AAOS Updates Clinical Practice Guidelines for Osteoarthritis of the Knee
The American Academy of Orthopaedic Surgeons (AAOS) recently released a revised clinical practice guideline for treatment of osteoarthritis of the knee with key changes to recommendations regarding the dosage of acetaminophen and use of intra-articular hyaluronic acid (HA).
David S. Jevsevar, MD, MBA, chair of the AAOS Evidence Based Quality & Value Committee and chair of the workgroup for OA of the knee, said the 2013 edition of the clinical practice guideline (CPG) contains 15 recommendations and replaces the first edition of the CPG, which had elicited some concern over the methodology employed in garnering some of the evidence … specifically that attached to the use of HA.
Jevsevar, a board-certified orthopedic surgeon at Intermountain Zion Orthopedics & Sports Medicine in St. George, Utah, said it is the policy of the AAOS to do all CPG data analysis in-house. However, the earlier guideline utilized synthesized data from three outside sources — the Agency for Healthcare Research and Quality, Osteoarthritis Research Society, and Cochrane Database of Systematic Reviews.
Both those who sell and manufacture HA, as well as a number of AAOS members, were specifically concerned about the issue of viscosupplementation, which garnered an ‘inconclusive’ recommendation in the first issue. Jevsevar said the committee was clear that a more vigorous internal review of the use of intra-articular hyaluronic acid could result in the same outcome, a stronger recommendation backing the use of HA … or a reversal recommending physicians not use the treatment option. “When we actually did the analysis, that’s what happened,” he said of the reversal, which resulted in a ‘cannot recommend’ designation for the use of HA for patients with symptomatic OA of the knee.
“When you use clinical significance as your bar for recommendation — and we took the 14 best studies out there — it really doesn’t support the use of viscosupplementation, or HA,” he said. “Although a few individual studies found statistically significant treatment effects, when combined together in a meta-analysis, the evidence did not meet the minimum clinically important improvement thresholds.”
Jevsevar went on to explain there is a difference in statistical significance and clinical significance. He noted that on the clinical pain analysis where 0 is no pain and 10 is the worst pain, having patients move from a 9 to an 8.8 after treatment could be considered statistically significant but wouldn’t feel much different to the person with OA. “We use the higher bar of clinical significance,” he continued. ”We feel that’s the one most important to patients.”
Perhaps not surprisingly, the strong recommendation against the use of HA has created some pushback from physicians. “They feel like we have very few treatments for osteoarthritis that work so they are always concerned when we take one away,” he said. However, Jevsevar continued, “Doing something that is expensive and hasn’t been proven isn’t the right thing either.”
He said it’s hard to gauge the true effectiveness of various treatments in the clinical setting for a couple of reasons. “Arthritis research is hard because osteoarthritis patients don’t have the same level of pain everyday,” he explained. “Many of those patients want to do anything but surgery, which is understandable,” Jevsevar continued. “They want the treatment to work, but that creates a placebo effect or bias for whatever is being used.” More research, he added, is certainly needed.
One concern for physicians using HA is that insurance companies will quit reimbursing for the treatment. “We synthesize the evidence, but we don’t make recommendations for insurance,” Jevsevar said. However, he admitted insurance companies could misapply the guidelines for financial purposes. Still, he noted, discontinuing reimbursement for viscosupplementation might not be to a payer’s benefit since it could drive more OA patients to opt for the much more expensive knee implant.
Furthermore, Jevsevar said treatment decisions should replicate the foundation of a three-legged stool — 1) the evidence, 2) physician expertise and experience, and 3) patient preferences and values. “You have to take all three into account when treating a patient. One doesn’t trump the other,” he said.
In addition to the controversial HA ‘no’ recommendation, the work group also reduced the recommended dosage of acetaminophen from 4,000 mg to 3,000 mg a day, which mirrors an overall change made by the Food and Drug Administration for individuals using acetaminophen for any purpose. In patients with symptomatic OA of the knee, Jevsevar said, “Actually, there’s not a lot of evidence to support the use of acetaminophen.”
Other important recommendations that remained the same in the revised guidelines included:
Patients who only display symptoms of OA and no other problems, such as loose bodies or meniscus tears, should not be treated with arthroscopic lavage.
Patients with a body mass index (BMI) greater than 25 should lose a minimum of five percent of their body weight.
Jevsevar noted telling patients to lose weight and get active are “tough discussions” to have but important ones. Low impact exercises including swimming, walking and using an elliptical machine have been proven effective to slow the progression of OA of the knee.
The work group strongly recommended against the use of glucosamine and/or chondroitin sulfate or hydrochloride and against the use of acupuncture. A “strong” strength of recommendation means the quality of the supporting evidence was high with an implication that practitioners should follow strong recommendations unless a clear and compelling rationale for an alternative approach exists. Jevsevar added the ‘no’ recommendations were based on a lack of efficacy rather than a potential for harm. The group also had a moderate recommendation against custom lateral wedge insoles. A moderate recommendation also is compelling, but the quality or applicability of the existing evidence is not as strong.
Due to a lack of research, the CPG was unable to recommend for or against the use of physical agents including electrotherapeutic modalities, manual therapy, bracing, growth factor injections and/or platelet rich plasma.
In the second edition, all included studies had to have a sample size of at least 30 participants and a follow-up period of at least four weeks. More than 10,000 separate pieces of literature were reviewed during the evidence analysis phase. When completed, Jevsevar said the updated OA knee CPG was subjected to the most extensive peer review to date for any AAOS CPG. Ultimately, 16 peer reviewers representing multiple specialty societies submitted formal reviews. “Each meticulously dissected the final recommendations of the document and, based on their well-informed and insightful comments, important changes were made to the final document,” Jevsevar said in an AAOS editorial.
For more information on the second edition OA knee CPG, go online to: www.aaos.org/research/guidelines/GuidelineOAKnee.asp