The robot recommended a cut. James Gholson, MD, knew something was wrong.
He was assisting a young orthopedic surgeon through her first robotic knee replacement. She had extensive training with the technology but relatively little experience performing the procedure manually. When the system recommended a bone cut that Dr. Gholson believed would create a problem requiring revision implants to correct, he stopped her.
“No, that’s not recoverable,” he recalled saying. “We’re going to have to turn the robot off.”
He completed the step manually, departing substantially from the system’s recommendation. Then he used the robot to assess the result.
It gave him a green check mark. “Perfect job! That was a perfect cut,” he recalled the system indicating.
For Dr. Gholson, an orthopedic surgeon at Mountain View, Calif.-based El Camino Health, the experience crystallized a concern that extends far beyond one operation. As robotics becomes more deeply embedded in joint replacement, surgeons are gaining access to increasingly sophisticated technology. But some may be losing the very skills needed to recognize when that technology is wrong.
The paradox is difficult to ignore: The more capable the machines become, the more important it may be for surgeons to know how to operate without them.
The skills at risk
Robotic-assisted surgery has transformed how many orthopedic surgeons approach joint replacement, offering detailed anatomical measurements, real-time feedback and guidance intended to improve surgical precision. Yet greater precision has not consistently translated into demonstrably better clinical outcomes, Dr. Gholson said.
He pointed to research, including the RACER-Knee trial, that has raised questions about whether the additional time and expense associated with robotic procedures produce meaningful benefits for patients.
That uncertainty has done little to diminish enthusiasm for the technology. Patient demand remains strong, and robotic platforms continue to become more prominent in orthopedic practice. Dr. Gholson, who specializes in hip and knee replacement, sees considerable potential in these systems. His concern is not that surgeons are using robots, but that some may become too dependent on them.
The implications are particularly significant for younger physicians whose training has centered on robotic techniques. Certain platforms perform much of the bone preparation while leaving delicate areas near ligaments and other structures for the surgeon to complete. Those remaining steps can require substantial technical skill, particularly when the surgeon must work around the system’s limitations.
“Without those manual skills, I find that these new surgeons aren’t very good at using a saw or other instruments to complete the robot’s work,” he told Becker’s.
The concern becomes more consequential when a system malfunctions or produces an unreliable recommendation. Dr. Gholson worries that surgeons who have performed relatively few conventional knee replacements may struggle to complete a procedure independently if robotic assistance becomes unavailable.
“I worry some of these surgeons might, frankly, have to call a more senior surgeon in to finish the operation,” he said. “That’s scary.”
It raises an uncomfortable question for orthopedic education: How much responsibility can technology assume before the skills it was designed to support begin to deteriorate?
A more demanding operating room
For all the attention paid to surgical accuracy, Dr. Gholson believes another consequence of robotics has received too little consideration: what it does to the surgeon’s experience. He describes conventional knee replacement as reading a book in a quiet library. An experienced surgeon follows a familiar sequence, concentrating on the operation with relatively few distractions. Robotic surgery, by comparison, feels like stepping into an IMAX theater.
“There’s another three or four screens flashing suggestions at you,” he said, comparing the experience to the early internet, when pop-up advertisements constantly competed for a user’s attention.
Each new measurement or recommendation requires interpretation. Surgeons must reconcile the system’s suggestions with what they see, feel and understand about the patient’s anatomy. Rather than simplifying the procedure, that additional information can introduce another layer of complexity.
“I would argue our current robots probably increase cognitive load and cognitive burden,” Dr. Gholson said.
He said robotic knee replacements take him longer to perform and do not reduce the mental demands of surgery. In some cases, they also introduce physical challenges.
Joint replacement is already a physically demanding specialty, with surgeons susceptible to occupational injuries involving their shoulders and hands. Dr. Gholson said working around robotic equipment can require awkward instrument positioning, particularly when correcting or completing portions of a procedure manually.
He believes the industry has largely overlooked an opportunity to develop technology that makes surgery less physically and mentally taxing. A robotic platform that reduces those burdens could help surgeons sustain their careers while maintaining their focus on patient outcomes.
Instead, he believes some current systems have moved in the opposite direction. “It’s not there right now,” he said. “I hope it is at some point.”
The problem behind the green check mark
One of the most consequential limitations of robotic surgery, Dr. Gholson argues, begins with the information the system receives. Robotic platforms rely on anatomical data collected through imaging, intraoperative registration or both. During registration, surgeons identify landmarks that help the technology construct a model of the patient’s anatomy and guide implant positioning.
Depending on the system, that process can involve manually identifying dozens of points around the exposed knee. The resulting digital model may appear precise, but its accuracy depends on the quality of those initial measurements.
Two experienced surgeons can identify slightly different landmarks on the same patient. Anatomical variations introduce further uncertainty.
“The quality of how you define the anatomy 100% defines the quality of what that robot does,” Dr. Gholson said.
That dependence on human input complicates the assumption that robotic recommendations are inherently more reliable than a surgeon’s judgment. Dr. Gholson described cases in which a system suggested removing more bone than he believed appropriate. After making a more conservative adjustment and reassessing the knee, he said the technology frequently indicated that the result was satisfactory.
The experience has made him especially cautious about accepting recommendations that conflict with his assessment of the joint. It also exposes a potential weakness in one of robotics’ central promises.
These systems are intended, in part, to help surgeons achieve more consistent results regardless of experience. But Dr. Gholson worries that recognizing an inaccurate recommendation may itself require considerable experience.
“That’s who everyone hoped the robots would help the most,” he said, referring to lower-volume surgeons.
For him, the surgeon’s role is therefore not becoming less important. It is changing from executing a familiar sequence of steps to continuously evaluating the recommendations of another source of information.
He compares even the best robotic systems to talented junior residents: capable of performing routine tasks and providing useful assistance, but unable to appreciate every variable an experienced surgeon encounters during an operation.
The surgeon must remain prepared to intervene.
What the next generation of robotics should solve
Despite his concerns, Dr. Gholson sees considerable potential for better technology, particularly systems capable of scanning a patient’s anatomy in real time rather than relying on manually identified landmarks.
Such advances could improve accuracy, account for soft-tissue balance and allow surgeons to adjust implant positioning as an operation progresses. He also wants robotics to reduce the physical and mental demands of surgery while delivering measurable improvements in patient outcomes and efficiency.
Whether those capabilities become reality remains uncertain. For now, Dr. Gholson believes technological sophistication should never come at the expense of surgical judgment.
“They have to embrace their role as the leader of the operation that now has a lot of extra data,” he said. As orthopedic surgery moves toward greater automation, that may be the most important skill to preserve.
At the Becker’s 32nd Annual Meeting: The Business and Operations of ASCs, taking place October 29-31 in Chicago, ASC leaders, surgeons and healthcare executives will explore strategies to drive growth, enhance operational performance, navigate reimbursement challenges and prepare for the future of ambulatory surgery. Apply for complimentary registration now.
