In the past ten years, I have become increasingly concerned as to where the practice of orthopaedics is headed. While at the Annual Meeting of the American Academy of Orthopaedic Surgeons in 2009, I had a long discussion with several of my peers from around the country about many of the problems we, as orthopaedic surgeons, are facing. The discussion included our thoughts regarding the use of technology in our profession. Although we all agreed that technology has benefited us in some ways, we also agreed that it has not necessarily made us better physicians. One of my peers stated, "I think I am a worse doctor now than I was twenty years ago." Increasing dependence on technology, although attractive, might be making us lose our ability to reason and, hence, be good physicians.
In my orthopaedic practice, I specialize in the treatment of knee injuries and I see many patients who are seeking a second opinion. In recent years, some of the trends I have observed regarding how other physicians arrive at a diagnosis have been alarming to me. I have seen some patients who had undergone a magnetic resonance imaging scan before they were seen for an initial evaluation. Some patients were told they needed a total knee arthroplasty, on the basis of radiographic findings, but they had never had their knees examined physically. Other patients were told that they needed an articular cartilage repair procedure after their physician had evaluated their radiographs and magnetic resonance imaging scan but had not examined the knee or had a conversation about pain and symptoms.
I have observed that some patients are surprised when they are asked to wear shorts so I can see their legs for the physical examination. Patients have been even more surprised when their knees were touched for the examination. These observations led me to conduct a survey of my patients to ascertain whether my observations were accurate and to determine what knee examination practices were used by the physicians who had initially evaluated these patients.
The purposes of the survey were to determine the current knee examination practices of physicians with regard to how many of them ordered radiographs and/or magnetic resonance imaging scans without having physically examined the knee, how many physicians adequately exposed the knee to allow for a thorough examination, and how many physicians examined the uninvolved knee to obtain a baseline for comparison with the involved knee.
Approval for the survey was obtained from the institutional review board at my institution. Patients who were seen as a new patient or for a new injury evaluation for a unilateral knee problem were asked whether they had seen a physician in the last six months for the same problem. If so, they were asked to participate in an online survey, which they completed in a private room. The survey was conducted at the time of the initial visit before a history was obtained or the physical examination was performed.
First, the patient was asked, "What type of doctor did you see?" The choices were orthopaedic surgeon, primary care physician, emergency department physician, or chiropractor, and the patient could check more than one if he or she had seen more than one doctor for the same problem in the past six months. If so, the patient would answer the following yes-or-no questions for each type of doctor seen: "Did the doctor have you wear shorts or a gown or have your knees exposed for the examination?" "Did the doctor physically touch your injured knee for the examination?" "Did the doctor physically touch your noninvolved knee for the examination?" "Did the doctor obtain x-rays?" And "Did the doctor obtain a magnetic resonance imaging scan?"
From March to November 2009, 900 patients were seen for a unilateral knee problem. Of those patients, 361 had seen another physician for the same problem within the previous six months and all 361 completed the survey. Sixty-seven patients had seen more than one physician for the same knee problem, and they completed the survey for each physician they had seen. The mean age (and standard deviation) of the patients at the time of evaluation in my office was 42.4 ± 18.7 years. Ninety-three percent of the patients were from the state of Indiana, and the remaining 7% were from seven other states around the U.S. The primary diagnosis was osteoarthritis for 24% of the patients; a meniscal tear for 26%; a knee ligament injury for 19%; patellar dislocation for 5%; knee stiffness, atrophy, or deconditioned knee for 19%; and another diagnosis for 7%.
Of the 428 prior knee examinations performed, 202 were by orthopaedic surgeons; 154, by primary care physicians; forty-four, by emergency department physicians; and twenty-eight, by chiropractors. Overall, only 54% of the physicians had their patients dressed so that the knees were exposed for the examination. Eighty percent of these patients stated they wore shorts, 15% said they wore a gown, and 5% said their knees were exposed in another manner. For the knee examination, 87% of the physicians physically touched the involved knee, 37% touched the noninvolved knee, 64% had radiographs made, and 51% acquired a magnetic resonance imaging scan (Fig. 1).
Among the 202 orthopaedic surgeons, only 63% had the knees exposed for the examination, 89% touched the involved knee, 37% touched the noninvolved knee, 76% had radiographs made, and 68% acquired a magnetic resonance imaging scan. Of twenty-two orthopaedic surgeons who did not touch the injured knee for the examination, sixteen ordered magnetic resonance imaging scans. Of seventy-five orthopaedic surgeons who did not have the knees exposed for the examination, fifty-nine (79%) evaluated the involved knee through clothing and thirteen (17%) evaluated the noninvolved knee.
The results of this survey confirmed the suspicions I had made regarding knee examination practices. On the one hand, it is reassuring to find that 89% of orthopaedic surgeons had examined the injured knee; however, it is a concern that 11% had not. Even more of a concern is that only 37% of the orthopaedic surgeons examined the noninvolved knee.
The survey was conducted so that patients reported what other physicians had done for their examination. It is possible that some patients did not remember correctly and that there was a margin of error in the findings. However, the questions of the survey were intentionally made to be simple so that patients would need to remember only some basic details of the examination. I did not attempt to quantify the quality of the knee examinations that were done. Instead, I only wanted to determine whether any examination had been performed, and I believe patients can easily remember whether they were asked to change clothes and whether they were touched.
To evaluate any knee problem or injury, physicians should perform a thorough examination on both knees in order to compare the findings of the symptomatic side with those of the contralateral side, and specific symptoms should be associated with findings on radiographic studies. Dargel et al.1 found a good correlation of knee morphology in a comparison of the right and left knees of the same subject. Most people have symmetrical knees, but when one knee becomes injured or begins to feel different from the other knee, the patient becomes concerned that something is wrong and seeks advice and treatment. When patients have unilateral knee pain, it is impossible to know whether the specific findings in the involved knee are the cause of a problem unless the noninvolved knee is evaluated to determine whether the same findings are present in that knee as well. The evaluation of knee range of motion and the determination of the presence of effusion, tenderness, or crepitus (the list is endless) cannot be done without knowing what those findings are in the noninvolved knee. The results of two separate studies found that a loss of even a few degrees of knee extension can adversely affect the patient's symptoms2-3. If range of motion and other physical tests are not performed on both knees, a physician cannot know what is normal for that patient.
Just as we routinely evaluate both lungs, both eyes, and both hands during a physical examination, the accepted standard for a thorough knee examination is an evaluation of both knees4. This is what is taught in medical school and residency. At what point during medical education or through one's medical practice is this basic concept abandoned? It takes time to talk to the patient about his or her problem, and it takes time to perform a thorough examination of both knees. This is time the patient needs with the physician to establish a rapport and feel comfortable with the care received.
In his book Doc: Then and Now with a Montana Physician5, Losee described three commandments for the clinician. The first is to listen (he credited Dr. Guy Liorzou for this commandment). If you take time to listen to the patient talk about his or her problem, not only will you obtain a pretty good idea as to the diagnosis, you will know the extent to which the problem limits the patient's lifestyle. The second commandment is to look at the function. For an examination of the knee, this means watching the patient walk, hop, and squat. The third commandment is to examine the patient. While these commandments appear rather simple, they obviously merit repeating. Following these initial steps leads the physician to a presumptive diagnosis.
I believe that, as technological advances have been made, our capacity for thought and reason is being used less and less. The results of this survey indicate that many knee conditions are being treated with surgery on the basis of the findings on radiographs and magnetic resonance imaging scans without a thorough examination of both knees being performed. Sixteen patients who had been seen by orthopaedic surgeons reported that the surgeon had acquired a magnetic resonance imaging scan but had not performed an examination of the injured knee. Several studies have shown how common it is to find abnormal signals in the meniscus or meniscal tears on magnetic resonance imaging scans of individuals who are asymptomatic, and these abnormalities become more common in older age groups6-8. Technology, which should be an assistive tool that helps us to confirm a suspected diagnosis, has become a crutch in our profession instead of a tool. In my opinion, a magnetic resonance imaging scan should not be performed unless a thorough examination of both knees has been done, the physician has a presumptive diagnosis, and the scan is needed to confirm the suspected diagnosis and/or to guide treatment. Sarmiento9 eloquently discussed how technology has been a major factor in the decline of respect for the discipline of orthopaedics when he stated, "It is disingenuous to argue that unnecessary surgery and abuse of nonessential expensive tests are not widespread and becoming more frequent."
I suspect that the problem I bring to light from this study is also present for other joints in the body. A new emphasis on educating physicians (of all kinds) about how to correctly and thoroughly examine the knee (and other joints) is needed. There are no shortcuts to providing appropriate patient care, and such care begins by listening to the patient talk about his or her problem and by performing a thorough physical examination. The results of a physical examination should supersede the findings on a magnetic resonance imaging scan or other diagnostic test when discrepancies are found. Clinical judgment should be paramount.