Osteoarthritis is classified as primary or secondary. The cause of primary osteoarthritis is idiopathic; no abnormality is the cause of changes in the joint [9]. Secondary osteoarthritis is the result of a known cause, most often trauma/injury or systemic diseases. Secondary osteoarthritis is most often found in the shoulder, elbow, and ankle and is more likely to become clinically apparent at a younger age than primary osteoarthritis [9,30,31,32]. A population-based study showed that secondary osteoarthritis related to trauma accounts for approximately 12% of the overall prevalence of symptomatic osteoarthritis of the knee, hip, or ankle [33]. Injuries sustained in sports activities comprise a large portion of post-traumatic osteoarthritis [34]. A wide variety of systemic diseases have been identified as frequent causes of secondary osteoarthritis; these conditions include metabolic diseases, endocrine disorders, bone dysplasias, and crystal deposition diseases (Table 1) [9,35].

Research has shown that the symptoms of osteoarthritis do not correlate well with its radiographic evidence [19,39,40,41]. According to a systematic literature review, radiographic evidence of osteoarthritis is found in 15% to 76% of individuals with pain, and 15% to 81% of individuals with radiographic evidence of disease have pain [39]. An estimated 40% of individuals with structural changes on radiographs are asymptomatic [39,40]. In addition, many individuals have joint-related symptoms and no radiographic evidence [5,9]. As a result of this discordance, the disease is defined as either radiographic (evidence on imaging studies) or symptomatic (frequent pain in a joint plus radiographic evidence of osteoarthritis in that joint) [42]. Total joint replacement is used as a surrogate measure of symptomatic end-stage osteoarthritis, as the procedure is the option chosen when nonoperative measures have failed to manage pain and improve function and mobility.

Data on the age at the time of diagnosis of osteoarthritis at other joints are limited. However, studies have indicated a younger age at the time of clinical presentation of elbow osteoarthritis (approximately 50 years) and ankle osteoarthritis (43 to 58 years) [32,58].

Studies have consistently shown that osteoarthritis of the knee is more prevalent in the Black population than the White population. Multivariable analysis of data from NHANES III showed significantly higher odds of radiographic knee osteoarthritis (Kellgren-Lawrence grade 2 or higher) among non-Hispanic Black participants (52%) compared with White (36%) or Mexican American (38%) participants [62,65]. Although the findings of the Johnston County Osteoarthritis Project also demonstrated that knee-related symptoms, radiographic knee osteoarthritis (mild), and symptomatic knee osteoarthritis were all more prevalent among Black individuals than White individuals, the difference was slight. However, the prevalence of moderate-to-severe radiographic osteoarthritis was significantly greater for both men and women in the Black population (11% vs. 5% for Black vs. White men and 16% vs. 8% for Black vs. White women) [45]. A study of more than 1,000 premenopausal and perimenopausal women demonstrated that early osteoarthritis changes were more prevalent in Black women than White women (23% vs. 9%) [66]. The prevalence of knee osteoarthritis has also been found to be higher in the Chinese population than in the White population [67].

Normal adult articular cartilage is made up of extracellular matrix (approximately 98% to 99%) and chondrocytes (1% to 2%) [75]. The chondrocytes secrete enzymes and cytokines that help regulate the normal cycle of degradation and repair of articular cartilage by inhibiting the production of proteoglycans and collagen, the two major components of the extracellular matrix [75]. Damage to the extracellular matrix interferes with its ability to bind or exclude water, resulting in edema and subsequent softening of the cartilage and expansion of the matrix, which makes the matrix vulnerable to further injury and breakdown of its components [9,76,77,78].

There is substantial heterogeneity in osteoarthritis across anatomic sites with regard to risk factors, clinical features, and outcomes, which has drawn some researchers to conclude that osteoarthritis of different joints are distinct clinical entities [84,85]. Some examples to support the concept of distinct disease entities include [31,32,36,86]:

Studies have indicated that there may be a genetic factor to the development of osteoarthritis, and the familial risk factor for osteoarthritis of the knee, hip, and hand has ranged from 27% to 60% [35,57,84]. It is thought that most genes related to osteoarthritis affect the development of the disease at any joint but that specific genes may also be involved at specific joints [35,84]. Over the past several years, a candidate gene study and several genome-wide association studies have collectively established 15 loci associated with knee or hip osteoarthritis that have been replicated with genome-wide significance, providing further evidence of joint-specific effects in osteoarthritis [19,84,85,90,91,92,93,94]. In 2019, researchers performed a genome-wide association study with more than 77,000 participants and identified 64 loci, 52 of them being novel. Of these 64 loci, therapeutics are currently available or in clinical trials for 10 of the effector genes, making them a future prospect for effective treatment of osteoporosis [95]. Despite the increased reports of potential risk loci for osteoarthritis, some research indicates that epigenetic changes may have a role in the pathogenesis of osteoarthritis [96].

Clinical studies have long demonstrated that the risk of osteoarthritis is higher for individuals who are overweight or obese, and obesity has been referred to as the most important modifiable risk factor for severe osteoarthritis of the knee and, to a lesser extent, of the hips [9,97,98,99]. In a meta-analysis, those who were obese or overweight were nearly three times as likely to report osteoarthritis of the knee [100]. Overweight as a risk factor is thought to be related to the increased load on weight-bearing joints; however, some studies have indicated an association between obesity and osteoarthritis of the hand and shoulder, which suggests factors other than joint overload [30,36,57]. Factors that have been proposed are a metabolic intermediary (such as diabetes or lipid abnormalities) or an increased production of humoral factors (produced by excess adipose tissue), which alters the metabolism of articular cartilage [9,101].

The data on osteoarthritis and overweight have been more consistent for osteoarthritis of the knee than for disease at other joint sites, and most studies have indicated that overweight/obesity is a greater risk factor for women [38,84,87,97,101,102,103,104,105]. In the Framingham Osteoarthritis Study, there was more than a 50% decrease in the risk among women who had a loss of approximately 11 pounds or a decrease in body mass index (BMI) of 2 or more [97]. Weight gain was also associated with an increased risk for osteoarthritis, but the difference was not significant [97]. In a population-based case-control study in England (525 men and women [45 years of age and older] with primary knee osteoarthritis and 525 matched controls), the risk of osteoarthritis increased progressively with higher BMI; compared with a BMI of 24.0–24.9, the risk was 0.1 for a BMI of less than 20 and 13.6 for a BMI of 36 or greater [99].

The differential diagnosis of osteoarthritis varies according to the anatomic site as well as such patient-related factors as age, gender, and history (Table 5) [30,32,38,40,170,171,172]. In general, the differential diagnosis includes infection, traumatic injuries, bursitis, other types of arthritis, and overuse syndromes [40]. In addition, clinicians should consider secondary osteoarthritis in patients who have metabolic bone disorders, endocrine diseases, and other systemic conditions, as described earlier [40]. Ancillary testing should be done for patients who have joint pain at night, who have progressive joint pain, or who have a strong family history of inflammatory arthritis [79]. Many features on clinical evaluation and imaging studies are characteristic of osteoarthritis, and some features differ according to joint site (Table 6) [30,31,38,58,171].

The differential diagnosis of osteoarthritis varies according to the anatomic site as well as such patient-related factors as age, gender, and history (Table 5) [30,32,38,40,170,171,172]. In general, the differential diagnosis includes infection, traumatic injuries, bursitis, other types of arthritis, and overuse syndromes [40]. In addition, clinicians should consider secondary osteoarthritis in patients who have metabolic bone disorders, endocrine diseases, and other systemic conditions, as described earlier [40]. Ancillary testing should be done for patients who have joint pain at night, who have progressive joint pain, or who have a strong family history of inflammatory arthritis [79]. Many features on clinical evaluation and imaging studies are characteristic of osteoarthritis, and some features differ according to joint site (Table 6) [30,31,38,58,171].

When considering patients' self-reports of pain and function, clinicians should understand that these self-reports can differ according to gender and race/ethnicity [48,181,182]. Self-reports of work or activity limitations or severe pain have been significantly more common among Black, Hispanic, and mixed-race individuals than among White individuals with osteoarthritis; the rate of self-reports for Asian/Pacific Islander and Alaska Native/American Indian populations have been similar to those for the White population [48]. Among participants in the Johnston County Osteoarthritis Project, total scores on the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and scores on the pain and function subscales were significantly worse for Black individuals than for White individuals with knee osteoarthritis. The total WOMAC scores were similar for the two racial groups among individuals who had only hip osteoarthritis or hip and knee osteoarthritis [182]. The researchers hypothesized that high BMI and frequent depressive symptoms in the Black population may have contributed to the racial/ethnic differences.

According to the EULAR guidelines on the diagnosis of knee osteoarthritis, a diagnosis can be made with 99% confidence when three symptoms and three signs are present [173]:

The strongest sign of hip osteoarthritis on physical examination is pain that is exacerbated by internal or external rotation of the hip with the knee in full extension [38,177]. Other signs include crepitus and gait abnormalities (resulting from alterations in walking to avoid pain) [186]. Deformity and instability are late signs of severe osteoarthritis, but they are uncommon [186]. Both hips should be examined if osteoarthritis is suspected, as the disease occurs bilaterally in approximately 20% of individuals [38].

Osteoarthritis of the hand is characterized by pain with use, which affects one or a few joints at any one time, and mild stiffness in the morning and/or after a period of inactivity [178]. The severity of osteoarthritis-related pain varies, and the pain may be intermittent. The joints most often affected are the distal and proximal interphalangeal joints and the base of the thumb [176,177,178]. Individuals who have evidence of osteoarthritis at several joints in the hand are at increased risk for generalized osteoarthritis, and clinicians should evaluate such patients as appropriate [178].

Osteoarthritis of the hand may be associated with substantial limitations in function, and the clinician should ask the patient whether he or she has difficulty with such tasks as dressing, eating, writing, handling or fingering small objects, and carrying or lifting 10 pounds [44,56]. Several validated questionnaires are available to assess function of the hand, and the choice of questionnaire depends primarily on the clinical question [171]. Individuals with symptomatic osteoarthritis of the hand also may have reduced maximal grip strength [44,56].

Individuals with osteoarthritis of the elbow typically have pain, stiffness, and weakness in the joint [31]. Later stage disease is associated with pain when carrying a heavy object at the side of the body with the elbow in extension. The history is important when evaluating symptoms related to the elbow because of the strong relationship between trauma or occupation with osteoarthritis, especially in individuals who are younger than 40 years of age [58]. Primary osteoarthritis of the elbow is often associated with osteoarthritis at another joint site, especially the second and third metacarpophalangeal joints, the knee, and the hip, and those joints should be evaluated as appropriate [190].

A history of ankle fracture or ligamentous injury is a hallmark feature of osteoarthritis of the ankle [32]. Diagnostic evaluation includes radiographs of the ankles made with the patient standing. MRI is also recommended, as it can provide evidence of osteonecrosis as well as indicate the amount of involvement, the extent of bone loss, and the size of subchondral cysts [32].

Given the strong correlation between overweight/obesity (defined as a BMI greater than 25) and osteoarthritis of the knee and hip, weight reduction and maintenance of a healthy weight are central to guidelines on the management of osteoarthritis at these sites [185,198,206,223,224]. A systematic review showed that a moderate weight-loss program (0.25% of body weight per week) can reduce pain and physical disability for individuals with osteoarthritis of the knee [225]. In its 2021 guideline for the treatment of osteoarthritis of the knee, the AAOS recommends weight reduction, specifically, achieving and/or maintaining a BMI ≤25 [198].

Some patients may fear that regular exercise will exacerbate pain, but a review of the literature has shown that moderate exercise does not increase the risk for progression of osteoarthritis, provided that care is taken to avoid injury [110,227]. The goal of an exercise program is to control pain, increase flexibility, and improve muscle strength and endurance [228]. The exercise program should be individualized to the patient, with consideration given to the patient's age, comorbidities, and mobility [229]. Guidelines suggest that exercise should be prescribed for all patients with osteoarthritis, regardless of age, severity of pain and disability, and comorbidity [199]. The American Geriatric Society notes that absolute contraindications to an exercise program include uncontrolled arrhythmias, third-degree heart block, changes on recent electrocardiography, unstable angina, acute myocardial infarction, and acute congestive heart failure [228]. Relative contraindications include cardiomyopathy, valvular heart disease, poorly controlled blood pressure, and uncontrolled metabolic disease [228].

The AAOS found inconclusive evidence for physical therapy as an effective treatment option for osteoarthritis of the glenohumeral joint and is unable to recommend for or against physical therapy as part of initial treatment of the condition [197]. Similarly, a supervised physical therapy program is not routinely a treatment approach for osteoarthritis of the ankle [32]. Physical therapy should begin in the early stages of osteoarthritis of the elbow (mild pain and loss of less than 15 degrees of motion) [31]. Strategies may include gentle range-of-motion exercises to maintain mobility and strength [31].

Some guidelines recommended acetaminophen as the initial analgesic for the management of mild-to-moderate pain related to osteoarthritis, but this recommendation has since been shown to be questionable [193]. A comparative effectiveness study conducted by the AHRQ found good evidence that acetaminophen is modestly inferior in efficacy compared with NSAIDs but has a lower risk of gastrointestinal complications [244]. An update to this study found that no currently available analgesic offered a clear overall advantage compared with the others [200]. Its original findings on acetaminophen remained the same, with the addition that acetaminophen poses a higher risk of liver injury [200]. Other research has shown that NSAIDs are more effective than acetaminophen for relieving osteoarthritis-related pain, especially moderate-to-severe pain [245]. The 2021 AAOS guideline provides a strong recommendation for oral acetaminophen to improve pain and function in the treatment of knee osteoarthritis when not contraindicated [198]. The working group noted that when oral acetaminophen was compared to NSAIDs, the use of oral NSAIDs provided a significant reduction in pain and improved function. As a result, providers may consider using oral NSAIDs instead of acetaminophen when a contraindication to oral NSAIDs does not exist [198]. NSAIDs should be prescribed at the lowest effective dose, and their long-term use should be avoided [193]. A COX-2 selective agent or an NSAID with a prescription for a gastroprotective agent (such as a proton-pump inhibitor) may be used for patients who have an increased risk for gastrointestinal complications [193].

There is good evidence that nonselective NSAIDs and COX-2-selective NSAIDs have comparable efficacy and that COX-2-selective agents are comparable to each other [200,246]. Although COX-2-selective agents have better tolerability in general compared with NSAIDs, there is considerable variability across individual drugs in terms of protection against serious gastrointestinal events [246]. In addition, some COX-2 selective NSAIDs have been associated with an increased risk of myocardial infarction, and these drugs should be used with caution in patients with cardiovascular risk factors [200,246].

Studies have found that opioids were more effective overall than control interventions with respect to pain relief and improved function, but the beneficial effects were small to moderate and were outweighed by a substantial increase in the risk of adverse events [247,248]. The authors of the review concluded that opioids should not be used routinely for individuals with osteoarthritis, even for severe pain. Some guidelines suggest the use of weak narcotics or opioids for pain that has been refractory to other pharmacologic agents; however, the guidelines note that strong opioids should be used sparingly [193]. The 2021 AAOS guideline on the treatment of knee osteoarthritis emphasizes the importance of removal of oral narcotics from the medications prescribed due to the rise of the opioid epidemic in the United States [198]. The ACR guidelines conditionally recommend against using opioid analgesics for osteoarthritis of the hand [185].

In reviewing the literature for its guidelines on the treatment of osteoarthritis of the glenohumeral joint, the AAOS was not able to find sufficient evidence to support several pharmacologic treatments, including acetaminophen, NSAIDs, opioids, or narcotics. As a result, the AAOS states it is unable to recommend for or against the use of any of these options for the initial treatment of patients with osteoarthritis of this joint [197].

Moderate-quality evidence indicates that compared with placebo, tramadol alone or in combination with acetaminophen probably has no important benefit on mean pain or function in people with osteoarthritis, although slightly more people in the tramadol group report an important improvement (defined as 20% or more). Moderate-quality evidence shows that adverse events probably cause substantially more participants to stop taking tramadol. The increase in serious adverse events with tramadol is less certain, due to the small number of events [249].

In reviewing the literature for its guidelines on the treatment of osteoarthritis of the glenohumeral joint, the AAOS was not able to find sufficient evidence to support several pharmacologic treatments, including acetaminophen, NSAIDs, opioids, or narcotics. As a result, the AAOS states it is unable to recommend for or against the use of any of these options for the initial treatment of patients with osteoarthritis of this joint [197].

Certain guidelines conditionally recommend intra-articular injection of corticosteroids into the knee or hip, especially after aspiration of fluid in patients who have signs of local inflammation with joint effusion [9,185,206]. For example, the ACR recommends this therapy for knee and hip osteoarthritis if the patient does not have satisfactory response to acetaminophen and topical NSAIDs and if there is a contraindication to oral NSAIDs. The AAOS provides a moderate recommendation for the use of intra-articular corticosteroid therapies. Extended release agents are recommended over immediate release to improve patient outcomes [198].]. Although the approach is otherwise widely recommended, it is acknowledged that intra-articular corticosteroids provide short-term relief only [35,253,254]. A meta-analysis of 28 trials (1,973 patients) of knee osteoarthritis showed a benefit of pain relief for two to four weeks, with no benefit in terms of functional improvement and no benefit in either pain or function beyond four weeks [253]. An update to the meta-analysis, which included 27 trials (1,767 patients), found that the overall quality of the evidence did not clearly support a benefit of intra-articular corticosteroid use after one to six weeks [254]. Despite the short-term benefit found in most studies, clinical experience has shown longer relief in many patients [35]. Because of the potential side effects of intra-articular injections, which include long-term damage to joint cartilage, flare after injection, and infection, most physicians do not recommend more than three to four injections per joint per year [9,35]. Intra-articular injection is more technically difficult in the hip joint than in the knee, and radiographic or ultrasonographic guidance has been suggested, although there are no comparative data to provide evidence that accuracy is increased with such guidance [9,223].

Certain guidelines conditionally recommend intra-articular injection of corticosteroids into the knee or hip, especially after aspiration of fluid in patients who have signs of local inflammation with joint effusion [9,185,206]. For example, the ACR recommends this therapy for knee and hip osteoarthritis if the patient does not have satisfactory response to acetaminophen and topical NSAIDs and if there is a contraindication to oral NSAIDs. The AAOS provides a moderate recommendation for the use of intra-articular corticosteroid therapies. Extended release agents are recommended over immediate release to improve patient outcomes [198].]. Although the approach is otherwise widely recommended, it is acknowledged that intra-articular corticosteroids provide short-term relief only [35,253,254]. A meta-analysis of 28 trials (1,973 patients) of knee osteoarthritis showed a benefit of pain relief for two to four weeks, with no benefit in terms of functional improvement and no benefit in either pain or function beyond four weeks [253]. An update to the meta-analysis, which included 27 trials (1,767 patients), found that the overall quality of the evidence did not clearly support a benefit of intra-articular corticosteroid use after one to six weeks [254]. Despite the short-term benefit found in most studies, clinical experience has shown longer relief in many patients [35]. Because of the potential side effects of intra-articular injections, which include long-term damage to joint cartilage, flare after injection, and infection, most physicians do not recommend more than three to four injections per joint per year [9,35]. Intra-articular injection is more technically difficult in the hip joint than in the knee, and radiographic or ultrasonographic guidance has been suggested, although there are no comparative data to provide evidence that accuracy is increased with such guidance [9,223].

It is difficult to determine the efficacy of hyaluronan because research evidence is confounded by different molecular weights of hyaluronan preparations, different dosing schedules, and poor trial design, and the level of evidence across studies has been low [223,255,256,259]. Most of the evidence available is related to osteoarthritis of the knee, with limited data available on use of the treatment for osteoarthritis of the hip, hand, or shoulder.

Since the publication of the 2000 ACR guidelines, certain studies and analyses have supported the efficacy of hyaluronan/hylan derivatives for relieving pain and improving function in patients with symptomatic osteoarthritis of the knee (compared with placebo), with the greatest benefit found in conjunction with less severe pain and disability at 5 to 13 weeks after injection [206,256,260,261]. However, researchers have noted that the effect size is small compared with placebo and that the effect may be overestimated as a result of publication bias [255,256,262]. When compared with NSAIDs, hyaluronan takes longer to relieve knee symptoms; additionally, the dosing schedule necessitates more office visits than intra-articular corticosteroids, creating inconvenience and increasing costs [206,223]. Uncontrolled and small studies of hyaluronic acid for hip osteoarthritis have shown pain reduction after treatment, but intra-articular corticosteroids were more effective in one small study [223,259,263].

In its 2019 recommendations, the ACR conditionally recommends against using intra-articular therapies for hand osteoarthritis [185]. This recommendation is based largely on the absence of evidence from randomized controlled trials to support the benefits as well as the potential for harm from such therapy [185].

In a systematic review undertaken to evaluate the effectiveness of 22 herbal medicinal products, there was some evidence of pain relief with topical capsaicin, avocado-soybean unsaponifiables, and SKI306X (a Chinese herbal mixture). However, none of the 22 products had proof of effectiveness beyond doubt [277]. According to a review of studies involving antioxidant and anti-inflammatory supplements, the following cannot be recommended for the treatment of osteoarthritis: vitamin E (alone); a combination of vitamins A, C, and E; ginger; turmeric; omega-3 fatty acids; or Zyflamend (an extract of 10 different herbs) [278]. Additional clinical trials are needed before alternative supplements can be recommended.

Because there are anatomic differences in joint structure and size between men and women, a gender-specific knee prosthesis was designed specifically for women [287]. Researchers believed that the better fit would lead to improvements in recovery and outcomes for women who had total knee arthroplasty. In one study, 85 women who received a standard joint in one knee and the gender-specific joint in the other knee were followed up for two years after the surgery [287]. Patient satisfaction, range of motion while lying, and WOMAC scores were similar for both prostheses. The researchers did note that the standard prostheses appeared to fit at the distal part of the femur better than the gender-specific type; furthermore, the small size of the gender-specific prosthesis exposed more bone and resulted in more bleeding immediately after surgery. Although the study concluded that there were no benefits to the use of gender-specific prostheses in women undergoing total knee arthroplasty, research evaluating long-term effects is necessary.

Postoperative rehabilitation is a necessary component of recovery after operative treatment of osteoarthritis and requires cooperation of the entire multidisciplinary team. In the case of total knee arthroplasty, patients should be guided on a postoperative exercise and rehabilitation plan that focuses on obtaining an acceptable level of joint function, range of motion, and quality of life (e.g., ability to perform activities of daily living unassisted). In some cases, a continuous passive motion device may be used. This device has been suggested as a means to obtain greater range of motion more quickly after surgery [288]. While this may be the case, no long-term benefits (e.g., ultimate range of motion) have been definitively proven, and evidence on the short-term effects are conflicting [286,288,289]. It is not a recommendation of the AAOS or the ACR at this time.

In general, the institution of a structured exercise plan, guided by the physician and physical therapist, will assist patients in regaining range of motion and return to performing daily activities. A daily physical therapy program after total knee arthroplasty should continue for four to six weeks, at which point the patient's needs will be reassessed. According to one study, the greatest improvements in lower-extremity functional status after total knee arthroplasty were demonstrated in the first 12 weeks, with little improvement noted after 26 weeks [290]. By the end of physical therapy, the patient should be able to perform activities of daily living and progress to ambulating on flat surfaces and stairs. Strengthening and stretching exercises focusing on the hamstrings and quadriceps should be incorporated into the program.

There is some debate regarding the importance of supervised outpatient physical therapy compared with exercise programs carried out in the patient's home. One meta-analysis of 10 randomized controlled trials found that supervised physical therapy provided no benefits for patients who were younger at the time of surgery and had few or no comorbidities [291]. However, the researchers noted that there is a lack of evidence regarding the use of outpatient physical therapy for older patients with comorbidities and those who have undergone more complicated surgeries.