Clinical Outcomes of Robotic Surgery Compared to Conventional Surgical Approaches (Laparoscopic or Open)

A Systematic Overview of Reviews

Hala Muaddi, MD, MSc; Melanie El Hafid; Woo Jin Choi, MD; Erin Lillie, MSc; Charles de Mestral, MD, PhD; Avery Nathens, MD, PhD; Therese A. Stukel, PhD; Paul J. Karanicolas, MD, PhD


Annals of Surgery. 2021;273(3):467-473. 

In This Article


Utilization of robotic surgery has rapidly increased over time. The robot enhances ergonomics and offers several technical advantages over laparoscopic surgery such as seamless motion, less surgeon fatigue, filtering tremor, 7 degrees of motion, and 3-dimensional vision.[5] These assets may overcome some of the limitations of laparoscopic surgery and partly explain rapid uptake in technology. Yet the benefits in clinically relevant patient-important outcomes such as disease-free survival and postoperative complications that affect quality of life rather than surrogate measures remain unclear and in some surgical specialties controversial.[4,40–43]

Herein we conducted an overview of reviews using a robust systematic approach to examine postoperative outcomes after robotic surgery compared to the alternative conventional approach; laparoscopic or open. There were 3 main findings from our overview. First despite the wide incorporation of robotic surgery, since January 2017 the systematic reviews combined identified only 18 RCTs across 18 different surgical procedures. Those include 1 RCT comparing robotic-assisted to open radical prostatectomy, 2 RCTs comparing robotic-assisted to laparoscopic radical prostatectomy, 4 RCT comparing robotic to laparoscopic hysterectomy, 5 RCTs comparing robotic to laparoscopic rectal surgery, 5 RCTs comparing robotic to laparoscopic cholecystectomy, 1 RCT comparing robotic to laparoscopic Roux-en-Y gastrin bypass. Future systematic reviews may identify other procedures and RCTs however our message remains the same; with such a few numbers of RCTs, it would be challenging to reach conclusions regarding the effectiveness of robotic surgery.

Second, many of the identified RCTs demonstrated no sustained differences in clinically relevant outcomes between the surgical approaches offered. For example, robotic-assisted radical prostatectomy may offer some improvement in quality of recovery and pain scores compared to open radical prostatectomy, but those differences were not observed after 6 weeks postoperatively.[13] Longer follow up of the RCT will provide further evidence to differences in patient survival and disease recurrence.[12] However when robotic was compared to laparoscopic radical prostatectomy there was preliminary evidence for improved urinary and sexual function.[15,16] There was no benefit for robotic rectal surgery, hysterectomy, cholecystectomy or roux-en-y gastric bypass compared to the alternative standard approach.[23,31,36,38] In fact, there were concerns for added harm with minimally invasive techniques for management of cervical cancer as these patients had higher disease recurrence and lower overall survival when compared to open radical hysterectomy.[28,29]

Third, the systematic reviews identified several retrospective and prospective non-randomized observational studies; however, these studies were limited by residual confounding bias, selection bias and observer bias. Furthermore, many of the systematic reviews are of low quality critically. Therefore, more rigorous methodology is required to evaluate patient-important outcomes after robotic surgery. Additionally, given that the presence of a financial conflict of interest maybe associated with these studies reporting improved outcomes with robotic surgery, a larger effort should be made by primary studies to report their conflict of interest and future systematic reviews should take these into consideration in their analysis and reporting.[44]

Implementation of surgical innovation introduces several challenges. In general, the regulatory standards for new devices are less rigorous than for new drug development, such that most devices receive regulatory approval when they demonstrate clinical safety.[45] This does not necessarily address the effectiveness of these devices in a clinical setting.[46] Subsequent to introducing the new devices into the clinical setting several strategies for post market surveillance of medical devices are in place to continue monitoring the novel technology.[47] Therefore a balance must be struck between premature adoption into a clinical setting, which may subject patients to potential harm, and delayed adoption, which may withhold potential benefits to patients.

Although this systematic review focuses on robotic surgery, the challenges of assessment of effectiveness of surgical innovation apply broadly to all areas of surgery. Early adoption of innovation may be related to patient demand, low cost for the surgeon to learn and use the technology, and manufacturer promotion.[48] It is difficult to reliably plan for a timely assessment of effectiveness through RCTs such that at an early stage, the innovative technology may not be perfected and surgeons may still be in the early phase of the learning curve.[48] At a later stage, if adoption is high, it may no longer be feasible to conduct a formal RCT assessment because the technology has already been integrated into the healthcare system.[48] Other challenges faced in establishing RCTs relate to surgeon and patient preferences to a specific treatment, infrequent outcomes and loss to follow up.[49] As a result, less than 10% of the published articles in surgical journals are classified as RCTs.[50] Nonrandomized studies are easier to conduct than RCTs particularly with the availability of electronic data collection and administrative databases.[49] However the lack of rigorous planning, poor data quality, and unmeasured confounders, undermine the validity of the results from those sources.[49]

The IDEAL framework was conceptualized to categorize surgical innovation into distinct stages with a structured process of investigation focused on evidence-based health care.[51,52] IDEAL comprises 5 unique stages: Idea (stage 1), Development (stage 2a), Exploration (stage 2b), Assessment (stage 3), and Long-term study (stage 4).[51,52] These stages of surgical innovation emphasize the need to achieve clinical evidence to justify further implementation of the innovation.[51] Because the development of this framework there has been significant improvement in reporting and conducting preliminary studies before initiating RCTs. However, there remains a persistent weakness of assessment of surgical innovation as demonstrated by the number of surgical procedures that had a widespread adoption without adequate evidence of effectiveness as was the case of robotic prostatectomy.[52] This is not limited to radical prostatectomy, such that most procedures also omitted investigating effectiveness of robotic surgery through stages 2a, 2b, and 3 of the IDEAL framework and focused heavily on stage 4 (long-term study) through observational non-randomized studies and implementation into surgical practice.

With the rapid expansion of robotic surgery and the paucity of data regarding clinical benefits relative to cost, the future applications of robotic surgery should be required to undergo rigorous evaluation at the development (stage 2a), exploration (stage 2b), and assessment (stage 3) stages as described by the IDEAL framework.[6,7,51] These goals are best achieved through prospectively designed national registries and RCTs.[48,51] In the case of robotic surgery a tipping point may have occurred where wide spread adoption took place without adequate evidence.[51] This poses further challenges for conducting RCTs because patients and surgeons have developed strong subjective preferences and may be reluctant to randomize.[51] There is a role for methodologically sound observational studies, however causal inferences established from nonrandomized studies are weaker than those identified in RCTs and must be interpreted with caution.[49] In reality, the stages of assessment and implementation of innovation are not linear, and the stages will often overlap.[51] Therefore future studies should focus on stages 2a, 2b, and 3 to examine clinically relevant patient-important outcomes rather than surrogate measures.[4] Furthermore, given the high costs associated with the purchase of the robot and the use of disposable instruments economic evaluation studies are needed to ascertain cost effectiveness of robotic surgery. This research is instrumental to inform future implementation of this technology and funding decisions.