When a patient has end-stage osteoarthritis (OA) of the knee, joint replacement is the ultimate treatment. But before a patient reaches that point, clinicians have numerous other treatment options—and many can be very helpful in reducing pain, improving function, and delaying or occasionally even precluding the need for arthroplasty.
Determining which of these other treatments would be appropriate—and when—isn’t always clear. The AAOS Clinical Practice Guideline (CPG) on the Treatment of OA Knee may be helpful, but can’t cover every clinical situation. At their meeting on Dec. 6, 2013, the AAOS Board of Directors approved another quality tool that can assist physicians in determining appropriate treatments for patients with OA knee: Appropriate Use Criteria (AUC) for Nonarthroplasty Treatment of OA Knee.
“This was a monumental project,” said Michael Heggeness, MD, who served as one of the discussion moderators for the voting panel. “It involved more than 40 physicians, as well as representatives from the Arthritis Foundation, supported by seven staff members from the AAOS.”
The AUC, which is based on a systematic review of the literature in conjunction with clinician expertise from multiple specialties, covers 10 treatments (Table 1). More than 500 different patient scenarios are presented, resulting in more than 5,700 different patient/treatment combinations. Even so, not all possible indications are covered.
“AUC are meant to augment—not supersede—clinician expertise and experience or patient preference,” said Kevin J. Bozic, MD, MBA, chair of the AAOS Council on Research and Quality. “The ultimate judgment regarding any specific criteria should address all circumstances presented by the patient and the needs and resources particular to the practice setting. The scope of this AUC includes nonpharmacologic and pharmacologic interventions for symptomatic OA of the knee as well as surgical procedures less invasive than arthroplasty.”
“The members of the writing panel really deserve gold medals because they put their hearts and souls into this project,” said Dr. Heggeness. “They held 20 phone calls and participated in separate discussions about all 5,700-plus scenarios. They put a lot of work, professionalism, and thoughtful effort into developing a useful tool for clinicians.”
“The panels put tremendous thought into this AUC,” agreed James O. Sanders, MD, who served as section leader and moderator during the AUC’s development. “They provided careful analysis of appropriate care through both the published literature and extensive clinical experience.”
Members of the writing panel developed clinical scenarios using the following guiding principles:
- Patient scenarios must include a broad spectrum of patients that may be eligible for management of OA of the knee; that is, they must be comprehensive.
- Patient indications must classify patients into unique scenarios that are mutually exclusive.
- Patient indications must consistently classify similar patients into the same scenario, creating reliable, valid indicators.
As a result, scenarios ranged from a younger patient who can no longer walk 18 holes without pain and has some mild joint space narrowing to an elderly woman with constant knee pain, especially at night, no significant instability, and moderate tibio-femoral joint space narrowing on the standing anteroposterior radiographs—and hundreds of examples in between.
The scenarios were analyzed and a description matrix was developed, focused on the clinical findings such as the following:
- function-limiting pain
- range of motion (extension/flexion)
- ligamentous instability (not to include antalgic giving way)
- pattern of arthritic involvement
- imaging (joint space in the most involved compartment)
- limb alignment
- mechanical symptoms
- patient age
- Help from an app
“The OA Knee AUC app (www.aaos.org/aucapp) enables the clinician who is examining a specific patient to reference the AUC document where someone has, in hypothetical terms, described an almost identical patient,” explained Dr. Heggeness. “The physician can literally push a button and see what the AUC voting panel thought about the appropriateness of available treatments. For example, a green checkmark indicates that the group thought a treatment would be appropriate; a red circle means the treatment would rarely be appropriate. For many scenarios, a spectrum of options might be appropriate (Fig. 1).
“This can facilitate a conversation between the physician and the patient about the benefits of a treatment or the potential timing of an intervention,” he continued. The app presents an indication profile so that the clinician can select the patient characteristics. It also includes a demonstration “tour,” definitions, background information, the literature review, and the list of contributors.
Out of more than 5,700 different patient/treatment combinations, more than half (53 percent) were rated as “appropriate,” 29 percent were rated as “may be appropriate,” and 18 percent were rated as “rarely appropriate” (Fig. 2).
Dr. Heggeness noted that patients who have OA of the knee may not see an orthopaedic surgeon initially, and that this AUC would be a valuable tool for primary care physicians, rheumatologists, or other doctors, enabling them to draw on the expertise of orthopaedic surgeons in treating these patients.
“Physicians are sometimes slow to change their practices, but the AUC app is so easy to use that even a 61-year-old fellow like me could be persuaded to try it,” he laughed. “I expect that appropriate use criteria will eventually be embraced and shared, just as clinical practice guidelines have been over the years.”
Disclosure information: Dr. Heggeness—K2M, Relievant Medsystems, Spine, Spine Journal, North American Spine Society. Dr. Bozic—AAOS, American Association of Hip and Knee Surgeons, American Joint Replacement Registry, American Orthopaedic Association, California Joint Replacement Registry Project, California Orthopaedic Association, Orthopaedic Research and Education Foundation
Mary Ann Porucznik is managing editor, AAOS Now. She can be reached at email@example.com
- The new appropriate use criteria (AUC) on nonarthroplasty treatment of patients with OA of the knee covers 10 different treatment options and more than 500 patient scenarios.
- A web-based application (www.aaos.org/aucapp) enables clinicians to submit a patient profile based on specific clinical findings and receive feedback on the appropriateness of various treatment options.
- Although the AUC addresses the most common clinical scenarios, it does not include all of the possible indications, and is not meant to supersede clinician expertise and experience or patient preference.
The Making of an AUC
Appropriate use criteria (AUC) are derivative products of Clinical Practice Guidelines (CPGs). CPGs are developed to determine what is effective; AUCs are developed to determine when, how and for whom medical and surgical procedures are appropriate.
AUCs are developed using the RAND/UCLA appropriateness method, which involves three separate panels, as follows:
The writing panel—seven clinicians who are experts in the topic under study—created a list of patient indications, assumptions, and treatments based on the evidence-based systematic review of the literature, augmented by their clinical judgment and expertise where evidence is lacking. The writing panel for the AUC on nonarthroplasty treatment of osteoarthritis of the knee (OAK) included representatives from The Knee Society, American Academy of Hip and Knee Surgeons, American Physical Therapy Association (APTA), American Academy of Physical Medicine and Rehabilitation (AAPMR), and the Arthritis Foundation (AF).
The review panel, a group of 16 clinicians, reviewed the materials and provided suggestions for improvement. The OAK AUC review panel included representatives from AAOS, AF, APTA, AAPMR, The Knee Society, and the Association of Bone and Joint Surgeons (ABJS).
The voting panel—a multidisciplinary 16-member group—used a review of the most current and relevant literature along with their expert clinical judgment to rate the appropriateness of each treatment for each of the various patient scenarios, using a 9-point scale (1–3, rarely appropriate; 4–6, may be appropriate; 7–9, appropriate). The OAK AUC voting panel included representatives from AAOS, ABJS, AF, AAPMR, the Knee Society, and the Musculoskeletal Infection Society. Two rounds of electronic voting were held, the second after a face-to-face meeting focused on the patient scenarios that resulted in initial disagreements among the panelists.