GM 43, March 2013

Up to 20 years ago osteoarthritis (OA) was described as “wear and tear” and although previous injury is a risk factor, it is now recognised that OA has an inflammatory component.

Joint pain due to OA accounts for 5% of GP consultations.1  In the UK 25% of the over 50 year old age group report knee pain lasting over three months (in the previous year) and 33% report pain at any site interfering with their activities.

Age is a risk factor for OA but the prevalence of OA is difficult to determine because x-ray changes and knee pain are weakly linked. In patients with knee pain x-ray changes occur in 15–76%. Conversely in patients with radiographic changes of knee OA, 15–81% had pain in various studies.2 Clearly patients may have OA changes on x-ray and have no symptoms, whilst others have severe debilitating pain and are prioritised for joint replacements.

To investigate the natural history of OA of the knee, x-rays were taken on a community cohort of women’s knees with a mean age of 53 years.3 At baseline 13% had x-ray evidence of knee OA and at year 15 this had increased to 48%, however half the women did not develop x-ray changes. By the time a patient is over 75 years, 80% will have x-ray changes of OA.4


In 2008, NICE published osteoarthritis national clinical guidelines for care and management in adults.”5 This offered evidence-based patient-centred pathways of care. NICE defines OA as persistent joint pain that is worse with use in a patient who is age 45 years old and over and there may be morning stiffness lasting no more than half an hour.

Other common complaints are stiffness from inactivity, sometimes called “gelling”, unsightly bony enlargement, instability, functional problems eg. in walking, kneeling and pincer grip.

Examination may reveal joint line tenderness, crepitus, a small effusion in the knee, deformity and loss of movement. Secondary nerve entrapment may occur, especially at the neck and wrist causing further symptoms and signs. The patient’s gait may be antalgic, that is a short stride with a limp to avoid weight bearing on the painful joint.

Typical deformities of OA are Heberden’s nodes at the DIP joints, Bouchard’s nodes at the PIP joints, squaring of the CMC joint, bony enlargement of the knees and bunion formation at the 1st MTP joint. The medial aspect of the knee bears most weight and as the joint narrows the patient develops a knocked knee appearance (genu valgum), which can secondarily affect weight loading at the foot and hip.

Differential diagnoses are inflammatory arthritis, trauma, fracture, sepsis, tumour, polymyalgia rheumatica hips, avascular necrosis hips, fibromyalgia, and Charcot’s joints.

Patients with chronic inflammatory arthritides may develop secondary OA of their joints. A difficult differential is chondrocalcinosis (also known as pseudogout or calcium pyrophosphate deposition disease), which can mimic OA or cause acute joint inflammation mimicking gout. Joint aspiration of effusions and x-ray changes of calcium pyrophosphate deposition in cartilage help differentiate it from OA.

In addition, patients may have more curable joint problems eg. cartilage or ligament damage or bursitis (hip bursitis, tailor’s or weaver’s bottom at the ischial bursa, clergyman’s or housemaid’s knees).

Risk factors are age, obesity, trauma, inheritance (particularly polyarticular OA with nodal hand OA), obesity, hypermobility, mechanical stress, repetitive movements, occupation, sports and leisure activities. Less commonly recognised are sickle cell anaemia, thalasaaemia, haemochromatosis, Charcot’s joints, congenital hip dislocation, Pagets, avascular necrosis, slipped femoral capital epiphysis hip, muscle weakness, past inflammatory or septic arthritis, acromegaly, alkaptonuria and Wilson’s disease.6

Mechanisms of OA and investigations

OA is a disease of synovial joints affecting the muscles, subchondral bone, capsule, articular cartilage, synovial fluid and membrane. It is characterised by the degenerative breakdown of the hyaline articular cartilage. Cartilage acts as a shock absorber, protecting the bone by distributing loading stress and reducing friction. Synoviocytes in the synovial membrane form synovial fluid, secreting hyaluronan and lubricin and also supply nutrients to the cartilage.

Over the past 30 years there have been advances in the understanding and manipulation of inflammation and immunomodulators. Cytokines are proteins used in intercellular communication and can be pro- or anti-inflammatory, also referred to as immunomodulators. GPs may recognise interferon (IF), interleukins (IL), tumour necrosis factor-a (TNF-a) and drugs like imatinib (Glivec) and infliximib (Remicade). Other inflammatory pathways work via COX-2 and production of inflammatory prostaglandin eg. PG-E2. In inflammation in early and late OA, synoviocytes secrete pro-inflammatory cytokines like IL- 1B, IL-6, TNF-a and there is increased production of proteinases and peptidases.7

Muscles secrete anti-inflammatory cytokines and muscle generated movement helps nutrient distribution to chondrocytes. In animal models mechanical loading of joints stimulates growth and development of articular cartilage and suppresses matrix proteinases up to a threshold  load when damage occurs.8 Joint motion suppresses inflammatory cytokines like IL-1B and reduces COX-2 activity. Quads weakness is a common feature of knee OA and this offers a biochemical explanation for exercise as a treatment in OA.

Early OA causes swelling of the cartilage and chondrocytes try to repair damaged cartilage by increased synthesis of proteoglycans6 but this can become hypertrophic. Later the cartilage softens and is less elastic. Flaking and clefts in the cartilage reduce the joint space, the bone is exposed and there is increased vascularity and inflammatory activity in attempts to repair the joint surface. Areas of bone develop sclerosis or cystic degeneration and subchondral cysts are a feature of OA on x-ray. Irregular areas of excessive bone, osteophytes, occur at joint margins and can reduce movement, cause nerve entrapment and fragment as loose bodies into the joint.

There are no diagnostic tests. ESR may be normal or slightly high. X-ray may show a narrow joint space, osteophytes, subchondral bone cysts and sclerosis. MRI (NMR) also shows damage to the articular cartilage. Meniscal body protrusions are highly prevalent in NMR scans in patients with tibiofemoral OA.9 In this study meniscal protrusion was more frequent in men who were older and in both men and women higher BMI was associated with higher medial meniscal protrusion. Meniscal protrusion was also associated with meniscal damage and OA on knee X-ray. Aspiration of an effusion may occasionally be needed to exclude other pathology.

Patient-centred treatment

The GP aims to give the patient understanding and education regarding OA and dispel abnormal theories and behaviour. GPs should also provide and signpost the patient to websites, leaflets and local organisations. The GP can improve patient function by encouraging them to build up muscle strength, lose weight and provide pain relief. They should be realistic about outcomes of interventions and refer the patient to a physiotherapist, occupational therapist and hospital services when required.

A study10 of British general practice did not find significant inequality in referral and management in primary care  on socioeconomic backgrounds of patients and found that GPs referred patients according to need and without bias. A further study11 revealed that patients complaining of knee pain continued to have knee pain or disability three years later whether they had consulted with a GP or not.

A main problem with medication is non adherence and lack of patient understanding about therapy. Some patients are concerned about addiction and under-medicate, others over-medicate so that they are on maximal therapy regardless of pain and activity. Others fail to appreciate that they need to anticipate and work around side-effects. Opiate use might be avoided if found to be a sedative but might be appropriate on another day or evening time. Patients need advice regarding taking laxatives if using opiates and experiencing constipation. Education may need to be repeated and enhanced through written directions and advice. GPs are in an enviable position in knowing their patients, discussing medication, side-effects, checking interactions and other medical conditions as well as increasing and reducing analgesia.

Non-medical treatment

Arthritis Research12 has a number of exercise sheets and video clips for patients to access, as does Arthritis Care.13 Patients and relatives can download them, though practices can provide the booklets from the charities for a small donation.

In patients with OA and knee pain, 33% will have a 30% reduction in pain at two years. With knee exercises an additional 12 patients (45% of people) will have less pain at two years1 so patients should be encouraged to exercise sensibly.

Weight reduction has been shown to reduce disability in patients, but not pain.

There are a number of aids and adaptations to keep active. A walking stick used in either hand improves confidence, stride length and walking rhythm. Footwear with a wide toe box prevents rubbing and ulceration of toes. Well-cushioned insoles promote comfort. Specialist running shoe fitters can offer older sports people trainers, which support and correct their stance. Part-time knee bracing (2–4 hours a day when exercising) in people with medial knee joint OA has been shown to improve pain, function and activity at 16 weeks.14

Hand splints can be useful for painful CMC joints. Tap turners, seat risers, wall handles, bathing aids etc. are all available from the occupational therapist who will undertake an assessment of the patient in their home.

TENS machines can reduce pain and stiffness in joints. Patients asking about products can be referred to This is a charity that gives consumer reviews for the older and disabled person for items such as powered scooters and a range of appliances. Patients should be given advice to prevent osteoporosis and osteomalacia.



It is not fully understood how paracetamol works, but it appears partially to be a COX inhibitor, reducing certain PG levels. In trials one in seven patients benefit from paracetamol use.1

Topical NSAIDs 

After paracetamol, NICE5 recommends adding a topical NSAID. Trials have shown benefit but most are only three months long. This allows some relaxation of the muscles through rubbing, which may encourage the patient to improve the movement of the joint. They also they have a low incidence of side-effects compared to oral NSAIDs. Ibuprofen 10% is of similar cost to 5% gel. Unfortunately some patients request bathfuls of topical NSAIDs and it is unlikely they are efficacious in those individuals. The GP needs to limit the amount of gel as part of gatekeeping NHS costs.

Capsaicin is derived from the chilli pepper but its irritant effect can be limiting. For some patients it acts as a local analgesic and is helpful. It is recommended at no more than a 45G tube per month and costs at present over twice that of topical ibuprofen.

Opiate analgesic 

Codeine is the most frequently prescribed and the patient can increase and reduce the dose according to their expected activity and side-effects if it is prescribed with paracetamol but not in a combination tablet. Side-effects are nausea 30%, constipation 23%, dizziness 20%, somnolence 18% and vomiting 13%.


Night pain

Longer acting medications to last through the night are dihydrocodeine modified released (MR) 60mg nocte or tramadol MR 100mg nocte. Tricyclic medications, amitriptyline or nortriptyline in low dose may alternatively be helpful. There is little evidence however for nortriptyline in peripheral joint arthritis.


Inhibition of COX-1 and COX-2 prevents PG synthesis. COX-2 inhibitors cause less GIT side-effects. They help pain and stiffness but do not modify the disease. In short term use for inflammatory flares they can be useful. GPs are well aware of the problems of stomach ulceration, hypertension, renal and heart failure and ankle oedema with NSAIDs. There are a number of other contraindications eg. not prescribing with warfarin, aspirin or clopidogrel. NICE recommends prescribing as short a course as possible at as low a dose as possible and co-prescribing a PPI. Long-term use of NSAIDs should be avoided.

Joint injections and surgery

Intra-articular steroids can lead to a small improvement in pain. A steroid injection in the knee improved pain at one week and lasted six weeks and in the hip pain relief lasted 12 weeks.6 Side-effects are small but an injection of steroid may damage the articular cartilage and so injections are limited to three a year. They may cause joint flare, introduce sepsis and cause local skin atrophy.

Joint replacement and PROMs

The NHS PROMs are patient recorded outcome measures introduced by the government in 2009 to measure outcomes of four common surgical procedures, including hip and knee replacements. Patients are asked about their function and general health through a number of measures eg. EQ-5D Index score, EQ-VAS score, Oxford hip and knee scores preoperatively and postoperatively. For instance the EQ-5D index score assesses mobility, self-care, usual activities, pain or discomfort and anxiety and depression. In 2009–2010, 95% of hip replacement and 91% of knee replacement respondents recorded improvements post operatively. Also 87% of hip and 78% of knee respondents recorded an increase in general health post operatively and 62% of hip and 50% of knee replacements recorded an increase in EQ-VAS.15

Knee and hip replacements involve replacing the joint surface and replacing it with a prosthesis secured by cement or bone in-growth. Prostheses usually last 8–15 years. 25% of patients with hip OA have a hip replacement in four years. Hip overall operative 30 day mortality is 0.5%.

Controversies in therapy

Glucosamine continues to be bought over the counter by patients. It is naturally present in the shells of shellfish, animal bones, cock combs and bone marrow; it is also present in some fungi, such as Aspergillus niger.16 It is an abundant monosaccharide and helps form glycosaminoglycans, a constituent of cartilage. It was not advised by NICE in 2008 as core GP treatment but a review of trial data suggests that some patients had mild-moderate pain relief over six months using glucosamine sulphate 1500mg/day. NICE therefore suggested that patients keen to try glucosamine sulphate could use this dose for three months to see if there is benefit. There were no significant adverse effects but the BNF suggests use in caution in patients with glucose intolerance and diabetes (monitor glucose control) and in those with heart disease (monitor cholesterol).

The price is vastly variable in the high street and online from £20 a month to £10 per annum. In 2009 a Cochrane review17 of glucosamine use reported that glucosamine use caused a small reduction in pain and increased function when using a Rotta preparation. Chondroitin and glucosamine with chondroitin did not show significant benefits to make NHS prescribing cost effective and are also not recommended by NICE in NHS prescription.

Intra-articular viscosupplementation

Sodium hyaluronate is injected into the joint and has been shown to be safe and effective, but like steroid injection it can cause joint flare and sepsis. Patients need to attend on multiple occasions and it has been shown to improve pain up to three months. It is not recommended by NICE due to the cost of NHS treatment with respect to benefit gained.


NICE do not recommend arthroscopic lavage and debridement in knee OA unless there is a history of knee locking. It still has a place in repair of menisci and removal of loose bodies causing locking.



Disease modifying drugs of OA are in research and early use but not yet  in the UK. The antibiotic, doxycyline, has been researched as a DMOAD as in animal trials. It reduces cartilage loss by inhibiting proteases.7 However human trials have not shown a significant improvement in symptoms.  Avocado-soya bean extracts have an anti-inflammatory action on chondrocytes to prevent cartilage breakdown. Statins have been shown to inhibit proteases. Calcitonin and vitamin D may have an action in OA.

There are a number of research projects looking at different methods of delivering drugs into the OA joint. Diacerein4 and its active derivative, rhein, are IL-1beta inhibitors derived from Cassia angustifolia shrubs from South India. The leaves contain this inhibitor which has moderate anti-inflammatory, analgesic and weak laxative effects (the leaves are also used to collect senna). Diacerein and rhein inhibit collagenase and reduce fibroblast activity in synovial fluid. They do not affect PG and so do not have gastric mucosal adverse effects. They are slow acting. A Cochrane review found studies of two months to three years in length, diarrhoea was the most common side effect and concluded that there was some small consistent benefit in OA with a possibility of slowing damage in the joints.

Metal on metal hip replacements have become  controversial over the last year. The British Orthopaedic Association18 has recommended that these hip replacements are stopped as some have failed more frequently than the traditional metal on polyethylene implants. In addition there are concerns about cobalt and chromium levels in patients’ blood, the significance of which is not clear.


Knowledge of the disease processes of OA have advanced in the past 30 years, but there are not yet the advances in therapeutics needed to keep the elderly with OA pain free and functioning optimally. In 2008 NICE produced its evidence based guide on best treatments and we await breakthroughs in therapy.


Conflict of interest: none declared



1.  Porcheret M, Healey E, Dziedzic K, et al. Osteoarthritis: a modern approach to diagnosis and management. Reports on the Rheumatic Diseases. Series 6. Autumn 2011. Hands On No 10

2. Bedson J, Croft PR. The discordance between clinical and radiographic knee osteoarthritis: a systematic search and summary of the literature. BMC Musculoskelet Disord 2008; 9: 116

3.  Leyland KM, Hart DJ, Javaid MK, et al. The natural history of radiographic knee osteoarthritis: A fourteen-year population-based cohort study. Arthritis & Rheumatism 2012 64: 2243–51 doi: 10.1002/art.34415

4.  Fidelix TS, Soares B, Fernandes Moça Trevisani V. Diacerein for osteoarthritis. Cochrane Database of Systematic Reviews 2006, Issue 1. Art. No.: CD005117. DOI: 10.1002/14651858.CD005117.pub2.

5. OSTEOARTHRITIS National clinical guideline for care and management in adults. 2008 full guideline.

6.  Lozada CJ, et al.  Osteoarthritis accessed online July 2012

7.  Rainbow R, Ren W, Zeng Li. Inflammation and Joint Tissue Interactions in OA: Implications for Potential Therapeutic Arthritis [2012, 2012: 741582]

8. Sun B, Cardoso HB, Yokota H. Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders 2011; 4 :65–70

9.  Svensson F, Felson DT, Guermazi A, et al. Meniscal protrusion on knee MRI in the general population: Association with age, sex ,body mass index and radiographic osteoarthritis. 2012 Osteoarthritis and Cartilage 2012; 20. 213–14

10.       Jinks C, Vohora K, Young J, et al. Inequalities in primary care management of knee pain and disability in older adults: an observational cohort study.Rheumatology (Oxford). 2011; 50(10): 1869–78

11.       Blagojevic M, Jinks C, Jordan KP. The influence of consulting primary care on knee pain in older people: a prospective cohort study. Ann Rheum Dis 2008; 67(12): 1702–9

12. professionals accessed July 2012.

13. accessed July 2012

14.       Müller-Rath R, Cho HY, Siebert CH, Miltner O. Clinical and gait analytical investigation of valgus knee bracing in therapy for medial degenerative joint disease of the knee. Z Orthop Unfall. 2011; 149(2): 160–65

15. Accessed February 2012

16.       Wikipedia glucosamine. Accessed February 2012

17.       Towheed T, Maxwell L, Anastassiades TP, et al. Glucosamine therapy for treating osteoarthritis. Cochrane Database of Systematic Reviews 2005, Issue 2.

18.       British Orthopaedic Association website accessed July 2012http:/