High financial costs and limited evidence restrict the adoption of robotic surgery, but new uses for robots and attempts to assess efficacy could widen access.

Seated in a high-back, cushioned chair, Varun Bathani resembles a video gamer more than a surgeon. He places his forearms on rests and uses hand controls to manipulate robotic arms fitted with surgical grabbers and clippers. Watching a high-resolution screen, Bathani, who is at St. Paul’s Hospital in Saskatoon, Canada, guides the tools to perform minimally invasive surgery for prostate cancer. “The visualization is incredible,” Bathani says. “That 3D view that you get makes surgery that much better, that much more precise.” Operating from outside the theatre also allows him to sit in a comfortable position, instead of being hunched over the patient.

Robotic surgery looks futuristic, but surgeons first used robotic arms in the mid-1980s¹. The momentum really increased in the late 1990s, when California-based Intuitive Surgical’s da Vinci surgical system came on the scene². Today, the da Vinci system commands 80% of the market. Although Intuitive Surgical dominates the field, other companies — including CMR Surgical, Johnson & Johnson and Medtronic — offer platforms for robotic surgery. “This has helped to foster innovation in the field, for example, with the creation of modular, open-console robots,” says Ara Darzi, an expert in minimal-access surgery at Imperial College London.

Robotic surgery can be carried out at a console while others watch on a monitor.

In 2000, Darzi participated in the first robotic surgery procedure in the UK, a gallbladder removal performed at St Mary’s Hospital. “Now there are more than 100 of these systems actively being used across the country,” Darzi says. “Unlike laparoscopy, which is one of the mainstays of minimally invasive surgery, the robotic system offers better visualization with stereoscopic views and improved ergonomics.”

After several decades of applications and improvements, though, a lingering question remains: is robotic surgery better surgery?

Robots cost money and time

Some experts believe that certain procedures benefit from using a robot. As Darzi says, “Robots are regularly being used in urology, gynecology, general surgery, and ear, nose and throat, where surgeons work in a narrow field, and where visualization is often sub-optimal.”

Still, it’s not clear to everyone in healthcare that robotic platforms are an improvement. “We still don’t actually have a lot of proof that they’re better than old-fashioned surgeons — at least old-fashioned surgeons doing minimally invasive surgery,” says Peter McCulloch, professor of surgery at the University of Oxford in the UK. Plus, adding a robot to surgery takes more than just installing it. “You’ve got to change your entire way of doing surgery,” McCulloch says. As one example, he notes that putting the surgeon behind a robot console “disrupts the traditional communication and teamwork within the surgical team.” This requires surgical teams to adjust to a different dynamic.

Money and time are also key considerations. Robotic surgery is expensive, with an average platform costing about €1.4 million, according to one market survey. “Then you need to factor in the cost of disposables, which can be a few hundred thousand pounds a year, and maintenance,” Darzi says. “This initial outlay can often preclude centers getting a robot, and only makes it viable if costs can be recouped across multiple fiscal cycles.”

In terms of time, several of the experts interviewed by Nature Medicine note that it takes longer to set up robotic surgery than a manual approach. “Some studies have demonstrated that median operating time is significantly longer for robotic compared to either laparoscopic or open procedures,” Darzi says. “This varies [depending] on the type of procedure being performed, the surgeon and the familiarity of the theatre team.”

Unclear benefits

Time in surgery, however, is not the only temporal metric that matters. As reported by Katherine Fay, a fellow in minimally invasive surgery at Emory University School of Medicine, and Ankit Patel, chief of surgery at Emory Saint Joseph’s Hospital, both in Atlanta: “Studies have shown that, across multiple specialties — including general surgery and surgical oncology — patients who undergo robotic-assisted surgery have… shorter postoperative length of stay than patients who undergo laparoscopic surgery”³.

Even with improved ergonomics and better visualization, today’s robots come with some limitations. As Fay and Patel noted: “One major shortcoming of the robotic platform compared to laparoscopy is the lack of tactile feedback.”

Robotic surgery comes with pros and cons, with some experts unclear about its impact. “How do we scientifically test robotic surgery?” McCulloch asks. “It’s much harder to test surgery than it is to test medicines.” In surgery, even the same procedure varies from one surgeon to the next, which is the case with or without a robot. By contrast, he says, “the drug is the drug.”

Making such a scientific assessment of robotic surgery also comes under the pressure of progress. “In the early days, there were a lot of efforts to do high-quality evaluations of robots,” McCulloch explains. “Frankly, they were kind of brushed aside by the stampede to start using robots, and surprisingly, after nearly 20 years, there’s a very small amount of evidence that robots are better than a human surgeon.” McCulloch does grant that robotic surgery could be a bit better in some operations, such as prostate surgeries.

McCulloch argues that more scientific tests of robotic surgery are needed. “If you’re going to spend all that money and change the direction, let’s make sure it’s the right thing,” he says. “Let’s have the scientists prove it.”

Competing with the human hand

Neurosurgery could be a final frontier for robots. Zimmer Biomet has developed the ROSA robotic arm for stereotactic neurosurgery. This robot can be used in various brain-related procedures, including implanting electrodes to treat epilepsy. Many more robotic options are available for neurosurgeons.

“In the past five years, we’ve had an explosion of robots for spine surgery,” says Peter Konrad, chair of neurosurgery at West Virginia University Health Sciences. As a neurosurgeon who also earned a PhD in biomedical engineering, he is concerned about the value of robots for neurosurgery but sees the benefits of using them in spine surgery. “That is a really good use of a robotic arm for the placement of screws and hardware into the spine,” he says. “It’s much more error-prone to do that by hand.”

For neurosurgery, Konrad only sees the value of robots for repetitive and standardized steps — so far. “The complexity of the [neurosurgery] cases we do at the large medical centers is so varied, and this is where robotics fails to help,” he says.

Beyond placing hardware in a patient’s spine, Konrad says, “The real advantages will be if we can get robots to do something we cannot physically do.” As an example, he mentions reconnecting axons to repair nerves. At just a few micrometers in diameter — sometimes even fractions of a micrometer — stitching axons together one by one is impossible for now. “That’s where the robotic development people should go,” he argues, “doing stuff that we can’t do.”

Robots struggle to compete with the dexterity of the human hand, says Konrad: “This is by far the most sensitive instrument we have, because your vision and your sense of touch make an incredible tool that is hard to improve on.”

Still, experts will keep trying to find niches for robotic surgery. Pierre Dupont, an expert in mechanical and biomedical engineering at Boston Children’s Hospital, and his colleagues are developing a two-armed endoscopic robot to help neurosurgeons work in tight spaces⁴. “The point is you’re doing two things: you’re bringing your imaging and your arms right down to a surgical site, like a tumor in a brain’s ventricular space,” Dupont explains. With this device, a neurosurgeon could work with two tools, controlled by joysticks.

Dupont and his colleagues developed a set of skill tests modeled after removing a brain tumor. When comparing surgeons completing these tests with and without a robot, Dupont says, “Using the robot was pretty consistently faster — even in the hands of the same clinician.” Plus, with the robot’s two “hands,” Dupont explained, “you can sort of walk down the edge of the tumor to separate it from the surrounding brain.”

Global inequities

Robotic surgery raises ethical issues because it is not available to everyone. More than 60% of the 7,544 da Vinci platforms around the world in 2022 were in the USA, and only about 2% were in Latin America, according to data from Intuitive presented by Nam Jin Kim at the 2023 Latin American Forum on Quality and Safety in Health. Kim, who is medical director of the surgical network of the Hospital Israelita Albert Einstein in São Paulo, Brazil, hopes to change that.

Since 2008, more than 14,000 robotic surgeries have been performed at Einstein, with a 72% increase between 2021 and 2022. “Robotic surgery is good for the patient — smaller incisions, reduced pain and discomfort, shorter hospital stays, faster recovery, less infection and better outcomes,” Kim says. He also points out the ergonomic benefits to surgeons.

Of the 150 da Vinci platforms in Latin America in 2022, 10 were at the Hospital Israelita Albert Einstein. “Many hospitals in Latin America are struggling to buy their first robotic platform,” Kim says. “It’s a shame [when] surgeons and patients still don’t have access to this technology.”

Once robotic platforms are acquired, surgeons must be trained to deliver consistent outcomes. In 2019, Einstein became an official Intuitive Training Centre, and part of the agreement included a da Vinci platform. At the time of writing, this center, which now has four da Vinci platforms, has trained more than 1,500 surgeons from 16 countries.

Outside the operating room

Robots might perform other clinical duties beyond surgery. Núria Vallès-Peris and Miquel Domènech of the Universitat Autònoma de Barcelona, in Spain, reviewed the use of robots in patient care⁵. These robots could be used in many ways, including feeding patients with severe mobility issues.

Progressing from idea to application with care robots is complicated. “Technical limitations remain very important, and, outside the laboratory or highly controlled robotic experiences, these applications do not always work as expected,” Vallès-Peris says. “Right now, talking about the benefits of such robots is more a matter of speculation than something we can verify empirically.”

Robots might also be used in rehabilitation⁶. José Lucas Oliveira de Sena, an occupational therapist at the Rehabilitation Center of the Hospital Israelita Albert Einstein, uses the ArmeoSpring robotic arm to help patients improve upper-limb motor skills damaged from a traumatic brain injury, such as that from a car accident, Parkinson’s disease, stroke or other neurological condition. As Sena says, “The robotic therapy complements conventional treatment protocols, enabling a virtual environment in which patients can hone the skills essential for occupational performance.”

The ArmeoSpring robotic arm serves as an exoskeleton that, Sena says, “is a powerful resource in rehabilitating the upper limbs of patients grappling with neurological disorders.” Combining this robot with games based on virtual reality helps patients “practice the repetitive movements that mirror daily routines,” Sena explains.

Much like care robots, however, Sena emphasizes: “While technological advancements and robotic devices herald new horizons for therapeutic interventions, we must tread with clarity.” In particular, he notes, “It’s essential to establish clear clinical guidelines for deploying these novel tools to ensure we retain and potentially enhance the effectiveness of treatments.”

AI-powered autonomous robots

Artificial intelligence (AI) promises to expand the capabilities of tomorrow’s surgical robots. As McCulloch says, “AI plus a robot does give you a kind of science fiction future possibility, and I don’t think it’s actually that far off.” That future could include autonomous robots that perform some procedures on their own. “I think the potential for autonomous robots in the future is very high, whether we want them or we don’t want them,” McCulloch says. “That’s a society issue.”

Some robotic devices being developed for surgery already use AI. For instance, Ron Alterovitz, professor of computer science at the University of North Carolina at Chapel Hill, says, “By leveraging the power of robotics and AI, I am developing new robots capable of autonomously moving through the patient’s anatomy to complete procedures with unprecedented accuracy and safety.” As one example, Alterovitz and his colleagues recently described a robot that can autonomously steer a needle through a patient⁷. “It’s akin to a self-driving car, but it navigates through lung tissue, avoiding obstacles like significant blood vessels as it travels to its destination,” Alterovitz explains.

Although Alterovitz points out that “medical procedures are an especially challenging domain for autonomous robots,” he believes that “autonomous medical robots have the potential to improve the accuracy, precision and safety of medical procedures and enable entirely new procedures not possible by instruments limited by human dexterity.”