Etiology

Two types of OA of the knee are recognized: the more prevalent primary OA of the knee is the result of the progressive destruction of joint cartilage over time, whereas secondary OA of the knee can be caused by trauma, inactivity, overweight, or a disease process such as rheumatoid arthritis. No evidence suggests that the two types are treated differently or respond differently to treatments.⁷ Therefore, the remainder of this report treats them as one entity.

Diagnosis

The clinical diagnosis of OA of the knee is typically based on presentation, including insidious onset of weight-bearing knee pain that is exacerbated by use of the joint and relieved by rest, and that tends to worsen over the course of the day. Radiographic evidence of OA may precede symptomatic OA but may not correlate with symptom severity. Radiologic severity can be estimated and expressed using the Kellgren and Lawrence (K-L) criteria. However, a number of versions of the criteria exist. In addition, at less severe grades, correlation with symptoms is poor,⁸ whereas at more sever grades, agreement tends to be higher. The primary impact of the different versions of the criteria may be the challenge that they create in trying to assess, compare, and pool the findings of research studies.⁸ Some longitudinal studies have even used different criteria at different time points within the same study. Because of the variation in scores for radiographic finding under various versions of the criteria (especially for individuals with less-advanced disease), stratification of findings by some other objective functional baseline criteria is important.

Some evidence suggests that among individuals with knee pain, MRI demonstrates physical signs of osteoarthritic changes in the knee before they are visible radiographically.⁹ However, the sensitivity and specificity of MRI in diagnosis and monitoring of progression have not yet been definitively demonstrated and are not yet used in clinical practice.

Treatment Strategies

The goals of treatment for OA of the knee include relief of pain and inflammation, reduction of stiffness and improvement or preservation of range of motion, and improvement in or maintenance of mobility, function (including ADLs), and health-related quality of life (HRQoL).

Treatment options for OA of the knee include those in the following list. Information on the U. S. Food and Drug Administration status, indications, and warnings for the treatments included in this review (indicated by boldface and italics in this list) is included in Appendix G.

• Analgesics (oral, intra-articular, or topical) and anti-inflammatory agents (nonsteroidal anti-inflammatory agents [NSAIDs], intraarticular corticosteroids);
• Dietary supplements (including glucosamine with or without chondroitin and herbal mixtures), variously proposed to control pain and possibly serve as disease-modifying agents;
• Ayurvedic preparations, Traditional Chinese Medicine preparations, and acupuncture, all aimed at analgesia;
• Physical treatments (including strength or aerobic exercise, physical therapy, stretching, heat, aqua-therapy, whole-body vibration, electrical stimulation therapies (neuromuscular electrical stimulation [NMES] and transcutaneous electrical nerve stimulation [TENS]), massage, and chiropractic manipulation), proposed to strengthen muscles that support the affected joints and to increase range of motion;
• Education in pain coping strategies, self-management, and activity modification;
• Orthoses (knee braces, shoe inserts, custom shoes, and canes, intended to slow progression by shifting the weight from the affected joint area or other adaptive equipment to improve patients’ environments;
• Weight loss to decrease the stress on the joint;
• Intraarticular viscosupplementation, which involves local injections of the natural joint lubricant, hyaluronic acid (HA),
• Biologic agents (antinerve growth factor antibodies or antitumor necrosis factor antibodies, which are used to treat rheumatoid arthritis, and may have some benefit for OA)
• Injections of platelet-rich plasma (PRP), plasma products, stem cells, and cartilage tissue, also aimed at reversing or slowing the progression of the disease.¹⁰
• Surgical procedures, including arthroscopy with lavage and/or debridement, and partial or total arthroplasty (knee replacement), which may be recommended for advanced cases if patients fail to obtain satisfactory relief from pain and improved function from the aforementioned treatments.

Numerous recent evidence-based treatment guidelines have been issued, including the 2012 American College of Rheumatology Guidelines¹⁰ and the 2013 American Academy of Orthopedic Surgeons guidelines for the treatment of OA of the knee. These guidelines are not in total agreement about the recommended treatments: For example the 2012 American College of Rheumatology (ACR) Guidelines conditionally recommend hyaluronic acid (HA), while the American Academy of Orthopaedic Surgeons (AAOS) guidelines recommend against its use to treat patients with symptomatic conditions.¹¹

Scope of the Review

Systematic reviews have been conducted on many of the interventions used to treat OA of the knee, including four reviews by Agency for Healthcare Research and Quality Evidence-based Practice Centers since 2007.^{7, 12–14} Uncertainty continues to surround the use of all treatments intended as disease-modifying agents (including HA and glucosamine and chondroitin), acupuncture, physical therapy, exercise, braces and orthotics, and arthroscopic lavage, as well as the comparative efficacy and safety of oral, topical, and intraarticular analgesics and anti-inflammatories.

This review aimed to update a systematic review that was conducted in 2007.⁷ That review assessed the efficacy and safety of HA, glucosamine and/or chondroitin, and arthroscopic surgery (the title of the original review, “Treatment of Primary and Secondary OA of the Knee: an Update Review,” was changed to “Treatment of OA of the Knee”). Prior to preparing this review, we conducted an updating surveillance assessment that comprised an environmental scan and consultation with a Technical Expert Panel (TEP) to assess the currency of the conclusions of the 2007 review.¹⁵ A document that summarized the findings of this bifurcated process was posted for public review.¹⁶ The treatment interventions selected for inclusion in this review reflect a combination of the findings of the environmental scan, the TEP for the Surveillance process, the public comments, and the TEP for the current review.

The TEP for the surveillance process uniformly advised us that the conclusions of the 2007 report for intraarticular HA, oral glucosamine chondroitin, and arthroscopic surgery remained current and did not need updating. Instead, they recommended reviewing cell-based therapies, physical interventions, SNRIs (serotonin–norepinephrine reuptake inhibitor), topical agents, weight loss, and acupuncture.

The environmental scan supported the TEP’s suggestion that the topics of intra-articular HA and arthroscopic surgery did not need updating. However, we identified several large recent trials on glucosamine-chondroitin that prompted us to want to update the review on this topic. In addition, we elected to review the topics of cell-based therapies and weight loss and to update a 2012 systematic review on physical interventions.¹³ Topics recommended by the TEP but not included in this report will be re-assessed for the need to update (or to conduct a new review) in a later surveillance period.

The included topics (interventions) are listed in the Population, Intervention, Comparison, Outcome, Timing, Study Design, and Setting (PICOTs) outline (Table 1).

Table 1

PICOTs for the review.

Key Questions

Based on the findings of the environmental scan, TEP assessments, and public comments, the Key Questions from the 2007 report were revised as follows.

Key Question 1a.

What is the clinical effectiveness of cell-based therapies, oral glucosamine and/or chondroitin, physical treatment interventions, weight loss, or home-based and self-management therapies in patients with OA of the knee, compared with appropriate placebo/sham controls or compared with other active interventions?

Key Question 1b.

How do the outcomes of each intervention differ by the following population and study characteristics: sex, disease subtype (lateral, patellofemoral), severity (stage/baseline pain and functional status), weight status (body mass index), baseline fitness (activity level), comorbidities, prior or concurrent treatments (including self-initiated therapies), and treatment duration or intensity?

Key Question 2a.

What harms are associated with each intervention in patients with OA of the knee?

Key Question 2b.

How do the harms associated with each intervention differ by the following population or study characteristics: sex, disease subtype (lateral tibiofemoral, patellofemoral), severity (stage/baseline pain and functional status), weight status (body mass index), baseline fitness (activity level), comorbidities, prior or concurrent treatments (including self-initiated therapies), and treatment duration or intensity?

Analytic Framework

The review was guided by the analytic framework shown in Figure 1.

Figure 1

Analytic framework for osteoarthritis of the knee. Figure notes: BMI = Body Mass Index; KQ = Key Question; OA = Osteoarthritis; OARSI = Osteoarthritis Research Society International; OMERACT = Outcome Measures in Rheumatology; TKR = Total Knee Replacement. (more...)