Residency · Residency · Rheumatology

Osteoarthritis - Evidence-Based Management

Introduction

Osteoarthritis is the most common joint disease worldwide, affecting approximately 300 million people and representing the leading cause of disability in older adults and the major driver of joint replacement surgery. OA is no longer considered simple "wear and tear" but is recognized as a complex disease involving inflammatory, metabolic, and biomechanical factors that collectively drive cartilage degradation and whole-joint pathology. No proven disease-modifying osteoarthritis drug exists, and management therefore focuses on symptom relief and functional optimization. The 2019 ACR/Arthritis Foundation, 2019 OARSI, and 2023 EULAR guidelines provide comprehensive evidence-based frameworks that emphasize non-pharmacologic interventions as the foundation of therapy.

Pathophysiology

Cartilage Biology

Articular cartilage is an avascular, aneural tissue composed of chondrocytes embedded in an extracellular matrix of type II collagen and aggrecan. Cartilage homeostasis depends on a balance between anabolic factors, including TGF-beta and IGF-1, and catabolic factors, including the matrix metalloproteinase MMP-13, the aggrecanase ADAMTS-5, IL-1, and TNF. Osteoarthritis disrupts this homeostatic balance, producing net cartilage degradation accompanied by subchondral bone changes and osteophyte formation. Chondrocyte senescence is an emerging concept in which aging chondrocytes produce increased levels of catabolic factors, and senolytic therapy aimed at clearing these senescent cells is under investigation as a potential disease-modifying approach.

Whole-Joint Disease Concept

Contemporary understanding recognizes OA as a disease of the entire joint rather than simply cartilage loss. All joint structures are affected, including cartilage, subchondral bone, synovium, menisci, ligaments, periarticular muscles, and the infrapatellar fat pad. Subchondral bone exhibits sclerosis, cyst formation, and bone marrow lesions, which on MRI predict cartilage loss and correlate with pain. Synovitis, a low-grade inflammatory process present in approximately 50 percent of OA joints, correlates with both pain and disease progression. Meniscal pathology, including tears and extrusion, contributes to altered compartmental loading and accelerated progression. Quadriceps weakness both precedes and perpetuates knee OA, establishing a vicious cycle that exercise therapy aims to interrupt.

OA Phenotypes (Emerging Concept)

The emerging concept of OA phenotyping recognizes distinct disease subtypes that may ultimately guide targeted therapy. The biomechanical or post-traumatic phenotype is injury-related, affects younger patients, and tends to be focal. The inflammatory phenotype features synovitis predominance, includes erosive hand OA, and is associated with elevated CRP. The metabolic syndrome-associated phenotype links OA to obesity, diabetes, and dyslipidemia through systemic low-grade inflammation. The minimal joint disease or pain-predominant phenotype has a significant central sensitization component. The aging or senescence-related phenotype is driven by chondrocyte senescence and tends toward widespread OA. While these phenotypes are not yet incorporated into clinical guidelines, they represent a promising framework for future personalized therapeutic approaches.

Clinical Features

Common OA Patterns

Knee OA is the most common form of symptomatic osteoarthritis, predominantly affecting the medial compartment. Major risk factors include obesity, prior knee injury involving the ACL or meniscus, and malalignment. Hip OA produces pain in the groin that may refer to the thigh or knee, with internal rotation being the first motion limited. The superior or superolateral pattern is most common. Hand OA characteristically involves the distal interphalangeal joints producing Heberden nodes, the proximal interphalangeal joints producing Bouchard nodes, and the first carpometacarpal joint at the thumb base. Nodal OA is more common in women. Erosive hand OA is a distinct inflammatory subtype affecting the DIP and PIP joints with inflammatory erosions producing a "gull-wing" deformity on radiographs and presenting with painful, red, swollen joints. Spinal OA encompasses facet joint osteoarthritis and disc degeneration producing cervical and lumbar spondylosis. Generalized OA involves 3 or more joint groups, is often familial, and most commonly combines hand and knee involvement.

Key Distinguishing Features from Inflammatory Arthritis

Several clinical features distinguish OA from inflammatory arthritis. Morning stiffness in OA lasts fewer than 30 minutes, compared to more than 1 hour in RA. Pain is usage-related, worsening with activity and improving with rest. Physical examination reveals bony enlargement from osteophytes rather than soft tissue synovitis. Systemic inflammation is absent, with normal ESR and CRP in most patients. RF and anti-CCP are negative. DIP involvement is common in OA but unusual in RA.

Risk Factors

Non-modifiable risk factors include age, which is the strongest overall risk factor, female sex, genetic susceptibility involving variants in GDF5, FRZB, and COL2A1, prior joint injury, and joint shape and alignment. Modifiable risk factors include obesity, which is the strongest modifiable risk factor through both mechanical and metabolic mechanisms, muscle weakness, occupational overuse, malalignment, and physical inactivity.

<image>A comparison illustration of normal joint anatomy versus osteoarthritic joint anatomy, shown as knee joint cross-sections side by side. Normal joint: Smooth articular cartilage covering femoral condyle and tibial plateau, intact menisci, thin synovial lining (1-2 cells thick), normal subchondral bone, healthy joint capsule. OA joint: Thinned and fibrillated cartilage with focal areas of complete cartilage loss (bone-on-bone), subchondral sclerosis with thickened trabeculae, subchondral cyst, marginal osteophytes at joint margins, thickened and inflamed synovium (synovitis), degenerated and extruded meniscus, loose bodies in joint space, bone marrow lesion (edema) in subchondral bone. Label all pathologic changes clearly with annotations explaining their clinical significance.</image>

Diagnosis

Clinical Diagnosis

Osteoarthritis is primarily a clinical diagnosis, with imaging serving a confirmatory rather than primary diagnostic role. The ACR clinical criteria for knee OA include age over 50, morning stiffness lasting fewer than 30 minutes, crepitus, bony tenderness, bony enlargement, and no palpable warmth. The knee examination should include gait assessment, alignment evaluation for varus or valgus deformity, range of motion testing, assessment for crepitus, and evaluation for effusion using ballottement and bulge sign techniques, along with patellar grind testing.

Imaging

Radiographs are the first-line imaging modality. The Kellgren-Lawrence grading system classifies OA from grade 0 (normal) through grade 1 (doubtful), grade 2 (minimal, with definite osteophytes and possible joint space narrowing), grade 3 (moderate, with moderate joint space narrowing and sclerosis), to grade 4 (severe, with large osteophytes, marked joint space narrowing, sclerosis, and deformity). Weight-bearing views are essential for knee imaging, including standing anteroposterior and PA flexion or Rosenberg views. A critically important clinical principle is that radiographic severity does not correlate well with symptoms, and treatment decisions should not be based solely on imaging findings. MRI is not routinely needed for diagnosis but demonstrates bone marrow lesions, synovitis, cartilage defects, and meniscal tears, and is indicated when diagnostic uncertainty exists or concurrent pathology such as meniscal locking, loose body, or avascular necrosis is suspected. Ultrasound can detect effusions, synovitis, osteophytes, and Baker cysts and is emerging as a useful clinical adjunct.

Laboratory

There is no diagnostic blood test for osteoarthritis. ESR and CRP are usually normal, though mild elevation may be seen in erosive OA or with large-joint effusion. RF and anti-CCP should be negative, and their testing serves to exclude RA. Synovial fluid analysis reveals a non-inflammatory pattern with fewer than 2,000 white blood cells per microliter, viscous and clear-to-yellow fluid, and absence of crystals, which excludes gout and CPPD.

Non-Pharmacologic Management (FIRST-LINE)

Exercise and Physical Therapy

Exercise has the strongest evidence of any OA intervention, with a number needed to treat of approximately 4 for meaningful pain reduction. Aerobic exercise including walking, cycling, and swimming improves pain, function, and cardiovascular fitness. Strengthening exercises, particularly quadriceps strengthening for knee OA and hip abductor strengthening, reduce pain by an amount equivalent to NSAIDs. Aquatic therapy is beneficial especially for patients unable to tolerate land-based exercise. Tai chi and yoga have moderate evidence supporting their use in knee OA.

Weight Management

Weight loss of 10 percent or more of body weight is associated with significant pain reduction in overweight and obese patients with knee OA. The IDEA trial demonstrated that diet plus exercise was superior to either intervention alone for knee OA outcomes. A useful clinical teaching point is that each pound of weight loss translates to a 4-pound reduction in load across the knee with each step.

Biomechanical Interventions

Lateral wedge insoles have mixed evidence and are not recommended by the 2019 OARSI guidelines. Knee braces, specifically valgus unloader braces, provide modest benefit for medial compartment OA. A cane held in the contralateral hand reduces ipsilateral knee load by approximately 25 percent. Patellar taping offers short-term pain relief in patellofemoral OA.

Patient Education and Self-Management

Self-management programs such as the Arthritis Self-Management Program improve self-efficacy and coping. Cognitive behavioral therapy is effective for OA-related pain catastrophizing. Addressing sleep disturbance through sleep hygiene interventions improves pain perception.

Pharmacologic Management

Topical Agents

Topical NSAIDs, including diclofenac gel 1 percent and diclofenac solution, are the first-line pharmacotherapy for knee and hand OA per ACR recommendations. They provide equivalent efficacy to oral NSAIDs for knee OA with significantly fewer systemic side effects and are applied four times daily with a 4- to 6-week trial recommended. Topical capsaicin offers modest benefit through depletion of substance P, though the burning sensation limits adherence.

Oral Analgesics

Acetaminophen is no longer recommended as first-line therapy based on the 2019 OARSI and ACR guidelines. The PACE trial demonstrated that acetaminophen was not superior to placebo for hip and knee OA pain. If used, the dose should not exceed 3 grams daily due to hepatotoxicity risk.

Oral NSAIDs are effective but should be used at the lowest effective dose for the shortest possible duration. COX-2 selective celecoxib at 200 milligrams daily is preferred for patients with gastrointestinal risk, and the PRECISION trial demonstrated that celecoxib was non-inferior to ibuprofen and naproxen for cardiovascular safety. Among non-selective NSAIDs, naproxen at 500 milligrams twice daily has the best cardiovascular safety profile. Gastrointestinal protection with a proton pump inhibitor should be added for patients with risk factors including age over 65, history of ulcer, or concurrent aspirin or anticoagulant use. Renal function monitoring is essential in elderly patients.

Duloxetine, an SNRI, is conditionally recommended for knee OA by the 2019 ACR guidelines. It addresses the central sensitization component of OA pain and is effective at 60 milligrams daily, with nausea, fatigue, and constipation as the main side effects. Tramadol, a weak opioid with SNRI activity, is conditionally recommended but only when other options have been exhausted. Opioids are strongly recommended against by the 2019 ACR guidelines due to their ineffectiveness for OA pain, harmful side effects, and dependency risk.

Intra-articular Therapies

Intra-articular glucocorticoids are conditionally recommended for knee and hip OA. Triamcinolone acetonide 40 milligrams injected into the knee provides 4 to 8 weeks of pain relief. The frequency should be limited to no more than 3 to 4 injections per year per joint due to concerns about accelerated cartilage loss with repeated use, as demonstrated in the McAlindon 2017 trial, which showed that repeated intra-articular triamcinolone every 12 weeks over 2 years resulted in greater cartilage volume loss compared to placebo, with no difference in pain outcomes.

Intra-articular hyaluronic acid, or viscosupplementation, remains controversial. The 2019 ACR guidelines conditionally recommend against its use for knee OA, while the 2019 OARSI guidelines give an uncertain recommendation. Meta-analyses show modest benefit of approximately 8 millimeters on a visual analog scale, of uncertain clinical significance. It may have a role in patients who have failed other therapies or are not candidates for surgery.

Intra-articular PRP is strongly recommended against by the 2019 ACR guidelines due to insufficient evidence and cost. Intra-articular stem cell therapy is not recommended due to insufficient evidence and lack of regulatory oversight.

Nutraceuticals and Supplements

Glucosamine and chondroitin are conditionally recommended against by the 2019 ACR guidelines. The GAIT trial found no significant benefit over placebo, with a possible exception for a moderate-to-severe pain subgroup. Colchicine is under investigation for the inflammatory OA phenotype but has limited evidence. Vitamin D supplementation showed no benefit for OA symptoms in the VIDEO trial.

Surgical Management

Arthroscopic surgery for OA alone, without true mechanical symptoms, is not recommended. The FIDELITY trial and the Moseley 2002 trial demonstrated no benefit over sham surgery. The exception is when true mechanical symptoms such as a locked knee from a loose body are present. Total joint arthroplasty is the most effective intervention for end-stage OA, indicated when there is persistent pain and disability despite optimal non-surgical management, significant radiographic disease, and inadequate quality of life. Patient satisfaction rates are 90 to 95 percent, with implant survival exceeding 95 percent at 15 to 20 years. Timing should be neither too early, to preserve implant longevity, nor too late, to avoid muscle atrophy and deconditioning. High tibial osteotomy is an option for younger patients with unicompartmental knee OA and malalignment. Partial knee replacement, or unicompartmental arthroplasty, is appropriate for isolated medial or lateral compartment OA and offers faster recovery than total knee arthroplasty.

<image>A stepped management pyramid for osteoarthritis, arranged from base (first-line) to apex (last resort). Base layer (all patients): Education, self-management, exercise (aerobic + strengthening), weight management (if overweight). Second layer: Topical NSAIDs (first-line pharmacotherapy), physical therapy, biomechanical aids (braces, walking aids). Third layer: Oral NSAIDs (lowest dose, shortest duration + PPI if needed), duloxetine, IA GC injections. Fourth layer: Tramadol (limited use), IA hyaluronic acid (controversial). Apex: Total joint arthroplasty (end-stage disease). On the side, mark interventions NOT recommended: opioids (red X), glucosamine/chondroitin (red X), arthroscopic debridement for OA alone (red X), PRP/stem cells (red X). Include evidence levels (strong/conditional) next to each recommendation.</image>

Emerging Therapies

Sprifermin, an intra-articular recombinant FGF-18, demonstrated cartilage thickness gain in the FORWARD trial but did not improve pain, and it remains unapproved. Lorecivivint, a Wnt pathway inhibitor, is under investigation for disease modification potential. Nerve growth factor inhibitors including tanezumab and fasinumab provided potent pain relief but raised FDA concerns about rapidly progressive OA and osteonecrosis. The STEP 1 and 2 trials showed tanezumab was effective for pain but increased rates of joint replacement and osteonecrosis, and these agents remain unapproved. Senolytic therapy with UBX0101, an MDM2 inhibitor, failed its Phase II trial, though the concept continues to be pursued. Gene therapy is under early investigation.

Key Clinical Pearls

• Exercise is the single most effective non-pharmacologic intervention for OA; NNT ~4 for clinically meaningful improvement
• Acetaminophen is no longer recommended first-line for OA; topical NSAIDs are preferred
• X-ray severity correlates poorly with symptoms; do not base treatment decisions solely on imaging
• IA glucocorticoids provide short-term relief but may accelerate cartilage loss with repeated use
• Arthroscopic surgery for OA alone (without true mechanical locking) has no benefit over sham surgery
• Weight loss of 10% significantly reduces knee OA pain; 1 pound of weight loss removes 4 pounds of joint load

References

1. Kolasinski SL, et al. 2019 ACR/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. Arthritis Care Res. 2020;72(2):149-162.
2. Bannuru RR, et al. OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis Cartilage. 2019;27(11):1578-1589.
3. McAlindon TE, et al. Effect of intra-articular triamcinolone vs saline on knee cartilage volume and pain in patients with knee osteoarthritis. JAMA. 2017;317(19):1967-1975.
4. Nissen SE, et al. Cardiovascular safety of celecoxib, naproxen, or ibuprofen for arthritis (PRECISION). N Engl J Med. 2016;375(26):2519-2529.
5. Hochberg MC, et al. Effect of intra-articular sprifermin vs placebo on femorotibial joint cartilage thickness in patients with osteoarthritis (FORWARD). JAMA. 2019;322(14):1360-1370.

Read this lecture as Markdown