the first evaluation of this research question. Second, we used several citation metrics in our analysis, including CPY and RCR, which overcome the bias of counting the total citations number, favoring older studies.