Robot-Assisted Surgery

Robot-assisted surgery is surgery in which a surgeon uses computer-controlled robotic equipment to manipulate instruments or assist with specific operative tasks. It includes console-operated systems for soft-tissue procedures and robotic devices that support image-guided positioning or bone preparation. Robotic assistance describes how surgical tools are controlled, rather than a single operation or treatment; it does not necessarily imply autonomous decision-making. Many applications extend minimally invasive surgery, including laparoscopy, while others assist orthopedic procedures. (fda.gov)

The technology brings together robotics, imaging, software, and surgical practice. Its significance lies in the ability to mediate instrument movement, provide visual guidance, and support work within restricted anatomical spaces. Clinical outcomes, however, must be assessed for particular operations and patient populations rather than inferred from technical capabilities alone. The field encompasses the development and operation of surgical systems, their specialty-specific uses, comparative clinical evidence, professional training, safety oversight, economic implications, and research into remote operation and supervised automation. (fda.gov)

Scope, Terminology, and Levels of Robotic Assistance

Robotic assistance is an operative technology used within surgery, not a separate surgical specialty or a complete description of an operation. Three attributes should be distinguished: the procedure, meaning the anatomical intervention; the access route, meaning how the operative site is reached; and the equipment, meaning the instruments and control system used. In a robot-assisted laparoscopic prostatectomy, for example, prostatectomy identifies removal of the prostate, laparoscopic identifies the access method, and robot-assisted identifies how instruments are controlled. The same technological category also supports procedures such as hysterectomy and gallbladder removal; it does not make these operations clinically equivalent. (fda.gov)

Computer-assisted surgery is the broader category. It includes preoperative planning, surgical navigation, and assistance with operative actions. Planning software can establish intended anatomical targets, while navigation can report an instrument’s position relative to the patient without physically moving it. Robotic systems add mechanical assistance with positioning or instrument manipulation. These functions can coexist within one platform, but displaying guidance and acting on tissue are distinct capabilities. Regulatory terminology may also be narrower than general technological usage: the US Food and Drug Administration’s September 25, 2026 draft guidance on robotically assisted surgical devices excludes preoperative planning and stereotaxic navigation systems from that document’s scope. (fda.gov)

In surgeon-controlled teleoperation, the surgeon supplies movement commands through an interface, and software translates them into movements of patient-side instruments. This leader–follower relationship can include motion scaling and filtering of physiological tremor without transferring operative decision-making to the machine. Teleoperation does not necessarily imply long-distance surgery: the console and patient-side equipment can occupy the same operating room. The FDA’s 2026 draft describes these systems as operating under a qualified practitioner’s direct control or supervision and distinguishes automated features from autonomy, which involves performing tasks and making decisions from sensed conditions without user intervention. (fda.gov)

Robotic guidance and bounded task autonomy represent different allocations of control. Guidance can constrain a surgeon-operated tool to a planned region while leaving its manipulation under continuous human control. Task autonomy instead permits a system to execute a specified action after human initiation, with the operator monitoring and intervening as necessary. In a six-level framework proposed by Guang-Zhong Yang and colleagues in 2017, teleoperation with motion scaling was classified as level 0, mechanical guidance under continuous human control as level 1, and task autonomy as level 2. Their illustrative level-2 example was suturing in which the surgeon specifies the suture location and the robot executes the task under supervision. These are categories of capability, not evidence that a system can independently conduct an entire operation. (biomechatronics.ucla.edu)

Robotic assistance and minimally invasive surgery therefore describe different dimensions of an intervention. Soft-tissue laparoscopic platforms manipulate instruments through small incisions, whereas equipment used in orthopedic surgery may principally support anatomical localization, implant planning, and bone preparation. The VELYS Robotic-Assisted Solution’s June 2024 FDA submission, for example, describes optical acquisition of anatomical landmarks and intraoperative planning of femoral and tibial implant positions. During bone preparation, the device constrains a surgeon-operated saw to planned resection planes and compensates for leg movement in real time. Here, robotic assistance concerns geometric guidance of bone cutting rather than laparoscopic access or autonomous soft-tissue dissection. (fda.gov)

Historical Development and Commercial Translation

During the 1980s, SRI International developed telepresence surgery under contract to the United States Army. Biomedical engineer Phillip S. Green and his colleagues combined stereoscopic displays, miniature cameras, remote-control electronics, and robotic manipulators holding surgical instruments. Their Green Telepresence System separated the surgeon’s workstation from a remote surgical unit: the surgeon viewed three-dimensional video and moved paired hand controls, whose inputs were transmitted to instrument manipulators. The research addressed the possibility of operating where an expert surgeon could not be physically present. Its central contribution was the integration of visualization and instrument control into a surgeon-operated system, rather than the replacement of surgical judgment by automation. (sri.com)

Commercial translation followed through Intuitive Surgical, established in 1995, which licensed SRI’s technology. The company subsequently developed the da Vinci surgical system and launched its initial commercial system in 2000. This chronology distinguishes the underlying research program from the later commercial product: da Vinci emerged from further development of telepresence concepts, not simply the renaming of an experimental apparatus. Intuitive’s historical account identifies a prototype surgeon console from 1996 and describes the period from 1995 to 2004 as the transition from prototypes to installed systems. Although remote operation had motivated the earlier research, the commercial platform brought surgeon-controlled robotic instrumentation into hospital operating rooms, including applications in minimally invasive surgery. (sri.com)

Long-distance operation received a separate demonstration on September 7, 2001, when Jacques Marescaux and colleagues performed a laparoscopic gallbladder removal on a patient in Strasbourg while controlling instruments from New York. Known as Operation Lindbergh, the procedure used Computer Motion’s ZEUS system, rather than da Vinci. France Télécom supplied a dedicated fiber-optic connection using asynchronous transfer mode technology; the investigators reported a mean delay of approximately 155 milliseconds between the surgeon’s action and the returned visual response. Telecommunications engineering was therefore integral to the achievement, alongside the surgical robot. The operation demonstrated transatlantic feasibility under specially arranged conditions, rather than establishing routine provision over ordinary hospital networks; the investigators noted that most hospitals then lacked the required network equipment. (pmc.ncbi.nlm.nih.gov)

Commercial diversification also occurred within surgeon-controlled soft-tissue systems. On October 13, 2017, the United States Food and Drug Administration cleared TransEnterix’s Senhance system for specified gynecological and colorectal procedures. Its 2021 regulatory documentation describes a multi-arm teleoperated platform in which console inputs control arms carrying laparoscopic instruments or an endoscope. Optional force feedback supplies tactile information at the handles, while eye tracking offers a method of directing the endoscope. Senhance thus represented another implementation of teleoperation, with different control features, rather than a transition to autonomous surgery. (accessdata.fda.gov)

Diversification extended beyond console-based laparoscopy into orthopedic surgery. FDA documentation from 2014 describes Blue Belt Technologies’ Navio as a navigation and bone-burring system using infrared tracking and a handheld instrument. Software could retract the bur into a guard, or stop its rotation, as it reached a planned bone surface. (accessdata.fda.gov) A different configuration appears in the VELYS documentation cleared on June 5, 2024: an operating-table-mounted robotic arm constrains a surgeon-manipulated saw to planned resection planes and compensates for leg movement. These products illustrate distinct commercial approaches to bone preparation, with procedure- and implant-specific indications, rather than interchangeable alternatives to soft-tissue teleoperation. (accessdata.fda.gov)

System Architecture, Imaging, and Instrument Control

A representative soft-tissue system separates the surgeon’s control station from the mechanisms that move instruments at the patient. The console provides a view of the operative field and interfaces for commanding instruments and the endoscope. A patient-side cart carries articulated manipulators—commonly three or four arms—with attachments for the endoscope and interchangeable surgical instruments. Supporting equipment, often organized on a separate tower, includes image-processing hardware, an endoscopic light source, an electrosurgical generator, and suction or irrigation equipment. Thus, the manipulators are only one part of an integrated surgical installation: visualization, mechanical movement, and delivery of surgical energy depend on distinct but connected subsystems. (fda.gov)

Instrument movement follows a leader–follower control architecture: the surgeon supplies commands at the console, and software translates those inputs into movements of the patient-side instruments. Translation need not reproduce hand displacement at identical scale. Motion scaling reduces the size of instrument movements relative to the operator’s movements, while tremor filtering attenuates unintended input frequencies without deliberately suppressing intended motion. These are control functions, not independent decisions about which tissue to cut or suture. The FDA’s September 25, 2026 draft guidance describes leader–follower systems that may incorporate automated functions while remaining under a qualified practitioner’s direct control or supervision. (fda.gov)

Stereoscopic visualization supplies depth information through separate optical views rather than merely enlarging a conventional laparoscopic image. In Intuitive’s da Vinci systems, the three-dimensional high-definition endoscope has two independent vision channels linked through image-processing electronics and software to separate displays in the console viewer. The paired views provide binocular depth perception; illumination makes the operative structures visible, while the endoscope determines the viewing position. Digital magnification can enlarge the displayed field without physically advancing the scope. Camera-control interfaces allow the surgeon to reposition the view using hand controls and foot controls, separating the task of looking around from the task of manipulating tissue. (isrg.intuitive.com)

Instrument design determines what the commanded movements can accomplish. Wristed tools provide articulation at the instrument tip, rather than relying solely on movement of a rigid shaft. For example, da Vinci Xi instruments incorporate distal bending and rotation, alongside motion filtering. Instruments and accessories nevertheless retain different physical functions: graspers hold tissue, scissors divide it, and electrosurgical instruments apply energy supplied by a generator. The console interface therefore coordinates several kinds of action—instrument positioning, camera movement, and instrument activation—rather than controlling a single interchangeable robotic tool. (intuitive.com)

Systems used in orthopedic surgery can have a substantially different architecture. The VELYS Robotic-Assisted Solution documentation associated with FDA clearance on June 5, 2024 describes an image-free system combining optical tracking, intraoperative anatomical landmark acquisition, implant-planning software, and a robotic cutting aid. Here, “image-free” does not mean that the system lacks optical sensing: landmarks acquired during surgery establish the anatomical references used to plan femoral and tibial implant positions. During bone preparation, the robotic device constrains the saw handpiece and blade to a planned resection plane, while the surgeon actuates and manipulates the saw within that plane. This arrangement assists execution of a geometrical bone-cutting plan, unlike a console-based soft-tissue system that translates hand movements into dissection or suturing movements. (accessdata.fda.gov)

Clinical Applications Across Surgical Specialties

In urology, robot-assisted radical prostatectomy is a laparoscopic approach to removing the prostate for prostate cancer: the operation remains a radical prostatectomy, while robotic assistance describes how the surgeon controls instruments through abdominal incisions. A camera supplies a three-dimensional view of the prostate and surrounding structures; instrument movements remain surgeon-controlled. Radical cystectomy is a separate operation involving bladder removal and urinary diversion, which may also be performed robotically. Reconstruction can be completed inside the abdomen rather than through an additional open incision. This distinction was incorporated into the 2022 iROC trial: 338 participants were randomized to robot-assisted cystectomy with intracorporeal urinary diversion or open cystectomy, illustrating that the reconstructive approach, not simply the presence of a robot, helps define the intervention. (cancer.gov)

General surgical applications include cholecystectomy, or gallbladder removal, and inguinal hernia repair. Colorectal applications include colon and rectal resections, in which robotic instruments assist tissue dissection and removal; pancreatic applications include pancreatectomy and pancreaticoduodenectomy, also called the Whipple procedure. These operations involve different anatomical targets and reconstructive requirements despite using related instrument-control technology. A rectal resection may require reconnection of bowel, whereas pancreatic surgery may require connections between the remaining pancreas, bile duct, and intestine. The underlying procedure therefore determines important clinical endpoints, including bowel-anastomotic leakage or pancreatic and biliary leakage; these are not interchangeable measures of “robotic” performance. (fda.gov)

In gynecology, hysterectomy—removal of the uterus—is a common application. Robotic assistance can support laparoscopic dissection and suturing, but it does not determine whether the operation is simple or radical, or whether additional tissues are removed. Other gynecologic contexts include myomectomy and pelvic reconstructive procedures such as sacrocolpopexy. Thoracic applications include lung lobectomy and mediastinal surgery. In cardiac surgery, robot-assisted mitral valve repair provides a distinct example: instruments enter through small chest incisions, but the repair still requires access to the heart and cardiopulmonary bypass. These applications differ substantially in operative anatomy and physiological support requirements. (fda.gov)

Head-and-neck applications include transoral robotic surgery, in which the surgeon controls instruments introduced through the mouth to reach selected areas of the mouth and throat. The National Cancer Institute describes its use for removing cancers in difficult-to-reach locations, with cameras providing three-dimensional visualization. Unlike abdominal robotic laparoscopy or thoracic surgery, this approach uses a natural opening rather than abdominal or chest ports; “robot-assisted” consequently does not specify a single access route. (cancer.gov)

In orthopedic surgery, robotic assistance in knee arthroplasty concerns implant planning and bone preparation rather than soft-tissue dissection and suturing. The Navio system cleared by the US Food and Drug Administration on July 2, 2014, used intraoperatively collected anatomical data to construct a model and plan implant placement for unicondylar and patellofemoral replacement. Optical tracking located the instruments relative to the bones. Software constrained a handheld cutting bur by retracting its tip into a guard or controlling its speed as the planned surface was approached. The surgeon thus manipulated the tool while the system limited bone removal according to the plan. (accessdata.fda.gov)

Device indications and clinical evidence remain procedure-specific. In the United States, FDA clearance of systems for specified surgical uses must not be interpreted as proof of improved cancer control. The FDA distinguishes authorization to assist surgical procedures from authorization specifically for cancer prevention or treatment. Likewise, the manufacturer’s precaution statement for da Vinci Xi/X systems notes that evaluations of representative procedures did not establish cancer-related outcomes such as overall survival, disease-free survival, or local recurrence. Evidence from one operation therefore does not establish oncologic effectiveness across specialties. (fda.gov)

Operating-Room Workflow and Team Competence

In console-based soft-tissue surgery, the operating surgeon controls the camera and robotic instruments, but bedside work remains essential. A bedside assistant exchanges instruments, provides suction and surgical exposure, and introduces sutures or other materials. Scrub personnel, circulating nurses, and anesthesia professionals retain distinct responsibilities within the operating-room team. Because the console surgeon is physically separated from the sterile field, bedside personnel also communicate conditions outside the endoscopic view. Robotic equipment therefore redistributes tasks rather than replacing the clinical team. The FDA explicitly includes other surgical staff, not only surgeons, in its expectations for device-specific training. (fda.gov)

Workflow encompasses equipment preparation, room layout, patient positioning, docking—the connection of patient-side robotic arms to the operative access equipment—and instrument deployment. These activities require coordination between sterile-field personnel and the anesthesia team, alongside arrangements for troubleshooting and emergency undocking. Communication is particularly important during instrument changes and interruptions: the University of California, San Francisco (UCSF) curriculum specifies closed-loop communication, in which messages are acknowledged rather than merely transmitted. In an October 2025 account published by the Association of periOperative Registered Nurses, Duke University Hospital robotics nursing program manager Stacy Williams described multidisciplinary simulation covering docking, equipment failures, and emergency undocking. Such exercises rehearse role allocation and coordinated responses without exposing patients to a training scenario. (surgicalskillslab.ucsf.edu)

Training pathways combine theoretical instruction, simulation, laboratory practice, observation, and progressively supervised operating. Theoretical content covers equipment functions and procedure-related considerations; simulation permits repeated practice of camera control, instrument manipulation, suturing, and energy-device use. Physical laboratory exercises add experience with equipment setup and troubleshooting. Clinical training then introduces bedside assistance and selected console tasks under an experienced surgeon’s supervision. These components are complementary: familiarity with the controls is distinct from competence in performing an operation. The Royal Australasian College of Surgeons’ June 2023 working-party report identified the need for appropriate curricula and assessment rather than a single approach applicable to every specialty and platform. (surgicalskillslab.ucsf.edu)

A concrete example is UCSF’s general-surgery curriculum, revised on August 23, 2023. It combines online modules, dry laboratories, scored simulation exercises, and supervised clinical experience. Its documented pathway to a manufacturer-issued equivalency certificate includes logs of 10 bedside-assistant cases and 20 console cases involving significant portions of the operation, verified by an attending surgeon, together with a program-director letter documenting experience and competence. These are requirements of that particular pathway, not universal thresholds for independent practice. The American College of Surgeons (ACS) likewise distinguishes case volume from demonstrated competence and emphasizes assessment of performance and outcomes. (surgicalskillslab.ucsf.edu)

Manufacturer training, professional qualifications, and institutional credentialing and privileging serve different purposes. Manufacturer instruction addresses use of a particular system; professional qualifications establish broader surgical education and specialty standing. Credentialing verifies qualifications, whereas privileging determines which procedures a clinician may perform at a particular institution. Completing a device course does not itself confer hospital privileges. In its April 2018 statement, the ACS described supervised introduction, proctored assessment, and monitoring during provisional practice before fully independent privileges are granted. Review includes patient outcomes and institutional resources, rather than technical proficiency alone; continued evaluation can draw on operative records, peer review, and risk-adjusted outcome data. (fda.gov)

Comparative Clinical Evidence and Outcome Measurement

Comparisons must identify the operation, patient population, surgical access route, and experience of the participating surgeons. Robotic assistance versus open surgery is a different comparison from robotic assistance versus conventional laparoscopy: the former changes both access and equipment, whereas the latter compares two minimally invasive approaches. Prior experience also matters because surgeons may have substantially different proficiency in the two techniques. Technical capabilities, such as facilitating dissection in confined spaces, are not themselves evidence of fewer complications or better recovery; those benefits require measurement in comparative clinical studies. (fda.gov)

Endpoints distinguish operative performance from patient outcomes. Conversion to open surgery records whether an intended minimally invasive operation requires an open approach; operating time measures procedural duration rather than recovery. Clinical endpoints include intraoperative injuries, postoperative complications, readmission, and mortality within a specified interval. Recovery can be assessed through hospital stay, return to activity, pain scores, and validated quality-of-life questionnaires, ideally with baseline measurements. Cancer studies additionally examine resection margins, recurrence, disease-free survival, and overall survival. These endpoints are not interchangeable: an improvement in operative completion or specimen quality does not by itself establish longer survival. ROLARR, for example, separately specified conversion, pathological findings, complications, functional outcomes, and longer-term disease outcomes. (njl-admin.nihr.ac.uk)

The 2017 ROLARR randomized controlled trial, reported by David Jayne and colleagues, enrolled 471 patients with rectal adenocarcinoma suitable for curative resection across 29 sites in 10 countries, involving 40 surgeons. Its primary endpoint was conversion to open laparotomy. Among patients with evaluable conversion data, conversion occurred in 19 of 236 robotic operations (8.1%) and 28 of 230 conventional laparoscopic operations (12.2%). The adjusted odds ratio was 0.61, with a 95% confidence interval of 0.31–1.21 and $P = .16$. Although the numerical difference favored robotic assistance, the interval included no difference and did not establish superiority. Prespecified secondary endpoints included circumferential resection-margin positivity, complications, 30-day mortality, and bladder and sexual function; these showed no statistically significant between-group differences in the report. Sensitivity analysis suggested that prior robotic experience could influence conversion outcomes, making the learning curve an important qualification rather than a basis for replacing the primary result. (jamanetwork.com)

The 2020 RIVAL pilot trial, reported by Ajita S. Prabhu and colleagues, analyzed 102 patients at six academic or affiliated sites: 48 underwent robotic and 54 conventional laparoscopic transabdominal preperitoneal inguinal hernia repair. Participants had unilateral primary or recurrent hernias, and approximately 90% were men. At 30 days, the study detected no significant differences in measured pain, health-related quality of life, mobility, wound morbidity, or cosmetic outcomes. Median operating time was nevertheless longer with robotic assistance: 75.5 versus 40.5 minutes. The findings therefore separated procedural duration from measured short-term patient benefit. The pilot sample, selected hernia population, and 30-day reporting period did not establish equivalence or resolve long-term recurrence and chronic pain; nor did they represent all forms of hernia repair. (jamanetwork.com)

Randomized trials and clinical registries provide complementary evidence. Randomization reduces systematic differences in treatment allocation, while registries can capture broader clinical populations, variation between institutions, and longitudinal outcomes in routine practice. Registry comparisons remain vulnerable to confounding, including differences in patient risk, case selection, and provider experience; incomplete follow-up can further distort estimates. Analyses therefore require clearly defined outcomes, relevant adjustment variables, and explicit handling of missing data. Surgical learning effects also require attention within randomized trials: a subsequent ROLARR analysis modeled prior operative experience because unequal expertise could distort treatment comparisons. Long-term, procedure-specific follow-up is necessary rather than extrapolating a short-term result from one operation to another. (fda.gov)

Safety, Device Failure, and Surveillance

A surgical complication and an equipment malfunction are different observations: patient harm does not by itself demonstrate device failure, and a malfunction need not cause injury. The United States Food and Drug Administration (FDA) identifies component breakage, mechanical problems, and image or display failures among reported problems with robotically assisted systems. Most reports it receives concern malfunctions, generally without associated injuries, although injuries and deaths have also been reported. Establishing causation therefore requires more than identifying that robotic equipment was used during an adverse event. (fda.gov)

A retrospective study by Homa Alemzadeh and colleagues examined 10,624 FDA reports concerning robotic systems and instruments submitted between January 2000 and December 2013. These included 144 reports of deaths, 1,391 of injuries, and 8,061 of malfunctions; these are reporting counts, not rates among operated patients. Documented failure mechanisms included broken or burnt instrument fragments falling into patients, electrical arcing, unintended instrument operation, system errors, and video or imaging problems. In 1,104 reported events, an operation was interrupted to restart equipment, change to a nonrobotic technique, or reschedule the procedure. Such interruptions distinguish a failure’s operational consequences from whether it produced a recorded injury. (arxiv.org)

Safety assessment also concerns the interaction between users, equipment, and the hospital. The FDA notes that system models differ in operation and functions, making model-specific training relevant to user–device interaction. The American College of Surgeons’ April 2018 statement identifies surgeon training, experience, institutional resources, and outcome measurement as determinants of surgical safety. It specifically includes operating-room equipment, intensive-care services, and blood-bank capabilities among resources shaping the complexity of care an institution can provide. Its framework calls for monitored introduction of new technologies and examination of risk-adjusted outcomes rather than reliance on case numbers alone: procedural volume can reflect the capabilities of an entire team, not simply an individual surgeon’s skill. (fda.gov)

Postmarket surveillance combines several information streams. In the United States, the FDA receives mandatory medical-device reports from manufacturers, importers, and device user facilities, alongside voluntary reports from clinicians, patients, caregivers, and consumers, including submissions through MedWatch. It also reviews published studies and case histories. Patient registries provide a complementary source of longitudinal information: the FDA identifies their potential to characterize learning curves, examine longer-term outcomes, and detect emerging problems. These sources address different questions; an individual malfunction report can identify a suspected failure, whereas systematic follow-up can examine patterns across patients and over time. (fda.gov)

Spontaneous reporting cannot independently establish complication rates or support straightforward comparisons between platforms. The FDA’s Manufacturer and User Facility Device Experience (MAUDE) reporting system can contain incomplete, inaccurate, unverified, or duplicated information. Underreporting leaves the number of events uncertain, while missing information about how frequently devices were used leaves the population at risk unknown. Consequently, dividing report counts by an assumed procedure total does not necessarily yield a reliable incidence estimate. A submitted report also does not establish that a device caused the event or was defective. Report-based safety signals require investigation and interpretation alongside other clinical evidence, rather than being treated as confirmed causal findings. (fda.gov)

Regulation, Consent, and Clinical Responsibility

In the United States, the Food and Drug Administration (FDA) regulates robotically assisted surgical devices for specified intended uses; authorization is not an endorsement of robotic surgery as a universally interchangeable technique. The relevant boundaries are those of the particular system, its instruments, and its authorized indications. For devices reviewed through the 510(k) pathway, manufacturers must establish substantial equivalence to a legally marketed predicate device, including consideration of technological differences. Authorization to assist with an operation does not itself establish every disease-related benefit: the FDA distinguishes clearance for procedures such as prostatectomy from marketing authorization specifically for preventing or treating cancer. (fda.gov)

On September 25, 2026, the FDA issued Robotically-Assisted Surgical Devices—Premarket Submissions as draft guidance, distributed for comment and marked as not for implementation. It is not a finalized requirement, and its recommendations are nonbinding. The draft addresses device architecture, compatible instruments, teleoperation, mechanical performance, visualization, software, cybersecurity, human factors, and biological safety. One concrete proposal is bench testing of positional accuracy, repeatability, and resolution using software-driven commands, independently assessing translational and rotational motion. It recommends testing at least three systems for these characteristics, with justification if fewer are used. It also proposes testing whether position-tracking errors trigger notification, movement limitation, stopping, or transition to a safe state. These are proposed evaluation methods, not numerical performance thresholds applicable to every surgical robot. (fda.gov)

Informed consent concerns the proposed operation, not simply permission to use a machine. The American College of Surgeons’ (ACS) Statements on Principles, updated in April 2016, specify discussion of the illness, consequences of no treatment, proposed operation, risks, expected recovery, alternatives—including nonoperative treatment—and participating clinicians’ roles. Applied to robotic assistance, this framework encompasses how the device will be used and how that approach differs from available alternatives, such as conventional laparoscopy or open surgery. The FDA separately identifies the risks and benefits of alternative treatments and the surgeon’s device-specific training and experience as relevant subjects for discussion. The ACS prohibits exaggerating benefits or making guarantees; uncertain procedure-specific advantages therefore cannot be presented as established merely because the equipment is authorized. (facs.org)

Clinical responsibility remains with identifiable professionals. Under ACS principles, the surgeon determines the need for and type of operation, obtains consent, and provides perioperative surgical care; the primary attending surgeon is personally responsible for the patient’s welfare during the operation. Participation by assistants or trainees requires disclosure of their roles and appropriate supervision. These professional responsibilities are distinct from the manufacturer’s responsibility for device design and performance. The FDA also assigns development and implementation of device training to manufacturers, physicians, and healthcare facilities rather than accrediting individual surgeons itself. (facs.org)

Institutional authorization to operate is a separate decision from device marketing authorization. The ACS’s April 2018 credentialing statement distinguishes verification of professional qualifications from privileging, which specifies the procedures a clinician may perform at a particular institution. For new technologies, it calls for assessment of additional training and experience, potentially including proctoring, together with institutional resources and outcome review. Manufacturer instruction consequently does not replace institutional assessment of clinical competence. These allocations describe regulatory and professional functions within the cited United States framework; they do not establish a jurisdiction-independent rule assigning legal liability whenever an adverse outcome involves robotic equipment. (facs.org)

Costs, Cost-Effectiveness, and Service Provision

Economic evaluation separates fixed investment from expenditure associated with each operation. Acquisition, installation, operating-room adaptation, sterilizing equipment, and maintenance contracts contribute to fixed costs; reusable instruments, disposable supplies, staff time, and operating-room occupancy contribute to per-case costs. Training adds expenditure during implementation and subsequent workforce development. Evaluation of the treatment episode also includes inpatient care, intensive care, complications, and readmissions, because savings after surgery may offset some additional operative expenditure. Capital costs require explicit assumptions about equipment lifespan and case volume: spreading the same investment across more operations reduces its allocated cost per case, without necessarily reducing consumable expenditure. A 2021 study at Sydney’s Royal Prince Alfred Hospital, for example, annualized the robotic system and sterilizing equipment over ten years and theatre refurbishment over twenty years. (link.springer.com)

Hospital costs, charges, and reimbursement measure different quantities. Costs represent resources consumed in providing care; charges are billed amounts; reimbursement is payment by a purchaser or insurer and need not equal either. Cost-effectiveness analysis instead relates differences in expenditure to differences in health outcomes. Its perspective—such as the hospital or the wider health service—determines which expenditures count, while its follow-up period determines how much subsequent recovery and care are captured. Cost-utility analysis commonly expresses outcomes as quality-adjusted life-years (QALYs), combining survival duration with health-related quality of life. An incremental cost-effectiveness ratio divides additional cost by additional health benefit; it does not establish that an intervention is affordable within a particular budget. (effectivehealthcare.ahrq.gov)

The 2020 RIVAL randomized pilot trial, led by Ajita S. Prabhu, illustrates a procedure-specific comparison with conventional laparoscopic surgery. Among 102 patients analyzed after unilateral inguinal hernia repair, median reported cost was US$3,258 for robotic repair and US$1,421 for laparoscopic repair; median operating time was 75.5 versus 40.5 minutes. Costing included operating-room time and disposable and reusable materials, but did not amortize capital equipment costs for either approach. The trial found no significant differences in its measured 30-day clinical outcomes. These figures therefore describe the studied operations and accounting boundary, rather than a universal price premium or a lifetime economic comparison. (jamanetwork.com)

The iROC economic evaluation, published in 2023 by Simon Dixon and colleagues, compared intracorporeal robot-assisted radical cystectomy with open radical cystectomy for bladder cancer. Its primary analysis included 305 patients from nine United Kingdom centres, adopted a National Health Service perspective, and covered 90 days. Reduced ward, intensive-care, and readmission expenditure partly offset higher equipment and theatre costs. Estimated additional cost was £1,124 per patient, with a 95% confidence interval from −£576 to £2,824; the estimated QALY gain was 0.01124. The resulting ratio was £100,008 per QALY gained, with a 16.3% probability of cost-effectiveness at a £20,000-per-QALY threshold. Unlike RIVAL, this analysis compared robotic with open surgery and incorporated recovery-related health benefits; the two results are not interchangeable. (bladder.uroonco.uroweb.org)

Service capacity depends on equipment distribution, trained surgeons and bedside teams, operating-room allocations, and opportunities for supervised training—not equipment ownership alone. The Royal Australasian College of Surgeons’ June 2023 report recorded 162 platforms across Australia and Aotearoa New Zealand as of May 2023: 136 in private hospitals and 26 in public hospitals, with substantial metropolitan concentration. Utilization also affects programme economics. At Royal Prince Alfred Hospital, 211 operations during August 2016–December 2018 represented 37% of estimated capacity within allocated sessions. The study modelled implementation costs falling from A$4,691 to A$1,730 per case at full utilization, and maintenance costs from A$2,290 to A$845. These were capacity-based estimates, not demonstrated improvements in patient access or outcomes. (surgeons.org)

Remote Operation, Automation, and Research Boundaries

Long-distance telesurgery places the operating surgeon and patient at separate sites while transmitting instrument-control commands and operative images through a telecommunications network. It differs from remote mentoring or teaching, in which a distant clinician provides guidance while the local surgeon retains instrument control. Jacques Marescaux and colleagues demonstrated transatlantic robot-assisted gallbladder removal in 2001, with surgeons in New York and the patient in Strasbourg. Their subsequent detailed report described a 54-minute operation using a dedicated asynchronous transfer mode network, with a mean transmission lag of 155 milliseconds. This established feasibility under a particular technical configuration, rather than proving that arbitrary long-distance connections could support equivalent performance. Communication delay separates a surgeon’s action from its visible consequences, making network performance part of the operative control problem rather than merely a requirement for videoconferencing. (pubmed.ncbi.nlm.nih.gov)

Research on automation also uses synchronized operative images and robot-motion records to develop and compare methods for recognizing tasks and reproducing movements. Irene Rivas-Blanco and colleagues’ ROSMA dataset, first described in a 2021 preprint and revised in 2023, contains 206 trials performed by 12 participants using the da Vinci Research Kit. Its three training exercises—post and sleeve, pea on a peg, and wire chaser—test object transfer, placement, and constrained manipulation. Each trial includes video and 154 recorded kinematic variables from the surgeon-side and instrument-side manipulators. Timestamping makes it possible to associate visible actions with recorded robot states, providing training data and common test material for machine-learning algorithms. These are laboratory skill exercises, not patient operations: performance on this dataset does not itself demonstrate safe tissue handling or clinical effectiveness. (arxiv.org)

A bounded example of supervised task autonomy is the Smart Tissue Autonomous Robot (STAR), described by Hamed Saeidi, Axel Krieger, and colleagues in a 2021 preprint on laparoscopic intestinal anastomosis, or joining bowel ends. The experiments included excised tissue and living porcine models. STAR combined a robotic suturing instrument with near-infrared imaging and three-dimensional tissue reconstruction to generate suture plans and track deformation. An operator selected and approved a proposed plan, authorized replanning, and monitored whether stitches needed correction. An assistant also cleared and tensioned loose suture. Thus, autonomous execution operated within a prepared, supervised task rather than encompassing an independently performed operation. The report described four in-vivo tests comprising 86 stitches; evaluated measures included stitch spacing, placement corrections, lumen patency, and leak pressure. (arxiv.org)

The STAR experiments illustrate the distinction between technical validation and clinical evidence. Consistent stitch placement or satisfactory leak pressure in an animal model evaluates specified aspects of an anastomosis; it does not establish outcomes across human patients, diseased tissues, or different operations. Likewise, recognizing a training exercise from video is different from reliably identifying the operative situation and selecting an appropriate clinical action. These studies support bounded research claims, not general effectiveness claims or authorization for autonomous human surgery. Regulatory permission for a device and intended use remains separate from publication of a successful experiment. (arxiv.org)

Evaluation of autonomous and remote capabilities is an active regulatory-development topic. The United States Food and Drug Administration has announced a public workshop for December 2–3, 2026, specifically addressing benefit–risk evaluation of robotic medical devices with autonomous or remote-teleoperation capabilities. Its stated purpose is to gather experience from manufacturers, healthcare professionals, patients, and other participants to inform assessment of safety and effectiveness. This engagement concerns how such capabilities should be evaluated; the workshop announcement is neither a clinical validation result nor authorization of a particular autonomous or remotely operated surgical system. (fda.gov)

References

1. Computer-Assisted Surgical Systemsfda.gov
2. Robotically-Assisted Surgical Devices - Premarket Submissionsfda.gov
3. VELYS Robotic-Assisted Solution: K240211accessdata.fda.gov
4. Medical robotics—Regulatory, ethical, and legal considerations for increasing levels of autonomybiomechatronics.ucla.edu
5. Telerobotic surgerysri.com
6. 75 Years of Innovation: The Robotic Surgeon (Teleoperator System and Method with Telepresence)sri.com
7. Intuitive Historyintuitive.com
8. Transcontinental Robot-Assisted Remote Telesurgery: Feasibility and Potential Applicationspmc.ncbi.nlm.nih.gov
9. 510(k) Premarket Notification: Senhance Surgical Robotic System, K171120accessdata.fda.gov
10. Senhance Surgical System: 510(k) Summary, K202166accessdata.fda.gov
11. Navio: 510(k) Summary, K140596accessdata.fda.gov
12. Intuitive Surgical Annual Report: 3DHD Vision System and Motion Scalingisrg.intuitive.com
13. Intuitive Surgical Annual Report: Camera Control and Instrument Movementsisrg.intuitive.com
14. Da Vinci Xi Robotic Surgical Systemintuitive.com
15. Prostate Cancer Treatment - NCIcancer.gov
16. Safety Information | Robotic Surgical Company | Intuitiveintuitive.com
17. Da Vinci Surgery | Colorectal Robotic Assisted Surgeryintuitive.com
18. Minimally invasive surgery - Care at Mayo Clinic - Mayo Clinicmayoclinic.org
19. Definition of TORS - NCI Dictionary of Cancer Terms - NCIcancer.gov
20. Robot-Assisted Surgery Working Party Final Report and Recommendationssurgeons.org
21. General Surgery Residency Program: Curriculum for Robotic Surgerysurgicalskillslab.ucsf.edu
22. Robotic Surgery Training: 5 Ways to Build Confidence and Communication in the ORaorn.org
23. Statement on Credentialing and Privileging and Volume Performance Issuesfacs.org
24. Exploring and adjusting for potential learning effects in ROLARR: a randomised controlled trial comparing robotic-assisted vs. standard laparoscopic surgery for rectal cancer resectionpmc.ncbi.nlm.nih.gov
25. Robotic-assisted surgery compared with laparoscopic resection surgery for rectal cancer: the ROLARR RCTnjl-admin.nihr.ac.uk
26. Use of Real-World Evidence to Support Regulatory Decision-Making for Medical Devicesfda.gov
27. Adverse Events in Robotic Surgery: A Retrospective Study of 14 Years of FDA Dataarxiv.org
28. About Manufacturer and User Facility Device Experience (MAUDE) Databasefda.gov
29. openFDA: MAUDEopen.fda.gov
30. Statements on Principlesfacs.org
31. Robot-Assisted Surgery Working Party Final Report and Recommendationssurgeons.org
32. Detailed cost of robotic-assisted surgery in the Australian public health sector: from implementation to a multi-specialty caseloadlink.springer.com
33. Public Reporting of Cost Measures in Health: An Environmental Scan of Current Practices and Assessment of Consumer Centerednesseffectivehealthcare.ahrq.gov
34. AHRQ Quality Indicators Toolkitahrq.gov
35. Cost-Effectiveness of Robot-Assisted Radical Cystectomy vs Open Radical Cystectomy for Patients With Bladder Cancerbladder.uroonco.uroweb.org
36. Transcontinental robot-assisted remote telesurgery: feasibility and potential applications.pubmed.ncbi.nlm.nih.gov
37. A surgical dataset from the da Vinci Research Kit for task automation and recognitionarxiv.org
38. A surgical dataset from the da Vinci Research Kit for task automation and recognitionarxiv.org
39. Breaking Barriers in Robotic Soft Tissue Surgery: Conditional Autonomous Intestinal Anastomosisarxiv.org
40. Robotic Medical Devicesfda.gov
41. Public Workshop: Evaluating Benefit-Risk in Robotic Medical Devices with Autonomous or Remote Teleoperation Capabilitiesfda.gov