Overview of the management of osteoarthritis

INTRODUCTION — Joint pain and functional impairment are the hallmarks of osteoarthritis (OA). OA-related pain results from a multifactorial biopsychosocial process in which non-cartilaginous structures, including the subchondral bone, synovium, and periarticular structures, are involved and are influenced by environmental and psychosocial factors. Peripheral and central sensitization of nociceptive pathways may ultimately drive the perpetuation of pain and play a role in the chronic aspects of the disease. OA-related pain has negative impacts on mood and sleep and frequently affects participation in occupational and recreational activities [1].

There are multiple components of the management of OA. They range from the approach to common OA-related conditions such as depression, sleeping disturbances, and social problems to joint-specific interventions including nonpharmacologic, pharmacologic, and surgical options. Ultimately, most are aimed at improving the pain and functional restriction that characterizes this prevalent disease. Despite numerous efforts, treatments to modify the course of the disease have not reached a threshold of efficacy to gain regulatory approval.

An overview of OA management as well as prognosis will be discussed here. Separate topic reviews on the pathogenesis, risk factors, clinical manifestations, diagnosis, and joint-specific treatment recommendations for OA include the following:

●(See "Pathogenesis of osteoarthritis".)

●(See "Epidemiology and risk factors for osteoarthritis".)

●(See "Clinical manifestations and diagnosis of osteoarthritis".)

●(See "Management of hand osteoarthritis".)

●(See "Management of knee osteoarthritis".)

●(See "Management of moderate to severe knee osteoarthritis".)

●(See "Management of hip osteoarthritis".)

GENERAL PRINCIPLES — The principles of chronic disease management apply to the care of patients with osteoarthritis (OA) and are based on the following: care should be continuous, tailored to patients according to individual needs, goals, and values, and be patient-centered; decision-making should be based on the best evidence available, with prioritization of the safety of the patient; information should be widely accessible to patients; and anticipation of needs should be prioritized over a reactive health service design [2].

Patient's adherence, optimal uptake of recommendations, and behavior modifications are key elements of OA treatment and can be optimized by OA and self-management education, establishing treatment goals, and periodic monitoring.

Education — Patients should be fully informed about the etiology of OA, risk factors (especially the ones that are modifiable and specific to the patient), and expected prognosis [3]. Clear information about the treatment options along with their benefits, harms, and costs should also be discussed. Providing this information helps to counter common misconceptions and direct the focus of the treatment to the patient, encouraging an active behavior in the management of their own diseases [4].

Patient education is an essential tool to optimize OA management [5]. A substantial part of the noncompliance with treatment, particularly when it comes to lifestyle changes, may occur due to the limited time that clinicians take to explain the purpose of the interventions and what the patient should expect in terms of pain relief.

Self-management — Self-management education is a complement to traditional patient education [6]. It aims at teaching patients problem-solving skills and involves the concept of self-efficacy. This in turn gives patients confidence in their capacity to carry out a particular behavior necessary to reach a desired goal. Self-management education can occur through several ways such as face-to-face meetings, group sessions, the Internet, and telephone-based sessions [7]. A Cochrane review comparing self-management education programs and standard care found small benefits in terms of self-management skills and OA symptoms [8].

Goal-setting — Setting goals helps the informed patient identify current issues, set priorities, and focus on specific changes. To develop a realistic plan, goals should be agreed upon with patients, considering their preferences and their biopsychosocial context. In addition, appropriate goals should be specific, timely, and measurable and should be reviewed periodically. Long-term goals should be broken up into short-term achievable steps (eg, start walking for 10 minutes on three days of the week to ultimately be able to walk for 30 minutes on three days of the week in three months' time) and achievements should be positively reinforced.

MONITORING AND ASSESSMENT — The management of patients with osteoarthritis (OA) should include a holistic assessment which considers the global needs of the patient. Monitoring of the patient's response to therapy should be done on a regular basis. While a variety of clinical tools have been developed for clinical symptom assessment and monitoring, the evaluation of treatment response is based primarily on the clinical evaluation.

Periodic monitoring — Periodic clinical assessments should be performed regularly (ideally every three months) to assess the effects of treatment on symptoms, functionality, and status, as well as quantify objective changes in metrics related to interventions such as weight and muscle strength. Assessing the patient periodically enables regular coaching and the reinforcement of the action plan. This also allows for monitoring of treatment effectiveness, side effects, and alterations to the management plan according to outcome.

Holistic assessment — A holistic assessment of the patient with OA facilitates a patient-professional partnership and collaborative care in which patients and health care professionals make decisions related to treatment together, building upon each other to improve outcomes [6]. OA is a multifaceted disease in which the structural evidence of joint damage frequently does not correlate with the presence and severity of joint pain and disability [9]. Moreover, patients with OA are predominantly older adults who frequently have different personal priorities and aspirations, which impacts the treatment choice.

The illness perception and the consequences of the disease vary substantially from person to person and are influenced by a number of factors that, if not identified, will hamper the development of an appropriate tailored plan and limit the success of treatment. Important factors that should be addressed by the clinician include:

●Previous knowledge of OA and its treatment options

●Impact of pain and functional impairment on activities of daily living and quality of life

●Participation restriction

●Recreational and occupational aspirations

●Mood disturbances and level of distress [10]

●Sleep disturbances and fatigue

●Falls risk assessment

●Presence of social support

●Presence of comorbidities [10-13]

●Expectations of treatment

●Modifiable risk factors, especially overweight/obesity, joint alignment, and injury/buckling

Patient preferences for certain types of therapies should also be assessed, as compliance and outcomes can be compromised if the care plan does not meet the patient's preferences and beliefs, modified or not by OA education. After addressing these factors, a personalized package of care can be developed and agreed upon with the patient.

Other assessment measures — Evaluation of treatment response in OA is based on clinical assessment, as there is no established role for laboratory or imaging tests in assessing disease activity/status in clinical practice. As a result, they are not required for OA diagnosis or disease monitoring. (See "Clinical manifestations and diagnosis of osteoarthritis".)

There are a variety of clinical tools aimed at evaluating the clinical status and patient-reported outcomes in OA that are mainly used in clinical research [14]. However, some clinicians use these tools in clinical practice. The approach in our interdisciplinary OA clinic is to use the Knee Injury and Osteoarthritis Outcome Score (KOOS) and the Hip Disability and Osteoarthritis Outcome Score (HOOS) to assess patient-reported measures in patients with knee and hip OA, respectively, and the six-minute walk test and timed up-and-go test to assess physical function based on patient's performance. We use the data from these instruments in conjunction with other clinical parameters, such as the patient's perceived improvement and overall satisfaction with their condition, to guide treatment decisions.

A few of the most commonly used instruments are described in more detail below:

●Knee and hip osteoarthritis – The Western Ontario and McMaster Universities (WOMAC) questionnaire has been widely used to assess pain, stiffness, and physical function in patients with hip and/or knee OA. It consists of 24 items rated on a 5-point Likert scale and is easy to complete [15]. The KOOS and the HOOS are also broadly studied and used questionnaires to assess patients with knee and hip OA, respectively [16,17]. They are composed of five domains including pain, other symptoms, difficult-to-perform activities of daily living, sports and recreational activities, and knee-related quality of life. These instruments are relatively easy and quick to complete.

The timed up-and-go and the 40-meter walk tests are good measures of physical function based on patient's performance and have adequate psychometric properties [18]. Other tests include 30-second chair test, stair-climb test, and the six-minute walk test.

●Hand osteoarthritis – The most commonly used tool to assess pain in patients with hand OA is the Australian/Canadian Hand OA Index (AUSCAN) pain subscale, in addition to the visual analog scale (VAS) [19]. The AUSCAN Index includes three subscales to assess pain, disability, and joint stiffness in hand OA, using 15 questions. It is brief and easy to score and has good psychometric properties [20]; however, it is not freely available. Function status is most frequently assessed using the AUSCAN function subscale, the Functional Index for Hand Osteoarthritis (FIHOA), and the Michigan Hand Outcomes Questionnaire [21,22], in addition to physical tests measuring grip and pinch strength [19].

OVERVIEW OF MANAGEMENT — The goals of osteoarthritis (OA) management are to minimize pain, optimize function, and beneficially modify the process of joint damage. The primary aim of clinicians should include targeting modifiable risk factors [23]. Although there are no approved disease-modifying OA drugs, a wide selection of interventions are available to address pain and function [7].

Due to the modest effects of the individual treatment options, a combination of therapeutic approaches is commonly used in practice and should prioritize therapies that are safer before considering drugs that can potentially cause harm (eg, opioids and oral nonsteroidal antiinflammatory drugs [NSAIDs]).

In addition to benefits, side effect profile, and patient-specific impairments and preferences, costs of interventions and local availability should also be considered. As an example, intraarticular hyaluronic acid injections are associated with high costs without clinically significant benefits over intraarticular placebo [7]. A discussion of the efficacy of hyaluronans is presented separately. (See "Management of knee osteoarthritis", section on 'Hyaluronans'.)

Management should also be individualized and target modifiable factors contributing to pain, particularly presence of joint malalignment, muscle weakness, overweight and obesity, and concurrent depression. The number of joints involved, presence of articular versus periarticular pain, and the degree of movement restriction and functional impairment should also guide the therapeutic plan.

Our management strategy is generally consistent with guidelines developed by professional organizations, including the American College of Rheumatology and Arthritis Foundation [24,25], the European Alliance of Associations for Rheumatology (EULAR; formerly known as European League Against Rheumatism) [26-29], the Osteoarthritis Research Society International [30], and the National Institute for Health and Care Excellence [5]. However, it is important to note that the overall quality of OA care has been found to be suboptimal in many countries, with less than 50 percent compliance with indicators of appropriate care, which is inferior to other chronic conditions such as diabetes and osteoporosis [31]. The particular domains in which OA care was poor were pharmacologic and nonpharmacologic treatments, with less than 40 percent of appropriateness [32]. (See 'Barriers to implementation' below.)

Nonpharmacologic therapy — Nonpharmacologic interventions are the mainstay of OA management and should be tried first, followed by or in concert with medications to relieve pain when necessary.

Nonpharmacologic therapies include weight management and exercises, braces and foot orthoses for patients suitable to these interventions, education, and use of assistive devices when required [33].

●Exercises have effects of similar magnitude on pain and function compared with NSAIDs [34]. A combination of aerobic and strengthening exercises is usually indicated to address the whole spectrum of disability associated with OA, but optimal prescription should be individualized [35,36]. (See "Management of knee osteoarthritis", section on 'Exercise'.)

●Loss of at least 10 percent of body weight through a combination of diet and exercises has been associated with a 50 percent reduction in pain scores in patients who are overweight or have obesity with knee OA after 18 months [37]. Weight loss can also be recommended for patients with hip OA, although there are no randomized trials assessing its impact on hip OA symptoms [30] and may be beneficial for patients with OA in the hands, in view of its increased risk in patients with higher body mass index (BMI) [38]. (See "Management of knee osteoarthritis", section on 'Weight loss'.)

●Walking aids and knee braces for patients with malalignment (tibiofemoral or patellofemoral OA) may improve pain and should be considered as adjunctive treatments [7]. In addition, splints are particularly recommended for the treatment of OA at the base of the thumb [39]. (See "Management of moderate to severe knee osteoarthritis", section on 'Knee braces' and "Management of moderate to severe knee osteoarthritis", section on 'Walking aids'.)

Pharmacologic therapy — We use pharmacologic agents for patients with symptomatic OA who have not responded adequately to initial nonpharmacologic measures or concomitantly with these interventions for those with more severe symptoms. Pharmacologic therapy should only be used during periods when symptoms are present, since none of the interventions have been shown to be disease-modifying. The main medications used in the pharmacologic management of OA include oral and topical NSAIDs, with topical capsaicin, duloxetine, and intraarticular glucocorticoids being other options depending on the clinical context, as discussed below [7]. The choice of pharmacologic agent used is influenced by the specific joint and number of joints involved, as well as the presence of certain comorbidities.

The following briefly describes our general approach to pharmacotherapy:

●In patients with one or a few joints affected, especially knee and/or hand OA, we initiate pharmacotherapy with topical NSAIDs due to their similar efficacy compared with oral NSAIDs and their better safety profile [40]. (See "Management of knee osteoarthritis", section on 'Topical NSAIDs'.)

●We use oral NSAIDs in patients with inadequate symptom relief from topical NSAIDs, symptomatic OA in multiple joints, and/or patients with hip OA. We use the lowest dose required to control the patient's symptoms on an as-needed basis. The use of NSAIDs in most patients is limited by the increased risk of serious gastrointestinal, cardiovascular, and renal complications. In patients with comorbidities such as diabetes, hypertension, and advanced age, a cyclooxygenase (COX)-2 selective NSAID or a nonselective NSAID associated with a proton-pump inhibitor should be used, though we prefer not to use these medications in patients with a high comorbidity risk (eg, previous gastrointestinal bleeding or chronic renal failure). (See "Comorbidities that impact management of osteoarthritis" and "NSAIDs (including aspirin): Primary prevention of gastroduodenal toxicity" and "NSAIDs (including aspirin): Pathogenesis and risk factors for gastroduodenal toxicity" and "NSAIDs: Acute kidney injury" and "NSAIDs: Adverse cardiovascular effects".)

●We use duloxetine for patients with OA in multiple joints and concomitant comorbidities that may contraindicate oral NSAIDs and for patients with knee OA who have not responded satisfactorily to other interventions. (See "Management of moderate to severe knee osteoarthritis", section on 'Duloxetine'.)

●Topical capsaicin is a treatment option when one or a few joints are involved and other interventions are ineffective or contraindicated; however, its use may be limited by common local side effects. (See "Management of knee osteoarthritis", section on 'Topical capsaicin'.)

●We do not routinely use intraarticular glucocorticoid injection due to the short duration of its effects (ie, approximately four weeks) and evidence that it may have deleterious effects on the hyaline cartilage [41] and may accelerate OA progression [42,43]. (See "Management of moderate to severe knee osteoarthritis", section on 'Limited role of intraarticular glucocorticoid injections'.)

Due to safety concerns pertaining to the use of acetaminophen (paracetamol) and increased awareness of its negligible and non-clinically significant effects on pain [44,45], this medication is no longer considered the first-line analgesic for the treatment of knee and hip OA by clinical guidelines and is no longer being initiated in our practice [5,30]. (See "Management of knee osteoarthritis", section on 'Acetaminophen'.)

The benefit of intraarticular hyaluronic acid (HA) is also controversial for knee and hip OA, and most evidence demonstrates only a small superiority over intraarticular placebo [46,47]. (See "Management of knee osteoarthritis", section on 'Hyaluronans'.)

We avoid prescribing opioids due to their overall small effects on pain over placebo and potential side effects (eg, nausea, dizziness, drowsiness), especially for long-term use and in the older adult population [7,39]. We consider opioids only for short-term use in patients with severe and disabling symptoms in whom other interventions have failed or are not appropriate. (See "Management of knee osteoarthritis", section on 'Opioids'.)

We do not routinely recommend nutritional supplements such as glucosamine, chondroitin, vitamin D, diacerein, avocado soybean unsaponifiables (ASU), and fish oil due to lack of clear evidence demonstrating a clinically important benefit from these supplements. Other supplements of interest that may have some effects on symptoms include curcumin (active ingredient of turmeric) and/or Boswellia serrata, but the data are limited [48-50]. A more detailed discussion on supplements is elsewhere. (See "Management of knee osteoarthritis", section on 'Nutritional supplements'.)

Surgery — Surgical treatment is dominated by total joint replacement, which is highly effective in patients with advanced knee and hip OA when conservative therapies have failed to provide adequate pain relief [51]. However, a systematic review of 14 studies demonstrated that about 20 and 9 percent of patients undergoing total knee and hip replacement, respectively, reported moderate to severe long-term pain postoperatively [52]. Preoperative levels of pain, presence of comorbidities and depression, and presence of concomitant pain at other joints are associated with an increased risk of an unfavorable pain outcome after surgery. This is potentially related to the involvement of alternative pain mechanisms contributing to the pain experience such as psychological factors and central sensitization [53,54], although there are limited data suggesting that knee replacement may be beneficial to reduce parameters related to central pain mechanisms [55]. In addition, patients may experience severe complications and require revision surgery, especially due to the increasingly younger age of patients in whom surgery has been performed [56].

Details on total knee and hip arthroplasty and alternative surgical approaches are discussed elsewhere. (See "Overview of surgical therapy of knee and hip osteoarthritis" and "Total knee arthroplasty" and "Total hip arthroplasty".)

Other alternative therapies — Other therapies that have been tried in the treatment of OA include acupuncture, traditional Chinese medicine, and transcutaneous nerve stimulation (TENS). (See "Acupuncture", section on 'Knee osteoarthritis' and "Management of knee osteoarthritis", section on 'Transcutaneous electrical nerve stimulation' and "Management of knee osteoarthritis", section on 'Nutritional supplements'.)

FACTORS AFFECTING RESPONSE TO THERAPY — The discordance between radiographic findings and pain supports the notion that the mechanisms of pain are complex and likely multifactorial. The placebo effect is also known to impact response to therapy.

Mechanisms of pain — The understanding of the mechanisms that lead to chronic pain in osteoarthritis (OA) has evolved, but several aspects remain unclear. These include the contribution of sensitization of nociceptive pathways and other external factors (ie, psychosocial) in different stages of OA, the role of neuropathic pain, and the influence of disease phenotypes in the patterns of pain [1].

Instead of a single structure or pathway responsible for pain onset or associated with its persistence, it is likely that the experience of pain has a multifactorial etiology [57]. In this regard, multiple joint structures (eg, periarticular muscles and ligaments, bone, and synovium) and altered modulation of stimulus transmission act in concert with psychosocial and environmental influences, as described below [1].

●Extraarticular factors – Extraarticular factors may influence the experience of pain and are usually unique to each patient, with significant between-person differences. They play an important role in the heterogeneity of clinical presentations among patients with similar disease severity [1]. These include:

•Demographic features (eg, ethnicity, sex)

•Genetics

•Personality

•Catastrophizing and coping skills

•Expectations of treatment

•Previous pain experiences

•Presence of comorbidities (eg, depression)

●Joint-related factors – There is evidence pointing to the importance of mediators released by the synovium and the subchondral bone in the articular pain. Changes in the size and severity of bone marrow lesions, joint effusion, and synovitis over time have been associated with fluctuations in pain, in addition to a higher prevalence of pain in knees in which these features were present [58]. Moreover, severe cartilage damage on magnetic resonance imaging (MRI) is also related to symptoms which probably occur through secondary mechanisms, since cartilage is aneural and unable to directly generate pain [59].

In addition, periarticular sources of pain such as pes anserine bursitis at the knee and trochanteric bursitis at the hip often coexist with OA and contribute to the symptoms [60].

●Sensitization of nociceptive pathways – A number of peripheral substances have been associated with OA-related pain and include inflammatory mediators (eg, prostaglandin E2, interleukin [IL]-1beta, IL-6, tumor necrosis factor [TNF]-alpha), nerve growth factor (NGF), and sodium and calcium ions channels [9]. These substances are involved in the activation of primary afferent nerves in response to inflammation or injury to the tissue and have been studied as potential targets for treatment of OA pain (see "Investigational approaches to the management of osteoarthritis", section on 'Inflammatory pathways'). When persistent, they lead to plastic changes in the cells that become more easily activated so that innocuous movements or loading of the joint evoke exaggerated responses (allodynia). Over time, persistent pain may occur even during resting or lying. This is the neurophysiological basis of the peripheral sensitization which occurs at the level of the joint.

Sensitization of the central nociceptive processing can also occur and involves changes at the spinal cord and cortical levels. In practice, central sensitization is also associated with a more diffuse and referred pattern of pain and with emotional consequences to pain through the activation of regions such as amygdala, thalamus, and cingulate cortex [9].

●Dissociation between symptoms versus radiographic disease – There is great discordance between joint symptoms and extent of radiographic disease. In patients with frequent hip pain, only 15.6 percent showed evidence of radiographic OA (Kellgren-Lawrence grade [KLG] ≥2), and 20.7 percent of hips with radiographic OA were symptomatic [61]. With regard to knee OA, a systematic review has shown that 15 to 81 percent of patients with radiographic OA had knee pain and 15 to 76 percent of those with knee pain had radiographic OA, with significant differences in pain definition, radiographic views, and demographic factors among studies [62].

On the other hand, studies correlating pain and OA severity have found a higher prevalence of knee pain in patients with KLG 4 compared with KLG 2 or 3 and in patients with greater rates of cartilage loss in the medial tibiofemoral compartment [63,64]. These findings indicate that despite the poor correlation between radiographic OA and joint symptoms, presence of pain may reflect activity of the disease not completely depicted by radiographs. (See "Clinical manifestations and diagnosis of osteoarthritis", section on 'Imaging'.)

Role of placebo effect — The placebo effect has a substantial implication in the treatment of patients with OA, often overcoming the actual effect sizes of individual treatments [65]. The following studies are illustrative:

●According to a meta-analysis of trials including a placebo group, the overall effect size estimate of placebo for pain (defined as change from baseline to endpoint) was 0.51 (95% CI 0.46 to 0.55) for all trials, though there was significant variation among distinct types of placebo [66]. Interventions delivered by invasive routes (ie, intraarticular injections and acupuncture) were associated with more robust placebo effects compared with oral or topical placebos. Moreover, the effect of several oral nonsteroidal antiinflammatory drugs (NSAIDs; eg, celecoxib, naproxen, ibuprofen, and diclofenac) was not significantly superior to intraarticular placebo, assessed by indirect comparison [47]. In addition to alleviating pain, improvements in other common clinical OA outcomes have also been observed with placebo such as stiffness and joint function [66].

●The landmark GAIT trial (Glucosamine/Chondroitin Arthritis Intervention) compared the effects of glucosamine, chondroitin sulfate, the combination of them, celecoxib, and oral placebo and has demonstrated that at least 20 percent improvement in pain assessed by the Western Ontario and McMaster Universities (WOMAC) questionnaire (primary outcome) was achieved by 60 percent of participants in the placebo group, while effects of active treatment groups ranged from 64 to 70 percent [67].

A number of determinants are able to evoke a placebo response. Among them, there are factors related to the intervention such as the route of delivery [68], frequency of administration, color and cost, how new the intervention is, and others related to the health professional-patient relationship (context effect). The latter involves the clinician's optimism and expectations regarding the treatment as well as the clinician's behavior towards providing the patient with a confident diagnosis and prognosis of the disease [65].

With this in mind and in line with the prerequisite of "do no harm," it is likely that patients with OA pain would benefit from clinicians who are able to optimize and use the placebo effect in clinical practice in the favor of their patients.

Barriers to implementation — There are a number of evidence-based guidelines providing recommendations to guide clinical management of OA, especially for hip and knee [2]. Yet implementation of these recommendations into practice has been suboptimal, and gaps and delays exist between best practice suggested by guidelines and the usual care for patients with OA [69]. Available data have revealed a suboptimal quality of OA care, with achievement of quality measures usually under 50 percent [70,71]. This rate is even lower than those for other common chronic diseases such as hypertension, depression, and osteoporosis [72].

Despite evidence of benefits of the nonpharmacologic interventions in OA, particularly weight loss and exercise, improving their implementation in practice remains a challenge [73]. Moreover, there is suboptimal referral to allied health professionals; inadequate uptake of safety recommendations for the prescription of OA medications, such as NSAIDs; inappropriate indication of non-beneficial surgical interventions such as arthroscopic debridement and lavage; and excessive use of imaging, such as MRI, which frequently does not beneficially change the decision-making.

The complex nature of OA may also pose a significant challenge in translating evidence into routine clinical practice. The extensive range of risk factors driving the disease, the heterogeneous clinical and radiological presentation of patients with OA, and the distinct clinical courses that can occur may make extrapolation of research findings to different populations problematic. Moreover, clinical trials frequently exclude participants with multiple comorbidities or with severe structural disease, which limits the generalization of the results for these groups of patients. Currently, there is no tool that allows clinicians to identify which patient to treat with which therapy for OA management and those who should be referred to specialist care.

PROGNOSIS — Although there is great variability among individuals and among different phenotypes of osteoarthritis (OA), courses of pain and physical functioning have been found to be predominantly stable, without substantial improvement or deterioration of symptoms over time [74-76].

Biomarkers (especially biochemical and imaging markers) have been studied to identify patients that are more likely to experience a progressive course of OA and to identify those that will most likely benefit of a given intervention [77]. However, none of them have been qualified for clinical use as yet [78].

Knee and hip osteoarthritis — In a cohort study of patients with radiographic knee OA (Kellgren-Lawrence grade [KLG] ≥2), patients experiencing distinct intensities of pain at baseline, ranging from absent to severe pain, tended to remain in the same group over the six-year follow-up, with no significant changes in baseline pain scores [75]. Moreover, radiographic progression has also been shown to occur at a slow rate [79]. These suggest that OA is predominantly characterized by minimum/slow rather than marked worsening over time. However, distinct rates of symptomatic and structural progression have been observed, and several risk factors have been associated with more rapid progression [75,80,81].

●Prognostic factors – Clinical factors that have been associated with pain and functional deterioration in patients with knee OA include [82]:

•Higher pain intensity at baseline

•Presence of depressive symptoms

•Presence of bilateral knee symptoms

Additional predictors of functional decline are:

•Worsening in radiographic OA

•Worsening of knee pain

•Presence of pain on patellofemoral joint compression

•Lower knee extension strength

•Higher comorbidity count

•Poor general health

•Psychosocial factors (lower vitality and poor mental health)

Body mass index (BMI) >25 has also been associated with trajectories with higher levels of pain in patients with knee and hip OA [83], and knee malalignment (varus or valgus) has been associated with structural progression of knee OA [84].

●Total joint replacement – Estimates of the burden of total joint replacement indicate that approximately one-third to -half of the patients with knee OA will undergo arthroplasty during their lives [56,85]. Annual incidence rates of total knee replacement in individuals from two longitudinal OA cohorts ranged from 1.6 to 11.9 percent in men and 2.0 to 10.9 percent in women [56]. Predictors of total knee replacement have been investigated and include worse pain scores [86] and radiographic severity, high BMI [85,87], limited knee range of motion [88,89], patient's willingness to undergo surgery [86], and structural features such as severity of MRI-detected medial compartment cartilage volume/thickness, bone marrow lesions, synovitis, and meniscal lesions [90].

Hand osteoarthritis — The concept of slow deterioration also exists in hand OA. However, the majority of patients experience radiographic progression after 10 years [91]. Symptomatic decline (pain and disability) occurs in around 50 percent of patients over three to eight years [92] and is more frequent in patients with higher levels of pain and number of joints affected at baseline (risk ratio 2.11, 95% CI 1.25 to 3.08, in patients with >8 joints affected compared with <4 joints) [93]. The erosive subset of hand OA has been positively associated with radiographic progression (relative risk [RR] 1.55, 95% CI 1.04-1.88) but not with symptomatic deterioration [93,94].

Mortality — An excess mortality due to all causes has been observed in patients with knee and hip OA [95-98]. Moreover, increased cause-specific mortality has been demonstrated particularly due to cardiovascular diseases (standardized mortality ratio 1.71, 95% CI 1.41 to 1.70) [95]. Patients with walking disability, impaired physical function, and associated comorbid conditions such as diabetes and other cardiovascular risk factors are at higher risk of mortality [95,96].

On the other hand, no relationship has been found between women with hand OA and mortality risk in a large prospective community-based cohort [99]. Yet data from the Framingham Heart Study have shown an increased risk of coronary heart disease events (myocardial infarction/coronary insufficiency syndrome) in patients with symptomatic hand OA (hazard ratio [HR] 2.26, 95% CI 1.22 to 4.18) after adjustment for use of nonsteroidal antiinflammatory drugs (NSAIDs) and several risk factors for cardiovascular disease [100].

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (See "Patient education: Osteoarthritis (The Basics)".)

●Beyond the Basics topics (See "Patient education: Osteoarthritis symptoms and diagnosis (Beyond the Basics)" and "Patient education: Osteoarthritis treatment (Beyond the Basics)".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Osteoarthritis".)

SUMMARY AND RECOMMENDATIONS

●General principles – General principles of osteoarthritis (OA) management include providing continuous care that is tailored to the patient according to individual needs, goals, and values and should be patient-centered. Treatment can be optimized by OA and self-management education, establishing treatment goals, and periodic monitoring. (See 'General principles' above.)

●Monitoring and assessment – The management of OA should include a holistic assessment which considers the global needs of the patient. Patient preferences for certain types of therapies should also be assessed, as compliance and outcomes can be compromised if the care plan does not meet the patient's preferences and beliefs. (See 'Monitoring and assessment' above.)

●Overview of management – The goals of OA management are to minimize pain, optimize function, and beneficially modify the process of joint damage. The primary aim of clinicians should include targeting modifiable risk factors. Due to the modest effects of the individual treatment options, a combination of therapeutic approaches is commonly used in practice and should prioritize therapies that are safer. (See 'Overview of management' above.)

●Nonpharmacologic therapy – Nonpharmacologic interventions are the mainstay of OA management and should be tried first, followed by or in concert with medications to relieve pain when necessary. Nonpharmacologic therapies including weight management and exercises, braces and foot orthoses for patients suitable to these interventions, education, and use of assistive devices when required. (See 'Nonpharmacologic therapy' above.)

●Pharmacologic therapy – The main medications used in the pharmacologic management of OA include oral and topical nonsteroidal antiinflammatory drugs (NSAIDs). Other options include topical capsaicin, duloxetine, and intraarticular glucocorticoids. Our general approach to pharmacotherapy is described below. More specific recommendations are included in separate topic reviews. (See 'Pharmacologic therapy' above.)

•In patients with one or a few joints affected, especially knee and/or hand OA, we initiate pharmacotherapy with topical NSAIDs due to their similar efficacy compared with oral NSAIDs and their better safety profile. (See "Management of knee osteoarthritis", section on 'Topical NSAIDs'.)

•We use oral NSAIDs in patients with inadequate symptom relief with topical NSAIDs, patients with symptomatic OA in multiple joints, and/or patients with hip OA. We use the lowest dose required to control the patient's symptoms on an as-needed basis. (See "Management of moderate to severe knee osteoarthritis", section on 'Oral NSAIDs'.)

•We use duloxetine for patients with OA in multiple joints and concomitant comorbidities that may contraindicate oral NSAIDs and for patients with knee OA who have not responded satisfactorily to other interventions. (See "Management of moderate to severe knee osteoarthritis", section on 'Duloxetine'.)

•Topical capsaicin is an option when one or a few joints are involved and other interventions are ineffective or contraindicated; however, its use may be limited by common local side effects. (See "Management of knee osteoarthritis", section on 'Topical capsaicin'.)

•We do not routinely use intraarticular glucocorticoid injections due to the short duration of its effects (ie, approximately four weeks). (See "Management of moderate to severe knee osteoarthritis", section on 'Limited role of intraarticular glucocorticoid injections'.)

•We avoid prescribing opioids due to their overall small effects on pain over placebo and potential side effects (eg, nausea, dizziness, drowsiness), especially for long-term use and in the older adult population. (See "Management of knee osteoarthritis", section on 'Opioids'.)

•We do not routinely recommend nutritional supplements such as glucosamine, chondroitin, vitamin D, diacerein, avocado soybean unsaponifiables (ASU), and fish oil due to a lack of clear evidence demonstrating a clinically important benefit from these supplements. Other nutritional supplements of interest that may have small effects on symptoms include curcumin (active ingredient of turmeric) and/or Boswellia serrata, but the data are limited. (See "Management of knee osteoarthritis", section on 'Nutritional supplements'.)

●Role of surgery – Surgical treatment is dominated by total joint replacement, which is highly effective in patients with advanced knee and hip OA when conservative therapies have failed to provide adequate pain relief. (See 'Surgery' above.)

●Factors affecting response to therapy – The discordance of radiographic findings to pain supports the notion that the mechanisms of pain are complex and likely multifactorial. The placebo effect is also known to impact response to therapy. (See 'Factors affecting response to therapy' above and 'Mechanisms of pain' above and 'Role of placebo effect' above and 'Barriers to implementation' above.)

●Prognosis – Although there is great variability among individuals and among different phenotypes of OA, courses of pain and physical functioning have been found to be predominantly stable, without substantial improvement or deterioration of symptoms over time. (See 'Prognosis' above.)