Osteoarthritis Treatment & Management

Updated: Nov 30, 2022
  • Author: Carlos J Lozada, MD; Chief Editor: Herbert S Diamond, MD  more...
  • Print

Approach Considerations

The goals of osteoarthritis treatment include alleviation of pain and improvement of functional status. Optimally, patients should receive a combination of nonpharmacologic and pharmacologic treatment. [72]

Nonpharmacologic interventions, which are the cornerstones of osteoarthritis therapy, include the following:

  • Patient education
  • Heat and cold
  • Weight loss [10]
  • Exercise
  • Physical therapy
  • Occupational therapy
  • Unloading in certain joints (eg, knee, hip)

A physiatrist may help in formulating a nonpharmacologic management plan for the patient with osteoarthritis, and a nutritionist may help the patient to lose weight. A referral to an orthopedic surgeon may be necessary if the osteoarthritis fails to respond to a medical management plan. Surgical procedures for osteoarthritis include arthroscopy, osteotomy, and (particularly with knee or hip osteoarthritis) arthroplasty.

Several organizations have issued guidelines on the treatment of osteoarthrits (see Guidelines), with recommendations keyed to the affected joints (ie, hand, knee, hip). An assessment of several treatments for osteoarthritis of the knee by the Agency for Healthcare Research and Quality (AHRQ) determined the following [73] :

  • Home-based exercise programs and tai chi show short- to medium-term benefits for symptoms (primarily pain, function, and quality of life) but data on long-term benefits are lacking
  • Strength and resistance training, pulsed electromagnetic field therapy, and transcutaneous electrical nerve stimulation show mostly short-term benefits, whereas agility training shows both short- and long-term benefits.
  • Weight loss and general exercise programs show medium- and long-term benefits.
  • Intra-articular platelet-rich plasma, balneotherapy, and whole-body vibration show medium-term benefits.
  • Glucosamine-chondroitin and glucosamine or chondroitin sulfate alone show medium-term benefits with no long-term benefits for pain or function.

Mesenchymal stem cell therapy continues to be a promising investigational approach to knee osteoarthritis. [74, 75]  However, the improvement reported with stem cell therapy has been modest, a placebo effect remains possible, and the quality of the supporting evidence has been questioned. [76, 77]  In addition, the variability in mesenchymal stem cell injection, including timing, frequency, and culturing mode,  warrant further research, as does the possible long-term risk. 


Pharmacologic Treatment

Analgesics and anti-inflammatory drugs

Begin treatment with acetaminophen for mild or moderate osteoarthritic pain without apparent inflammation. If the clinical response to acetaminophen is not satisfactory or if the clinical presentation of osteoarthritis is inflammatory, consider using a nonsteroidal anti-inflammatory drug (NSAID). Use the lowest effective dose or intermittent dosing if symptoms are intermittent, then try full doses if the patient’s response is insufficient.

However, a study in patients with knee osteoarthritis found no evidence that long-term NSAID use reduces inflammation or slows the progression of osteoarthritis. On magnetic resonance imaging (MRI) scans taken at baseline and after four years in 277 patients taking NSAIDs and 793 controls, biomarkers of inflammation and cartilage quality were worse at baseline in the participants taking NSAIDs, compared with the control group, and worsened at four-year follow-up. [59]

Topical NSAID preparations, particularly diclofenac, are available. These preparations can be particularly useful in patients with symptomatic disease that is limited to a few sites or in patients who are at increased risk for adverse events with systemic NSAIDs.

In patients with highly resistant pain, consider the analgesic tramadol. However, in an observational study of more than 88,000 patients, an initial prescription of tramadol was associated with a significantly higher rate of mortality over 1 year of follow-up compared with commonly prescribed NSAIDs in patients aged 50 years and older. [78]  The data from this study suggest an unfavorable safety profile of tramadol and require replication. Differences between the study groups (eg, cancer) that were not adjusted away using propensity score matching may account for part or even all of the difference in mortality rates. [79]

Similarly, a population-based study in British Columbia found that within 1 year of starting treatment, death rates were higher in patients taking tramadol than in those taking naproxen, diclofenac, or a selective cyclooxygenase (COX)-2 inhibitor. However, the all-cause mortality rate was lower with tramadol than with codeine. [80]

Options in patients at an elevated risk for GI toxicity from NSAIDs include the addition of a proton-pump inhibitor or misoprostol to the treatment regimen. Clinicians may also consider prescribing the COX-2 inhibitor celecoxib instead of a nonselective NSAID.

In 2018, the US Food and Drug Administration (FDA) approved the combination of the calcium channel blocker amlodipine with celecoxib (Consensi), for patients for whom treatment with amlodipine for hypertension and celecoxib for osteoarthritis are appropriate. Lowering blood pressure reduces the risk of fatal and nonfatal cardiovascular events, primarily strokes and myocardial infarctions, which is elevated in patients taking celecoxib. [81]

Hydroxychloroquine, which is approved for use in rheumatoid arthritis, has been used off-label in patients with hand osteoarthritis, on the supposition that it might help by relieving synovitis. However, a randomized study in 210 patients found that hydroxychloroquine does not improve symptoms of hand osteoarthritis. [82]

Study patients had experienced insufficient response to, or adverse effects from, analgesics such as acetaminophen, NSAIDs, or opioids, and were randomized to receive hydroxychloroquine (usually 300 mg/day) or placebo for 12 months, in addition to their usual analgesic medications. At 6 months, there were no significant differences between hydroxychloroquine and placebo recipients in hand pain—the primary outcome—or in grip strength and structural damage.

Agency for Healthcare Research and Quality findings

A 2011 comparison of analgesics for osteoarthritis carried out by the Agency for Healthcare Research and Quality (AHRQ) found that “no currently available analgesic reviewed in this report offers a clear overall advantage compared with the others.” [83]  The choice of analgesic for an individual patient should take into account the trade-off between benefits and adverse effects, which differs across analgesics. Patient age, comorbid conditions, and concomitant medication are key considerations.

The AHRQ comparison found that acetaminophen was modestly inferior to NSAIDs in reducing osteoarthritic pain but was associated with a lower risk of GI adverse effects. [83] On the other hand, acetaminophen poses a higher risk of liver injury.

AHRQ findings on adverse effects included the following:

  • Selective NSAIDs as a class were associated with a lower risk of ulcer complications than were the nonselective NSAIDs naproxen, ibuprofen, and diclofenac

  • The partially selective NSAIDs meloxicam and etodolac were associated with a lower risk of ulcer-related complications and symptomatic ulcers than were various nonselective NSAIDs

  • The risk of serious GI adverse effects was found to be higher with naproxen than with ibuprofen

  • Celecoxib and the nonselective NSAIDs ibuprofen and diclofenac were associated with an increased risk of cardiovascular adverse effects when compared with placebo

  • The nonselective NSAIDs ibuprofen and diclofenac, but not naproxen, were associated with an increased risk of heart attack when compared with placebo

The AHRQ noted that topical diclofenac was found to have efficacy similar to that of oral NSAIDs in patients with localized osteoarthritis. No head-to-head trials compared topical salicylates or capsaicin with oral NSAIDs for osteoarthritis. [83]

All NSAIDs had deleterious effects on blood pressure, edema, and kidney function. However, the AHRQ found no consistent clinically relevant differences between celecoxib, partially selective NSAIDs, and nonselective NSAIDs with regard to the risk of hypertension, heart failure, or impaired kidney function. [83]


The selective serotonin-norepinephrine reuptake inhibitor duloxetine has been found to be effective in treating osteoarthritis pain. [84] For example, in patients with knee osteoarthritis who had persistent moderate pain despite optimized NSAID therapy, a randomized, double-blind trial found significant additional pain reduction and functional improvement with duloxetine as compared with placebo. [85]

However, duloxetine was also associated with significantly more nausea, dry mouth, constipation, fatigue, and decreased appetite than placebo was. [85] To date, trials of duloxetine in osteoarthritis have been short in duration (10-13 weeks), and studies comparing duloxetine directly with other therapies have not been performed.

Intra-articular injections

Intra-articular pharmacologic therapy includes injection of a corticosteroid or sodium hyaluronate (ie, hyaluronic acid [HA] or hyaluronan) or biologic agent (ie, platelet-rich plasma [PRP]), which may provide pain relief and have an anti-inflammatory effect on the affected joint. [86, 87] Ultrasound guidance can facilitate arthrocentesis and injection and is increasingly being adopted by physicians such as rheumatologists and physiatrists for this purpose.


After the introduction of the needle into the joint and before steroid administration, aspiration of as much synovial fluid as possible should be attempted. Aspiration often provides symptomatic relief for the patient and allows laboratory evaluation of the fluid, if necessary. Infected joint fluid and bacteremia are contraindications to steroid injection.

In patients with osteoarthritic knee pain, steroid injections generally result in clinically and statistically significant pain reduction as soon as 1 week after injection. The effect may last, on average, anywhere from 4 to 6 weeks per injection, but the benefit is unlikely to continue beyond that time frame. [88]  

However, in a randomized trial by McAlindon et al comprising 140 patients with symptomatic knee osteoarthritis with synovitis, intra-articular injections of steroid (40 mg triamcinolone, every 12 weeks for 2 years) resulted in significantly greater cartilage volume loss and no significant difference in knee pain, compared with placebo injections of saline. The authors concluded that their findings do not support the use of intra-articular steroid injections for symptomatic knee osteoarthritis. [89]

Similarly, in two studies of patients with knee osteoarthritis, comparison of imaging studies performed at the time of injection and after 2 years in one study and 3 years in the other showed that corticosteroid injections were significantly associated with overall progression of osteoarthritis in the joint. In comparison, intra-articular injection of hyaluronic acid was associated with reduced progression of knee osteoarthritis. [90]

In 2017, the FDA approved triamcinolone acetonide extended-release injectable suspension (Zilretta) for intra-articular treatment of osteoarthritic knee pain. Approval was based on data from a randomized, double-blind international phase III trial in which 484 patients were treated and followed for up to 24 weeks. Patients receiving Zilretta reported a statistically significant reduction in the weekly mean of the average daily pain intensity scores (ADP) from baseline to week 12. [91]

For hip osteoarthritis, a randomized, placebo-controlled study confirmed the effectiveness of corticosteroid injection, with benefits often lasting as long as 3 months. [92]

Some controversial evidence exists regarding frequent steroid injections and subsequent damage to cartilage (chondrodegeneration). Accordingly, it is usually recommended that no more than three injections per year be delivered to any individual osteoarthritic joint. Systemic glucocorticoids have no role in the management of osteoarthritis.

Local anesthetics are often injected along with corticosteroids, to provide immediate pain relief (which also supports the diagnosis of intra-articular pathology), dilute the steroid preparation, and moderate or eliminate the postinjection flare. However, chondrotoxicity (eg, chondrolysis) is a potential drawback. [93, 94]

A review by Kompel et al of intra-articular corticosteroid injections of the hip and knee describes four main adverse events that may occur in injected joints: accelerated osteoarthritis progression; subchondral insufficiency fracture; complications of osteonecrosis; and rapid joint destruction, including bone loss.  These authors recommend that, although high-quality evidence is lacking, certain patient characteristics, including but not limited to acute change in pain not explained by using radiography and no or only mild osteoarthritis at radiography, should lead to careful reconsideration of a planned injection. In such cases, MRI may be diagnostically helpful. [94]

Patients with subchondral insufficiency fracture typically present with acute pain in a weight-bearing joint, despite no identifiable trauma; the pain gradually worsens for weeks. On plain x-rays, the condition may be subtle or occult; MRI may provide more definitive evidence. Identification of a subchondral insufficiency fracture before intra-articular corticosteroid injection is clinically important, as the steroids may inhibit the healing process. Instead, primary treatment is conservative and includes protected weight-bearing or non–weight-bearing activities. Some authors have proposed adjunctive treatment with bisphosphonates or prostacyclin analogs, but little evidence supports these approaches. [94]

In patients whose x-ray shows no osteoarthritis or only mild osteoarthritis, the authors recommend closely scrutinizing the indication for intra-articular steroid injection, as these patients are at increased risk for developing rapid progressive joint space loss or destructive osteoarthritis. [94]

For more information, see Corticosteroid Injections of Joints and Soft Tissues.

Sodium hyaluronate

Intra-articular injection of sodium hyaluronate, also referred to as viscosupplementation, has been shown to be safe and possibly effective for symptomatic relief of knee osteoarthritis. [95, 96] In the United States, intra-articular HAs are classified as medical devices rather than as drugs. [97]

Intra-articular HAs approved by the FDA for the treatment of osteoarthritic knee pain include the naturally extracted, non–cross-linked sodium hyaluronate products Hyalgan, [98] Supartz, Orthovisc, and Euflexxa, as well as the cross-linked sodium hyaluronate product known as hylan G-F 20 (Synvisc).

Euflexxa is derived from a fermentation process (Streptococcus), whereas the source material for the other products listed is chicken combs. At present, no distinct advantage or disadvantage has been associated with any particular source of HA.

Some differences between the viscosupplements do exist in the FDA-approved prescribing information. For example, whereas Hyalgan and Synvisc have been established as safe for repeat treatment, the safety and efficacy of other products for repeat treatment have not been established.

The exact mechanisms of action through which HAs provide symptomatic relief are unknown. Possible mechanisms include direct binding to receptors (CD44 in particular) in the synovium and cartilage that can lead to several biologic activation pathways. [99, 100]

The HA class in general has demonstrated a very favorable safety profile for chronic pain management in knee osteoarthritis, with the most common adverse event being injection-site pain. Although any intra-articular injection (whether of HAs or of steroids) may elicit an inflammatory response and possible effusion, only the cross-linked hylan G-F 20 product has been associated with a clinically distinct acute inflammatory side effect (ie, severe acute inflammatory reaction [SAIR] or HA-associated intra-articular pseudosepsis).

Platelet-rich plasma

Platelet-rich plasma (PRP) is defined as a volume of plasma with a platelet concentration higher than the average in peripheral blood (150,000–350,000 platelets/μl). [101]  A meta-analysis of 10 randomized controlled trials (RCTs) with a total of 1069 patients found that intra-articular PRP injection may have more benefit in pain relief and functional improvement than hyaluronic acid or saline in patients with symptomatic knee osteoarthritis at 1 year postinjection. [102]

Because of its high platelet concentration, PRP contains hyperphysiologic levels of clotting and growth factors, including insulinlike growth factor 1 (IGF-1), transforming growth factor  beta (TGF-beta), and platelet-derived growth factor (PDGF), among others. [103] Increased cell proliferation, collagen synthesis, and vascularity has been attributed to the activity of PRP. [103] However, concerns have been raised regarding the clinical efficacy of PRP, mainly due to the heterogeneity of preparation methods and resulting products, the scarceness of high-quality RCTs, and the contradictory results that have been found so far. [104]  

The FDA has cleared PRP preparation systems as a device "substantially equivalent" to another device previously cleared. However, the clearance applies only to the device and its intended use in an operative setting and makes no claim about its effectiveness for a particular indication. Similarly, in the European Union, only the preparation procedure, and not the product itself, is regulated, and the regulation does not include any requirements about the composition or effectiveness of PRP. [101]


In prolotherapy, small volumes of an irritant solution are injected at ligament and tendon insertions and in adjacent joint spaces over several treatment sessions. In a randomized, controlled trial of 90 adults with painful knee osteoarthritis who were randomized to either dextrose prolotherapy, saline injections, or at-home exercise, the patients on prolotherapy experienced significantly greater improvement in pain, function, and stiffness over the other 2 groups. Injections were administered at 1, 5, and 9 weeks, with additional injections provided as needed at weeks 13 and 17. [105]

Additional pharmacologic agents

Muscle relaxants may benefit patients with evidence of muscle spasm. Judicious use of narcotics (eg, oxycodone and acetaminophen with codeine) is reserved for patients with severe osteoarthritis.

Glucosamine and chondroitin sulfate have been used in Europe for many years and continue to be popular with patients worldwide. In the United States, however, the glucosamine/chondroitin arthritis intervention trial (GAIT) reported, at best, limited benefit from glucosamine (500 mg 3 times daily), chondroitin sulfate (400 mg 3 times daily), or the combination of the 2 in patients with knee osteoarthritis. [106, 107]

In GAIT patients overall, glucosamine and chondroitin sulfate alone or in combination did not reduce pain effectively at 24 weeks, but in patients with moderate-to-severe pain at baseline, the rate of response was significantly higher with combined therapy than with placebo (79.2% vs. 54.3%). [107] At 2 years, no treatment achieved a clinically important difference in loss of joint-space width, though treatment effects on Kellgren-Lawrence grade 2 knees showed a trend toward improvement relative to the placebo group. [106]

The AHRQ comparison found no clear difference between glucosamine or chondroitin and oral NSAIDs for relieving pain or improving function. [83] However, the AHRQ observed that most trials showing therapeutic benefits from glucosamine used pharmaceutical-grade glucosamine that is not available in the United States, noting that the trial findings may therefore be inapplicable to currently available over-the-counter preparations.

Another agent, S-adenosylmethionine (SAM-e), is a European supplement receiving significant attention in the United States. A systematic review of SAM-e found that the evidence was inconclusive, with a number of small trials of questionable quality; the authors concluded that the effects of SAM-e on pain and function may be potentially clinically relevant but are expected to be small. [108]

Modest relief of osteoarthritic knee pain in adults has been reported with use of Curcuma longa extract (turmeric). Curcumin, the main component of the turmeric plant, has anti-inflammatory and analgesic properties. In a randomized trial in 70 adults with painful knee osteoarthritis and ultrasound-defined effusion–synovitis, relief of knee pain after 12 weeks was significantly greater with turmeric extract (two 500-mg capsules daily) than with placebo (mean reduction of 24 mm vs 15 mm on a 100-mm visual analog scale). However, change in effusion–synovitis volume on MRI did not differ significantly between the groups. [109]

Chondroprotective drugs (ie, matrix metalloproteinase [MMP] inhibitors and growth factors) are being tested as disease-modifying drugs in the management of osteoarthritis. For example, MMP-13 is specifically expressed in the cartilage of individuals with osteoarthritis but not in the cartilage of normal adults. [110] German researchers reported on the synthesis and biologic evaluation of an MMP-13 selective inhibitor that has demonstrated efficacy as a disease-modifying intra-articular injection for osteoarthritis. [111]

MIV-711, a potent selective cathepsin K inhibitor, is potential disease-modifying agent. Athough MIV-711 was found to be no better than placebo in relieving pain in knee osteoarthritis, it has shown promise in reducing bone and cartilage damage. In a phase 2a study in 244 patients with primary knee osteoarthritis, no statistically significant change in pain score occurred among patients randomly allocated to receive MIV-711 100 or 200 mg daily, or placebo for 26 weeks. However, patients in the treatment groups demonstrated reductions in biomarkers of bone resorbption (CTX-I) and cartilage loss (CTX-II). [112] Hence, additional research is needed regarding the overall and long-term effects and possible benefits of this therapy.

Other investigational agents include monoclonal antibodies that inhibit nerve growth factor (NGF), such as tanezumab. Phase III trials of subcutaneously administered tanezumab in patients with hip or knee osteoarthritis have reported statistically significant improvements in joint pain, physical function, and patient global assessment of osteoarthritis. [113, 114]


Lifestyle Modification, Physical/Occupational Therapy, and Other Nonpharmacologic Measures

Lifestyle modification, particularly exercise and weight reduction, is a core component in the management of osteoarthritis. [115, 116]  

There is substantial evidence regarding the short-term benefits of therapy on pain and function, with long-term effectiveness improved with adherence to home-based exercise programs. However, Bennell et al reported that in 74 patients with medial knee osteoarthritis who completed a 12-week physiotherapist-supervised exercise trial, the addition of two 30-minute physiotherapy booster sessions had no significant influence on pain, physical function outcomes, or measures of home exercise adherence. [117]

Instruct the patient to avoid aggravating stress to the affected joint. Implement corrective procedures if the patient has poor posture.

Weight reduction relieves stress on the affected knees or hips. The benefits of weight loss, whether obtained through regular exercise and diet or through surgical intervention, may extend not only to symptom relief but also to a slowing in cartilage loss in weight-bearing joints (eg, knees). [118] In addition, weight loss lowers levels of the inflammatory cytokines and adipokines that may play a role in cartilage degradation. [119]

Some patients with osteoarthritis benefit from heat placed locally over the affected joint. A minority of patients report relief with ice. [120]

In a study of 26 patients with painful and deforming hand osteoarthritis, rigid, custom splints worn nightly on one arthritic finger joint per patient for three months significantly reduced pain in 17 of the 23 patients (74%) who completed the study. Average pain remained significantly lower in splinted joints compared with non-splinted joints three months after patients stopped using the splints. [121, 122]

Physical activity

Although people with osteoarthritis tend to avoid activity, exercise is an effective treatment for this condition, producing improvements in pain, physical function, and walking distance. Long-term walking and resistance-training programs have been shown to slow the functional decline seen in many patients with osteoarthritis, including older patients. [119]

In a systematic review and meta-analysis of 48 randomized controlled trials, Juhl and colleagues found that the optimal exercise program for reducing pain and patient-reported disability in knee osteoarthritis should have a single aim, which can be improving aerobic capacity, strengthening the quadriceps muscle, or improving lower extremity performance. For best results, the exercise program should be supervised and performed 3 times weekly. [123]

Osteoarthritis of the knee may result in disuse atrophy of the quadriceps. Because these muscles help protect the articular cartilage from further stress, quadriceps strengthening is likely to benefit patients with knee osteoarthritis. Stretching exercises are also important in the treatment of osteoarthritis because they increase range of motion.

In a study of patients with knee osteoarthritis, Jan et al found that in most respects, non–weight-bearing exercise was as therapeutically effective as weight-bearing exercise. [124] After an 8-week exercise program, the 2 types of exercise resulted in equally significant improvements in function, walking speed, and muscle torque. However, patients in the weight-bearing group demonstrated greater improvement in position sense, which may help patients with complex walking tasks, such as walking on a spongy surface.

Chaipinyo and Karoonsupcharoen found no significant difference between home-based strength training and home-based balance training for knee pain caused by osteoarthritis. However, greater improvement in knee-related quality of life was noted in the strength-training group. [125]

The importance of aerobic conditioning, particularly low-impact exercises (if osteoarthritis is affecting weight-bearing joints), should be stressed as well. Swimming, especially the aerobic aquatic programs developed by the Arthritis Foundation, can be helpful.

The benefits of exercise have been found to decline over time, possibly because of poor adherence. Factors that determine adherence to exercise have not been carefully studied in patients with osteoarthritis. In a review of this topic, Marks and Allegrante concluded that interventions to enhance self-efficacy, social support, and skills in the long-term monitoring of progress are necessary to foster exercise adherence in people with osteoarthritis. [126]

Tai chi

A prospective, single-blind, randomized, controlled study by Wang et al suggested that tai chi is a potentially effective treatment for pain associated with osteoarthritis of the knee. [127] In this trial, 40 patients with symptomatic tibiofemoral osteoarthritis who performed 60 minutes of tai chi twice weekly for 12 weeks experienced significantly greater pain reduction than did control subjects who underwent 12 weeks of wellness education and stretching.

The mean difference in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain scores was −118.80 mm. [127] The tai chi cohort also had significantly better WOMAC physical function scores, patient and physician global visual analog scale scores, chair stand times, Center for Epidemiologic Studies Depression Scale scores, self-efficacy scores, and Short Form 36 physical component summaries.

A subsequent trial by Wang et al that compared tai chi (2 times per week for 12 weeks) with standard physical therapy (2 times per week for 6 weeks, followed by 6 weeks of monitored home exercise), reported substantial and comparable reductions in WOMAC scores in both patient groups, as well as smilar clinically significant improvement in most secondary outcomes, and the benefits were maintained up to 52 weeks. Furthermore, the tai chi group had significantly greater improvements in depression and the physical component of quality of life. [128]

A systematic review and meta-analysis concluded that research results are encouraging and suggest that tai chi may be effective in controlling pain and improving physical function in patients with knee osteoarthritis. [129] The researchers noted, however, that the strength of the evidence is limited by the small number of randomized, controlled trials with a low risk of bias.

Assistive devices

The use of assistive devices for ambulation and for activities of daily living (ADLs) may be indicated for patients with osteoarthritis. Braces  may also be of some use. A cane can be used in the contralateral hand for hip or knee osteoarthritis. The patient can be taught joint-protection and energy-conservation techniques.

In a randomized study of 126 patients with painful knee osteoarthritis, wearing a slip-on knee brace for a median of 7.35 hours a day for 6 weeks reduced pain and bone marrow lesions. Pain scores during activity declined about 18 points in patients using the brace, but showed almost no change in those not wearing the brace. MRIs revealed that bone marrow lesion volumes in the patellofemoral joint, which were nearly identical at baseline in both groups, had decreased by 25% after 6 weeks in patients in the brace group. [130]

The ACR/Arthritis Foundation strongly recommends hand orthoses for patients with osteoarthritis of the first carpometacarpal joint and conditionally recommends hand orthoses for patients with osteoarthritis in other joints of the hand. The guideline conditionally recommends kinesiotaping for patients with osteoarthritis of the first carpometacarpal joint. [131]

For knee osteoarthritis, guidelines recommend appropriate footwear as part of self-management. However, a randomized trial by Hinman et al that compared shoes designed to unload the knee versus new conventional shoes found that both groups showed comparable improvement in pain with walking as well as other benefits. In the trial, the intervention group received walking shoes with triple-density, variable-stiffness midsoles and mild lateral-wedge insoles designed to unload the medial knee and worn daily, while the control group received supportive lace-up walking shoes. [132]

Occupational therapy and physical therapy

Occupational adjustments may be necessary for some patients with osteoarthritis. An occupational therapist can assist with evaluating how well the patient performs ADLs, as well as with retraining of the patient as necessary. Joint-protection techniques should be emphasized. Physical therapy modalities, especially those aimed at deconditioned patients, can be helpful, particularly in patients with hip or knee involvement.

Electromagnetic field stimulation and TENS

A pulsed electromagnetic field stimulation device (Bionicare) has been approved by the US Food and Drug Administration (FDA) for use in patients with knee osteoarthritis. Pulsed electromagnetic field stimulation is believed to act at the level of articular cartilage by maintaining the proteoglycan composition of chondrocytes through downregulation of its turnover. [133]

A multicenter, double-blind, randomized, placebo-controlled 4-week trial in 78 patients with knee osteoarthritis found improved pain and function in those who were treated with the device. [134] A double-blind, placebo-controlled 3-month trial in 58 patients with moderate-to-severe knee osteoarthritis showed that the use of a highly optimized, capacitively coupled, pulsed electrical stimulus device yielded significant symptomatic and functional improvement. [135]

Another randomized clinical trial demonstrated that pulsed short-wave treatment was effective in relieving pain and improving function and quality of life in women with knee osteoarthritis on a short-term basis; additional studies are needed to validate the 12-month follow-up. [136]

Transcutaneous electrical nerve stimulation (TENS) may be another treatment option for pain relief. To date, however, there is only limited evidence that TENS is beneficial in this setting. A systematic review could not confirm that TENS is effective for pain relief in knee osteoarthritis. [137] A randomized controlled trial found that TENS applied in conjunction with therapeutic exercise and daily activities increased quadriceps activation and function in patients with tibiofemoral osteoarthritis. [138]


Acupuncture is becoming a more frequently used option for treatment of the pain and physical dysfunction associated with osteoarthritis. Some evidence supports its use. For example, a review article of randomized, controlled trials reported that the level of pain persisting after acupuncture was significantly lower than the level of pain persisting after control treatments. [139]

A study evaluating 442 patients with knee osteoarthritis found that intensive acupuncture (3 times weekly for 8 weeks) resulted in less pain and better function at week 8, compared with sham acupuncutre, with persistence of those benefits through week 26. Response rates at week 8 were 60.3% (91 of 151 patients) for electroacupuncture versus 47.3% (69 of 146 patients) for sham acupuncture. Manual acupuncture was not significantly better than sham acupuncture at week 8 (58.6%; 85 of 145) but demonstrated significantly higher response rates at weeks 16 and 26. [140]

Several groups have issued guideline recommendations regarding acupuncture for knee osteoarthritis. The American Academy of Orthopaedic Surgeons offers limited support, stating that acupuncture may improve pain and function in patients with knee osteoarthritis. [141] The ACR/Arthritis Foundation guidelines conditionally recommend acupuncture for patients with knee, hip, and/or hand osteoarthritis, while noting that the greatest number of positive trials with the largest effect sizes have been carried out in knee osteoarthritis. [131]



A procedure of low invasiveness and morbidity, arthroscopy will not interfere with future surgery. However, a randomized, controlled trial in patients with moderate-to-severe osteoarthritis found that arthroscopic surgery for osteoarthritis of the knee provided no additional benefit beyond that afforded by optimized physical and medical therapy. [142]

Arthroscopy is indicated for removal of meniscal tears and loose bodies; less predictable arthroscopic procedures include debridement of loose articular cartilage with a microfracture technique and cartilaginous implants in areas of eburnated subchondral bone (see the images below). These treatments have varying success rates and should be performed only by surgeons experienced in arthroscopic surgical techniques. [142, 143] Overall, arthroscopy is not recommended for nonspecific “cleaning of the knee” in osteoarthritis.

Arthroscopic view of a torn meniscus before (top) Arthroscopic view of a torn meniscus before (top) and after (bottom) removal of loose meniscal fragments.
Arthroscopic view of an arthritic knee. Arthroscopic view of an arthritic knee.
Arthroscopic view of a knee after the removal of l Arthroscopic view of a knee after the removal of loose fragments of articular and meniscal cartilage.
Arthroscopic view of the removal of cartilaginous Arthroscopic view of the removal of cartilaginous loose body.

Patients who undergo arthroscopy usually require a period of crutch use or exercise therapy. This period typically lasts days but sometimes extends for weeks.



Osteotomy is used in active patients younger than 60 years who have a malaligned hip or knee joint and want to continue with reasonable physical activity. [144] The principle underlying this procedure is to shift weight from the damaged cartilage on the medial aspect of the knee to the healthy lateral aspect of the knee. Osteotomy is most beneficial for significant genu varum, or bowleg deformity. (The effectiveness of osteotomy for genu valgum is not highly predictable.)

Osteotomy often can help individuals avoid requiring a total knee replacement until they are older. It can lessen pain, but it can also lead to more challenging surgery if the patient later requires arthroplasty.

A follow-up study of 147 opening-wedge high tibial osteotomies in a consecutive series of patients affected by varus knee malalignment with isolated medial compartment degenerative joint disease reported good or excellent results in 94% of cases. Follow-up averaged 9.5 years, with a range of 7 to 12 years. [145]

Contraindications for osteotomy are as follows:

  • Knee flexion of less than 90°
  • A flexion-extension contracture of more than 15°
  • Varus over 15°-20°
  • Instability from previous trauma or surgery
  • Severe arterial insufficiency
  • Bicompartmental involvement

Patients undergoing osteotomy require partial weight-bearing until bony healing occurs. Afterward, exercise is indicated.



Arthroplasty consists of the surgical removal of joint surface and the insertion of a metal and plastic prosthesis (see the images below). The prosthesis is held in place by cement or by bone ingrowth into a porous coating on the prosthesis. The use of cement results in faster pain relief, but bone ingrowth may provide a more durable bond; accordingly, prostheses with a porous coating are used in younger patients.

Anteroposterior radiograph shows knee replacement Anteroposterior radiograph shows knee replacement in 1 knee and arthritis in the other, with medial joint-space narrowing and subchondral sclerosis.
Anteroposterior radiograph of the pelvis and hips Anteroposterior radiograph of the pelvis and hips shows an arthritic hip not treated surgically and a total hip replacement.
Anteroposterior radiograph obtained after knee rep Anteroposterior radiograph obtained after knee replacement.
Lateral radiograph obtained after knee replacement Lateral radiograph obtained after knee replacement (same patient as in the above image).

Arthroplasty is performed if all other modalities are ineffective and osteotomy is not appropriate or if a patient cannot perform ADLs despite maximal therapy. [146, 147] This procedure alleviates pain and may improve function. At a minimum, 10-15 years of viability are expected from joint replacement in the absence of complications.

Infection is a particular postoperative concern in cases of total joint replacement. This complication is now rare, however, especially with the use of perioperative antibiotics.

Prevention of thrombophlebitis and resultant pulmonary embolism is important in patients who undergo lower-extremity arthroplasty procedures for osteoarthritis. The surgeon must use all means available to prevent these complications. Early motion and ambulation, when possible, are of particular importance. The use of low-molecular-weight heparin or warfarin is also indicated.

After joint replacement, patients require partial weight-bearing, which progresses to full weight-bearing in 1-3 months; range-of-motion and strengthening exercises are started within a few days after joint-replacement surgery and continued until the patient has good range of motion and strength. After resection arthroplasty of the hip, patients require instruction in the use of crutches or a walker, which are usually needed permanently.

For more information, see the following articles:


Fusion and Joint Lavage

Fusion consists of the union of bones on either side of the joint. This procedure relieves pain but prevents motion and puts more stress on surrounding joints. Fusion is sometimes used after knee replacements fail or as a primary procedure for ankle or foot arthritis.

Observational studies suggested a benefit for joint lavage. However, sham-controlled trials yielded conflicting results, and a meta-analysis concluded that joint lavage does not result in pain relief or improvement of function in patients with knee osteoarthritis. [148]



Overweight patients who have early signs of osteoarthritis or who are at high risk should be encouraged to lose weight. Recommend quadriceps-strengthening exercises in patients with osteoarthritis of the knees, except in those with pronounced valgus or varus deformity at the knees. (See Lifestyle Modification, Physical/Occupational Therapy, and Other Nonpharmacologic Measures.)

It has been proposed that low vitamin D levels may play a role in the development and progression of osteoarthritis; however, studies of vitamin D status and osteoarthritis have produced conflicting results. [149, 150]

A systematic review found no convincing evidence that selenium, vitamin A, or vitamin C is effective for the treatment of osteoarthritis. [151] A prospective cohort study also found no evidence that vitamin C supplementation slowed the progression of knee osteoarthritis; however, it did find that patients who reported taking vitamin C were 11% less likely to develop knee osteoarthritis. [152]


Stem Cell Therapy

Mesenchymal stem cells (MSCs) are an investigational treatment of osteoarthritis. The International Society for Cellular Therapy (ISCT) defines MSCs as mesenchymal “stromal” cells having the attributes of being plastic-adherent culture-expanded cells without hematopoietic cell markers that express specific cell surface markers (ie, CD73, CD90, and CD105) and that show the ability to differentiate into osteoblasts, adipocytes, and chondrocytes in vitro. [104]  MSCs can secrete bioactive molecules that stimulate angiogenesis and tissue repair, and reduce T-cell response and inflammation. MSCs, via cytokines and growth factors such as vascular endothelial growth factor (VEGF) and transforming growth factor beta (TGF-beta), can migrate to regions of cartilage ischemia and have been reported as immunomodulatory. [153]

MSCs can be derived from a variety of tissues, but for therapeutic purposes they are collected from bone marrow or adipose tissue. Autologous MSCs are easily harvested and used therapeutically, but allogeneic MSCs have been utilized. Delivery methods have included implantation and microfracture, but the current focus is on intra-articular injection of suspensions containing large numbers of cells. [74, 75]

In clinical trials, placement of MSCs into the knee joint has proved an effective treatment for osteoarthritis, with no significant increases in adverse events. [74, 75, 11, 154, 153] However, no MSC therapies have yet been cleared by the FDA for human clinical application to musculoskeletal diseases. [104]

A meta-analysis comparing the conditions of patients with knee osteoarthritis before and after treatment with MSCs demonstrated continual efficacy for at least 24 months, with greater pooled effect size (pain and functional changes) at 12 and 24 months than summed effect sizes at 3 months. However, a dose-responsiveness association was not demonstrated in the MSC numbers. Notably, MSC therapy appeared more beneficial in patients with early osteoarthritis than in those with advanced disease. [74]

Two systematic reviews have questioned the quality of the studies of stem cell therapy for knee osteoarthritis, citing high risk of bias, low quality of evidence, only modest improvement reported, and possible placebo effect. [76, 77]  Additional studies are needed with standardization of methdology, efficacy reporting, and safety. Nevertheless, a survey identified hundreds of clinics in the United States that offer stem cell therapy for knee arthritis, at prices ranging from $1150 to $12,000 for a unilateral injection. In many cases these injections would consist of centrifuged blood or bone marrow aspirate rather than cultivated stem cells. [155]