- Email: [email protected]
The Role of Intra-articular Hyaluronic Acid in Symptomatic Osteoarthritis of the Knee
PM R 7 (2015) 995-1001
www.pmrjournal.org
Point/Counterpoint
Guest Discussants: Thiru M. Annaswamy, MD, MA, FAAPMR, Erika V. Gosai, MD, David S. Jevsevar, MD, MBA Feature Editor: Jaspal Ricky Singh, MD
The Role of Intra-articular Hyaluronic Acid in Symptomatic Osteoarthritis of the Knee CASE SCENARIO A 62-year-old female nursing supervisor with a history of diabetes mellitus (hemoglobin A1C score: 7.8) presents to your office with a 5þ-year history of right knee pain. She does not report any traumatic events or injuries and states that her right knee is progressively getting more painful with walking and especially when descending stairs. She has participated in both land-based and aquatic therapy in the past with limited improvement. She also reports that she received an ultrasound-guided intra-articular corticosteroid injection 8 weeks ago with tremendous improvement in her pain and function, but now the pain has returned. Examination reveals that she is obese, with a body mass index (BMI) of 34.5, and that she has moderate tenderness over the medial and lateral right knee joint lines. Her range of motion is normal with minimal swelling around the knee joint. Her right knee radiograph reveals Kellgren-Lawrence grade 3 osteoarthritis (OA). She has recently read the guidelines for treating knee OA put forth by the American Academy of Orthopaedic Surgeons (AAOS), in which viscosupplementation is no longer recommended. However, she has many family members and friends who received “Synvisc” in the past and experienced significant improvement in pain and function, and she would like to try this treatment option. Drs Thiru Annaswamy and Erika Gosai will argue that, despite the new AAOS guidelines, viscosupplementation should be offered to the patient to treat the symptoms caused by knee degenerative joint disease. Dr David Jevsevar will argue that there is very little evidence that viscosupplementation is effective for knee OA and that either alternative conservative treatments or surgery should be recommended. Thiru M. Annaswamy, MD, MA, FAAPMR, and Erika V. Gosai, MD, Respond The question before us is how to best manage this patient’s symptomatic knee OA given her medical and prior treatment history. Optimal nonoperative management of OA of the knee includes behavioral and lifestyle changes [1], which include weight loss, aerobic exercise, and a strengthening program. Patients who are clinically defined as obese (BMI >30) are 4 times more likely to have knee OA than are persons with a BMI <25. Every 5 kg of weight gain increases the risk of knee OA by 36% [2]. In addition, higher BMI levels in patients with OA are associated with increased pain and poorer self-reported and performance-based measures of function [3]. It would certainly be in the best interest of this patient to participate in an exercise and weight loss
program. For every pound of weight lost, she would experience a 4-fold reduction in the load exerted on the knee, per step. The loss of only 5% of body weight can provide some pain relief, and a 10% weight loss can provide significant pain reduction [2]. Furthermore, 18 months of aerobic exercise for 1 hour daily, 3 times weekly, can improve self-reported physical disability, knee pain, and performance on physical tasks by up to 12% [3]. Once an appropriate weight has been reached, 80 minutes of moderate exercise or 35 minutes of vigorous exercise per day is necessary to prevent weight gain [2]. However, the patient’s current weight and stage of arthritis (Kellgren-Lawrence grade 3 OA) would likely prevent her from being able to participate in an exercise program vigorous enough to provide any
1934-1482/$ - see front matter ª 2015 by the American Academy of Physical Medicine and Rehabilitation http://dx.doi.org/10.1016/j.pmrj.2015.08.002
996
Intra-articular Hyaluronic Acid and Knee OA
benefit. The Centers for Disease Control and Prevention [4] reported that “.adults with obesity and arthritis were 44% more likely to be physically inactive compared with persons with obesity but without arthritis.” Thus, without further intervention to decrease her pain level again (after it was initially decreased by a corticosteroid injection), it is highly unlikely that this patient would be compliant with an exercise program. Therefore, we need to focus on choosing the most effective pain-relieving treatment we can offer. Nonsteroidal anti-inflammatory drugs (NSAIDs) have long been the mainstay of arthritis treatment, but they come with inherent risks. NSAIDs are widely known to increase the risk of gastrointestinal toxicity and contribute to chronic kidney disease (CKD), and they are linked to vascular events including myocardial infarction, stroke, and cardiovascular mortality [5]. The American College of Rheumatology (ACR) recommends against the use of NSAIDs in patients with CKD stage IV or greater, and careful consideration of the risks and benefits for patients with CKD stage III [6]. Given the patient’s obesity and diabetes, it is possible that she has some level of kidney disease. Therefore, NSAIDs should be used with caution in this patient. Is this patient’s only feasible next option, then, a total knee arthroplasty (TKA)? TKA typically is reserved for patients with refractory pain and disability. Prior to surgical intervention, patients typically have tried 3 or more conservative management options, including medications, intra-articular injections, and physical therapy [7]. It is likely that this patient has already failed to respond to other conservative treatments such as physical modalities and local agents or topical medications, given that she recently received an intraarticular corticosteroid injection. Intra-articular steroids can be effective for short-term relief, with improvement in pain that usually lasts 2-3 weeks [8]. However, repeating the corticosteroid injection only 8 weeks after her previous injection is not advisable. In this patient, who has a history of diabetes, another corticosteroid injection can result in elevated blood pressure and/or blood glucose levels, with effects lasting up to 7 days. In addition, there is a risk of a postinjection flare of pain within the first 24-48 hours as a result of a chemical synovitis from the corticosteroid crystalline deposits [8]. The Osteoarthritis Research Society International guidelines suggest that, with regard to intra-articular steroids, for longer durations of pain relief, clinicians should consider other treatment options [9]. Considering the aforementioned factors, at this point, a repeat corticosteroid injection is not advisable in this patient. Therefore, to achieve effective pain control and better enable this patient to exercise, an intra-articular hyaluronic acid (IAHA) injection may be a suitable alternative, especially compared with both NSAIDs and a repeat intra-articular steroid injection. In a
meta-analysis of 5 trials comparing IAHA and NSAIDs, no statistically significant difference in change in pain scores from baseline to last follow-up was found. NSAIDs have a small positive, though modest, effect on pain. The effect of IAHA was found to be comparable to that of NSAIDs at 4 and 12 weeks, with an effect size of 0.2 [5]. Unlike NSAIDs, however, IAHA preparations are generally considered to be safe, with infrequent local reactions and a relative paucity of systemic adverse events [5]. The most common adverse reaction to IAHA is an inflammatory response at the injection site characterized by local swelling and pain and, rarely, pseudosepsis [8]. However, we need to adequately address the patient’s valid concern of the AAOS guidelines that do not recommend the use of IAHA. In 2013, AAOS released guidelines on the treatment of OA of the knee, in which Recommendation 9 stated, “cannot recommend using hyaluronic acid for patients with symptomatic osteoarthritis of the knee.” It is important to note that this recommendation was based on their determination that IAHA lacked efficacy and was not based on safety concerns [10]. In the same guidelines, even though AAOS could not recommend for or against the use of intra-articular corticosteroids, they cited a study that found steroid injections to be inferior to IAHA injections [10]. Interestingly, guidelines other than those of AAOS have reached slightly different conclusions regarding the use of IAHA. The ACR guidelines could not make a recommendation regarding the use of IAHA in the initial management of knee OA [6]. The Osteoarthritis Research Society International guidelines concluded, with good quality of evidence, that the use of IAHA is uncertain in OA affecting only the knee and not appropriate in multiple-joint OA. The guidelines stated that IAHA has a small but significant efficacy in reducing pain by week 4, with a peak at week 8 (with moderate clinical significance) and residual benefits that lasted until 24 weeks. They further noted that whereas intra-articular steroids provided greater benefit for pain 2 weeks after injection, IAHA provided greater benefit at 12 and 26 weeks [9]. The conclusion of the AAOS guideline that IAHA should not be used was determined on the basis of the controversial use of minimum clinically important improvement (MCII) values. The use of this one metric to determine their recommendation fails to acknowledge multiple studies showing statistically and clinically significant improvements in symptoms after treatment. Although the MCII can be a useful tool, others have pointed out that “it should not be used as a cornerstone of clinical decision making” [11]. Individual clinical decisions need to be made using the concept called the “evidence-based triad” [12]. This concept provides a guide for deciding on the applicability of evidence to individual patients by seeking common ground in the triad of (1) clinical impression based on individual clinical expertise, (2) best available external evidence,
T.M. Annaswamy et al. / PM R 7 (2015) 995-1001
and (3) the patient’s unique values and expectations. In our patient’s case, 2 out of 3 factors in the triad are in favor of using IAHA injections to treat this patient’s pain: the clinical impression and clinical expertise of the treating physician, and the patient’s unique values and expectations that she “would like to try this treatment option.” The second factor, best available clinical evidence, is, according to the AAOS guidelines, marginal. As Dr Jevsevardthe Chair of the AAOS volunteer workgroup that developed the clinical practice guideline (CPG), and the author of the counter argument in this Point-Counterpoint articledhimself noted when discussing the AAOS Knee CPGs, “Evidence, whether strong or inconclusive, is never sufficient to make important clinical decisions. Individual values and preferences must balance the evidence to achieve optimal shared decision making” [13]. Therefore, considering the overall parameters of this case, and practicing sound evidence-based medicine, it is our opinion that a trial of IAHA injections in this patient is reasonable and appropriate.
References 1. Bert JM, Bert TM. Nonoperative treatment of unicompartmental arthritis: From bracing to injection. Clin Sports Med 2014;33: 1-10. 2. Bliddal H, Leeds AR, Christensen R. Osteoarthritis, obesity and weight loss: Evidence, hypotheses and horizonsda scoping review. Obes Rev 2014;15:578-586.
997
3. Miller GD, Rejeski WJ, Williamson JD, et al. The Arthritis, Diet and Activity Promotion Trial (ADAPT): Design, rationale, and baseline results. Control Clin Trials 2003;24:462-480. 4. Centers for Disease Control and Prevention. Arthritis as a potential barrier to physical activity among adults with obesitydUnited States, 2007 and 2009. MMWR Morb Mortal Wkly Rep 2011;60:614-618. 5. Bannuru RR, Vaysbrot EE, Sullivan MC, McAlindon TE. Relative efficacy of hyaluronic acid in comparison with NSAIDs for knee osteoarthritis: A systematic review and meta-analysis. Semin Arthritis Rheum 2014;43:593-599. 6. Hochberg MC, Altman RD, April KT, et al. American College of Rheumatology 2012 recommendations for the use of nonpharmacologic and pharmacologic therapies in osteoarthritis of the hand, hip, and knee. Arthritis Care Res 2012;64:465-474. 7. Klett MJ, Frankovich R, Dervin GF, Stacey D. Impact of a surgical screening clinic for patients with knee osteoarthritis: A descriptive study. Clin J Sport Med 2012;22:274-277. 8. Hameed F, Ihm J. Injectable medications for osteoarthritis. PM R 2012;4(5 suppl):S75-S81. 9. McAlindon TE, Bannuru RR, Sullivan MC, et al. OARSI guidelines for the non-surgical management of knee osteoarthritis. Osteoarthritis Cartilage 2014;22:363-388. 10. Jevsevar DS, Brown GA, Jones DL, et al. The American Academy of Orthopaedic Surgeons evidence-based guideline on treatment of osteoarthritis of the knee, 2nd edition. J Bone Joint Surg Am 2013; 95:1885-1886. 11. Bannuru RR, Vaysbrot EE, McIntyre LF. Did the American Academy of Orthopaedic Surgeons osteoarthritis guidelines miss the mark? Arthroscopy 2014;30:86-89. 12. Glasziou P, Guyatt GH, Dans AL, Dans LF, Straus S, Sackett DL. Applying the results of trials and systematic reviews to individual patients. ACP J Club 1998;129:A15-A16. 13. Jevsevar DS. AAOS approves updated OA knee CPG. AAOS Now 2013;7:1-6. Available at: http://www.aaos.org/news/aaosnow/ jun13/cover1.asp. Accessed August 11, 2015.
David S. Jevsevar, MD, MBA, Responds This case scenario presents a patient with a history that all of us involved in the care of patients with knee OA commonly encounter. Her long-standing symptomatic moderate to severe joint degeneration (based on the Kellgren-Lawrence grade) is now bothersome enough that she is seeking treatment. Her response to an intraarticular corticosteroid injection, although transient, confirms that her current symptoms are related to her knee disease. Other beneficial information pertaining to her history would include her response to NSAIDS, her ability to participate in exercise programs, the quality and timing of her symptoms, and the presence of mechanical symptoms. Clinical findings that would aid in her care might include her range of motion (especially the presence of a significant flexion contracture), the presence/absence of an effusion, and assessment of muscle strength (especially the quadriceps). In my practice, we use shared decision-making tools to aid in the discussion of treatment alternatives for patients with knee OA. Whatever final treatment course is pursued, the patient should be given every opportunity to be involved in her care. The integration of patient preferences, physician clinical expertise and experience, and
the best available evidence should be considered a fundamental tenet of care for all patients, but especially in preference-based decisions like those encountered with patients who have symptomatic knee OA [1]. We also use the concept of an arthritis ladder, starting with the most conservative, least-invasive treatment alternatives and progressing to more aggressive interventions with surgery as the final end point of the ladder. Lastly, we also try to offer interventions to our patients that have been proven to provide clinically significant improvement in their pain or stiffness. Clinical significance generally refers to an improvement (or worsening) of a patient’s condition or symptoms that is important to the patient. For this specific patient, a discussion of weight loss and exercise is indicated. Although we tend to underestimate the value of weight loss and exercise, a solid evidence base exists to support these interventions [2-4]. Weight loss likely has an effect by reducing the load through the joint during daily activities of living. Exercise and strengthening aim to reduce joint forces by increasing strength in the surrounding musculature. In the scenario provided, there are no apparent contraindications to trying a course of NSAIDs. Multiple
998
Intra-articular Hyaluronic Acid and Knee OA
well-conducted randomized controlled trials have demonstrated improvement in knee pain and stiffness [5]. NSAIDs have also been shown to be superior to acetaminophen in the treatment of knee OA. For this patient, ensuring normal renal function is an important caveat to starting NSAIDs. In general, response to narcotic pain medications is not predictable, and the dependency potential and other adverse consequences of long-term use should be considered. Although the existing literature does not provide specific dosing and timing of intra-articular corticosteroids, a number of randomized controlled trials (RCTs) demonstrate a clinically significant improvement in pain and stiffness [6]. Most physicians space intra-articular steroids over periods greater than 2 months, and the patient’s elevated hemoglobin A1C score is a relative contraindication to another injection. More robust studies are needed to determine better dose response and utilization intervals. A number of other treatment modalities have been described for the treatment of patients with symptomatic knee OA. Shoe inserts, knee braces, and acupuncture have all been studied and generally have been found inadequate for symptomatic relief [7,8]. In general, the evidence does not support the use of any of these interventions, although their utilization persists in the United States. In 2009, AAOS published a CPG on the Treatment of Knee Osteoarthritis [9]. In that CPG, the workgroup could not recommend for or against the use of IAHA, which is commonly referred to as viscosupplementation. That recommendation was based on analysis of clinical significance culled from a literature review provided by the Agency for Healthcare Research and Quality (AHRQ). Representatives of the manufacturers of IAHA expressed concerns about the inconclusive recommendation, citing a number of studies that the AHRQ analysis had failed to include. As such, they requested a new analysis of IAHA, including all appropriate literature as evidence. The AAOS Evidence Based Practice Committee agreed that several key studies were not included in the AHRQ dataset and appointed a new workgroup to create an updated and inclusive review and analysis of all of the literature regarding the treatment of knee OA. CPGs are “systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances” [10]. AAOS had concluded that the veracity of the evidence-based program would benefit from internal evidence searches and analysis. AAOS made a significant investment, hiring research analysts and staff to internally perform evidence search, grading, and analysis. AAOS CPGs meet all of the criteria for trustworthy guidelines established by the Institute of Medicine [10]. To date, the AAOS has produced 17 CPGs on a variety of topics within musculoskeletal care. CPGs identify diagnostic or therapeutic interventions that are effective; the AAOS Appropriate Use Criteria then determine in which patients and when these interventions are appropriate.
The Treatment of Knee Osteoarthritis (2nd edition) was published in 2012 [5]. It was created using the internal AAOS process for CPGs, but included representatives from many nonorthopedic professional medical societies, including the ACR, the American Academy of Family Physicians, the American Academy of Physical Medicine and Rehabilitation, and the American Physical Therapy Association. Each member of the workgroup had no reported conflicts of interest. The CPG has 15 recommendations, but Recommendation 9, which addresses the use of IAHA in patients with knee OA, has been the most controversial. Recommendation 9 states, “We cannot recommend using hyaluronic acid for patients with symptomatic osteoarthritis of the knee.” The AAOS CPG process is transparent, and clinically significant results are used preferentially and in addition to the presence of statistical significance. In the knee OA literature, the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) was frequently used to measure the outcomes of treatment [11]. Minimally clinically important difference (MCID), a measure of clinical significance, has been described and validated in patients with knee OA. AAOS considers the use of anchor-based MCID to be preferential to other measures of clinical significance such as IMMPACT, which arbitrarily define clinically important results based on physician consensus rather than actual patient response. For knee OA, 2 studies report on MCID from the WOMAC in patients with knee OA [12,13]. The MCIDs reported in these studies were used as the bar for determining clinical significance in the CPG. The AAOS CPG program also utilizes the “best evidence” approach to systematic analysis and review, as described by Slavin [14]. This methodology acknowledges that higher quality studies should possess greater weight than low-quality or unpublished data. Well-performed, higher quality studies should reflect the real effect of any given intervention and treatment. The “best evidence” methodology also tends to reduce the degree of heterogeneity, or differences in study design, which strengthens the ability to make inferences from the published literature. For Recommendation 9 in the CPG, 14 high-quality studies were included as the basis for the final recommendation. Nine of these high-quality studies provided enough data to perform a meta-analysis of the IAHA literature. Because these high-quality studies were RCTs comparing IAHA with placebo (intra-articular saline injection), meta-analyses for IAHA were performed in 3 domains: WOMAC Pain, WOMAC Stiffness, and WOMAC Global. Calculation of the pooled mean effect size for both IAHA and placebo were then performed for the metaanalyses and subsequent Forest Plots. These plots may look somewhat different from those usually seen because the workgroup applied the concept of MCID. When compared with placebo, IAHA does have a statistically significant effect in patients with knee OA. However, the
T.M. Annaswamy et al. / PM R 7 (2015) 995-1001
analyses reveal that IAHA does not meet the established measure of clinical significance when compared with placebo. The evidence reinforced the recommendation that the AAOS CPG could not support the use of IAHA in patients with knee OA. Both NSAIDs and intra-articular corticosteroids do meet the bar of clinical significance, and the evidence supports their recommendation for use. The AAOS CPG, and especially Recommendation 9, have been subjected to a great deal of criticism. Some critics have argued against the use of clinical significance measures such as MCID, but AAOS strongly believes that patient outcomes are the most important. Although used previously in peer-reviewed literature, another criticism has been the application of group mean pooled effect sizes to MCID values. Although this approach has flaws, the exact same argument applies to the use of these same group effect sizes to statistical significance. Practically speaking, very little of our published, peer-reviewed literature addresses individual effect size or the proportion of patients achieving these desired clinically significant effects, yet we routinely apply the mean effect sizes reported in the literature to our individual patients. Thus this critique rings somewhat hollow and arbitrary. Other persons have argued that intra-articular saline injections are not an inert placebo, but saline solution has been used routinely as the placebo of choice for intra-articular injections. The lack of the use of ultrasound-guided injections is another argument against the existing literature on IAHA. This argument can be dismissed because determinations can only be made from the existing evidence, and it would seem unlikely from a probability point of view that more IAHA injections were inappropriately placed than the saline injections. Lastly, it is important to note that this recommendation has been controversial because IAHA is a treatment modality that many of us have used for years. Our tendency toward confirmation bias leads us to affirm that which we do over any evidence that confronts our actions. Recommendation 9, as indicated in the CPG, is not to be used as the basis for insurance coverageebased determinations, which generally also require an analysis of cost-effectiveness, risks/harms, current practice, and patient burden. Significant variation in utilization of IAHA also likely reflects factors other than effectiveness in its routine use. The U.S. health care system has had unsustainable growth in spending as part of our national gross domestic product, and this growth
999
is in part due to the utilization of treatments that do not have proven clinical efficacy. As stewards of our health care system, it is a fundamental principle that we should critically assess all of our interventions to ensure that they provide a measurably beneficial effect to our patients. Currently, the existing evidence does not support the routine use of IAHA. References 1. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine: What it is and what it isn’t. BMJ 1996; 312:71-72. 2. Coleman S, Briffa NK, Carroll G, Inderjeeth C, Cook N, McQuade J. A randomised controlled trial of a self-management education program for osteoarthritis of the knee delivered by health care professionals. Arthritis Res Ther 2012;14:R21. 3. Focht BC, Rejeski WJ, Ambrosius WT, Katula JA, Messier SP. Exercise, self-efficacy, and mobility performance in overweight and obese older adults with knee osteoarthritis. Arthritis Rheum 2005;53:659-665. 4. Miller GD, Nicklas BJ, Davis C, et al. Intensive weight loss program improves physical function in older obese adults with knee osteoarthritis. Obesity (Silver Spring) 2006;14:1219-1230. 5. American Academy of Orthopaedic Surgeons. Treatment of Osteoarthritis of the Knee: Evidence-Based Guideline. 2nd ed. Available at: http://www.aaos.org/Research/guidelines/TreatmentofOsteoarthritis oftheKneeGuideline.pdf. Accessed August 11, 2015. 6. Jones A, Doherty M. Intra-articular corticosteroids are effective in osteoarthritis but there are no clinical predictors of response. Ann Rheum Dis 1996;55:829-832. 7. Bennell KL, Bowles KA, Payne C, et al. Lateral wedge insoles for medial knee osteoarthritis: 12 month randomised controlled trial. BMJ 2011;342:d2912. 8. Kirkley A, Webster-Bogaert S, Litchfield R, et al. The effect of bracing on varus gonarthrosis. J Bone Joint Surg Am 1999;81:539-548. 9. American Academy of Orthopaedic Surgeons. Treatment of Osteoarthritis of the Knee: Evidence-Based Guideline. 1st ed. Rosemont, IL: American Academy of Orthopaedic Surgeons; 2009. 10. National Research Council. Clinical Practice Guidelines We Can Trust. Washington, DC: National Academies Press; 2011. 11. Stucki G, Sangha O, Stucki S, et al. Comparison of the WOMAC (Western Ontario and McMaster Universities) osteoarthritis index and a self-report format of the self-administered Lequesne-Algofunctional index in patients with knee and hip osteoarthritis. Osteoarthritis Cartilage 1998;6:79-86. 12. Angst F, Aeschlimann A, Stucki G. Smallest detectable and minimal clinically important differences of rehabilitation intervention with their implications for required sample sizes using WOMAC and SF36 quality of life measurement instruments in patients with osteoarthritis of the lower extremities. Arthritis Rheum 2001;45:384-391. 13. Tubach F, Wells GA, Ravaud P, Dougados M. Minimal clinically important difference, low disease activity state, and patient acceptable symptom state: Methodological issues. J Rheumatol 2005;32:2025-2029. 14. Slavin RE. Best-evidence synthesis: An alternative to meta-analytic and traditional reviews. Educational Researcher 1986;15:5-11.
Thiru M. Annaswamy, MD, MA, FAAPMR, and Erika V. Gosai, MD, Respond We are in full agreement with Dr Jevsevar regarding the basic principles of shared decision making, use of the arthritis ladder, and specific recommendations of weight loss, exercise, and use of NSAIDs. We also agree
that it is inadvisable to administer a repeat intraarticular corticosteroid injection this early. However, at this decision point in the care of this patient, best practice dictates the use of the “evidence-
1000
Intra-articular Hyaluronic Acid and Knee OA
based triad” [1] as mentioned in our initial argument. It is only by considering all 3 facets of this triadd(1) clinical impression based on individual clinical expertise, (2) best available external evidence, and (3) the patient’s unique values and expectationsdthat we can determine the best treatment course for our patients. To practice medicine based on only one aspect of the triad (eg, relying on only one clinical practice guideline) would be a disservice to our patients. Our patient has already expressed her interest in receiving IAHA for management of her symptoms, satisfying one of the facets of the triad. What does the best available external evidence show? Our colleague mentions that the AAOS put together a committee to address concerns that the guidelines published in 2009 were based on the AHRQ literature review, which was not inclusive of the most recently available literature. The AHRQ subsequently published another review of IAHA in 2014 that was inclusive of articles not represented in the 2009 AAOS guideline. In this review, the conclusion was that “the literature suggests a small role, of unclear importance, for HA in improving function among older individuals and an equally small role in reducing pain among adults, with few serious adverse events” [2]. This statement is in line with the 2012 guidelines, in which the ACR conditionally recommended IAHA for patients who had an inadequate response to initial therapy [3]. It differs, however, from the 2013 AAOS guidelines that strongly recommended against the use of IAHA [4]. Taking all the aforementioned, and somewhat conflicting, guidelines into account, the best summarization of the “best available and recent evidence” would seem to indicate that, although not appropriate for everyone, some patients might benefit from the use of IAHA in the management of symptomatic knee OA. Optimal patient selection might come down to specific medical history or patient characteristics. While reviewing the “high-quality” source articles that formed the basis of Recommendation 9 in the 2013 AAOS guidelines, we found that none
of them included patients with a BMI in our patient’s range in their study. In addition, none of the studies mention the presence or absence of other medical diagnoses such as diabetes. Therefore, our patient may not be similar to the population studied in the articles that form the basis of Recommendation 9. In summary, our opinion is that the best available external evidence is neutral regarding the use of IAHA in our patient. Our colleague also cautions against the trap of confirmation bias. We urge him to practice the same caution, for it is once a firm position on an issue has been taken that a person’s primary purpose becomes one of defense and justification [5]. Given the uncertainty of the best available external evidence, we must rely on our own clinical experience and the expectations of our patients. To facilitate this patient’s ability to exercise and lose weight, and to overcome the barrier of her current symptoms to pursue those goals, we reaffirm our position that a trial of IAHA would be a reasonable next step in the clinical management of this patient. References 1. Glasziou P, Guyatt GH, Dans AL, Dans LF, Straus S, Sackett DL. Applying the results of trials and systematic reviews to individual patients. ACP J Club 1998;129:A15-A16. 2. RAND Southern California Evidence-Based Practice Center. Systematic Review for Effectiveness of Hyaluronic Acid in the Treatment of Severe Degenerative Joint Disease (DJD) of the Knee. Available at: http://ryortho.com/wp-content/uploads/ 2014/12/December-15-2014-AHRQ-Report-titled-djdkneepain-draft. pdf; November 26, 2014. Accessed August 11, 2015. 3. Hochberg MC, Altman RD, April KT, et al. American College of Rheumatology 2012 recommendations for the use of nonpharmacologic and pharmacologic therapies in osteoarthritis of the hand, hip, and knee. Arthritis Care Res (Hoboken) 2012;64:465-474. 4. American Academy of Orthopaedic Surgeons. Treatment of Osteoarthritis of the Knee: Evidence-Based Guideline. 2nd ed. Available at: http://www.aaos.org/Research/guidelines/ TreatmentofOsteoarthritisoftheKneeGuideline.pdf. Accessed August 11, 2015. 5. Nickerson RS. Confirmation bias: A ubiquitous phenomenon in many guises. Rev Gen Psychol 1998;2:175-220.
David S. Jevsevar, MD, MBA, Responds Drs Annaswamy and Gosai present cogent arguments for care of this patient; however, I do take issue with a few of the rationales for the treatment they have proposed. While making a strong case for the importance of weight loss and exercise, they contradict themselves with the statement, “the patient’s current weight and stage of arthritis (Kellgren-Lawrence grade 3 OA) would likely prevent her from being able to participate in an exercise program vigorous enough to provide any benefit.” Patient activation is an important aspect of care for patients with knee OA, and an attempt at intervening in the areas of weight loss and exercise are worthwhile [1]. It is certainly worth exploring the patient’s renal function because NSAIDs are an effective intervention in patients with symptomatic knee OA.
Drs Annaswamy and Gosai rely heavily on the metaanalysis comparing NSAIDs to IAHA [2]. A number of criticisms have been directed at this study, many of which the authors themselves acknowledge. The first is the relative paucity of RCTs and the overall quality of these RCTs. Although the meta-analysis found no difference in pain and stiffness between NSAIDs and IAHA, the authors comment on some of the potential bias created by placebo effect. I believe, however, that their argument is incomplete. They appropriately point out their inability to complete a true network metaanalysis, with NSAID-placebo and IAHA-placebo as the intermediate comparators. The authors then note that this type of comparison might not be appropriate,
T.M. Annaswamy et al. / PM R 7 (2015) 995-1001
because the placebo effect of an intra-articular placebo is significantly larger. I would apply this same argument to the findings of the direct comparison of NSAIDs to IAHA. Because the placebo effect of intra-articular injections is greater, the finding of no difference in effect between NSAIDs and IAHA may truly reflect greater benefit of NSAIDs when placebo effect is considered. Lastly, I strongly disagree with the description of clinical significance by Drs Annaswamy and Gosai. They describe the use of clinical significance, in this case MCII, as controversial. I find it interesting that the concept of an outcome from the patient’s point of view would be considered controversial. In fact, I would argue it is the most important and perhaps only outcome that we should be measuring. The authors then cite an opinion-based commentary by Bannuru and colleagues to support this statement [3]. Without a measure of clinical importance to the patient, we are left with evaluating the differences we find statistically significant. Taken to its logical end point, this argument suggests that each individual physician can independently
1001
evaluate that significance and then make an individual determination of whether that effect is important. Measures of clinical importance, like MCII, attempt to mitigate this apparent arbitrariness of clinical decision making that has led to the significant variation in utilization that we see in our health care system. Lastly, substituting physician opinion for the existing strong evidence is not appropriate, and IAHA injections are not supported by the current evidence.
References 1. Tzeng A, Tzeng TH, Vasdev S, et al. The role of patient activation in achieving better outcomes and cost-effectiveness in patient care. JBJS Rev 2015;3:e4. 2. Bannuru RR, Schmid CH, Kent DM, Vaysbrot EE, Wong JB, McAlindon TE. Comparative effectiveness of pharmacologic interventions for knee osteoarthritis: A systematic review and network meta-analysis. Ann Intern Med 2015;162:46-54. 3. Bannuru RR, Vaysbrot EE, McIntyre LF. Did the American Academy of Orthopaedic Surgeons osteoarthritis guidelines miss the mark? Arthroscopy 2014;30:86-89.
Disclosure T.M.A. PM&R Service, VA North Texas Health Care System, Dallas, TX; Department of PM&R, UT Southwestern Medical Center, Dallas, TX Disclosure: nothing to disclose E.V.G. Department of PM&R, UT Southwestern Medical Center, Dallas, TX Disclosure: nothing to disclose
J.R.S. Physical Medicine and Rehabilitation, Weill Cornell Medical College, 525 E 68th Street, Baker 16, New York, NY 10065. Address correspondence to: J.R.S.; e-mail: [email protected] Disclosure outside this publication: consultancy, Kimberly Clark (money to author)
D.S.J. Department of Orthopaedics, Dartmouth-Hitchcock Medical Center, Lebanon, NH Disclosure: nothing to disclose
Web Poll Question For the case scenario presented in this Point/Counterpoint, which approach would you take? a. offer viscosupplementation b. offer either conservative treatment or surgery To cast your vote, visit www.pmrjournal.org
Results of June and July* Web Polls For the case scenario presented in Platelet-Rich Plasma Versus an Eccentric Exercise Program for Recalcitrant Lateral Elbow Tendinopathy, which approach would you take? 29% eccentric exercise strengthening program 71% platelet-rich plasma For the case scenario presented in Conservative Versus Operative Management for Lumbosacral Radiculopathy With Motor Deficit, what do you recommend? 33% surgery 67% continued nonsurgical care *Due to low polling response for July, results are not statistically valid.
www.pmrjournal.org
Point/Counterpoint
Guest Discussants: Thiru M. Annaswamy, MD, MA, FAAPMR, Erika V. Gosai, MD, David S. Jevsevar, MD, MBA Feature Editor: Jaspal Ricky Singh, MD
The Role of Intra-articular Hyaluronic Acid in Symptomatic Osteoarthritis of the Knee CASE SCENARIO A 62-year-old female nursing supervisor with a history of diabetes mellitus (hemoglobin A1C score: 7.8) presents to your office with a 5þ-year history of right knee pain. She does not report any traumatic events or injuries and states that her right knee is progressively getting more painful with walking and especially when descending stairs. She has participated in both land-based and aquatic therapy in the past with limited improvement. She also reports that she received an ultrasound-guided intra-articular corticosteroid injection 8 weeks ago with tremendous improvement in her pain and function, but now the pain has returned. Examination reveals that she is obese, with a body mass index (BMI) of 34.5, and that she has moderate tenderness over the medial and lateral right knee joint lines. Her range of motion is normal with minimal swelling around the knee joint. Her right knee radiograph reveals Kellgren-Lawrence grade 3 osteoarthritis (OA). She has recently read the guidelines for treating knee OA put forth by the American Academy of Orthopaedic Surgeons (AAOS), in which viscosupplementation is no longer recommended. However, she has many family members and friends who received “Synvisc” in the past and experienced significant improvement in pain and function, and she would like to try this treatment option. Drs Thiru Annaswamy and Erika Gosai will argue that, despite the new AAOS guidelines, viscosupplementation should be offered to the patient to treat the symptoms caused by knee degenerative joint disease. Dr David Jevsevar will argue that there is very little evidence that viscosupplementation is effective for knee OA and that either alternative conservative treatments or surgery should be recommended. Thiru M. Annaswamy, MD, MA, FAAPMR, and Erika V. Gosai, MD, Respond The question before us is how to best manage this patient’s symptomatic knee OA given her medical and prior treatment history. Optimal nonoperative management of OA of the knee includes behavioral and lifestyle changes [1], which include weight loss, aerobic exercise, and a strengthening program. Patients who are clinically defined as obese (BMI >30) are 4 times more likely to have knee OA than are persons with a BMI <25. Every 5 kg of weight gain increases the risk of knee OA by 36% [2]. In addition, higher BMI levels in patients with OA are associated with increased pain and poorer self-reported and performance-based measures of function [3]. It would certainly be in the best interest of this patient to participate in an exercise and weight loss
program. For every pound of weight lost, she would experience a 4-fold reduction in the load exerted on the knee, per step. The loss of only 5% of body weight can provide some pain relief, and a 10% weight loss can provide significant pain reduction [2]. Furthermore, 18 months of aerobic exercise for 1 hour daily, 3 times weekly, can improve self-reported physical disability, knee pain, and performance on physical tasks by up to 12% [3]. Once an appropriate weight has been reached, 80 minutes of moderate exercise or 35 minutes of vigorous exercise per day is necessary to prevent weight gain [2]. However, the patient’s current weight and stage of arthritis (Kellgren-Lawrence grade 3 OA) would likely prevent her from being able to participate in an exercise program vigorous enough to provide any
1934-1482/$ - see front matter ª 2015 by the American Academy of Physical Medicine and Rehabilitation http://dx.doi.org/10.1016/j.pmrj.2015.08.002
996
Intra-articular Hyaluronic Acid and Knee OA
benefit. The Centers for Disease Control and Prevention [4] reported that “.adults with obesity and arthritis were 44% more likely to be physically inactive compared with persons with obesity but without arthritis.” Thus, without further intervention to decrease her pain level again (after it was initially decreased by a corticosteroid injection), it is highly unlikely that this patient would be compliant with an exercise program. Therefore, we need to focus on choosing the most effective pain-relieving treatment we can offer. Nonsteroidal anti-inflammatory drugs (NSAIDs) have long been the mainstay of arthritis treatment, but they come with inherent risks. NSAIDs are widely known to increase the risk of gastrointestinal toxicity and contribute to chronic kidney disease (CKD), and they are linked to vascular events including myocardial infarction, stroke, and cardiovascular mortality [5]. The American College of Rheumatology (ACR) recommends against the use of NSAIDs in patients with CKD stage IV or greater, and careful consideration of the risks and benefits for patients with CKD stage III [6]. Given the patient’s obesity and diabetes, it is possible that she has some level of kidney disease. Therefore, NSAIDs should be used with caution in this patient. Is this patient’s only feasible next option, then, a total knee arthroplasty (TKA)? TKA typically is reserved for patients with refractory pain and disability. Prior to surgical intervention, patients typically have tried 3 or more conservative management options, including medications, intra-articular injections, and physical therapy [7]. It is likely that this patient has already failed to respond to other conservative treatments such as physical modalities and local agents or topical medications, given that she recently received an intraarticular corticosteroid injection. Intra-articular steroids can be effective for short-term relief, with improvement in pain that usually lasts 2-3 weeks [8]. However, repeating the corticosteroid injection only 8 weeks after her previous injection is not advisable. In this patient, who has a history of diabetes, another corticosteroid injection can result in elevated blood pressure and/or blood glucose levels, with effects lasting up to 7 days. In addition, there is a risk of a postinjection flare of pain within the first 24-48 hours as a result of a chemical synovitis from the corticosteroid crystalline deposits [8]. The Osteoarthritis Research Society International guidelines suggest that, with regard to intra-articular steroids, for longer durations of pain relief, clinicians should consider other treatment options [9]. Considering the aforementioned factors, at this point, a repeat corticosteroid injection is not advisable in this patient. Therefore, to achieve effective pain control and better enable this patient to exercise, an intra-articular hyaluronic acid (IAHA) injection may be a suitable alternative, especially compared with both NSAIDs and a repeat intra-articular steroid injection. In a
meta-analysis of 5 trials comparing IAHA and NSAIDs, no statistically significant difference in change in pain scores from baseline to last follow-up was found. NSAIDs have a small positive, though modest, effect on pain. The effect of IAHA was found to be comparable to that of NSAIDs at 4 and 12 weeks, with an effect size of 0.2 [5]. Unlike NSAIDs, however, IAHA preparations are generally considered to be safe, with infrequent local reactions and a relative paucity of systemic adverse events [5]. The most common adverse reaction to IAHA is an inflammatory response at the injection site characterized by local swelling and pain and, rarely, pseudosepsis [8]. However, we need to adequately address the patient’s valid concern of the AAOS guidelines that do not recommend the use of IAHA. In 2013, AAOS released guidelines on the treatment of OA of the knee, in which Recommendation 9 stated, “cannot recommend using hyaluronic acid for patients with symptomatic osteoarthritis of the knee.” It is important to note that this recommendation was based on their determination that IAHA lacked efficacy and was not based on safety concerns [10]. In the same guidelines, even though AAOS could not recommend for or against the use of intra-articular corticosteroids, they cited a study that found steroid injections to be inferior to IAHA injections [10]. Interestingly, guidelines other than those of AAOS have reached slightly different conclusions regarding the use of IAHA. The ACR guidelines could not make a recommendation regarding the use of IAHA in the initial management of knee OA [6]. The Osteoarthritis Research Society International guidelines concluded, with good quality of evidence, that the use of IAHA is uncertain in OA affecting only the knee and not appropriate in multiple-joint OA. The guidelines stated that IAHA has a small but significant efficacy in reducing pain by week 4, with a peak at week 8 (with moderate clinical significance) and residual benefits that lasted until 24 weeks. They further noted that whereas intra-articular steroids provided greater benefit for pain 2 weeks after injection, IAHA provided greater benefit at 12 and 26 weeks [9]. The conclusion of the AAOS guideline that IAHA should not be used was determined on the basis of the controversial use of minimum clinically important improvement (MCII) values. The use of this one metric to determine their recommendation fails to acknowledge multiple studies showing statistically and clinically significant improvements in symptoms after treatment. Although the MCII can be a useful tool, others have pointed out that “it should not be used as a cornerstone of clinical decision making” [11]. Individual clinical decisions need to be made using the concept called the “evidence-based triad” [12]. This concept provides a guide for deciding on the applicability of evidence to individual patients by seeking common ground in the triad of (1) clinical impression based on individual clinical expertise, (2) best available external evidence,
T.M. Annaswamy et al. / PM R 7 (2015) 995-1001
and (3) the patient’s unique values and expectations. In our patient’s case, 2 out of 3 factors in the triad are in favor of using IAHA injections to treat this patient’s pain: the clinical impression and clinical expertise of the treating physician, and the patient’s unique values and expectations that she “would like to try this treatment option.” The second factor, best available clinical evidence, is, according to the AAOS guidelines, marginal. As Dr Jevsevardthe Chair of the AAOS volunteer workgroup that developed the clinical practice guideline (CPG), and the author of the counter argument in this Point-Counterpoint articledhimself noted when discussing the AAOS Knee CPGs, “Evidence, whether strong or inconclusive, is never sufficient to make important clinical decisions. Individual values and preferences must balance the evidence to achieve optimal shared decision making” [13]. Therefore, considering the overall parameters of this case, and practicing sound evidence-based medicine, it is our opinion that a trial of IAHA injections in this patient is reasonable and appropriate.
References 1. Bert JM, Bert TM. Nonoperative treatment of unicompartmental arthritis: From bracing to injection. Clin Sports Med 2014;33: 1-10. 2. Bliddal H, Leeds AR, Christensen R. Osteoarthritis, obesity and weight loss: Evidence, hypotheses and horizonsda scoping review. Obes Rev 2014;15:578-586.
997
3. Miller GD, Rejeski WJ, Williamson JD, et al. The Arthritis, Diet and Activity Promotion Trial (ADAPT): Design, rationale, and baseline results. Control Clin Trials 2003;24:462-480. 4. Centers for Disease Control and Prevention. Arthritis as a potential barrier to physical activity among adults with obesitydUnited States, 2007 and 2009. MMWR Morb Mortal Wkly Rep 2011;60:614-618. 5. Bannuru RR, Vaysbrot EE, Sullivan MC, McAlindon TE. Relative efficacy of hyaluronic acid in comparison with NSAIDs for knee osteoarthritis: A systematic review and meta-analysis. Semin Arthritis Rheum 2014;43:593-599. 6. Hochberg MC, Altman RD, April KT, et al. American College of Rheumatology 2012 recommendations for the use of nonpharmacologic and pharmacologic therapies in osteoarthritis of the hand, hip, and knee. Arthritis Care Res 2012;64:465-474. 7. Klett MJ, Frankovich R, Dervin GF, Stacey D. Impact of a surgical screening clinic for patients with knee osteoarthritis: A descriptive study. Clin J Sport Med 2012;22:274-277. 8. Hameed F, Ihm J. Injectable medications for osteoarthritis. PM R 2012;4(5 suppl):S75-S81. 9. McAlindon TE, Bannuru RR, Sullivan MC, et al. OARSI guidelines for the non-surgical management of knee osteoarthritis. Osteoarthritis Cartilage 2014;22:363-388. 10. Jevsevar DS, Brown GA, Jones DL, et al. The American Academy of Orthopaedic Surgeons evidence-based guideline on treatment of osteoarthritis of the knee, 2nd edition. J Bone Joint Surg Am 2013; 95:1885-1886. 11. Bannuru RR, Vaysbrot EE, McIntyre LF. Did the American Academy of Orthopaedic Surgeons osteoarthritis guidelines miss the mark? Arthroscopy 2014;30:86-89. 12. Glasziou P, Guyatt GH, Dans AL, Dans LF, Straus S, Sackett DL. Applying the results of trials and systematic reviews to individual patients. ACP J Club 1998;129:A15-A16. 13. Jevsevar DS. AAOS approves updated OA knee CPG. AAOS Now 2013;7:1-6. Available at: http://www.aaos.org/news/aaosnow/ jun13/cover1.asp. Accessed August 11, 2015.
David S. Jevsevar, MD, MBA, Responds This case scenario presents a patient with a history that all of us involved in the care of patients with knee OA commonly encounter. Her long-standing symptomatic moderate to severe joint degeneration (based on the Kellgren-Lawrence grade) is now bothersome enough that she is seeking treatment. Her response to an intraarticular corticosteroid injection, although transient, confirms that her current symptoms are related to her knee disease. Other beneficial information pertaining to her history would include her response to NSAIDS, her ability to participate in exercise programs, the quality and timing of her symptoms, and the presence of mechanical symptoms. Clinical findings that would aid in her care might include her range of motion (especially the presence of a significant flexion contracture), the presence/absence of an effusion, and assessment of muscle strength (especially the quadriceps). In my practice, we use shared decision-making tools to aid in the discussion of treatment alternatives for patients with knee OA. Whatever final treatment course is pursued, the patient should be given every opportunity to be involved in her care. The integration of patient preferences, physician clinical expertise and experience, and
the best available evidence should be considered a fundamental tenet of care for all patients, but especially in preference-based decisions like those encountered with patients who have symptomatic knee OA [1]. We also use the concept of an arthritis ladder, starting with the most conservative, least-invasive treatment alternatives and progressing to more aggressive interventions with surgery as the final end point of the ladder. Lastly, we also try to offer interventions to our patients that have been proven to provide clinically significant improvement in their pain or stiffness. Clinical significance generally refers to an improvement (or worsening) of a patient’s condition or symptoms that is important to the patient. For this specific patient, a discussion of weight loss and exercise is indicated. Although we tend to underestimate the value of weight loss and exercise, a solid evidence base exists to support these interventions [2-4]. Weight loss likely has an effect by reducing the load through the joint during daily activities of living. Exercise and strengthening aim to reduce joint forces by increasing strength in the surrounding musculature. In the scenario provided, there are no apparent contraindications to trying a course of NSAIDs. Multiple
998
Intra-articular Hyaluronic Acid and Knee OA
well-conducted randomized controlled trials have demonstrated improvement in knee pain and stiffness [5]. NSAIDs have also been shown to be superior to acetaminophen in the treatment of knee OA. For this patient, ensuring normal renal function is an important caveat to starting NSAIDs. In general, response to narcotic pain medications is not predictable, and the dependency potential and other adverse consequences of long-term use should be considered. Although the existing literature does not provide specific dosing and timing of intra-articular corticosteroids, a number of randomized controlled trials (RCTs) demonstrate a clinically significant improvement in pain and stiffness [6]. Most physicians space intra-articular steroids over periods greater than 2 months, and the patient’s elevated hemoglobin A1C score is a relative contraindication to another injection. More robust studies are needed to determine better dose response and utilization intervals. A number of other treatment modalities have been described for the treatment of patients with symptomatic knee OA. Shoe inserts, knee braces, and acupuncture have all been studied and generally have been found inadequate for symptomatic relief [7,8]. In general, the evidence does not support the use of any of these interventions, although their utilization persists in the United States. In 2009, AAOS published a CPG on the Treatment of Knee Osteoarthritis [9]. In that CPG, the workgroup could not recommend for or against the use of IAHA, which is commonly referred to as viscosupplementation. That recommendation was based on analysis of clinical significance culled from a literature review provided by the Agency for Healthcare Research and Quality (AHRQ). Representatives of the manufacturers of IAHA expressed concerns about the inconclusive recommendation, citing a number of studies that the AHRQ analysis had failed to include. As such, they requested a new analysis of IAHA, including all appropriate literature as evidence. The AAOS Evidence Based Practice Committee agreed that several key studies were not included in the AHRQ dataset and appointed a new workgroup to create an updated and inclusive review and analysis of all of the literature regarding the treatment of knee OA. CPGs are “systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances” [10]. AAOS had concluded that the veracity of the evidence-based program would benefit from internal evidence searches and analysis. AAOS made a significant investment, hiring research analysts and staff to internally perform evidence search, grading, and analysis. AAOS CPGs meet all of the criteria for trustworthy guidelines established by the Institute of Medicine [10]. To date, the AAOS has produced 17 CPGs on a variety of topics within musculoskeletal care. CPGs identify diagnostic or therapeutic interventions that are effective; the AAOS Appropriate Use Criteria then determine in which patients and when these interventions are appropriate.
The Treatment of Knee Osteoarthritis (2nd edition) was published in 2012 [5]. It was created using the internal AAOS process for CPGs, but included representatives from many nonorthopedic professional medical societies, including the ACR, the American Academy of Family Physicians, the American Academy of Physical Medicine and Rehabilitation, and the American Physical Therapy Association. Each member of the workgroup had no reported conflicts of interest. The CPG has 15 recommendations, but Recommendation 9, which addresses the use of IAHA in patients with knee OA, has been the most controversial. Recommendation 9 states, “We cannot recommend using hyaluronic acid for patients with symptomatic osteoarthritis of the knee.” The AAOS CPG process is transparent, and clinically significant results are used preferentially and in addition to the presence of statistical significance. In the knee OA literature, the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) was frequently used to measure the outcomes of treatment [11]. Minimally clinically important difference (MCID), a measure of clinical significance, has been described and validated in patients with knee OA. AAOS considers the use of anchor-based MCID to be preferential to other measures of clinical significance such as IMMPACT, which arbitrarily define clinically important results based on physician consensus rather than actual patient response. For knee OA, 2 studies report on MCID from the WOMAC in patients with knee OA [12,13]. The MCIDs reported in these studies were used as the bar for determining clinical significance in the CPG. The AAOS CPG program also utilizes the “best evidence” approach to systematic analysis and review, as described by Slavin [14]. This methodology acknowledges that higher quality studies should possess greater weight than low-quality or unpublished data. Well-performed, higher quality studies should reflect the real effect of any given intervention and treatment. The “best evidence” methodology also tends to reduce the degree of heterogeneity, or differences in study design, which strengthens the ability to make inferences from the published literature. For Recommendation 9 in the CPG, 14 high-quality studies were included as the basis for the final recommendation. Nine of these high-quality studies provided enough data to perform a meta-analysis of the IAHA literature. Because these high-quality studies were RCTs comparing IAHA with placebo (intra-articular saline injection), meta-analyses for IAHA were performed in 3 domains: WOMAC Pain, WOMAC Stiffness, and WOMAC Global. Calculation of the pooled mean effect size for both IAHA and placebo were then performed for the metaanalyses and subsequent Forest Plots. These plots may look somewhat different from those usually seen because the workgroup applied the concept of MCID. When compared with placebo, IAHA does have a statistically significant effect in patients with knee OA. However, the
T.M. Annaswamy et al. / PM R 7 (2015) 995-1001
analyses reveal that IAHA does not meet the established measure of clinical significance when compared with placebo. The evidence reinforced the recommendation that the AAOS CPG could not support the use of IAHA in patients with knee OA. Both NSAIDs and intra-articular corticosteroids do meet the bar of clinical significance, and the evidence supports their recommendation for use. The AAOS CPG, and especially Recommendation 9, have been subjected to a great deal of criticism. Some critics have argued against the use of clinical significance measures such as MCID, but AAOS strongly believes that patient outcomes are the most important. Although used previously in peer-reviewed literature, another criticism has been the application of group mean pooled effect sizes to MCID values. Although this approach has flaws, the exact same argument applies to the use of these same group effect sizes to statistical significance. Practically speaking, very little of our published, peer-reviewed literature addresses individual effect size or the proportion of patients achieving these desired clinically significant effects, yet we routinely apply the mean effect sizes reported in the literature to our individual patients. Thus this critique rings somewhat hollow and arbitrary. Other persons have argued that intra-articular saline injections are not an inert placebo, but saline solution has been used routinely as the placebo of choice for intra-articular injections. The lack of the use of ultrasound-guided injections is another argument against the existing literature on IAHA. This argument can be dismissed because determinations can only be made from the existing evidence, and it would seem unlikely from a probability point of view that more IAHA injections were inappropriately placed than the saline injections. Lastly, it is important to note that this recommendation has been controversial because IAHA is a treatment modality that many of us have used for years. Our tendency toward confirmation bias leads us to affirm that which we do over any evidence that confronts our actions. Recommendation 9, as indicated in the CPG, is not to be used as the basis for insurance coverageebased determinations, which generally also require an analysis of cost-effectiveness, risks/harms, current practice, and patient burden. Significant variation in utilization of IAHA also likely reflects factors other than effectiveness in its routine use. The U.S. health care system has had unsustainable growth in spending as part of our national gross domestic product, and this growth
999
is in part due to the utilization of treatments that do not have proven clinical efficacy. As stewards of our health care system, it is a fundamental principle that we should critically assess all of our interventions to ensure that they provide a measurably beneficial effect to our patients. Currently, the existing evidence does not support the routine use of IAHA. References 1. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidence based medicine: What it is and what it isn’t. BMJ 1996; 312:71-72. 2. Coleman S, Briffa NK, Carroll G, Inderjeeth C, Cook N, McQuade J. A randomised controlled trial of a self-management education program for osteoarthritis of the knee delivered by health care professionals. Arthritis Res Ther 2012;14:R21. 3. Focht BC, Rejeski WJ, Ambrosius WT, Katula JA, Messier SP. Exercise, self-efficacy, and mobility performance in overweight and obese older adults with knee osteoarthritis. Arthritis Rheum 2005;53:659-665. 4. Miller GD, Nicklas BJ, Davis C, et al. Intensive weight loss program improves physical function in older obese adults with knee osteoarthritis. Obesity (Silver Spring) 2006;14:1219-1230. 5. American Academy of Orthopaedic Surgeons. Treatment of Osteoarthritis of the Knee: Evidence-Based Guideline. 2nd ed. Available at: http://www.aaos.org/Research/guidelines/TreatmentofOsteoarthritis oftheKneeGuideline.pdf. Accessed August 11, 2015. 6. Jones A, Doherty M. Intra-articular corticosteroids are effective in osteoarthritis but there are no clinical predictors of response. Ann Rheum Dis 1996;55:829-832. 7. Bennell KL, Bowles KA, Payne C, et al. Lateral wedge insoles for medial knee osteoarthritis: 12 month randomised controlled trial. BMJ 2011;342:d2912. 8. Kirkley A, Webster-Bogaert S, Litchfield R, et al. The effect of bracing on varus gonarthrosis. J Bone Joint Surg Am 1999;81:539-548. 9. American Academy of Orthopaedic Surgeons. Treatment of Osteoarthritis of the Knee: Evidence-Based Guideline. 1st ed. Rosemont, IL: American Academy of Orthopaedic Surgeons; 2009. 10. National Research Council. Clinical Practice Guidelines We Can Trust. Washington, DC: National Academies Press; 2011. 11. Stucki G, Sangha O, Stucki S, et al. Comparison of the WOMAC (Western Ontario and McMaster Universities) osteoarthritis index and a self-report format of the self-administered Lequesne-Algofunctional index in patients with knee and hip osteoarthritis. Osteoarthritis Cartilage 1998;6:79-86. 12. Angst F, Aeschlimann A, Stucki G. Smallest detectable and minimal clinically important differences of rehabilitation intervention with their implications for required sample sizes using WOMAC and SF36 quality of life measurement instruments in patients with osteoarthritis of the lower extremities. Arthritis Rheum 2001;45:384-391. 13. Tubach F, Wells GA, Ravaud P, Dougados M. Minimal clinically important difference, low disease activity state, and patient acceptable symptom state: Methodological issues. J Rheumatol 2005;32:2025-2029. 14. Slavin RE. Best-evidence synthesis: An alternative to meta-analytic and traditional reviews. Educational Researcher 1986;15:5-11.
Thiru M. Annaswamy, MD, MA, FAAPMR, and Erika V. Gosai, MD, Respond We are in full agreement with Dr Jevsevar regarding the basic principles of shared decision making, use of the arthritis ladder, and specific recommendations of weight loss, exercise, and use of NSAIDs. We also agree
that it is inadvisable to administer a repeat intraarticular corticosteroid injection this early. However, at this decision point in the care of this patient, best practice dictates the use of the “evidence-
1000
Intra-articular Hyaluronic Acid and Knee OA
based triad” [1] as mentioned in our initial argument. It is only by considering all 3 facets of this triadd(1) clinical impression based on individual clinical expertise, (2) best available external evidence, and (3) the patient’s unique values and expectationsdthat we can determine the best treatment course for our patients. To practice medicine based on only one aspect of the triad (eg, relying on only one clinical practice guideline) would be a disservice to our patients. Our patient has already expressed her interest in receiving IAHA for management of her symptoms, satisfying one of the facets of the triad. What does the best available external evidence show? Our colleague mentions that the AAOS put together a committee to address concerns that the guidelines published in 2009 were based on the AHRQ literature review, which was not inclusive of the most recently available literature. The AHRQ subsequently published another review of IAHA in 2014 that was inclusive of articles not represented in the 2009 AAOS guideline. In this review, the conclusion was that “the literature suggests a small role, of unclear importance, for HA in improving function among older individuals and an equally small role in reducing pain among adults, with few serious adverse events” [2]. This statement is in line with the 2012 guidelines, in which the ACR conditionally recommended IAHA for patients who had an inadequate response to initial therapy [3]. It differs, however, from the 2013 AAOS guidelines that strongly recommended against the use of IAHA [4]. Taking all the aforementioned, and somewhat conflicting, guidelines into account, the best summarization of the “best available and recent evidence” would seem to indicate that, although not appropriate for everyone, some patients might benefit from the use of IAHA in the management of symptomatic knee OA. Optimal patient selection might come down to specific medical history or patient characteristics. While reviewing the “high-quality” source articles that formed the basis of Recommendation 9 in the 2013 AAOS guidelines, we found that none
of them included patients with a BMI in our patient’s range in their study. In addition, none of the studies mention the presence or absence of other medical diagnoses such as diabetes. Therefore, our patient may not be similar to the population studied in the articles that form the basis of Recommendation 9. In summary, our opinion is that the best available external evidence is neutral regarding the use of IAHA in our patient. Our colleague also cautions against the trap of confirmation bias. We urge him to practice the same caution, for it is once a firm position on an issue has been taken that a person’s primary purpose becomes one of defense and justification [5]. Given the uncertainty of the best available external evidence, we must rely on our own clinical experience and the expectations of our patients. To facilitate this patient’s ability to exercise and lose weight, and to overcome the barrier of her current symptoms to pursue those goals, we reaffirm our position that a trial of IAHA would be a reasonable next step in the clinical management of this patient. References 1. Glasziou P, Guyatt GH, Dans AL, Dans LF, Straus S, Sackett DL. Applying the results of trials and systematic reviews to individual patients. ACP J Club 1998;129:A15-A16. 2. RAND Southern California Evidence-Based Practice Center. Systematic Review for Effectiveness of Hyaluronic Acid in the Treatment of Severe Degenerative Joint Disease (DJD) of the Knee. Available at: http://ryortho.com/wp-content/uploads/ 2014/12/December-15-2014-AHRQ-Report-titled-djdkneepain-draft. pdf; November 26, 2014. Accessed August 11, 2015. 3. Hochberg MC, Altman RD, April KT, et al. American College of Rheumatology 2012 recommendations for the use of nonpharmacologic and pharmacologic therapies in osteoarthritis of the hand, hip, and knee. Arthritis Care Res (Hoboken) 2012;64:465-474. 4. American Academy of Orthopaedic Surgeons. Treatment of Osteoarthritis of the Knee: Evidence-Based Guideline. 2nd ed. Available at: http://www.aaos.org/Research/guidelines/ TreatmentofOsteoarthritisoftheKneeGuideline.pdf. Accessed August 11, 2015. 5. Nickerson RS. Confirmation bias: A ubiquitous phenomenon in many guises. Rev Gen Psychol 1998;2:175-220.
David S. Jevsevar, MD, MBA, Responds Drs Annaswamy and Gosai present cogent arguments for care of this patient; however, I do take issue with a few of the rationales for the treatment they have proposed. While making a strong case for the importance of weight loss and exercise, they contradict themselves with the statement, “the patient’s current weight and stage of arthritis (Kellgren-Lawrence grade 3 OA) would likely prevent her from being able to participate in an exercise program vigorous enough to provide any benefit.” Patient activation is an important aspect of care for patients with knee OA, and an attempt at intervening in the areas of weight loss and exercise are worthwhile [1]. It is certainly worth exploring the patient’s renal function because NSAIDs are an effective intervention in patients with symptomatic knee OA.
Drs Annaswamy and Gosai rely heavily on the metaanalysis comparing NSAIDs to IAHA [2]. A number of criticisms have been directed at this study, many of which the authors themselves acknowledge. The first is the relative paucity of RCTs and the overall quality of these RCTs. Although the meta-analysis found no difference in pain and stiffness between NSAIDs and IAHA, the authors comment on some of the potential bias created by placebo effect. I believe, however, that their argument is incomplete. They appropriately point out their inability to complete a true network metaanalysis, with NSAID-placebo and IAHA-placebo as the intermediate comparators. The authors then note that this type of comparison might not be appropriate,
T.M. Annaswamy et al. / PM R 7 (2015) 995-1001
because the placebo effect of an intra-articular placebo is significantly larger. I would apply this same argument to the findings of the direct comparison of NSAIDs to IAHA. Because the placebo effect of intra-articular injections is greater, the finding of no difference in effect between NSAIDs and IAHA may truly reflect greater benefit of NSAIDs when placebo effect is considered. Lastly, I strongly disagree with the description of clinical significance by Drs Annaswamy and Gosai. They describe the use of clinical significance, in this case MCII, as controversial. I find it interesting that the concept of an outcome from the patient’s point of view would be considered controversial. In fact, I would argue it is the most important and perhaps only outcome that we should be measuring. The authors then cite an opinion-based commentary by Bannuru and colleagues to support this statement [3]. Without a measure of clinical importance to the patient, we are left with evaluating the differences we find statistically significant. Taken to its logical end point, this argument suggests that each individual physician can independently
1001
evaluate that significance and then make an individual determination of whether that effect is important. Measures of clinical importance, like MCII, attempt to mitigate this apparent arbitrariness of clinical decision making that has led to the significant variation in utilization that we see in our health care system. Lastly, substituting physician opinion for the existing strong evidence is not appropriate, and IAHA injections are not supported by the current evidence.
References 1. Tzeng A, Tzeng TH, Vasdev S, et al. The role of patient activation in achieving better outcomes and cost-effectiveness in patient care. JBJS Rev 2015;3:e4. 2. Bannuru RR, Schmid CH, Kent DM, Vaysbrot EE, Wong JB, McAlindon TE. Comparative effectiveness of pharmacologic interventions for knee osteoarthritis: A systematic review and network meta-analysis. Ann Intern Med 2015;162:46-54. 3. Bannuru RR, Vaysbrot EE, McIntyre LF. Did the American Academy of Orthopaedic Surgeons osteoarthritis guidelines miss the mark? Arthroscopy 2014;30:86-89.
Disclosure T.M.A. PM&R Service, VA North Texas Health Care System, Dallas, TX; Department of PM&R, UT Southwestern Medical Center, Dallas, TX Disclosure: nothing to disclose E.V.G. Department of PM&R, UT Southwestern Medical Center, Dallas, TX Disclosure: nothing to disclose
J.R.S. Physical Medicine and Rehabilitation, Weill Cornell Medical College, 525 E 68th Street, Baker 16, New York, NY 10065. Address correspondence to: J.R.S.; e-mail: [email protected] Disclosure outside this publication: consultancy, Kimberly Clark (money to author)
D.S.J. Department of Orthopaedics, Dartmouth-Hitchcock Medical Center, Lebanon, NH Disclosure: nothing to disclose
Web Poll Question For the case scenario presented in this Point/Counterpoint, which approach would you take? a. offer viscosupplementation b. offer either conservative treatment or surgery To cast your vote, visit www.pmrjournal.org
Results of June and July* Web Polls For the case scenario presented in Platelet-Rich Plasma Versus an Eccentric Exercise Program for Recalcitrant Lateral Elbow Tendinopathy, which approach would you take? 29% eccentric exercise strengthening program 71% platelet-rich plasma For the case scenario presented in Conservative Versus Operative Management for Lumbosacral Radiculopathy With Motor Deficit, what do you recommend? 33% surgery 67% continued nonsurgical care *Due to low polling response for July, results are not statistically valid.