Section 1: Epidemiology and aetiology
Knee pain is common and becomes more common with advancing age. About one in five people with knee pain also complain of hip pain.
Most of these patients have osteoarthritis and develop a varus deformity of the knee. A valgus deformity tends to be associated with rheumatoid patients. Occasionally patients suffer from other conditions such as chronic gout or previous knee sepsis.
Some patients have a history of previous injury or surgery to the knee and this may predispose the patients to later degenerative change. Many former footballers have a short incision at the side of the knee where their medial meniscus has been removed. These patients sometimes require a knee replacement at a younger than average age.
Knee pain is more directly related to obesity than hip, ankle or back pain. Advising a patient to lose weight will delay the need for surgery.
Many of my patients claim that glucosamine tablets have relieved their pain. Given that the absorption of oral glucosamine is difficult to detect from a blood sample, it is hard to imagine that such medication can do any harm.1
Some patients seem to experience improvement with various commercially available knee braces and these are not associated with any side-effects.
Since osteoarthritis tends to represent a progressive degenerative condition, when all else fails, arthroscopic surgery and/or joint replacements are still viable options.
National joint register
In recent years, the UK has acquired a national joint register in which details of all hip and knee replacements are logged in a central database.
Whilst there are some problems with achieving 100 per cent submission of cases, this might provide independent data on the success or failure of some of the more controversial procedures now being performed.
Section 2: Making the diagnosis
Knee pain may originate from three places: the knee, the hip or the spine.
Pain from the ball and socket joint of the hip is classically felt in the groin although many patients also experience hip pain in the front of the thigh or knee. Isolated knee pain may be derived purely from the hip and in cases where patients appear to have ipsilateral hip and knee arthritis the hip is usually replaced first as this may well solve both problems.
Pain from the facet joints at the back of the lumbar spine tends to be felt in the low back and radiates down the buttocks and the back of the thighs to (but not below) the knee. Pain that shoots down the whole length of the limb into the foot is usually nerve root pain (sciatica).
Knee pain that is more pronounced walking up and down stairs or rising out of a chair is likely to originate in the patello-femoral joint (the PFJ). The pain is more severe rising out of a low chair or, worse of all, rising from a squatting position.
Plain X-rays of the knee should be performed standing as this reveals loss of joint space on the film. When referring a patient for an X-ray, stipulate 'standing AP and lateral' otherwise the radiographers usually take the film with the patient lying flat.
Sky line views are useful if you suspect patellofemoral pain and an AP pelvis is worth requesting if you think the pain may originate in the hip.
Most patients have diffuse wear and tear to their knees. It is difficult to correlate the degree of damage seen on arthroscopy or X-ray to the degree of pain reported by the patient.
If this seems surprising, remember articular cartilage itself is aneural - i.e. the pain the patient feels in osteoarthritis is not coming from the degenerating articular cartilage itself.
MRI is used after acute injuries to the knee in younger patients where soft tissue damage is suspected. MRI is a good imaging modality for confirming anterior or posterior cruciate ligament ruptures or meniscal tears.
Section 3: Managing the condition
Simple analgesics are highly effective for many patients, limited only by side-effects.
Knee arthroscopy, which is usually performed as a day case, can assess the degree of damage to the knee and subsequently repair or debride various tissues. Loose bodies can usually be removed. Arthroscopic wash-out of the knee can suppress the patient's pain for several months. It is often used as a treatment although a recent study has questioned its use.2
Knee replacements were introduced later than hip replacements but we now perform more knee replacements than hips (about 60,000 knees versus 50,000 hips a year in the UK).
Total knee replacement is effective in eliminating pain.3 The range of movement in a stiff knee can also be improved, although in practice stiffness can be a more stubborn problem than pain.
About 1 per cent of patients will develop a deep infection post surgery. Deep infection is a major problem that may be solved by repeated arthroscopic wash-out and irrigation; however, one or two stage revision surgery is often required.
Thromboembolic disease is a recognised complication after knee surgery.
Knee replacement is also associated with blood loss despite being performed under tourniquet control. Postoperative acute coronary syndromes have also been reported.
It should be remembered that the majority of patients receiving surgery is elderly with a high prevalence of comorbidities.
Patients may also experience nerve injuries. In a valgus knee, there is an increased risk of common peroneal nerve injury with subsequent foot drop such as procedure will lengthen the lateral side of the joint.
If both knees need to be replaced, or indeed both knees and both hips then postoperative risks multiply. Some units have attempted to perform bilateral knee replacements using one anaesthetic although most would try to space the procedures by two or three months.
About 1 per cent of patients will require further surgery to their knee each year.
Total knee replacement involves five to seven days in hospital. The patients have a mid line incision that often knocks out a superficial nerve, leaving them with a numb patch on the front of their knee.
Whilst this complication is rarely emphasised in the literature, quite a high proportion of patients seem to be concerned about it in clinic and it is best to warn them of this in advance.
New types of replacement
As the technology has evolved, a host of new knee replacements have been developed to solve specific problems.
There are three compartments in the knee: two main compartments and the PFJ where the back of the patella articulates with the femur.
Osteoarthritis usually begins in the medial compartment and as the joint space on plain X-ray narrows, the knee acquires a varus deformity.
Unicompartmental knees have been developed for those patients where the other two compartments are believed to be well preserved (about 10 per cent).
These patients have a shorter scar and, often, recovery period than total knee replacements.4
In carefully selected patients, surgeons in Oxford have actually performed the 'Oxford knee replacement' as a day case. Their knee preserves both cruciate ligaments and it has been claimed that this leaves the patient with a more normal knee movement and position.
PFJ knee replacements
Attempts to replace the lateral compartment in isolation have had more mixed results. Where only the PFJ is affected specific PFJ knee replacements have been developed, with good results from the pioneering centres concerned.
The Avon Orthopaedic Centre in Bristol has developed such a device and some of their results are impressive. If and when these prosthesis fail they can be revised to a total knee replacement, although this does represent a revision procedure.
Computer navigated hip and knee replacements have now been developed and some units appear to achieve good results with this technique.
Section 4: Prognosis
Pain in osteoarthritis of the knee is difficult to predict. In some patients the pain mysteriously seems to wax and wane. In my own practice, I see pain as the main indication for surgery although I would be reluctant to operate on a patient with pain but no radiological changes.
If conservative measures succeed and surgery can be avoided, then I usually encourage patients to lose weight, take painkillers, and consider physiotherapy or a walking stick. Some patients also try using glucosamine.1
Because of the age at which osteoarthritis manifests, many patients are in a position to change their working practises and take on a less physically active role in the workplace.
Nowadays, surgery is considered in much younger patients although the implications of surgery should be discussed in detail.
Super specialist centres are performing procedures such as autogenous cartilage implantation in which attempts are made to recreate articular cartilage using cartilage cells from the knee.5
This is experimental and very expensive. It is probably suitable for very carefully selected younger patients who have suffered sporting injuries.
In the longer term, attempts will be made to re-grow hyaline cartilage using stem cells, although it will be many years before this becomes available.
1. Chard J, Dieppe P. Glucosamine for osteoarthritis: magic, hype, or confusion? It's probably safe - but there's no good evidence that it works. BMJ 2001; 322(7300): 1439-40.
2. Moseley JB, O'Malley K, Petersen NJ et al. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med 2002; 347(2): 81-8.
3. Kim TH, Lee DH, Bin SI. The NexGen LPS-flex to the knee prosthesis at a minimum of three years. J Bone Joint Surg Br 2008; 90(10): 1304-10.
4. Pandit H, Jenkins C, Barker K et al. The Oxford medial unicompartmental knee replacement using a minimally-invasive approach. J Bone Joint Surg Br 2006; 88(1): 54-60.
5. Glaser C, Tins BJ, Trumm CG et al. Quantitative 3D MR evaluation of autologous chondrocyte implantation in the knee: feasibility and initial results.Osteoarthritis Cartilage 2007 Jul; 15(7): 798-807.
Cutts S, Edwards A, Prince R. Primary Care Orthopaedics. RCGP, 2004.
- This topic falls under section 15.9 of the RCGP curriculum 'Rheumatology and conditions of the musculoskeletal system'