While orthopedists tend to be a convivial bunch, like any human, talk to them of their failures and they may turn sour, show you the door, change the subject, etc.
Such is the uphill battle faced by some in the field, including Dr. Frederick Matsen, Professor of Orthopaedics and Sports Medicine at the University of Washington in Seattle.
Each failure is a chance to learn, ” says Dr. Matsen. “My analogy: We don’t need to worry about the planes that landed safety…it’s when the 737 falls out of the sky that we have to analyze what went wrong and learn to prevent future crashes. But this approach—learning from failure identification and analysis—is not a common practice in medicine.
Understanding Patients’ Expectations
Failure is not just a case of “oops, wrong knee.” Dr. Matsen: “While there are a multitude of perspectives on what failure actually is, the simplest definition is perhaps the best: the patient didn’t get the result she’d hoped for. It puts the focus on the ultimate judge of the effectiveness of treatment: the patient. While this approach is often criticized as being ‘subjective’, it does establish patient satisfaction as the Gold Standard. The biggest benefit patients get from any treatment is subjective.”
“Take knee replacement, continues Dr. Matsen. “Someone goes through that because he wants to have a better feeling about his health. He or she is not particularly interested in the specifics of how their range of motion has increased from 82 to 113 degrees of flexion—they want to walk without pain, sleep comfortably, and to be able to get in and out of their car easily. To determine if these goals were achieved, we have only to ask the patient.”
So if Mr. Smith isn’t smiling at the conclusion of treatment, it may be because the patient had unrealistic expectations. “If the patient asks, ‘Will my shoulder be normal after surgery?, ’ it is tempting for the surgeon to say ‘Sure.’ That, however, creates an unrealistic expectation because normal shoulders haven’t had surgery. The patient may not recall exactly what the surgeon outlined beforehand, i.e., ‘your shoulder will be stable, but it will not have a normal range of motion.’ Now, after surgery, the shoulder no longer dislocates, but the patient can’t throw like he would like to. Or, the surgeon may delight in great looking x-rays after surgery, but the patient says ‘my shoulder is too weak for me to do what I need to do.’”
Dr. Matsen adds, “Often ‘failure’ is equated to ‘complication’, such as injury to a blood vessel or nerve. More often, however, there is a mismatch between the patient, the problem, and the treatment. Let’s say you have two men with identical knee Xrays showing the same degree of arthritis; one was hurt on the job, is depressed, and is taking a lot of pain medications; the other is the same weight and height as the first guy, but was injured while playing football, isn’t depressed, and is not on medications. Chances are of course better that the latter person will have a successful surgical outcome.
Dr. William Osler made this point years ago: ‘it is much more important to know what sort of a patient has a disease than what sort of a disease a patient has.
“It is also important, ” states Dr. Matsen, “to recognize that the ‘surgeon is the method.’ A hip replacement by one surgeon is not the same as a hip replacement with the identical prosthesis by a different surgeon. Training, experience, patient selection, volume of cases per year, the anesthesiology and nursing teams, and the post operative care team working with a surgeon all help determine the quality of the result perceived by the patient. Importantly, it is not a part of current orthopedic culture to make these failures visible, to conduct a thorough analysis of these failures, and to share the resulting learning with others in the field. This is in contrast to what happens in mountaineering, for example.”
Dr. Matsen, a passionate fan of mountain climbing, notes, “Interestingly, we orthopedists could learn much from The American Alpine Club, a group that publishes an annual report entitled, ‘Accidents in North American Mountaineering.’ They analyze and describe in painstaking detail the reasons for the accidents, be it inexperience, getting lost, rock fall, snow bridge collapse, improper clothing, etc. The idea is to focus on sharing lessons learned.”
Reaching for 100% Satisfaction
As for a view from another mountaintop—the summit of the American Academy of Orthopaedic Surgeons (AAOS)—there is the current president, Dr. Joseph Zuckerman. Chair of the Department of Orthopaedic Surgery at NYU Hospital for Joint Diseases, Dr. Zuckerman has a history of failure analysis. Along with Dr. Matsen, in 1984 he published a study on complications associated with screws and staples in glenohumeral joint surgery…and he’s been pounding the patient drum ever since. “It all starts with impressing upon residents and fellows the importance of ensuring that the patient is satisfied. If there are 100 operations with a 91% success rate, there must be some discussion of the 9% that didn’t go well.”
And there is an established forum for this, says Dr. Zuckerman. “The classic approach to failure analysis is the biweekly morbidity and mortality conference. This review of complications is our quality assurance process, but there is room for improvement, of course. All programs are placing an increased emphasis on quality and the need for more robust systems to identify problems early on. The reality, however, is that it is uncomfortable to look at the cases that went wrong. This is especially true because we live in a litigious society. I do see a bit of a shift away from that as a concern, however.”
Failures become real when they are examined. They become really real when put in black and white. Dr. Zuckerman: “Joint registries are important because you get a large volume of data that you wouldn’t get otherwise, something that enables clinicians to make more informed treatment decisions. AAOS has been working toward this for years now and we’re now close to the finish line. Pooling data will go a long way toward helping us understand what takes place across larger studies.”
Adding Precision to Failure Analysis
Dr. Vivek Mohan, an attending orthopedic surgeon with Kaiser Permanente-Orange County in Anaheim, California, is also an advocate for the practical nature of failure analysis…and for registries. He explains, “To me, failure analysis is a way of saying, ‘How can I do better next time?’ Several years ago I began using our Kaiser Permanente total joint registry to mine data and examine surgical failures. We focused on failures of failures, i.e, re-revisions, and presented two papers at the AAOS convention: one on aseptic multiply revised total knees and the other on total hips.”
“Taking the former as an example, we found that at five years there was a 98-99% survivorship rate. Sounds good? Time to go home? No. It was time to ask, ‘Why did that 1-2% fail? Where was the error? What about the surgical technique, patient issues, implant issues, etc.?’ We found that with multiply revised total knee replacements, the time for subsequent failures accelerates. This means that it’s important to correctly identify and aggressively treat the failure mode the first time—something that’s more likely to happen if you are in the habit of examining your failures.”
Failing to examine your failures, says Dr. Mohan, introduces an element of guesswork into an otherwise precise field. “The takeaway message here may be that we’re getting by with xyz, and it may work rather than it will work. With each subsequent reoperation we are losing host bone such that you need longer and longer stems and increased constraints in the implants, and are digging more into the bony canal. This is more invasive, and is giving you less and less fixation.”
Dr. Mohan also uses his data as an educational and counseling tool for patients. “Let’s say I have a 350 lb patient who is 45 years old and needs a knee replacement. Intuitively, I know that the implant is going to fail due to wear related issues—and then he will need another surgery. Or, take a 49 year old with a body mass index (BMI) of 41 who is a poorly controlled diabetic. In either case I show them the survival analysis curves and say, ‘Our KP Registry data suggests that anyone who gets a knee or hip replacement and who has a BMI that is similar to yours will have X chance of failure within five years. You are down here and people with a BMI of 30 are up here.’ They see it and that really helps them understand the need to do something. We can then lay out a plan for controlling the diabetes, weight loss, etc, thereby truly optimizing them for the best outcome.”
Taking this work into the high-tech arena, the Orthopaedic Division at Kaiser-Orange County has developed a “failure calculator.” “We asked the head of our registry, Liz Paxton, to examine our data and come up with software that predicts failure. Winding up the development phase, this calculator is like the FRAX tool, which was developed by the World Health Organization to evaluate the risk of fracture. It takes into account age, height weight, and other variables. So if you plug in all the data, you can a priori assess their risk of failure and thus counsel them accordingly.”
Dr. Mohan concludes,
For me, the fundamental issue is, ‘How can I make these implants live longer and better? Before I operate the first time, how can I make it the last?
Drawing a line in the sand, Dr. Matsen states, “We still need to focus on the admonition that E. A. Codman nicely stated almost 100 years ago as he proposed the End Result Idea: “The common sense notion that every hospital should follow every patient it treats, long enough to determine whether or not the treatment has been successful, and then to inquire, ‘If not, why not?’ with a view to preventing similar failures in the future.”
