Osteoarthritis
Red Flags β investigate before managing as OA
| Red flag | Why dangerous | Action |
|---|---|---|
| Hot, red, swollen joint + fever + systemically unwell | Septic arthritis β joint-destroying emergency; destroys joint within 24β48 hrs if untreated; mortality if missed | Same-day hospital / A&E |
| Stiffness >60 min + multiple small joints + symmetrical + RF positive | Rheumatoid arthritis β early DMARD treatment prevents joint destruction; do not manage as OA | Urgent 2-week rheumatology |
| Night pain + weight loss + known cancer + age >50 | Bone metastases β X-ray + CT urgently; must be excluded | Urgent investigations + 2WW |
| Locked joint with acute severe pain | Loose body, meniscal tear, or avascular necrosis β do not manage as OA flare | Urgent orthopaedic assessment |
- Not asking about morning stiffness duration β the key OA vs inflammatory discriminator
- Not addressing the surgical expectation β patient leaves disengaged from conservative management
- Not asking about drug history for NSAID safety screening (amlodipine = CVD risk = naproxen preferred)
Immediate Action
Act now- Hot, red joint + feverSeptic arthritis β same-day hospital; joint aspiration
- Stiffness >60 min + symmetrical small jointsRA β urgent 2-week rheumatology
- Night pain + weight loss + cancerBone metastases β urgent 2WW
- Locked jointMeniscal tear/loose body β urgent orthopaedics
Failed Conservative Tx
When indicated- NRS β₯7 + failed exercise + physio + NSAIDs + IA steroidOrthopaedics for surgical consideration
- Significant functional impairment β cannot perform ADLsDocument all failed conservative interventions
OA Management
GP practice- Typical OA (age β₯45, activity-related, stiffness <30 min)Clinical diagnosis; no imaging; exercise + weight loss first-line
- Pharmacotherapy if neededTopical NSAID β oral NSAID + PPI β IA corticosteroid
- Depression comorbidityPHQ-9; duloxetine (NICE NG226 evidence)
- Surgical referral at first presentation without conservative management trial
- Not red-flag screening before assuming OA
- Not measuring weight/BMI β misses the most important modifiable factor
- Not checking for inflammatory signs β risk of labelling RA as OA
- Ordering routine X-ray for typical OA β not indicated per NICE NG226
- Not checking eGFR before prescribing oral NSAID in a patient with hypertension (CKD risk)
"Osteoarthritis is what we call wear and tear in the joint β the cartilage that cushions the bones gradually wears thinner, and the bones start to grind together more than they should. This causes pain, stiffness, and swelling. Despite what many people believe, there is a lot we can do before surgery. Exercise is actually the most effective treatment β not because it fixes the cartilage, but because strong muscles act as shock absorbers and take load off the joint. And weight matters enormously: every kilogram of body weight puts about 4 kilograms of force through the knee joint with each step. Losing 5 kilograms would take 20 kilograms of force off your knee β that's the kind of change that makes a real difference to the pain."
"My father had a knee replacement β I assumed that's what I'd need too."
"Surgery is absolutely a real option, and I'm not dismissing it. But I want to share what the evidence actually shows. Studies comparing structured exercise and weight loss programmes against knee replacement found that a significant proportion of people waiting for surgery got equivalent benefit from the programme alone. Surgery works very well when it's needed β but it has real recovery demands, usually 3 to 6 months, and its own risks. If we can get you back to playing golf without an operation, that's a much better outcome. Let's give the programme a proper 3-month try β if you're still significantly impaired after that, I will support a surgical referral."
Osteoarthritis (NICE NG226)
Age β₯45 + activity-related pain + stiffness <30 min. Clinical diagnosis. Exercise + weight loss. Topical NSAID if pharmacotherapy needed. No routine X-ray.
Gout (crystal arthropathy)
Acute monoarthritis + hyperuricaemia risk factors. Colchicine or NSAID acutely. Allopurinol after acute attack resolved (not during). Common OA coexistence.
Rheumatoid arthritis
Stiffness >60 min + symmetrical small joints + RF/anti-CCP positive. Urgent rheumatology. Early DMARD prevents joint destruction.
OA requiring surgery
NRS β₯7 + failed conservative management. Orthopaedic referral with full conservative management documentation.
Septic arthritis
Hot, red, swollen joint + fever. Same-day hospital. Joint aspiration. Destroys joint within 24β48 hrs untreated.
Bone metastases
Night pain + weight loss + known cancer. Urgent X-ray + CT. 2WW if new suspected malignancy.
- Not quantifying the weight loss benefit β vague "lose some weight" does not motivate
- Dismissing surgery entirely β patient needs to know the pathway if conservative fails
- Not checking understanding after the explanation
Validate
The expectation of surgery is entirely understandable if a family member has had a replacement. Acknowledge this explicitly before presenting the evidence.
"I completely understand why you'd expect a replacement β your father had one and got excellent results. That's not an unreasonable expectation at all."Explain
Share the evidence honestly with specific numbers. Exercise and weight loss produce significant pain reduction, often comparable to surgery.
"The research shows that a structured exercise programme and losing just 5β10% of your body weight reduces knee pain by about 50% in most people β comparable to what surgery achieves, without the 3β6 month recovery. I want to make sure you get that benefit first."Negotiate
Always leave with a specific agreement and goal. Connect the plan to returning to golf.
"Could we agree on a 3-month structured programme β exercise, weight loss target, and a physio referral. If you're still significantly impaired after that, I'll support a surgical referral with full documentation. And let's use getting back to golf as our specific goal."Strong muscles act as shock absorbers, reducing articular cartilage loading. Exercise also reduces central sensitisation and improves joint lubrication. Exercise does NOT accelerate cartilage loss β it protects it.
Low-impact: swimming, cycling, walking, aqua exercise. Quadriceps strengthening: straight leg raises, mini-squats, step-ups. Physio referral for supervised programme. Reassure: pain during exercise is acceptable; pain worsening after exercise should prompt review.
Each kg of body weight exerts 4 kg of force through the knee joint. Weight loss reduces articular loading and has an anti-inflammatory effect through reduced adipokine secretion.
Set a specific target: "5 kg in 3 months" is more motivating than "lose some weight." Structured lifestyle programme referral. Mediterranean diet evidence for OA. Commercial weight loss programme if BMI >30.
Targeted muscle strengthening, biomechanical correction, pacing strategies. Patient education changes illness beliefs β particularly fatalism β the most important determinant of long-term function in OA.
Refer all new OA diagnoses. ESCAPE Pain group exercise programme. SELF-management education. Walking aids if indicated (reduces loading 15β20%).
Heat: warm bath, heat pack for morning stiffness and post-exercise. Cold: ice pack (wrapped in cloth) 15β20 min after activity or during flares. TENS has evidence for short-term relief.
Depression prevalence ~20% in OA. Bidirectional: chronic pain β depression; depression β amplified pain. Depression predicts poor surgical outcomes. PHQ-9 at every review.
Duloxetine 30β60 mg OD: NICE NG226 evidence for OA pain + comorbid depression. NHS Talking Therapies referral. CBT for chronic pain.
Well-fitting cushioned sports shoes. Avoid high heels. Lateral wedge insoles for medial compartment knee OA. Refer to podiatry or orthotics if appropriate.
- Topical diclofenac 1% gel: apply 3β4Γ/day to affected joint
- NICE NG226 first-line pharmacological choice for peripheral OA
- Less systemic absorption β fewer GI and CVD side effects
- No PPI required β key advantage over oral NSAIDs
- Review at 4 weeks; if inadequate β add oral NSAID + PPI
- Naproxen 250β500 mg twice daily: preferred in high CVD / hypertension (lower thrombotic risk)
- Ibuprofen 400 mg three times daily: lower CVD risk patients
- Celecoxib 100β200 mg OD: high GI risk; caution in significant CVD
- PPI MANDATORY β omeprazole 20 mg OD or lansoprazole 15 mg OD always co-prescribed
- Lowest effective dose, shortest duration; review at 4 weeks
- IA triamcinolone 40 mg: acute flare; relief 4β8 weeks; max 3Γ/year; not if prosthetic joint
- Duloxetine 30β60 mg OD: NICE NG226 evidence for OA pain; central sensitisation or comorbid depression
- NOT recommended: paracetamol routinely (NICE NG226, 2022); opioids; glucosamine; acupuncture
Select characteristics β see drug cards below
"Apply a small amount directly over the knee joint 3 to 4 times a day β let it dry before covering. The advantage over tablets is much less gets into the bloodstream. You don't need a stomach-protecting tablet with this one."
In SCA: prescribing oral NSAID before topical for peripheral OA = NICE NG226 error. The mark: topical diclofenac first + no PPI needed. Jumping straight to oral NSAID loses this mark.
"Take twice daily with food β never on an empty stomach. Also take the stomach-protecting tablet I'm prescribing alongside it, every day without fail. If you develop stomach pain, dark stools, or feel very unwell, stop the tablet and contact us."
In SCA: two specific marks β (1) naproxen preferred for CVD risk patients, not diclofenac or ibuprofen; (2) PPI always co-prescribed. Both must be stated explicitly. "I'll also prescribe a stomach-protecting tablet alongside this."
"Three times a day with food. Always take the stomach-protecting tablet as well β that's not optional. Stop and contact us if you develop stomach pain or notice dark stools."
"This injection reduces the joint inflammation. You may have 24β48 hours of worsening pain initially β that's a normal reaction. After that, most people get significant relief for 4β8 weeks. We can repeat this up to 3 times a year."
In SCA: key knowledge points β max 3 IA injections/year; must not inject if septic arthritis suspected; warn diabetics about blood glucose; not for prosthetic joints.
"This tablet works on pain pathways in the brain and spine β it's not a conventional painkiller, but it changes how pain is processed. Start at 30 mg for 2 weeks, then increase. Allow 2β4 weeks to see the pain effect. Don't stop abruptly β we'd taper it."
In SCA: duloxetine for OA is a NICE NG226 (2022) recommendation β a specific knowledge point. Knowing it is indicated for central sensitisation and comorbid depression demonstrates up-to-date guideline knowledge.
"For the first 2β4 weeks you'll feel a burning sensation at the site β this is it working, not a problem. Stick with it; after a few weeks that reduces and the pain relief builds. Apply 3β4 times a day consistently."
Loss of Valued Activities
OA causes patients to give up activities central to their identity β golf, gardening, dancing, walking the dog. These losses cascade into depression, reduced activity, weight gain, and worsening OA β a vicious cycle.
Addressing the specific activity loss as a treatment goal breaks this cycle. "Getting back to golf" is more powerful than "reduce NRS by 3."
"Getting you back to golf is our goal β let's be specific about what you'd need to be able to do that."Depression and Chronic Pain
Depression prevalence in OA is ~20%. Depression amplifies pain and reduces exercise engagement β bidirectional worsening. PHQ-9 at every OA review. Duloxetine treats both simultaneously.
"How has your mood been β living with this level of pain day to day can really take a toll. I want to make sure we're not missing anything there."Fatalism and Self-Efficacy
"There's nothing that can help short of surgery" is the most common belief barrier to OA management. Directly preventing engagement with exercise and weight loss.
Motivational interviewing, specific evidence with numbers, and connecting treatment to meaningful goals are the most effective approaches.
"I want to challenge one thing β many people believe exercise makes it worse. The research actually shows the opposite β let me explain why."Work and Productivity
Manual workers with OA (this patient: retired builder) often have occupational history as a contributing cause. OA is a leading cause of early retirement. Occupational therapy referral may be appropriate.
"Has the pain been affecting your ability to do things day-to-day β I want to make sure we factor that into the plan."Sleep Disruption
Chronic pain disrupts sleep. Poor sleep amplifies pain perception and reduces physical function β bidirectional. Evening duloxetine (if prescribed) may help sleep. Sleep hygiene advice and CBT for pain/sleep comorbidity.
"Is the pain affecting your sleep? This is something we can address specifically and it makes a real difference to how you experience the pain during the day."Exercise Fear
Many patients fear that exercise will accelerate joint damage. This misbelief must be directly addressed. Exercise does NOT wear the joint out more β it protects it through muscle strengthening and load redistribution.
"I need to address something important: exercise does not wear the joint out more. The muscles become shock absorbers β the joint actually gets more protection, not less."4β6 weeks β Exercise and topical NSAID review
Engaging with exercise? Topical diclofenac adequate? Weight change documented. PHQ-9. If topical inadequate: add naproxen + PPI (for this patient with hypertension). Physiotherapy referral confirmed.
3 months β Golf goal review
Have we met the golf goal or progressing? Weight change. NRS recorded. If exercise + topical insufficient: oral NSAID + PPI. Acute flare: IA corticosteroid. Begin surgical referral documentation if NRS β₯7 despite optimised management.
6 months β Escalation or surgical referral
NRS β₯7 despite optimised conservative management β surgical referral with full documentation. PHQ-9 repeated. Duloxetine if central sensitisation or comorbid depression.
Annual β Long-term OA review
Weight; pain NRS; functional status; NSAID safety review (BP, eGFR, GI symptoms); PHQ-9; red flag rescreen. Review whether ongoing NSAID needed or topical alone now adequate.
Every 3β6 months β NSAID safety monitoring
Blood pressure (NSAIDs raise BP β important in this hypertensive patient), eGFR, FBC. Confirm PPI taken consistently. Reinforce lowest effective dose for shortest time.
β Three scenario-specific safety-net phrases
- Oral NSAID before topical first β NICE NG226 error
- Oral NSAID without PPI β prescribing error; never omit
- Diclofenac oral for hypertensive patient β naproxen preferred
- Paracetamol prescribed routinely β NICE NG226 (2022) no longer recommends this
- Surgical expectation not acknowledged and addressed
- Functional goal (golf) not set
- Closing question absent
- Clinical OA diagnosis without X-ray (NICE NG226)
- Topical diclofenac as Step 1 pharmacological treatment
- Naproxen (not diclofenac) for hypertensive patient + CVD risk
- PPI co-prescribed with oral NSAID β always
- No paracetamol routinely (NICE NG226, 2022)
- Physiotherapy referral made
- Weight loss quantified: 5 kg = 20% pain reduction / 4 kg per 1 kg lost
- Septic arthritis safety-net: hot joint + fever = A&E
- NSAID GI safety-net: dark stools = stop + A&E
- Surgical expectation validated then reframed with evidence
- Exercise mechanism explained: muscles as shock absorbers
- Weight loss mechanism: 4 kg per kg lost β specific numbers
- Golf as specific functional goal committed to
- Understanding checked
- Closing question asked
Who you are
David Chen, 62 years old, retired builder. Bilateral knee pain for 2 years, worse on the right. Pain worst on walking (stairs, slopes), at the end of the day. Morning stiffness about 15 minutes β loosens up as he moves. Has tried OTC ibuprofen occasionally with partial benefit. Stopped playing golf 6 months ago β used to play weekly with friends. Takes amlodipine 5 mg for hypertension. BMI 31 (96 kg). Non-smoker. No inflammatory signs β no small joint pain, no symmetrical pattern. No cancer history. No weight loss.
Hidden agenda β surgical expectation
David's father had a knee replacement at 60 and was "back to full function" β David assumes he needs the same and is expecting a referral today. He will not be hostile about not getting a referral if the GP explains properly. He needs: (1) the surgical expectation validated before being reframed; (2) exercise mechanism explained (muscles as shock absorbers); (3) weight loss benefit quantified with specific numbers; (4) golf specifically set as the 3-month goal. If the GP says "try some exercises" without quantifying the benefit, he mentally dismisses it. If the GP says "losing 5 kg would take 20 kg of force off your knee β the research shows about 50% pain reduction," he becomes genuinely engaged.
If asked about stiffness
- "About 10β15 minutes in the morning β then it loosens up as I move around"
- No small joint pain (hands, wrists)
- No constant night pain β pain resolves with rest
- No fever, no hot/red joint
If asked about functional goals
- "Golf β I used to play every week with my friends. I really miss it."
- "Walking the dog is also getting hard β we used to go an hour, now 20 minutes is painful"
- If golf is set as a specific target: "Do you really think I could get back to that?"
Resolution criteria: David accepts the conservative plan if: (1) surgical expectation validated ("I understand why you'd think that β your father had excellent results"); (2) exercise mechanism explained (muscles as shock absorbers β "this doesn't wear the joint out more"); (3) weight loss quantified with numbers; (4) golf set as specific 3-month goal; (5) surgical pathway not closed ("if this doesn't give sufficient benefit in 3 months, I will support a referral").
β NICE NG226: topical diclofenac is Step 1; no PPI needed for topical
β PPI is mandatory with every oral NSAID; never omit
β Naproxen preferred in CVD risk patients (lower thrombotic risk)
β NICE NG226 (2022): no longer recommended for OA
β Patient leaves disengaged from conservative management
β "Lose some weight" does not motivate; "4 kg per kg lost" does
β The single most important OA vs inflammatory discriminator
β NICE NG226: not indicated; clinical diagnosis
β Always end with "Is there anything else on your mind?"