New Technology: Safety, Efficacy, and Learning Curves

The subspecialty of arthroplasty experiences the mixed blessing that is new technology: new imaging approaches, new bearing surfaces, new navigation systems, new implant designs, and the list goes on. While these technologies may benefit patients, they almost always come with a substantial rise in cost compared existing approaches, and often hit the market with little or no documented clinical benefits. We must consider rising healthcare costs with important practical and ethical considerations.

I have invited two well-known experts in the field to discuss their thoughts on integrating new technology into practice. Paul J. Duwelius MD is a well-respected private-practice surgeon in Portland, OR with extensive experience in adult reconstruction and orthopaedic traumatology. During his years in practice, he developed contoured plates for periarticular fractures, as well as a modular total hip system. Javad Parvizi MD is a recognized thought leader on many related subjects, including joint reconstruction outcomes research, prosthetic joint infections, and hip conditions affecting the young patients. I am most grateful to them for their participation here.

Laura Matsen Ko MD: Let us begin by considering how we decide when the incremental value of a new technology (defined as increase in benefit over the increase in cost) merits widespread adoption. Can you offer some examples where you have made this calculation?

Paul J. Duwelius MD: As surgeons we want to develop better methods of helping patients, and this often leads us to consider new implant designs. As you mentioned, I have had the opportunity to participate in the development of anatomically contoured plates, as well as a modular hip stem, with a major goal of improved recreation of the femoral head center. But the incremental value, as you define it, is a challenge to measure.

Precontoured plates seem to make surgery easier in cases of difficult periarticular fractures, and they allow the choice of nonlocked, locking or lag screws, which can improve fracture fixation. Although it remains to be proven whether superior clinical outcomes justify the increased cost, I suspect they will.

By contrast, with respect to the modular stem, there are problems with corrosion, fretting, and mechanical issues with the trunion [1, 11] Additionally, the improvement in head center placement was not tremendous, and might not have even been clinically important [4]. In this case, the increased benefit did not justify the incremental expense, especially when considering the cost of the revisions. As this instance shows, demonstrating increased value is challenging because some of the costs (those associated with complications and revisions) are not evident until months or years later [3, 12–14].

We need to recognize the balance that exists between the desire for an early general release of a new product, and the need to make sure that the product adds substantial benefit to the patient in the long run — an assessment than requires long-term followup in well-conducted clinical trials before general release.

Javad Parvizi MD: Introduction of a new technology must be developed around the solution to a specific problem. I agree with Dr. Duwelius that striking the right balance in terms of timing the release of a particular technology into the community is difficult.

It is my view that prior to introduction, any innovation or new technology needs to demonstrate its intended benefits in both the preclinical and clinical setting. There are numerous technologies that have been introduced with the goal of solving a particular issue, but lacked concrete data to demonstrate that this was done — or worse, in situations where the new technology introduced a new problem with disastrous consequences. Examples of this include the introduction of metal-on-metal (MOM) bearing surfaces, modular femoral stems for primary hip arthroplasty, and the two-incision technique for THA.

Dr. Matsen Ko: Your last example, Dr. Parvizi, leads nicely into my next question. We know that the effectiveness and the safety of a technology are related to the surgeon’s experience in using it. What is the surgeon’s obligation to disclose to the patient where the surgeon is on the learning curve for the new technology? It seems relevant to a patient’s decision whether the surgeon is the first patient to have a particular procedure or the hundredth. Is it?

Dr. Duwelius: We are obligated to disclose all relevant risks to our patients, and certainly there are many — even without considering the increased risks inherent to new technology. On top of these risks is the added risk of the learning curve — the incremental risk of applying a new technique or a new implant before one is entirely facile with the new tool. The learning curve applies to every surgeon who attempts a new method. Even with extensive educational courses, dry labs, and cadaveric sessions, surgeons experience unique challenges when they try a new technique in the operating room.

We do not know how long it takes to get “on top” of a new technique — is it 10 cases or 50? This probably depends on other factors, such as the complexity of the new approach and the surgeon’s experience with similar techniques. Do we get better with each case without limit (is our 200th safer and better than our 100th)? In any event, we all have seen complications when surgeons try a new method. It is our duty to disclose our best estimate of this risk in our own hands.

On a personal note, when I adopted the two-incision technique for THA, I experienced a steep learning curve. I had a few periprosthetic fractures and I was devastated. I also noted that the lateral femoral cutaneous nerve has a variable course and so is at risk; patients do not like having loss of sensation to their thighs. I then performed a study and compared the two-incision to the posterior approach to a total hip and found that there was no difference in the outcomes such as Harris Hip Scores, MOS SF-36 Physical Health and Physical Function scores, as well as similar improvement in WOMAC pain, stiffness, and physical function [5]. The cost of the learning curve was not worth it.

I now feel comfortable with the methods I am using. As I move forward in my practice, prior to considering a new technique or a new implant, I would need to be convinced that it has the potential to be a lot better than my status quo. I would then be sure to disclose to my patients that, while I have done many thousand hip and knee replacements, I am not sure of the benefit or the risks of this new method in my hands. Our patients need to know and hear the true risks from us as surgeons. We have to be the champions who dispel nebulous Internet claims or blatant advertisements or marketing statements that have no merit at the very least, and sometimes even are false or misleading.

Dr. Parvizi: I agree with Dr. Duwelius. Any novel surgical technique introduced will provide new challenges to the surgeon, even the most skilled. It is the duty of the surgeons embracing a new technique not only to disclose the issue of learning curve to their patients, but also to take concrete steps to reduce the risk to the patients. This includes attending and performing cadaveric surgery, observing other surgeons who have overcome the learning curve, and learning from each case performed.

The issue of the learning curve not only applies to any new technique, but also to any surgeon in the beginning of his or her career. Surgeons who have performed a particular procedure numerous times and have a high volume of practice are likely to have a lower complication rate and better results than those with less experience. The latter not only applies to surgery but to any other task that requires skill. As Malcolm Gladwell states, the transition from less experienced to refined may be a long time in coming — he suggests for complex tasks, it may take 10,000 hours of practice or more [6].

Dr. Matsen Ko: We can all think of examples of technologies that were released without adequate study, and in some of those instances, patients paid a big price for the problems we discovered later. How has our specialty learned from these events, and how can we improve the process so that the risk of such “surprises” is minimized?

Dr. Parvizi: Medicine is a truly fluid scientific field. The flow is created by the efforts of clinicians and scientists whose mission is to enhance patient care. This mission has been accomplished in numerous areas; for example life expectancy in the United States has doubled in a little more than a century.

Advances in medicine have come from two main sources: Observation and innovation. Observations by Louis Pasteur, Ignaz Philipp Semmelweis, Alexander Fleming, and Joseph Lister lead to extraordinary discoveries and transformed medicine. The observations of Marshal Urist and the discovery of bone morphogenetic proteins are an orthopaedic example. Innovation is the other source of advancement in the field. The innovations in minimally invasive surgery, for example, have transformed the patient experience and outcome of patients undergoing cholecystectomy, abdominal aortic aneurysm, and heart valve surgery in the modern era of medicine. Orthopaedic surgery also has benefitted from these innovations. Innovations in our specialty include the introduction of arthroscopic surgery, and advances in pain management. But advances can also expose patients to risk; mitigating that risk is critical.

The Food and Drug Administration in the United States, and similar organizations in other countries, seek to do this. In my opinion, and based on the available data, the FDA has been for the most part successful in preventing harm to patients. The problem arises when the data presented to the FDA or to the orthopaedic community on a particular innovation appear to be compelling based on laboratory or preclinical studies, or on the experience of innovator-surgeons, but the innovation does not deliver the same results in the general orthopaedic community. Although one may be tempted to place the blame on the regulatory approval process, the fact is that any innovation is likely to bring with it a potential exposure to risks. This may never be totally avoidable.

Strategies to minimize these risks may involve introduction of a specific technology or innovation through a more systematic approach, in which the innovation is made available to a select group of surgeons who will monitor the outcome in their patients very closely prior to introduction into the general community. Postmarket surveillance of a particular technology that is being increasingly mandated by the FDA is another strategy that may limit the risks. Finally, there are many authorities among us who believe that registries will be able to limit this exposure by detecting early failures of a particular device.

Dr. Duwelius: Dr. Parvizi cites great examples of medical advances that have improved health. Orthopaedic advancements must continue and I believe they will, as our field continues to attract the best minds in the field of medicine. A recent example of an important advance was Dr. Parvizi’s study [8] that proved Foley catheters for routine total joints were not indicated. This randomized prospective study convincingly demonstrated the superiority of one approach over another. This level of scientific rigor, unfortunately, is not the norm.

Total joint surgery has come a long way since Charnley’s pioneering work. Yes, new technologies should be subjected to sound outcomes-research at a few centers before being brought to the larger orthopaedic community. This would minimize the risk to the general population. Even brief reflection on the experience with MOM hips would emphasize this point. But importantly, some problems may not become evident until many years down the line, and so typical postmarketing surveillance probably is not enough for our specialty. Thus, total joint registries may be the best way to conduct surveillance on new technologies in the hands of multiple surgeons. I strongly support the formation of the American Total Joint Registry for this reason.

A disturbing trend in this country is the direct to consumer marketing claims being made by industries without scientific merit. Typically these are attempts to attract patients to demand new technologies before they have stood the test of time. Again, new innovations should have demonstrated increased value (greater benefit to the patient divided by the cost) before they are released for widespread use. Otherwise, healthcare becomes more expensive without improving the quality of care of our patients.

Dr. Matsen Ko: In that bundled payments may soon be upon us, how should new technologies be introduced and managed in a bundled payment reimbursement system, given that these technologies almost always come with increased costs of development, introduction, and unanticipated complications?

Dr. Parvizi: Because innovation does come at an added cost, and the time horizons of the benefits may be long while the costs of bundled payments will be evaluated in the short-term, this issue promises to be a real challenge. One example that comes to mind is the introduction of 90-day episode of care bundling. It will be difficult for an accountable care organization to justify paying a premium for a bearing surface that may improve longevity of a hip arthroplasty, with the benefit of the technology not being recognized for decades, when all financial modeling will concentrate on short-term savings for that episode of care. We will be under pressure to demonstrate that the premiums paid for the technology will eventually return dividends. In my opinion, moving forward, we will witness a real change in the manner in which technologies are introduced. Industry recognizes the challenging times ahead, and will need to restrategize their approach to the introduction of technologies. The medical community will share the challenges that lie ahead. It is possible that our patients may be deprived of benefits of innovations that come at added cost, but fail to demonstrate short-term gains.

Dr. Duwelius: Bundled payments represent an attempt to change the way we deliver healthcare from fee for service model to an episode of care practice. I think we as surgeons can see this as either another outside influence burdening our practice or as a tremendous opportunity to be part of the solution to our impending healthcare crisis. Between 2005 and 2030, primary knee arthroplasties in the United States are projected to grow by 3.48 million procedures annually [7]. In 2005, medical expenditures for the treatment of arthritis hit USD 353 billion, and they are expected to rise because of increases in the number of people with osteoarthritis, due in part to the aging of the U.S. population [2, 9, 10].

When we think about having to care for this volume of patients, and incorporating innovation (and its costs) into a bundled payment system, life gets complicated. My feeling is that bundled payments will force the innovation advocates to demonstrate increased value before the bundled payment will be increased to cover its cost. An example might be the new highly cross-linked polyethylene with enhanced vitamin E. Is this product better? How does it justify a higher expense? How will we learn of the long-term benefit to the patient?

The principal accountable care provider for the hip and knee replacement surgical episode is the orthopaedic surgeon who controls the procedure. In our hospital, we are looking at the value of targeting same-day or next-day discharge for healthy patients having hip and knee replacement. This requires consistent patient education, consistent messaging in physical/occupational therapy, advance discharge planning, day-of-surgery physical therapy, and rapid rehabilitation protocols. We hope this will reduce the cost of the care episode without reducing the benefit to the patient. Through these optimizations, we may improve value without depending on the creation of any new technology; rather it is the result of doing a better job with the current technology.