Osteoarthritis (OA) is the most common joint disorder in the United States and is a leading cause of disability.1,2 Approximately 10% of men and 13% of women aged ≥60 years have symptomatic knee OA.1 The number of patients diagnosed with symptomatic knee OA has been steadily increasing over the past several decades as a result of longer life expectancy, an increasing number of patients with obesity, and other modifiable factors (eg, physical inactivity, modern diets).3,4 Because of this high prevalence, symptomatic knee OA poses clinical and economic burdens to patients as well as to the healthcare system.3
Pain is a major driver for treatment decisions in patients with OA of the knee and is frequently correlated with decreased functionality.5 Therefore, the alleviation of pain and improvement in daily function are 2 main goals of therapy.6 Currently, there is no cure for OA, but several nonpharmacologic and pharmacologic options are available to help patients manage the disease.
Nonpharmacologic options for knee OA, such as bracing, physical therapy, and holistic approaches, such as tai chi and yoga, can be used and are acknowledged in current treatment guidelines.7 Common pharmacologic interventions that are used to help alleviate the symptoms of knee OA include nonsteroidal anti-inflammatory drugs (NSAIDs), intra-articular corticosteroids, intra-articular hyaluronic acid, as well as opioids.7
Pharmacologic agents for OA are widely used, and each has a different benefit-risk profile. NSAIDs have historically been the most often used medication; however, the long-term use of NSAIDs is limited, particularly in elderly patients with multiple comorbidities because of the associated gastrointestinal, renal, cardiac, and hematologic adverse events linked to this class of medication.7 NSAIDs now carry a black box warning for cardiovascular and gastrointestinal adverse events.8
Intra-articular corticosteroids have been used for decades for pain relief associated with knee OA, and provide short-term relief. The effectiveness of intra-articular corticosteroids reported in the literature varies.9-12 Multiple studies have been conducted with differing methodologies, resulting in contradictory results that make it difficult to reach a consensus about the role of intra-articular corticosteroids in the treatment of knee OA.9-12 Similarly, data regarding the use of intra-articular hyaluronic acid have been considered inconsistent, and outcomes have been noted to differ based on the molecular weight of the hyaluronic acid.10
Although opioids are generally not recommended as a result of the lack of clinical data supporting their efficacy and/or optimal benefit-risk ratio, they are still prescribed in patients with knee OA.10,13 Greater scrutiny is now also being applied to long-term opioid prescribing, given the propensity for their misuse and abuse.4,14
Nonpharmacologic and pharmacologic interventions are successful in providing symptomatic and functional relief to many patients; however, total knee arthroplasty is necessary in some patients with severe disease who have not responded adequately to nonsurgical interventions. Revision surgery for failed total knee arthroplasty is required for some patients, with infection and mechanical loosening as the most common causes for revision total knee arthroplasty procedures.15
In a survey of 27,372 patients who had knee arthroplasty, 17% of patients expressed overall dissatisfaction with their outcome after having knee arthroplasty, and the rate of dissatisfaction increased to 41% among the 2097 patients who underwent knee arthroplasty revision.16 Patient satisfaction is lower among patients undergoing total knee arthroplasty than in patients who have had other orthopedic procedures, because physical function may not be completely restored to pre-OA levels.17
Treatment guidelines for knee OA are available from multiple organizations, with the most frequently cited sources being the American Academy of Orthopaedic Surgeons (AAOS),10 the American College of Rheumatology (ACR),7 OA Research Society International (OARSI),13 and the American Medical Society for Sport Medicine (AMSSM).18 All these guidelines support exercise and weight loss, but they differ regarding options such as bracing.7,10,13,18 The recommendations for prescription pharmacologic therapy vary across the guidelines, particularly for intra-articular corticosteroids and intra-articular hyaluronic acid.7,10,13,18
Given the multiple guidelines with different recommendations, we hypothesized that treatment patterns among patients with knee OA were also variable. The purpose of this analysis was to compare descriptively the treatment patterns in a real-world setting and the varied guidelines recommendations in patients with newly diagnosed knee OA.
In this retrospective analysis we used data from the electronic health record (EHR) system of the Geisinger Health System between January 1, 2010, and December 2018. Geisinger Health System is an integrated health services organization serving residents throughout Pennsylvania and New Jersey. The Geisinger EHR system includes data of approximately 1.6 million patients, including, but not limited to, inpatient and outpatient visits from primary and specialty care, laboratory testing, medications, procedures, diagnoses, and patient demographics (ie, age, race, and sex).
Patients with knee OA were identified using data in the Geisinger EHR system based on internal Geisinger codes mapped from International Classification of Diseases Ninth Revision (ICD-9) and ICD Tenth Revision (ICD-10) codes (ICD-9: 715.16, 715.26, 715.36, 715.96; ICD-10: M17xx). The first diagnosis for knee OA was established as the index diagnosis if there were no codes in the 6-month period before the index diagnosis.
To be included in the analysis, patients could not have had another claim with a knee OA diagnosis and could not have received treatment with intra-articular corticosteroids, intra-articular hyaluronic acid, opioids, or prescription NSAIDs in the 6-month pre-index period. Patients were then followed for a variable length of time after the index knee OA diagnosis, with a requirement to have observable data for at least 6 months after the initial knee OA diagnosis.
The treatments of interest that were evaluated in the follow-up period included intra-articular corticosteroids, opioids, sodium hyaluronate intra-articular biofermentation-derived hyaluronic acid, prescription NSAIDs, physical therapy, total knee arthroplasty, and bracing (Figure 1).
These treatments were included because they are common nonpharmacologic, pharmacologic, and surgical interventions that are documented in EHR data. The specific intra-articular biofermentation-derived hyaluronic acid used here was Euflexxa, which was selected for this analysis because it is a high-molecular-weight linear hyaluronic acid,19-22 which is believed to have a more favorable safety profile and improved outcomes.19,23,24 This agent was the most frequently (>75%) prescribed hyaluronic acid in the current study population.
Patient demographics and characteristics were captured at baseline during the 6-month prestudy, including sex, age, and comorbidities using the Charlson Comorbidity Index.
The patients who met the study eligibility criteria were evaluated for patterns of select prescription medication use and nonpharmacologic therapy after the index knee OA diagnosis. The percentage of patients receiving each therapy was captured for the first and second treatments. The pharmacologic agents that were evaluated in the study are listed in Table 1.
In instances where patients received more than 1 treatment on the same day, the following hierarchy was used to determine the treatment category order: (1) total knee arthroplasty; (2) intra-articular bioengineered hyaluronic acid; (3) intra-articular corticosteroids; (4) opioids; (5) NSAIDs; (6) physical therapy; (7) bracing.
A total of 8776 patients had an initial (ie, index) diagnosis of knee OA during the study period. The median age was 63 years, and 63% of the patients were female. Additional patient characteristics, such as comorbid conditions, are described in Table 2.
Of the patients who were diagnosed with knee OA during the study period, 7737 (88.2%) received 1 of the evaluated therapies, whereas 1039 (11.8%) patients received none of the 7 therapies that were evaluated. Among those who received a first treatment, 5726 patients received a second treatment.
The most frequently prescribed first treatment option was intra-articular corticosteroids, which was received by approximately 26% of the patients who were newly diagnosed with knee OA. The second most common first treatment received was an opioid (17.6% of patients), followed by intra-articular bioengineered hyaluronic acid (14.9%), and NSAIDs (11.9%).
A total of 5726 patients received a second treatment, and 2011 (22.9%) patients received only 1 treatment during the study period, which may indicate that such patients might have maintained the therapy they received as a first treatment, might have stopped the first therapy and did not start a second treatment during the analysis period, or might not have been able to be monitored to a second treatment.
Opioids were the most frequently (15.8%) prescribed second-line therapy of the treatments that were evaluated in the second-line setting. A total of 2931 (33.4%) patients had exposure to opioids as their first or second treatment. In addition, 1307 patients received total knee arthroplasty as their first treatment (7.2%) or second treatment (7.7%). The distribution of therapies used as first- and second-line treatment is outlined in Figure 2.
These results show that there is a high degree of variability—even within an integrated delivery network—in the treatment choices and the order of therapies for patients newly diagnosed with knee OA. Of the 8776 patients with a knee OA diagnosis, 11.8% received none of the 7 evaluated therapies (some of which are recommended by current guidelines), and 14.9% of the patients had total knee arthroplasty as the first or second treatment, indicating a potential broad range of severity of illness at diagnosis (Figure 2).
Although it is possible that some patients might have had a previous diagnosis of knee OA that was not documented in the EHR data, total knee arthroplasty occurred early in the treatment sequence for several patients.
Opioid prescribing was the second highest category for first-line treatment, and the highest category of pharmacologic agent prescribed for second-line treatment, despite the increased risk associated with opioid use in patients with chronic pain.14 We identified extensive use of all therapeutic options, with the exception of bracing, which was used in less than 1% of patients.
The guideline recommendations for pharmacologic therapy for knee OA vary among the organizations (Table 3).7,10,13 The AAOS treatment guidelines do not recommend for or against the use of intra-articular corticosteroids,10 whereas the ACR and the OARSI guidelines conditionally recommend the use of intra-articular corticosteroids, although they do not mention specific conditions.7,13 Of note, our analysis shows that intra-articular corticosteroids were used in the greatest percentage (25.5%) of patients as a first-line treatment (Figure 2).
The use of opioids is not recommended for or against in the AAOS guidelines,10 is conditionally recommended against in the ACR guidelines,7 and is strongly recommended against in the OARSI guidelines.13 In our current analysis, opioid use was prevalent as a first- or second-line treatment.
Recommendations regarding intra-articular hyaluronic acid range from cannot recommend to conditionally against or conditional, depending on the guideline. The AAOS recommendation is largely attributable to inconsistencies in the clinical evidence for intra-articular hyaluronic acid.10 The AAOS guidelines do note that statistically significant outcomes have been associated with high-molecular-weight viscosupplementation, but that clinical efficacy could not be inferred because of a lack of minimum clinically important improvement for the overall group.10
NSAIDs are the only pharmacologic therapy that we evaluated that received a recommended status by current guidelines, but NSAIDs were used less often as first-line therapy than other pharmacologic therapies. It should be noted that our analysis was focused on prescription NSAID use only; thus, it is possible that patients were receiving over-the-counter NSAIDs.
The guidance for bracing varies by guideline.7,10,13 Only a small number of patients in our analysis were noted to have bracing as a first or second therapy. Although our analysis captured physical therapy specifically, some of the guidelines do not address physical therapy as a modality and only refer to exercise (Table 3).
Other associations have provided additional opinions regarding the evaluated therapies in the current analysis. For example, the AMSSM issued a scientific statement on viscosupplementation injections for knee OA in 2016.18 The group conducted a network meta-analysis and showed that patients who received intra-articular hyaluronic acid were 15% more likely to respond to treatment than patients who received intra-articular corticosteroids, and 11% more likely to respond to therapy compared with patients who receive intra-articular placebo (P <.05 for each). That consensus statement concluded that there is high-quality evidence supporting the use of viscosupplementation injections for Kellgren-Lawrence grade 2 or 3 knee OA in patients aged >60 years, and there is moderate quality evidence supporting the use for knee OA in patients aged <60 years.18 The panel also cautions about the consequences of guidelines that recommend against the use of hyaluronic acid, because individual patients may benefit from treatment with intra-articular hyaluronic acid, as was demonstrated in their meta-analysis.18
Clinical practice reflected the same utilization patterns.25 A survey conducted among physicians in the American Association of Hip and Knee Surgeons that assessed compliance with the AAOS guidelines for treatment of knee OA showed that of 345 responders, in a series of 5 clinical vignettes posed to the physicians, the most frequently selected intervention for patients with OA of the knee in stages 2 and 3 was hyaluronic acid, which aligns with the AMSSM recommendation.25
Because many of the therapies used to manage knee OA have been available for clinical use for decades, new clinical evidence is not likely to be forthcoming in this patient population. Current guidelines also indicate that there is often not enough strong clinical evidence to form an opinion or to make a recommendation for several of the therapies.7,10,13
Payer organizations often use clinical guidelines to inform coverage policies, and inconsistent guideline recommendations may lead to reduced patient access to appropriate therapy.18 This presents a conundrum for providers and for patients, because some therapies may be appropriate on an individual patient basis.
This study has several limitations. Patients were identified based on the identification of a diagnosis code for knee OA. As with any retrospective analyses, the possibility for miscoding or misclassification exists.
Pain measures and functional status are not captured consistently in the EHR data and were missing for many patients. Therefore, our study did not focus on the impact of therapy on pain relief, and functional improvement could not be assessed.
Information regarding the severity of OA was also not available as part of the EHR data and was therefore not captured in the study, nor was information pertaining to relevant history (eg, trauma, occupational hazards, propensity for repetitive use). The impact of disease severity and comorbidities on treatment decisions may be a topic to be explored in data sources that can elucidate the relationships.
This Geisinger Health System database contains information on prescribed medications, but it does not capture the use of nonprescription or over-the-counter medications, or the dosage and duration of long-term prescription medications. Therefore, the current analysis focused on patterns of prescribed therapies, but was not able to determine whether the patient received or continued to use the therapies.
For the same reason, the analysis was able to identify when the second treatment was prescribed, but was unable to determine whether it was used as a monotherapy or as an add-on therapy.
Finally, all patients in this analysis were members of the Geisinger Health System, with operations primarily in Pennsylvania and New Jersey and, therefore, the study findings may not be generalizable to the broader US population.
The results of this analysis shed light on the need for recommended treatment algorithms based on clinical experience and real-world data for patients with knee OA. The guidelines for the treatment of knee OA vary, potentially leading to variable real-world practice patterns. The results of this descriptive analysis suggest that providers are making use of a range of available therapeutic options.
Our results can be used to help inform and guide future research. In addition, there are several areas that could be optimized to improve patient care, including ensuring that total knee arthroplasty is performed at the proper time in the appropriate patient, minimizing the use of opioids and carefully monitoring patient response to manage pain and improve outcomes, and choosing therapies based on patient factors. The preservation of a broad range of therapeutic options will ensure providers’ ability to individualize treatment plans.
Funding for this study was provided by Ferring Pharmaceuticals, the manufacturer of the intra-articular hyaluronic acid used in this study.
Author Disclosure Statement
Dr Dysart is on the Speakers Bureau of Ferring Pharmaceuticals and a consultant to Pacira Biosciences; Dr Nelson, Dr Utkina, and Dr Stong were employees of Ferring Pharmaceuticals during the study; Dr Niazi is an employee of Ferring Pharmaceuticals; and Dr Sacks and Ms Healey are employees of Precision HEOR, which received financial support for this study from Ferring Pharmaceuticals.
- Zhang Y, Jordan JM. Epidemiology of osteoarthritis. Clin Geriatr Med. 2010;26:355-369. Erratum in: Clin Geriatr Med. 2013;29:ix.
- Murphy L, Helmick CG. The impact of osteoarthritis in the United States: a population-health perspective: a population-based review of the fourth most common cause of hospitalization in U.S. adults. Orthop Nurs. 2012;31:85-91.
- Wallace IJ, Worthington S, Felson DT, et al. Knee osteoarthritis has doubled in prevalence since the mid-20th century. Proc Natl Acad Sci USA. 2017;114:9332-9336.
- Mora JC, Przkora R, Cruz-Almeida Y. Knee osteoarthritis: pathophysiology and current treatment modalities. J Pain Res. 2018;11:2189-2196.
- Hunter DJ, Bierma-Zeinstra S. Osteoarthritis. Lancet. 2019;393:1745-1759.
- Newberry SJ, FitzGerald J, SooHoo NF, et al. Treatment of Osteoarthritis of the Knee: An Update Review. Comparative Effectiveness Reviews, No 190. Agency for Healthcare Research and Quality; May 2017. www.ncbi.nlm.nih.gov/books/NBK447543/pdf/Bookshelf_NBK447543.pdf. Accessed May 28, 2021.
- Kolasinski SL, Neogi T, Hochberg MC, et al. 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. Arthritis Rheumatol. 2020;72:220-233.
- WCG FDAnews. FDA releases black-box template for NSAIDs. June 27, 2005. www.fdanews.com/articles/73674-fda-releases-black-box-template-for-nsaids. Accessed March 20, 2020.
- Martin CL, Browne JA. Intra-articular corticosteroid injections for symptomatic knee osteoarthritis: what the orthopaedic provider needs to know. J Am Acad Orthop Surg. 2019;27:e758-e766.
- American Academy of Orthopaedic Surgeons. Treatment of Osteoarthritis of the Knee—2nd Edition: Evidence-Based Clinical Practice Guideline. May 18, 2013. www.aaos.org/globalassets/quality-and-practice-resources/osteoarthritis-of-the-knee/osteoarthritis-of-the-knee-2nd-editiion-clinical-practice-guideline.pdf. Accessed June 2, 2021.
- Chao J, Wu C, Sun B, et al. Inflammatory characteristics on ultrasound predict poorer longterm response to intraarticular corticosteroid injections in knee osteoarthritis. J Rheumatol. 2010;37:650-655.
- Raynauld JP, Buckland-Wright C, Ward R, et al. Safety and efficacy of long-term intraarticular steroid injections in osteoarthritis of the knee: a randomized, double-blind, placebo-controlled trial. Arthritis Rheum. 2003;48:370-377. Erratum in: Arthritis Rheum. 2003;48:3300.
- Bannuru RR, Osani MC, Vaysbrot EE, et al. OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis Cartilage. 2019;27:1578-1589.
- Dowell D, Haegerich TM, Chou R. CDC Guideline for Prescribing Opioids for Chronic Pain—United States, 2016. MMWR Recomm Rep. 2016;65:1-49. MMWR Recomm Rep. 2016;65:295.
- Delanois RE, Mistry JB, Gwam CU, et al. Current epidemiology of revision total knee arthroplasty in the United States. J Arthroplasty. 2017;32:2663-2668.
- Robertsson O, Dunbar M, Pehrsson T, et al. Patient satisfaction after knee arthroplasty: a report on 27,372 knees operated on between 1981 and 1995 in Sweden. Acta Orthop Scand. 2000;71:262-267.
- Varacallo M, Chakravarty R, Denehy K, Star A. Joint perception and patient perceived satisfaction after total hip and knee arthroplasty in the American population. J Orthop. 2018;15:495-499.
- Trojian TH, Concoff AL, Joy SM, et al. AMSSM Scientific Statement Concerning Viscosupplementation Injections for Knee Osteoarthritis: Importance for Individual Patient Outcomes. Clin J Sport Med. 2016;26:1-11.
- Altman RD, Bedi A, Karlsson J, et al. Product differences in intra-articular hyaluronic acids for osteoarthritis of the knee. Am J Sports Med. 2016;44:2158-2165.
- Kirchner M, Marshall D. A double-blind randomized controlled trial comparing alternate forms of high molecular weight hyaluronan for the treatment of osteoarthritis of the knee. Osteoarthritis Cartilage. 2006;14:154-162.
- Smith MM, Ghosh P. The synthesis of hyaluronic acid by human synovial fibroblasts is influenced by the nature of the hyaluronate in the extracellular environment. Rheumatol Int. 1987;7:113-122.
- Altman RD, Manjoo A, Fierlinger A, et al. The mechanism of action for hyaluronic acid treatment in the osteoarthritic knee: a systematic review. BMC Musculoskelet Disord. 2015;16:321.
- Schiavinato A, Finesso M, Cortivo R, Abatangelo G. Comparison of the effects of intra-articular injections of hyaluronan and its chemically cross-linked derivative (hylan G-F20) in normal rabbit knee joints. Clin Exp Rheumatol. 2002;20:445-454.
- Yoshioka K, Katayama M, Nishiyama T, et al. Biocompatibility study of different hyaluronan products for intra-articular treatment of knee osteoarthritis. BMC Musculoskelet Disord. 2019;20:424.
- Carlson VR, Ong AC, Orozco FR, et al. Compliance with the AAOS guidelines for treatment of osteoarthritis of the knee: a survey of the American Association of Hip and Knee Surgeons. J Am Acad Orthop Surg. 2018;26:103-107.