Osteoarthritis
The most common joint disease and one of the leading causes of disability, yet the treatment most residents reach for first is the one with the weakest evidence. Exercise outperforms every drug in the algorithm, imaging correlates poorly with symptoms, and the supplements patients are already taking do not work. The diagnosis is clinical; the job is function, not radiographs.
🔍 Overview
What It Is (and what it is not)
Osteoarthritis is whole-joint failure -cartilage loss with subchondral bone remodeling, osteophytes, synovial inflammation and periarticular muscle weakness -not simple "wear and tear," which is why load, alignment, obesity and muscle strength all modify it and why exercise treats it. Risk rises with age, obesity, prior joint injury, female sex, and joint-loading occupations.
The Clinical Signature
- Activity-related pain that improves with rest, worsening through the day -the inverse of inflammatory arthritis.
- Morning stiffness under 30 minutes (inflammatory arthritis runs longer, typically over an hour) and brief gelling after inactivity.
- Bony enlargement, crepitus, restricted range, and little or no warmth -a hot, boggy, swollen joint is not plain OA and needs another explanation.
- Typical joints: knees, hips, first CMC (thumb base), first MTP, DIPs (Heberden nodes) and PIPs (Bouchard nodes), and the cervical and lumbar spine.
- Joints that argue against OA: MCPs, wrists, elbows, ankles and shoulders -disease there suggests inflammatory arthritis, hemochromatosis, or CPPD, and should prompt a rethink rather than an OA label.
Imaging and symptoms diverge in both directions, which is the single most useful thing to tell a patient. Many people with severe radiographic changes have no pain, and many with disabling pain have modest films. That is why radiographs do not grade the problem, guide most decisions, or need repeating, and why "bone on bone" as a phrase does more harm than good: it convinces patients their joint is destroyed and that movement will damage it further, which is precisely backwards.
🧪 Diagnosis
Diagnose Clinically
- In a patient over 45 with activity-related joint pain and morning stiffness under 30 minutes, the diagnosis can be made without any imaging or bloodwork. Tests are for excluding alternatives, not for confirming OA.
- Radiographs when the diagnosis is uncertain, the presentation is atypical, or surgical referral is being considered (the surgeon needs films, the diagnosis does not). Weight-bearing views for knees; expect joint space narrowing, osteophytes, subchondral sclerosis and cysts.
- MRI is almost never indicated for suspected OA -it finds meniscal tears and cartilage lesions that are near-universal with age and rarely change management, while generating anxiety and low-value surgery.
- Labs are normal in OA and are ordered only to chase an alternative: ESR/CRP, RF/anti-CCP, uric acid where the story fits. A normal-appearing inflammatory marker panel does not need repeating in established OA.
Do Not Miss
- Septic arthritis -a hot, acutely swollen joint with fever or systemic illness is an aspiration, not an OA visit. This is the emergency hiding in a "flare."
- Crystal disease -gout and CPPD both present as acute monoarthritis in joints that also get OA; aspirate when the joint is acutely inflamed.
- Inflammatory arthritis -prolonged morning stiffness, small-joint symmetry, MCP or wrist involvement, systemic symptoms; get RF/anti-CCP and refer.
- Referred and regional causes -hip OA presents as groin pain, and knee pain can be referred from the hip, so examine the joint above and below. Trochanteric pain is usually gluteal tendinopathy or bursitis, not hip OA.
- Avascular necrosis and fracture -consider with steroid use, alcohol, sickle cell disease, or pain disproportionate to findings.
🚨 Management
Core Treatment: the Non-Drug Measures Come First Because They Work Best
- Exercise is the most effective intervention in osteoarthritis, drug or otherwise -strengthening, aerobic and range-of-motion work all help, and the best exercise is the one the patient will keep doing. Counsel explicitly that activity does not damage the joint: fear of "wearing it out" drives the deconditioning that worsens pain.
- Weight loss for overweight patients with knee or hip OA -benefit is dose-dependent, with roughly 5% producing symptom improvement and larger losses producing more, partly because each pound removed subtracts several pounds of force across the knee.
- Physical therapy for a supervised program, gait and strengthening; walking aids used on the opposite side; braces for unicompartmental disease; appropriate footwear.
- Self-management education and thermal modalities round out the core; tai chi has good evidence in knee and hip OA.
Drugs, in the Order the Evidence Supports
- Topical NSAIDs are first-line for knee and hand OA -comparable analgesia to oral NSAIDs at the knee with a fraction of the systemic GI, renal and cardiovascular risk, which makes them the obvious opener in older patients. Underprescribed almost everywhere.
- Oral NSAIDs -most effective oral analgesic class, used at the lowest effective dose for the shortest period, with explicit attention to GI risk (add a PPI in high-risk patients), renal function, blood pressure and heart failure, and the interaction with anticoagulants.
- Acetaminophen -modest, arguably minimal, benefit in OA; reasonable as an adjunct or where NSAIDs are contraindicated, but it should not be presented to the patient as the main therapy.
- Intra-articular glucocorticoid injection -short-term relief measured in weeks, useful for a flare or to enable a rehabilitation program. Repeated frequent injections in the same knee are discouraged.
- Duloxetine -helpful in chronic OA pain, particularly with widespread pain or coexisting depression.
- Opioids are not recommended for chronic OA pain -small analgesic benefit, substantial harms, and no functional advantage over NSAIDs; tramadol carries the same concerns plus serotonergic and seizure risk.
What Does Not Work (say it kindly, but say it)
- Glucosamine and chondroitin -not recommended; trials show no meaningful benefit over placebo. Patients are usually already taking them, so the conversation is unavoidable.
- Hyaluronic acid injections -not recommended for knee OA in current guidance; benefit is at best marginal.
- Platelet-rich plasma and stem cell injections -not recommended; heavily marketed, poorly supported, and often expensive out of pocket.
- Arthroscopic surgery for degenerative meniscal tears and knee OA -no benefit over sham or conservative care, and it remains a common low-value referral.
When to Refer for Surgery
- Total joint arthroplasty for knee or hip OA when pain and functional limitation persist despite an adequate trial of the measures above, and quality of life is meaningfully impaired -it is among the most effective operations in medicine for the right patient.
- The decision is driven by symptoms and function, not by the radiograph, which is why "the X-ray looks bad" is not by itself a referral criterion.
- Optimize first: weight, glycemic control, smoking cessation and prehabilitation improve outcomes and reduce complications.
🏥 Rounds
Pimp Questions
❓ Why is exercise the first-line treatment for osteoarthritis when the cartilage is already damaged?
Because OA is whole-joint failure, not just worn cartilage: periarticular muscle weakness, poor proprioception and deconditioning all contribute to pain and disability, and all are modifiable by exercise. Strengthening the quadriceps offloads and stabilizes the knee, aerobic conditioning improves function and mood, and activity does not accelerate joint damage. The belief that movement "wears the joint out" is the main barrier, so correcting it explicitly is part of the prescription -it drives the very deconditioning that makes the pain worse.
❓ A patient's knee X-ray shows severe joint space narrowing but they have mild symptoms. What do you do?
Treat the patient, not the film. Radiographic severity and symptom severity correlate poorly in both directions -severe changes with no pain and disabling pain with modest films are both common. Management is driven by pain and function, so mild symptoms get core non-drug therapy and topical NSAIDs, not a surgical referral. Avoid the phrase "bone on bone": it convinces patients the joint is destroyed and that movement will worsen it, producing exactly the deconditioning and catastrophizing that make outcomes worse.
❓ Which joints, if involved, should make you doubt the diagnosis of osteoarthritis?
MCPs, wrists, elbows, ankles and shoulders. Primary OA favors knees, hips, first CMC, first MTP, DIPs and PIPs, and the spine. Disease in the "wrong" joints suggests inflammatory arthritis (RA involves MCPs and wrists), hemochromatosis (classically second and third MCPs), or CPPD (wrists, knees). Combined with morning stiffness over an hour, symmetric small-joint swelling or systemic features, it should trigger inflammatory serologies and referral rather than an OA label.
❓ Why are topical NSAIDs preferred over oral NSAIDs for knee OA in an older patient?
Comparable analgesia at the knee with far lower systemic exposure, so the GI bleeding, renal impairment, blood pressure elevation, heart failure decompensation and anticoagulant interaction that make oral NSAIDs risky in older patients are largely avoided. That is why current guidance places topical NSAIDs first-line for knee and hand OA. They work best in superficial joints (knee, hand) and less well in the hip, which is deep. Despite the favorable profile, they remain substantially underprescribed.
❓ A patient asks about glucosamine, hyaluronic acid injections, and PRP. What is the evidence-based answer?
None are recommended. Glucosamine and chondroitin do not outperform placebo in good-quality trials. Hyaluronic acid injections are not recommended for knee OA in current guidance, with marginal benefit at best. PRP and stem cell injections lack supporting evidence and are frequently costly out of pocket. The productive move is to redirect: name what does work (exercise, weight loss, topical NSAIDs, PT) rather than only listing what does not, since patients arrive already invested in these products.
Sample Presentation
Mrs. Iqbal is a 63-year-old woman with BMI 32 and 18 months of right knee pain, worse with stairs and by the end of the day, better with rest. Morning stiffness lasts about 15 minutes. Exam: bony enlargement, crepitus, mild effusion without warmth, and quadriceps atrophy. No fever. She takes ibuprofen most days, has hypertension and CKD stage 3, and takes glucosamine bought online. An X-ray done elsewhere was reported as "bone on bone," which has frightened her into avoiding walking.
Key Points: (1) This is clinically classic knee OA -no further imaging or serology needed. (2) The most important intervention is the conversation: reframe "bone on bone," reassure that activity does not damage the joint, and prescribe exercise and PT as therapy, with a weight-loss target where each pound removed subtracts several pounds of knee force. (3) Her daily ibuprofen is actively dangerous with CKD 3 and hypertension: switch to a topical NSAID, which gives comparable knee analgesia with minimal systemic exposure, and stop the glucosamine while naming what does work instead. Reserve an intra-articular steroid for a flare that blocks her rehab, and refer for arthroplasty on symptoms and function, not on the radiograph.
Clinic Checklist
- Diagnosis made clinically -imaging ordered only for uncertainty, atypia, or surgical planning; no MRI for routine OA.
- Alternatives excluded: hot joint aspirated, inflammatory pattern serologies sent, hip examined when the knee hurts.
- Exercise prescribed as therapy, with explicit reassurance that activity does not damage the joint.
- Weight-loss target given where relevant, with the load-multiplier explanation.
- Topical NSAID tried before oral for knee and hand disease, especially in the elderly and the renally impaired.
- Oral NSAID risk assessed (GI, renal, cardiovascular, anticoagulant) and gastroprotection added where indicated.
- Supplements and injections addressed honestly, redirecting to what works.
- Surgical referral driven by function, with weight, glycemia and smoking optimized first.
📋 Summary
Summary
Diagnose It Clinically
Over 45, activity-related pain that eases with rest, morning stiffness under 30 minutes, bony enlargement and crepitus without warmth -no imaging or labs required. Tests exist to exclude alternatives, not to confirm OA. Radiographs only for uncertainty, atypical presentations, or surgical planning; MRI is almost never indicated and mostly finds age-universal changes that generate anxiety and low-value surgery.
Imaging Does Not Equal Symptoms
Severe films with no pain and disabling pain with mild films are both common, so management follows function, not radiology. Retire the phrase "bone on bone": it persuades patients the joint is destroyed and that movement is dangerous, driving the deconditioning and fear-avoidance that worsen outcomes.
Check the Joints Involved
OA favors knees, hips, first CMC, first MTP, DIPs and PIPs, and the spine. MCPs, wrists, elbows, ankles and shoulders argue against it and toward inflammatory arthritis, hemochromatosis or CPPD. Add prolonged morning stiffness, symmetric small-joint swelling or systemic features and the answer is serologies and referral, not an OA label.
Exercise Beats Every Drug
Exercise is the single most effective intervention, because OA is whole-joint failure with periarticular weakness and deconditioning, all modifiable. The best program is the one the patient sustains. Say explicitly that activity does not damage the joint -that belief is the main barrier and it produces the weakness that worsens pain.
Weight Loss Is Mechanical Therapy
Roughly 5% loss improves symptoms in knee and hip OA, with more benefit at larger losses, partly because each pound removed subtracts several pounds of force across the knee. Pair it with PT, walking aids used on the opposite side, braces for unicompartmental disease, and self-management education.
Topical NSAIDs Before Oral
First-line for knee and hand OA: comparable analgesia to oral NSAIDs with a fraction of the systemic risk, which matters most in exactly the older, renally impaired, anticoagulated patients who have OA. They work best in superficial joints and less well at the hip. Oral NSAIDs remain the most effective oral option but need lowest-dose, shortest-duration discipline plus GI, renal, blood pressure and heart failure assessment.
Know What Does Not Work
Glucosamine and chondroitin, hyaluronic acid injections, PRP and stem cell injections are not recommended, and arthroscopy for degenerative meniscal tears or knee OA is no better than conservative care. Patients often arrive already committed to these, so the effective conversation names what does work rather than only listing what does not. Opioids are not recommended for chronic OA pain.
Refer on Function, Not Film
Total joint arthroplasty is among the most effective operations in medicine for persistent pain and functional limitation despite adequate conservative therapy -but the trigger is impaired quality of life, not the radiograph. Optimize weight, glycemic control and smoking first, since prehabilitation improves outcomes and lowers complications. Steroid injections buy weeks and are best used to unlock a rehab program.
