Editor’s Spotlight/Take 5: CORR Synthesis: What Is the Role of Robotic-assisted Technology in Knee Arthroplasty?

Randall E. Marcus MDFor this Editor’s Spotlight commentary, I’m going to try something different. There will be no references in the text, other than to the article I’m commenting on [3]. It’s not that we lack good studies on the topic in question, quite the contrary—it’s that those studies don’t seem to be influencing the conversation to any great degree, so let’s dispense with them altogether. Let’s talk about our feelings. The topic is robotic-assisted joint replacement, but of course the larger issue—which will make what follows relevant to all surgeons, not just arthroplasty subspecialists—is how, when, and why we integrate new technology and tools into our practices. Surgeons who spend a lot of time with noses in clinical research journals point to the near-complete lack of high-quality evidence suggesting that robotic-assisted arthroplasty delivers benefits that patients can perceive. The wide adoption in practice of these tools, when held against the available evidence in favor of them, causes those surgeons to feel a dissonance. But it’s a two-way street. When I speak with surgeons who perform robotic-assisted arthroplasty, they consistently say that it has improved their surgical skills. They say that the experience of using a robot teaches the surgeon subtleties about the anatomy and joint kinematics that would be hard otherwise to perceive but are nonetheless important. Speaking of feelings, those surgeons also say that the joints “feel” better—as in more stable hips and more balanced knees—and their patients’ recoveries are quicker and more complete. Those surgeons feel a dissonance because these things seem obvious, and in some instances even dramatic, but these benefits haven’t turned up in the randomized clinical trials (RCTs) performed thus far on the topic. There is no doubt in my mind that these surgeons’ feelings are authentic and deeply held. Importantly, they may also be correct. One way or the other, two groups of surgeons seem to be talking past each other. Let’s see if we can find a way to talk with one another. In this month’s Editor’s Spotlight/Take 5 article, “CORR Synthesis: What Is the Role of Robotic-assisted Technology in Knee Arthroplasty?” [3], the authors from Case Western Reserve University in Cleveland, OH, USA, performed a thorough vetting of the available RCTs on the topic and found that knee alignment parameters often improved with use of robotic surgical systems, but those improvements seldom were associated with clinically important differences in endpoints patients can perceive, like patient-reported outcomes or complications. Only one study followed patients long enough to evaluate survivorship differences, but at 10 years, it found no benefit to robotic TKA in this endpoint. The cliché of “the absence of evidence isn’t evidence of absence” may be relevant here, but perhaps equally applicable is the old saw, “the plural of anecdote isn’t data.” Either way, this topic prompts tough questions in both directions, even as it may inflame the passions. It’s the kind of thing, though, that sensible people ought to be able to work out in conversation, so let’s have one. And, again, this conversation should be of interest to people who don’t do joint replacement as well as those who do, since the themes relate broadly to the question of when to adopt new approaches, which is a question that is of interest to all of us. As an aside, the senior author on the article I’m covering here, Randall E. Marcus MD, has had a long career in this subspecialty and a long tenure as the chair of a large department. Importantly, he’s also been a thoughtful observer of trends in our profession for 45 years. He’s great at seeing things from both sides, so he is the right guy to explore big questions with. Before we go to that interview, a quick plug: If you haven’t seen the CORR Synthesis article type before, have a look at Dr. Marcus’s article [3] in this issue (it follows directly after the interview in the Tabe of Contents). If you have an idea for a topic you’d like to cover in a CORR Synthesis piece, please visit our “Instructions for Authors” page on our website to learn how to write for this section of Clinical Orthopaedics and Related Research®, or send me an email at [email protected]. Now, join me in the Take 5 interview that follows, and if I missed any big questions that you’d like to ask, feel free to write us a letter to the editor. Take 5 Interview with Randall E. Marcus MD, senior author of “CORR Synthesis: What Is the Role of Robotic-assisted Technology in Knee Arthroplasty?” Seth S. Leopold MD: Congratulations on this stimulating article [3]. I’m expecting (and hoping for) a lot of letters to the editor about it. I have a sense for how you feel about the role of robotic tools in this setting from reading your article, but put yourself in the shoes of a surgeon who feels that the technology really is making a difference for his or her patients. We’ve all had that feeling about something that we do, and it’s often hard to prove. To what degree might real, important differences not show up in the kinds of RCTs you covered in your paper? Randall E. Marcus MD: Thank you, Dr. Leopold, for your kind words and the opportunity to publish our work in Clinical Orthopaedics and Related Research®. A couple of years ago, our medical center in Cleveland purchased three robotic arms for TKA surgery and several of our surgeons began using them. There is little doubt that these surgeons and others who use this new technology “feel” that robotic technology is making a difference for their patients undergoing TKA. It’s been said that for many orthopaedic surgeons, their feelings regarding improvements in technology are based on the last case they performed. I would suggest that the best way for us to make these decisions is to review high-quality data on the results of any new treatments. John Adams, the second president of the United States, said it best in defense of the British soldiers during the Boston Massacre trial in 1770: “Facts are stubborn things; and whatever may be our wishes, our inclinations, or the dictates of our passion, they cannot alter the state of facts and evidence” [1]. Adams’s brilliant words apply equally to the adaptation of new treatments for our patients, especially in the current era of skyrocketing healthcare costs. When my coauthor, Alexander Richards, and I began this study [3], we were very curious as to what the data would reveal regarding clinical, meaningful outcomes when comparing the results of robotic-assisted TKA and manual TKA. We know the use of robotic technology for knee replacement surgery is a very controversial and emotional subject for some knee surgeons. But the question we all want to answer is: Will the use of this technology for knee replacement change the outcomes for our patients, which include the 10% to 20% of patients with persistent pain, functional limitations, and some degree of dissatisfaction following TKA? We limited our systematic search to RCTs evaluating robotic-assisted TKA systems that utilize preoperative, CT-based imaging for surgical planning. It’s possible, but unlikely, that larger trials might reveal important differences that we are not seeing in our study. It’s conceivable that larger multicenter RCTs could theoretically reveal meaningful improvements in robotic-assisted TKA. Potentially, there may be specific subgroups of surgeons and/or patients who might benefit from robotic-assisted TKA technology. But as of now, based on the best evidence currently available, we were not able to find any meaningful differences between the two TKA surgical techniques. Dr. Leopold:Let’s say that the benefits to any particular patient from having robotic-assisted surgery are small, but that using robotic assistance and navigation indeed helps the surgeon become a more aware, more capable, and even more intuitive surgeon…which might help a lot of patients. Users of this kind of technology have told me again and again that these tools “teach” you about the joints you’re working on. Yes, the systems have costs, but we bear all kinds of other costs in efforts to make surgeons better; training is expensive. Why isn’t improving our skills a reasonable enough rationale to adopt these tools, or is it? Dr. Marcus: There is no doubt that surgical education is important for good patient outcomes. Robot-assisted TKA is an expensive undertaking, with a million-dollar price tag up front for purchasing the robotic arm plus six-figure annual service contracts. That would make the purchase of this technology a very expensive educational tool. A department chair would want to have data that would confirm the educational benefits for surgeons and their relationship to improved patient outcomes prior to purchasing a robotic-assist TKA for educational purposes. Dr. Leopold:Many widely used patient-reported outcome (PRO) scores suffer from ceiling effects, such that if a new surgical tool took nice results and made them really fantastic, we might not detect that improvement with the relatively crude PRO scores that we use. How should the desire to gain incremental but potentially meaningful benefits for active patients influence the way we think about adopting new implants, tools, and devices? Dr. Marcus: You correctly point out the problems with PROs. Effect size is an important measurement for interpreting subjectively reported outcomes, and it can be summarized in terms of minimum clinically important differences (MCIDs). But only four studies that fit the inclusion criteria for our study [3] reported MCIDs. One found no clinically important differences between robotic-assisted TKA and manual TKA. Another study reported improved total WOMAC scores in male and elderly patients at 6 months that exceeded the MCID without further follow-up reported. One study found no differences in OKS between the two patient groups at one year; however, a subgroup analysis from this study found improved OKS in patients who were men who underwent robotic assisted-TKA at 6 months, and that difference exceeded the MCID. A fourth study that reached the MCID for the vitality component of the SF-36 found robotic-assisted TKA improved the patient’s vitality and emotional health. There were no other differences in the two patient arms of the study. These data parallel what we found in PROMs across all studies with follow-up of at least 2 years: no meaningful evidence that shows robotic-assisted TKA improves PROMs. I would suggest that as surgeons who wish to improve TKA in ways that matter to patients, our decisions must be based on data and not industry marketing. In the case of robotic-assisted TKA, manufacturers have invested billions into developing the technology and hospitals have spent many millions buying it. From that standpoint, both companies and hospitals have investments to recoup. That leaves physicians—us—in the position of having to discern fact from hyperbole so that we can guide our patients in evidence-based ways. While surgeons are always looking for tools that can differentiate their practices—and robotic-type systems do appeal to patients—surgeons must be aware of aggressive marketing campaigns by manufacturers and weigh these messages against data-driven evidence when adopting new technology. Dr. Leopold:There is a large gulf between surgeons who use and those who don’t use these kinds of tools, and we seem to be talking past each other. It’s almost like each group “has their own facts.” What would help us communicate better as a profession, to come to a mutual understanding? Dr. Marcus: Your point regarding communication within our profession is very important. We are all seeking the same goal, which is to improve care for our patients. In the case of joint replacement, I believe that cooperating in registry-type studies would go a long way to providing good data and improvements in patient care. The adoption of new and improved technologies and treatments is important to our profession as long as they are driven by data, not marketing. Without strong evidence that expensive new innovations improve patient outcomes and safety, those innovations only add additional, unnecessary expense to the spiraling cost of healthcare for our patients. Dr. Leopold:From where I sit, the issue of helping patients with emotional distress prior to offering them elective orthopaedic surgery may be the single largest opportunity before us to improve our results. We know this factor is associated with persistent pain and clinical dissatisfaction in almost every setting in which the question has been explored, but when a patient manifests emotional distress, it seems rarely to cause the surgeon to tap the brakes. You’ve been at this longer than I have, though. What do you think is the single most important thing that surgeons can do differently to get better results in his or her practice? I’m guessing you’re not going to say, “Adopt a robot…” Dr. Marcus: There is tremendous pressure on surgeons to increase productivity. Unfortunately, this makes it difficult for surgeons to get to know their patients and develop a clear understanding of their social and psychological issues, which “may be the more important determinants of the results of surgery than surgical technique alone” [3]. Sir William Osler said it best: “The good physician treats the disease; the great physician treats the patient who has the disease” [2]. This is as true today as it was when Dr. Osler said it more than 100 years ago. Regardless of the daily pressures we all feel in the practice of medicine, in order to obtain the best results from our treatments, we must take the time to listen carefully to our patients in order to truly understand their problems and help them heal.