This guide is part of Upwell's evidence-based musculoskeletal content cluster. After reading, explore the related guides below.
Related reading from Upwell Health:
• Hip Pain in Runners: Gluteal Tendinopathy, FAI & Labral Tears — hip abductor weakness is one of the primary drivers of medial compartment knee OA progression. If your knee OA is accompanied by lateral hip pain or a history of running injuries, this guide explains the connection.
• Lower Back Pain: Australia's Most Comprehensive Guide — knee OA and lower back pain frequently coexist in older active adults. The pain science, exercise evidence, and management frameworks overlap significantly.
• ACL and Osteoarthritis: The Long Game Nobody Is Talking About — if your knee OA followed an ACL injury, this guide covers the post-traumatic OA pathway and what the evidence says about slowing progression.
• Knee Arthritis Treatment: What Really Works (and What Doesn’t) — a complementary read covering treatment decision-making in knee arthritis.
• Why 'Bone-on-Bone' Doesn’t Mean What You Think — one of the most-shared pieces on the Upwell blog. The imaging evidence explained for patients who have been told their knee is beyond exercise.
• How Can Physiotherapy Help Treat Osteoarthritis? — the specific physiotherapy modalities and their evidence base for knee OA management.
• Why Clinical Pilates Should Be Your Go-To Workout — clinical Pilates is particularly valuable for knee OA patients who cannot yet tolerate conventional loaded exercise. This guide explains why the Reformer equipment matters.
• The Complete Runner’s Guide — if you have mild to moderate knee OA and want to keep running, this guide covers load management, cadence optimisation, and why running is not universally contraindicated.
Updated May 2026. Written by the Upwell Health Collective clinical team. Clinically reviewed May 2026. Next review November 2026. For educational purposes only.
Knee osteoarthritis is one of the most misunderstood conditions in musculoskeletal medicine. The misunderstanding starts with the words used to describe it. “Wear and tear.” “Degenerative joint disease.” “Your cartilage is wearing away.” These phrases are not just imprecise — they are clinically harmful. They suggest that the joint is a passive structure grinding itself down with use, that damage is inevitable and irreversible, and that activity is the enemy. Every one of these implications is contradicted by the current evidence.
The modern scientific consensus is that describing OA as “wear and tear” has “hampered research progress and drug development” (Frontiers in Immunology, 2021). OA is a complex, whole-joint disease involving articular cartilage, subchondral bone, synovial membrane, meniscus, ligaments, the infrapatellar fat pad, and the peri-articular muscles. It is driven by a convergence of biomechanical, inflammatory, metabolic, and post-traumatic processes. It involves low-grade systemic inflammation. The evidence-based narrative: OA is a chronic condition highly responsive to treatment that does not follow a predetermined downward trajectory. Many people with knee OA remain active and functional for decades. Many improve measurably with the right management.
Also read: Why 'Bone-on-Bone' Doesn't Mean What You Think and How Can Physiotherapy Help Treat Osteoarthritis?
According to the AIHW (2024), approximately 2.1 million Australians — 8.3% of the population — were estimated to be living with osteoarthritis in 2022. Knee OA is the most common subtype. A Global Burden of Disease analysis found knee OA prevalence increasing by 126% from 1990 to 2019. Arthritis Australia projects the total burden to reach 5.39 million by 2040, with over 2 million of working age. In 2020–21, an estimated $4.3 billion was spent on the treatment and management of osteoarthritis.
Articular cartilage has no blood supply and no nerve supply — which is why it cannot directly cause pain and why it heals so poorly once damaged. Cartilage loss does not cause pain. A person with severe cartilage loss on imaging can have minimal symptoms. A person with modest radiographic changes can have debilitating pain. The correlation between cartilage appearance on MRI or x-ray and the patient’s pain level is poor. This is one of the most clinically important facts in OA management.
Subchondral bone is richly innervated — bone marrow lesions (marrow oedema) are one of the strongest correlates of knee pain in OA. The synovium becomes inflamed (synovitis) in OA, producing elevated levels of pro-inflammatory cytokines (IL-1β, IL-6, TNF-α) that drive further cartilage degradation and contribute directly to pain. The meniscus shows degenerative changes in early OA, often before radiographic cartilage loss. The infrapatellar fat pad produces pro-inflammatory adipokines. Peri-articular muscle weakness — particularly the quadriceps — both results from and contributes to OA progression. This is why strengthening is not just symptom management — it addresses the pathophysiology.
Knee OA pain arises from richly innervated structures: the subchondral bone, joint capsule, synovium, ligaments, and peri-articular muscles. In persistent OA pain, the nervous system undergoes changes that amplify pain signals — central sensitisation — which explains why some patients with mild radiographic OA have severe pain, and why purely structural interventions (including surgery) sometimes fail. Multiple studies confirm poor correlation between radiographic severity (KL grade) and pain intensity or function. See also: Why Your Pain Won't Go Away: Central Sensitisation Explained.
The x-ray is not the whole story. Your symptoms are not fully explained by your imaging.
A clinician who looks only at your imaging and says there’s nothing more to do has not fully assessed you. Your strength, movement patterns, nervous system sensitisation, psychological factors, body weight, and activity levels all matter as much or more to your outcome.
Weight-bearing anteroposterior x-ray is the standard first-line imaging. The Kellgren-Lawrence (KL) grading system classifies OA severity from grade 0 to grade 4. KL grade correlates poorly with pain and function — use it to confirm diagnosis and guide extreme-end decisions (like surgery timing), not to predict symptoms or prognosis. MRI findings must be interpreted with extreme caution — incidental degenerative findings are common in people over 40 with no knee symptoms at all.
What you should not do based on imaging alone: do not stop exercising because an x-ray shows bone-on-bone; do not consent to surgery because an MRI shows degenerative meniscal changes; do not adjust your activity expectations based solely on a KL grade. Also read: The Imaging Paradox: Why Your MRI Doesn’t Tell the Full Story.
Non-modifiable: Age, sex (women at higher rates particularly post-menopause), genetics, previous knee injury (ACL rupture carries the highest post-traumatic OA risk of any common knee injury — see ACL and Osteoarthritis: The Long Game Nobody Is Talking About), joint malalignment.
Modifiable: Obesity and overweight (each kilogram adds approximately 4 kilograms of force per step), quadriceps and hip weakness, sedentary behaviour, occupation, metabolic syndrome.
Exercise is the most evidence-supported, most effective, safest, and most broadly applicable treatment for knee osteoarthritis. Not a complementary add-on. The treatment. This is the position of every major clinical guideline: the ACSQHC (2024), ACR, EULAR, OARSI.
A network meta-analysis published in the BMJ in October 2025 (Yan et al.) reviewed 217 randomised clinical trials from 1990 to 2024 involving 15,684 participants. Aerobic exercise consistently showed the highest probability of being the best treatment across all outcomes — pain, function, and quality of life (mean SUCRA 0.72). All exercise modalities were found to be safe. No clear differences in adverse events between exercise interventions and control. Exercise is not risky for knee OA. Inactivity is risky.
Related: Exercise Is Medicine: 7 Evidence-Based Ways Movement Fights Chronic Pain.
A clinician who tells a patient with knee OA to “rest it” or “avoid anything that hurts” is causing harm. They are reinforcing fear-avoidance, allowing deconditioning to progress, and denying the patient access to the most effective treatment available. See: Why Rest Makes It Worse: The Deconditioning Spiral.
The quadriceps is the primary shock-absorber of the knee. A 30–40% gain in quadriceps strength has been proposed as the threshold for clinically meaningful improvement in pain and disability. Sustained, progressive loading is required — a short-term, low-intensity programme will not achieve this.
Beyond the quadriceps: the hip complex — gluteus medius, gluteus maximus, external rotators — is equally critical. Hip abductor weakness allows valgus collapse during loading activities, concentrating stress on the medial compartment. In many patients with medial compartment OA, hip weakness is the primary biomechanical driver. See: Hip Pain in Runners: The Hip-Knee Connection.
Evidence-based dosage: 2 to 3 sessions per week (minimum effective dose: 2), 2 to 4 sets of 8 to 15 repetitions, minimum 8 weeks (12 to 16 optimal), progressive load increase throughout.
Core programme: Seated knee extension, terminal knee extension, wall squats, step-ups, leg press (quadriceps); sidelying hip abduction, lateral band walks, single-leg stance (hip abductors); hip thrusts, Romanian deadlift, Bulgarian split squat (gluteus maximus); clamshells, monster walks (external rotators); calf raises progressing to single-leg.
Each kilogram of body weight adds approximately 4 kilograms of compressive force across the tibiofemoral joint with each step. A systematic review and network meta-analysis (Shahid et al., Osteoarthritis and Cartilage, 2025) of 13 RCTs involving 2,800 participants found that 5 to 10% weight loss was required to significantly improve disability, pain, and quality of life in knee OA. A 2025 network meta-analysis established a therapeutic threshold: approximately 7 to 10% weight loss is necessary for meaningful symptom improvement. Weight loss exceeding 7.5% substantially reduces the risk of requiring knee replacement.
The relationship between obesity and knee OA is not purely mechanical. Adipose tissue produces pro-inflammatory adipokines that promote synovitis and cartilage degradation — which is why obese individuals develop hand OA (a non-weight-bearing joint) at higher rates too. Diet combined with exercise produces the best outcomes across pain and function measures.
The STEP-9 trial (NEJM, 2024) — a large RCT of once-weekly semaglutide in persons with obesity and knee OA — found semaglutide produced significant weight loss and pain reduction compared to placebo. Improvements in KOOS pain scores, physical function, and body weight substantially exceeded the placebo group. This was the first high-quality RCT to demonstrate a GLP-1 agonist improves knee OA outcomes.
A March 2026 Cell Metabolism study demonstrated that semaglutide may have direct chondroprotective effects independent of weight loss, reducing cartilage degeneration, osteophyte formation, synovial lesion, and pain sensitivity through a metabolic restoration mechanism. The Arthritis Research Canada review (Nature Reviews, October 2025) found GLP-1 medications may benefit OA through systemic inflammation reduction, improved glycaemic control, reduced adipokine load, and possible direct cartilage-protective effects via GLP-1 receptors on chondrocytes.
Important clinical context on GLP-1s and knee OA.
GLP-1 receptor agonists are not currently approved as a first-line OA treatment. The evidence is promising but early. For patients who are obese and have a concurrent indication for GLP-1 therapy (type 2 diabetes, obesity management), the evidence increasingly supports that GLP-1s will produce meaningful OA benefit. This is a rapidly evolving area. Discuss with your GP or specialist.
Cortisone: Provides rapid short-term relief through anti-inflammatory effects on the synovium. A 2024 systematic review (EFORT Open Reviews, Bensa et al.) of 35 RCTs involving 3,300+ patients found that in the short term, corticosteroids and PRP performed similarly. Beyond the initial window, PRP outperformed corticosteroids at every follow-up point. Accumulating evidence suggests repeated cortisone injections may accelerate cartilage loss. Current guidance generally limits to 3 to 4 per year.
PRP: A 2025 comprehensive narrative review (Journal of Clinical Medicine) of 40 high-quality studies found PRP — particularly leukocyte-poor PRP — demonstrates superior pain relief and functional improvement compared to hyaluronic acid and corticosteroids, especially in KL grades I to III. Approximately 60% of patients experience clinically relevant improvement at 6-month follow-up. Not Medicare-funded in Australia.
Hyaluronic acid: Evidence is contested. Some meta-analyses find modest benefit over placebo; others find no clinically significant difference. The combination of PRP + HA has shown some superiority over PRP alone in recent meta-analyses.
Injection decision framework (2025 evidence):
• Acute flare: Cortisone. Effective short-term. Limit frequency. Combine with exercise continuation.
• Mild to moderate OA not responding to exercise alone: PRP is now the superior long-term option (KL grades I–III).
• What injections are not: A substitute for exercise. Injections without a concurrent exercise programme produce inferior outcomes.
• Advanced OA (KL grade IV): Injections may reduce symptoms but cannot address end-stage structural destruction. Surgical discussion appropriate when symptoms are severe and function significantly impaired despite optimal conservative management.
NSAIDs (ibuprofen, naproxen, celecoxib) are the most effective oral medications for knee OA pain. Topical NSAIDs (diclofenac gel) have comparable efficacy to oral forms with significantly lower systemic risk and are preferred in older adults. Paracetamol has been significantly downgraded — recent meta-analyses find inferior evidence compared to NSAIDs. Glucosamine and chondroitin have not demonstrated clinically meaningful benefit over placebo in adequately powered trials. Omega-3 fatty acids have anti-inflammatory properties and more promising emerging evidence.
A December 2025 cross-sectional study (Acta Orthopaedica, Kise et al.) of 653 patients with knee OA found that fear avoidance of physical activity and pain catastrophising had the strongest associations with poorer quality of life — stronger than age, sex, BMI, or radiographic severity. A 2025 systematic review and meta-analysis (Musculoskeletal Care, Phelps et al.) found that psychological interventions added to standard care improve both pain and function outcomes in knee OA.
Fear avoidance in knee OA works through a self-perpetuating cycle: pain → fear of movement → avoidance → deconditioning → increased pain sensitivity → more fear. Breaking it requires deliberate clinical attention. Related: The Fear-Avoidance Trap: How Fear of Pain Becomes More Disabling Than Pain Itself and Pain Is Not Damage.
OA symptoms fluctuate. Flares are part of the natural history and do not necessarily indicate disease progression. Common triggers: unusual activity level, cold or wet weather (perception-related, not structural), psychological stress, poor sleep, deconditioning, weight gain.
PEACE & LOVE framework for flares: Protect (modify, don't stop), Elevate (reduce inflammatory oedema), Avoid aggressive anti-inflammatory modalities early, Compress (sleeve for swelling), Educate (flare is temporary), Load (gradually return as it settles), Optimism (positive expectations improve recovery speed), Vascularisation (gentle aerobic movement reduces inflammation), Exercise (return to full programme as the flare resolves).
Related: The Flare-Up Survival Guide: What to Do When Pain Spikes.
Activity modification in knee OA does not mean becoming less active. It means distributing load intelligently. Running on hard surfaces: not universally contraindicated — transition to softer surfaces. High-impact sports: manageable with appropriate strength; may require modification during flares. Swap toward: cycling, swimming, hydrotherapy, elliptical, walking on flat surfaces and grass.
For runners specifically, see: The Complete Runner’s Guide for load management and surface selection guidance.
Current evidence-based criteria for TKA consideration: moderate to severe radiographic OA (KL grade 3 to 4), significant functional limitation and pain substantially interfering with daily life, pain of at least 3 months despite optimal management, genuine failure of adequate conservative management.
Knee replacement surgery is elective. It should only be considered after genuine failure of optimal conservative management.
A patient who has been told to “rest it,” given anti-inflammatory tablets, and waited has not had optimal conservative management. Multiple studies confirm that 20 to 30% of TKA patients have persistent pain and functional dissatisfaction post-operatively — substantially higher in patients with pre-operative central sensitisation or unaddressed psychological comorbidities. Surgery does not address sensitised nervous systems, fear avoidance, deconditioning, or obesity.
Arthroscopic surgery for degenerative meniscal tears in the context of OA: not recommended. Multiple high-quality RCTs (METEOR, FIDELITY, FIRST) found arthroscopy produces outcomes no better than sham surgery or physiotherapy. If you have been recommended arthroscopy for a degenerative meniscal tear, seek a second opinion. Also read: Do I Need ACL Surgery? How to Make the Right Decision for the broader surgery decision framework.
Clinical Pilates addresses the proximal contributors to poor knee mechanics: gluteus medius activation during single-leg loading, pelvic stability, lumbopelvic control, hip flexor/extensor balance. The Reformer equipment allows precise load calibration. Patients with significant pain who cannot tolerate conventional loaded exercise can often engage with clinical Pilates at significantly lower pain levels.
Aquatic exercise significantly improves pain, function, and quality of life in knee OA (Journal of Clinical Medicine, 2025). Buoyancy reduces effective body weight by 50 to 90%. Most effective as part of a progressive programme that transitions to land-based loading. Related: Why Clinical Pilates Should Be Your Go-To Workout.
Flat, flexible shoes reduce the knee adduction moment compared to structured, heeled shoes. High heels significantly increase medial compartment load. For everyday activity, flat, cushioned footwear with adequate width is the evidence-supported choice for medial knee OA. Lateral wedge insoles have inconsistent evidence of clinical benefit — discuss with a podiatrist in the context of specific gait assessment. Related: What Are Orthotics and How Do They Work?
OA progression is not inevitable or linear. The patients at highest risk of rapid progression are those who are obese, physically inactive, have poorly controlled diabetes or metabolic syndrome, have significant malalignment, and have early bone marrow lesions on MRI. The protective habits: regular strength training (2 to 3 sessions per week, year-round), regular aerobic exercise (at least 150 minutes per week), weight management, metabolic health optimisation, adequate sleep, regular clinical review. Related: The Sleep–Pain Connection.
Upwell Health Collective at 436 Burke Road, Camberwell manages knee OA at every stage — from early-onset OA in active people in their 40s who want to keep running, cycling, and playing sport, to severe OA in patients considering joint replacement.
Book an assessment online or contact our team directly.
Q: My x-ray says I have bone-on-bone. Is exercise safe?
A: Yes, for the vast majority of people. The 2025 BMJ network meta-analysis of 15,684 participants found no clear increase in adverse events with exercise across all OA severity levels. Loading is protective, not destructive, in the vast majority of OA presentations. Also read: Why 'Bone-on-Bone' Doesn't Mean What You Think.
Q: Should I rest when my knee is painful?
A: Rest is not the treatment for knee OA. During a flare, modify your activity — switch to cycling or swimming, reduce distance. But maintain movement. Return to your full programme as the flare settles.
Q: Will I definitely need a knee replacement?
A: Not necessarily. Many people with knee OA, including those with severe radiographic changes, manage their condition effectively for years or decades without surgery.
Q: Is cortisone injection bad for my knee in the long term?
A: Accumulating evidence suggests repeated injections may accelerate cartilage loss. An occasional injection to manage a severe flare or enable exercise engagement is a different scenario from regular injections as primary management.
Q: Is PRP worth it?
A: For patients with mild to moderate OA (KL grades I to III) who have not responded adequately to exercise alone, PRP has the strongest long-term evidence of the available injection options. It outperforms cortisone beyond the first 6 to 8 weeks. Not Medicare-funded in Australia.
Q: Can losing weight cure knee OA?
A: Weight loss cannot cure OA — structural changes are not reversible. But 7 to 10% weight loss produces significant, clinically meaningful improvements in pain and function and substantially reduces the risk of requiring knee replacement.
Q: I've been told I have a 'degenerative meniscal tear'. Do I need surgery?
A: Almost certainly not. Degenerative meniscal tears in the context of OA are overwhelmingly best managed conservatively. Multiple high-quality RCTs have found arthroscopy produces outcomes no better than sham surgery or physiotherapy.
Q: Can I still run with knee OA?
A: Often yes. The evidence suggests recreational running may not increase OA risk. High-volume running on hard surfaces warrants discussion with a sports physiotherapist about surface selection, load management, and footwear. Also read: The Complete Runner’s Guide.
Q: How long will it take to see improvement from exercise?
A: Most patients notice meaningful improvement in pain and function within 6 to 12 weeks of a consistent, progressive exercise programme. Significant gains continue over 12 to 24 weeks.
This article is for educational purposes only and does not substitute for individual clinical assessment. Information last reviewed May 2026. For personalised assessment, book with Upwell Health Collective at 436 Burke Road, Camberwell VIC 3124.