The use of soft tissue and orthopedic robotic surgery systems has increased exponentially over the past decade, sometimes without clear-cut evidence of better outcomes for patients.
New research released at a recent meeting of orthopedic surgeons has rekindled the debate over which is the best approach to knee replacement: the conventional way, or with the help of a robot?
For now, the answer appears to be both.
The study, presented at the American Academy of Orthopaedic Surgeons meeting in February, found that patients who received a cementless total knee replacement with robotic assistance were as likely to need revision surgery within two years as those whose procedures were performed with a manual method. Both groups had similar rates of implant loosening and infection.
The research underscores the limitations of what a robot can do and helps explain why orthopedic surgeons aren’t moving all of their cases over to the platforms just yet.
“What this study shows is that having the knee replaced with robotic assistance does not substantially change the likelihood that people need an early revision. Some knee replacements fail early, and they fail early if you use a robot or if you don’t,” said study author Lucas Nikkel, an assistant professor of orthopedic surgery at Johns Hopkins Medicine.
Joint replacement is one of the most common surgeries in the U.S., as patients seek to improve their mobility and find relief from chronic pain. Total knee arthroplasty is also one of the most expensive surgical procedures, with average total costs ranging from $31,558 to $37,370, according to one study.
In a total knee arthroplasty, cartilage and bone that is damaged due to arthritis or injury is replaced with an implant. More than 700,000 total knee replacements happen annually in the U.S., while more than 450,000 total hip replacements are performed each year, according to the AAOS.
Increasingly, those surgeries are done with robotic assistance. Stryker, one of the largest makers of artificial knees, has said that 60% of its knee implants in the U.S. are now performed using its sector-leading Mako robot.
The company maintains that its robotic system allows surgeons to operate with greater precision, resulting in less damage to the soft tissue and a faster recovery. Stryker has published more than 400 peer-reviewed studies in support of the Mako platform. Rival orthopedic device makers Zimmer Biomet, which makes the Rosa system, and Johnson & Johnson, which has Velys, similarly argue that their robots can help improve patient outcomes.
“We were a little surprised that there wasn’t at least some indication of better outcomes.”
Lucas Nikkel
Assistant professor of orthopedic surgery at Johns Hopkins Medicine
Nikkel and colleagues looked at 9,220 cementless total knee arthroplasty procedures, also referred to as TKA, in the American Joint Replacement Registry performed from January 2017 to March 2020. The study found robotic assistance did not reduce the odds of patients needing another surgery in two years.
“Our findings add to previous studies suggesting that robotic-assisted TKA does not improve implant survival in cementless total knee arthroplasty. In light of this, arthroplasty surgeons must continue to consider how best to integrate robotics in TKA,” the researchers wrote.
The study built on an earlier registry analysis by Nikkel’s group that also showed no difference in revision rates at two years between robot-assisted and conventional knee replacement.
With as many as 20% of people who get a total knee arthroplasty reporting dissatisfaction with their new knees, according to data cited by Stryker, doctors and orthopedic device companies are focused on improving the procedure so that more patients have successful results.
Robot-assisted knee surgery has the potential to achieve more precise, even bone cuts and more accurately match implant dimensions, the researchers said. Cementless implants, meanwhile, hold the promise of longer-lasting biological fixation between implant and bone, which led the team to zero in on this subset of patients.
“Cementless technology is increasing, and certainly the use of robotics is increasing. So we were looking at two technologies that are increasing in popularity and use, and looking at when you combine those two, are things better?” Nikkel said. “We were a little surprised that there wasn’t at least some indication of better outcomes.”
Nikkel, who uses a robot in about 5% to 10% of his own cases, said the technology is very accurate at doing what the surgeon tells it to do. “I’m not anti-robotic knee replacement,” he added.
Rather, Nikkel thinks the robot of the future could help improve patient outcomes by leveraging the systems’ potential to gather large amounts of information to guide decisions. That capacity could help surgeons solve challenges involving alignment, for example, when choosing targets for how knee replacements should be done.
“I’m hopeful that because the robots are collecting such good data, they’ll help get us there,” Nikkel said.
Cory Calendine, an orthopedic surgeon with the Bone and Joint Institute of Tennessee, argued that robotics already provides a better tool for allowing surgeons to achieve their goals. Calendine, who has been using robotic total knees exclusively since about 2016, said Stryker’s Mako system is based on a preoperative CT scan.
“You create a custom 3D model of the patient and then build your plan on top of that,” Calendine said. “That’s what everyone is drawn to: better information, better execution.”
Calendine was not surprised by the results of the revision study, noting surgeons must make decisions based on bone quality, and a uniform criteria for that variable does not yet exist.
“The challenge that we have now is, what is the patient’s function?” Calendine said.
“The robot is definitely here to stay.”
James Huddleston
Professor of orthopedic surgery at Stanford Medicine
Two other registry studies that tracked knee implant survivorship have shown improvement with a navigation system compared to conventional instrumentation over longer time periods, said James Huddleston, professor of orthopedic surgery at Stanford Medicine and chair of the American Joint Replacement Registry steering committee.
“The survivorship delta continues to increase as you have longer follow-up. So with this current study at two years, it’s not surprising that there wasn’t a difference in survivorship and specifically for revision for aseptic loosening in robotic performed knees,” Huddleston said. “As time goes on and the follow-up duration increases, we would expect the curves to diverge, and the robotic knees would show improved revision rates.”
Huddleston is a longtime user of surgical navigation technology but does not use a robot, in part due to cost. “The value proposition for the robot has not really panned out yet,” said Huddleston, who has helped design a navigation system with advanced ligament balancing throughout the arc of motion.
“In terms of the value of that, we certainly expect that to be a much higher proposition than what the robot has shown so far,” Huddleston said. “So that’s what I’m sticking to for now.”
The surgeon experience and a growing preference for the robot is driving wider clinical use, Huddleston acknowledged. “There is no question that surgeons like the robot,” said Huddleston. “The robot is definitely here to stay.”
Zimmer Biomet spokesperson Heather Zoumas-Lubeski said robot-assisted joint replacement can affect a range of outcomes beyond reduction in revision rates, such as procedure reproducibility and the patient’s recovery, satisfaction and quality of life.
“When assessing the value to a new technology, it’s important to take a holistic view of its impact on the surgeon, the patient and health economics,” Zoumas-Lubeski said in an email. An accurate cost comparison to manual surgery depends on factors including duration of the procedure, length of hospital stay and post-operative recovery.
“Ultimately, the decision on whether to use a robotic surgical assistant versus conventional manual instrumentation is in the hands of the surgeon. Robotic surgical assistants, like our ROSA platform, are meant to empower surgeons by supporting their preferred surgical approach and enhancing their surgical technique,” Zoumas-Lubeski wrote.
In a commentary published in the journal Clinical Orthopaedics and Related Research, Robert Sershon, a joint replacement surgeon with the Anderson Orthopaedic Research Institute in Virginia, said more robust evidence is needed to determine the right approach for each patient, with machine learning hopefully facilitating more individualized solutions over the next several decades.
“If (or when) the time comes for my knee replacement in 30 years, I hope my surgeon has a more solid rationale about how they perform their knee replacements than our current standard of: ‘This is how I prefer to do my knees,’” wrote Sershon.