Added to Saved items
This page has been archived. It has not been updated since 29/08/2014. External links and references may no longer work.
This article is for Medical Professionals

Professional Reference articles are designed for health professionals to use. They are written by UK doctors and based on research evidence, UK and European Guidelines. You may find the Knee Replacement article more useful, or one of our other health articles.

Read COVID-19 guidance from NICE

Treatment of almost all medical conditions has been affected by the COVID-19 pandemic. NICE has issued rapid update guidelines in relation to many of these. This guidance is changing frequently. Please visit to see if there is temporary guidance issued by NICE in relation to the management of this condition, which may vary from the information given below.

Editor's note

Dr Sarah Jarvis, 20th May 2022

This article has been archived and has not been fully updated since it was last reviewed. In 2022, the National Institute for Health and Care Excellence (NICE) issued new Quality Standards on joint replacement (primary): hip, knee and shoulder.[1] Please read the guidance for more details.

Knee replacement surgery has improved dramatically over the period of a few decades because of improved techniques, improved materials and improved understanding of knee physiology and function.[2] More successful and durable knee replacement has had a big impact on clinical practice. GPs need to know something of the advances that have been made and also how clinical practice has changed as a result, particularly in the area of patient assessment and selection for surgery. GPs should be able to distinguish normal from abnormal progress after surgery.

Editor's note

Dr Sarah Jarvis, 30th August 2019

Unicompartmental vs total knee replacement in medial compartment osteoarthritis

In a new study[3] in which 528 patients with medial compartment osteoarthritis were randomised to unicompartmental or total knee replacement and followed up five years later, the two groups showed similar outcomes in respect of:

  • Pain and function as measured by the Oxford knee score.
  • Likelihood or requiring further surgical intervention.
  • Incidence of complications.

On the basis of these findings, and taking into account the lower costs and better cost-effectiveness with partial knee replacement, the authors recommend that this should be considered the first-choice treatment for patients with late-stage isolated medial compartment osteoarthritis.

There are, broadly speaking, three types of knee replacement:

Unicompartmental knee replacement

  • This involves replacement of the medial or lateral tibiofemoral compartment only with components as below.
  • It has largely replaced osteotomy.
  • It is usually used for osteoarthritis (OA).
  • It is not suitable for patients with rheumatoid arthritis (RA).

Unconstrained bicompartmental knee replacement

  • The most common form of total knee replacement.
  • The femoral component is metal (and anatomically similar to the bone it replaces, with a groove for the patella).
  • The tibial component is a flat metal platform with a plastic (polyethylene) cushion and a stem extending into the tibia.
  • The patella component (button) is a dome-shaped piece of polyethylene anchored to a flat metal plate.

Constrained bicompartmental knee replacement

  • These are used much less often when considerable bone loss is involved - for example, with bone tumours.
  • It is not possible to preserve the more physiological joint action with preservation of supporting ligaments, etc and the joint works like a fixed hinge.
  • It is much more prone to loosening.

There are over 150 knee replacement designs on the market. Choice of prosthesis depends on many factors (including age, weight, level of activity, health, cost of prosthesis and experience of the surgeon).

The aim of knee replacement is to replace the deficient joint surfaces (condyles of the femur and top surface of the tibia) with low-friction artificial surfaces (including occasionally the patellofemoral joint if affected). The best function and outcome are, in general, achieved by restoring mechanical alignment and soft tissue balance. These, with newer designs, mean taking advantage of support from posterior cruciate ligament and collateral ligaments. Components are designed so that metal (titanium or cobalt/chromium-based alloys) articulates with plastic (ultra high-density polyethylene). A complete prosthesis now weighs only about 500 to 600 g.

Osteoarthritis (OA)

The most common reason for knee joint replacement is OA. This may be:

  • Primary OA. Some 40% of 40-year-olds will have radiographic evidence of OA, of which half will have symptoms. There is a genetic risk for this, with siblings of patients having joint replacement being up to five times more likely to need similar surgery.
  • Secondary OA. This is secondary to mechanical derangement (for example, meniscal or cruciate ligament injuries), infection, instability, and fracture into the joint.

Other causes of cartilage destruction

  • Rheumatoid arthritis (RA).
  • Haemophilia.
  • Seronegative arthritis.
  • Avascular necrosis. This can be idiopathic or steroid-induced.
  • Gout and other crystal deposition diseases.
  • Rare causes, which include bone dysplasias and bone cancers (eg, osteosarcoma). The bone destruction involved usually requires the use of more unusual 'fixed-hinge' or constrained knee joint prostheses.
  • Obesity.[4]


The basic indications are pain and loss of function. However, the assessment of these is complex and it is difficult to define a clear consensus between surgeons.[5] Selection for surgery should take into account severity of symptoms, the combination or integrated effect of these symptoms and the patient's motivation to proceed.

NICE guidance suggests: 'referral for joint replacement surgery should be considered for people with osteoarthritis who experience joint symptoms (pain, stiffness and reduced function) that have a substantial impact on their quality of life and are refractory to non-surgical treatment. Referral should be made before there is prolonged and established functional limitation and severe pain.'[6] . Broadly, a holistic approach to management is encouraged by NICE.

Factors to be considered include:

  • Age. Surgeons are less likely to operate on young patients.
  • Pain. This is predominantly on weight-bearing initially and then becomes more constant. It may be generalised or localised to one compartment (usually the medial compartment). Pain can be graded either from 1 to 10, or as mild, moderate or severe. Severe daily pain with rest pain on several days per week and daily pain on transferring would help persuade surgeons to operate.
  • Functional impairment considerations:
    • How are activities of daily living affected?
    • Are walking devices or aids used?
    • What is the maximum walking distance?
    • Can the patient climb stairs?
    • Can the patient dress and self-manage?
    • What other activities requiring knee bending are impaired?
  • Sleep. Sleep may be disrupted by pain. This can lead to depression and mood disturbance.
  • Sexual and social functioning.
  • Comorbidity. OA and RA are perceived as having a good outcome from surgery.[7]
  • Technical difficulties.
  • Patient motivation.
  • Employment and maintenance of independent living.

Scoring systems have been used in research in an attempt to quantify disability and measure outcomes. These include measures of general health status and specific knee scoring systems (eg, via the Knee Society).[8, 9]


Examination should incorporate:

  • General health and assessment of any medical conditions (for example, hypertension, heart disease, diabetes, etc).
  • Knee examination:
    • Inspection. Look for scars, skin sepsis and soft tissue defects around the knee. Skin defects or scarring may affect healing.
    • Palpation. Check pulses and ensure there is no peripheral arterial disease. Arrange for further investigation if peripheral arterial disease is suspected.
    • Look for wasting of the quadriceps and check there is good extensor function.
    • Measure the range of joint movement and any fixed flexion deformity (ideally using a goniometer) and record findings. The best predictor of range of movement after operation is the range before operation.


  • X-ray (posteroanterior (PA), anteroposterior (AP), lateral, skyline view of the patella, 45° PA or Rosenberg view). These should correlate with the clinical findings. Radiographic destruction of joint space, together with symptoms, are required findings by most surgeons. Cysts, subchondral sclerosis and osteophytes are also seen in OA.[10] . The Knee Society has also endorsed a scoring system for X-rays after arthroplasty.[11]
  • MRI scan. This may be done particularly to assess cruciate ligaments and the patellofemoral compartment further.
  • Indium white cell scanning and other techniques are rarely needed before operation. In complicated revision arthroplasty, where infection needs differentiating from aseptic loosening, various techniques can be used.[12]
  • Bone densitometry. This is indicated where there are risk factors for osteoporosis.
  • Arthroscopy. This has often been performed on younger patients to evaluate symptoms, injury, etc.

Measures to delay or avoid knee replacement

Conservative measures to be considered before surgery include:

  • Adequate pain management. This should include oral analgesics. Intra-articular viscosupplementation is not recommended by NICE.[6]
  • Weight loss.
  • Physiotherapy to help with muscle strengthening, gait, patient education, joint mobility, and walking aids.
  • Occupational therapy to aid with activities of daily living: bath aids, toilet aids, grab rails, etc.
  • Walking aids (shoe adjustments, bracing of joints, etc).
  • Exercise, including muscle strengthening procedures.[13]

Absolute contra-indications include:

  • Knee sepsis.
  • Any untreated source of infection.
  • Poor knee extensor function.
  • Significant arterial disease.
  • Genu recurvatum knee deformity (secondary to muscular weakness).
  • Presence of a good knee arthrodesis.

Other conditions which cause significant difficulty and are considered relative contra-indications include:

  • Any condition increasing anaesthetic risk and good recovery significantly.
  • Skin disorders around the line of incision.
  • Past history of osteomyelitis.
  • Obesity.
  • Neurological disorders affecting musculature or joint.

It may useful for the GP to have details about the operation where they have a bearing on postoperative progress and possible complications. Such detail is seldom routinely provided but could include:

  • Method of anaesthetic. Often epidurals are used.
  • Tourniquet time.
  • Operative approach. This is usually anterior through a medial para-patellar approach (may be lateral or subvastus). Mini-incision techniques may be used. Guidance from NICE has been produced.[14]
  • Whether good mechanical alignment was achieved with good patellofemoral alignment.
  • Whether knee ligaments were significantly contracted and, if so, whether knee ligament release was necessary.
  • Whether patellofemoral surgery was required (release, reefing or even resurfacing with polyethylene).
  • Detail of the prosthesis and whether cemented or uncemented.
  • In hospital, postoperatively, early knee movement within the first 24 hours is encouraged with:
    • Good analgesia. Often patient-controlled methods. Epidural methods are also commonly used.
    • Physiotherapy. A continuous passive motion machine may be used. Exercises are taught. Most patients walk on the second postoperative day.
    • Cryotherapy, which may be used to reduce swelling.
  • Early discharge is encouraged at between five days and two weeks, depending on progress; however, but only if:
    • Wound healing is satisfactory.
    • Mobility is satisfactory.
    • Knee flexion of 90° is achieved.
    • No complications have been identified.
  • Orthopaedic follow-up is usually at about six weeks in outpatients.
  • Review in general practice is likely to be for:
    • Analgesia.
    • Wound care.
    • Reassurance.
    • Identification of complications.
    • Review of thromboembolism prophylaxis. This is often continued at home.
    • Review for return to normal activities:[15]
      • Driving: if the left knee has been replaced and the vehicle is automatic, then driving can be resumed as soon as one week after surgery. If the right knee has been replaced, 4-6 weeks off driving is to be expected.
      • Return to work. This depends on the type of work, but may take up to eight weeks.
      • Travelling. Measures to prevent thromboembolic complications are recommended. Airport metal detectors may be triggered and a medical emergency identification bracelet or similar can be useful.
      • Other activities: swimming is recommended, usually from the time the sutures are removed, and the wound has healed fully (about six weeks). Dancing, golf (with spikeless shoes and cart), and cycling (level ground) are also acceptable. Activities which stress the joint should be avoided (for example, tennis, squash, jumping, ski-ing, and jogging).
  • Thromboembolic complications, including deep vein thrombosis and pulmonary embolus.
  • Infection.
  • Patellofemoral instability and other complications.
  • Vascular complications.
  • Neurological complications. Peroneal nerve palsy is the most common neurological problem, occurring most often when correcting fixed valgus and flexion deformities. 50% recover spontaneously.
  • Aseptic loosening.
  • Fractures. Periprosthetic fractures, especially of the femur (supracondylar) can (rarely) occur.
  • Arthrofibrosis or restrictive scarring can impair knee movement.

New developments include:

  • Use of uncemented designs allowing bone to grow into a porous prosthesis and give 'biologic' fixation.
  • New joint surfaces (eg, bioactive surfaces like hydroxyapatite).
  • Mobile-bearing knee replacements are being improved. In a mobile-bearing knee replacement both the femoral and tibial components move across a polyethylene insert, creating a dual surface articulation and helping to reduce wear and loosening.
  • Improved kinematics.
  • Improved fixation.
  • Use of navigation-guided surgery. Mini-incision for knee joint replacement is another term for use of navigation-guided instruments which allow smaller incisions and less tissue destruction. NICE guidance from 2010 supports use of the procedure but only by suitably trained surgeons and only with various measures in place to audit results.[14] The prostheses used are essentially the same.

If carefully and correctly selected for surgery, a good outcome is expected. There should be relief of pain and restoration of function, even allowing patients to participate in some gentle sporting activity. Longer-term studies are showing low failure rates of around 5% after 12-15 years. Cementless prostheses have not been followed up for as long.

Editor's note

Dr Sarah Jarvis, February 2019

A new pooled registry data study in the Lancet reveals that advances in technology and surgical techniques have extended the life of both unicompartmental knee replacements (UKRs) and total knee replacements (TKRs). You can advise patients that in the region of 82% of TKRs and 70% of UKRs will last for 25 years.[16]

Reducing the need for knee replacement will depend largely on progress in the prevention of OA and perhaps an holistic approach to management of those patients with OA.[6]

Knee joint replacement has been performed for more than 50 years. Although it was attempted in the 1860s, the first artificial implants were not tried until the 1940s. Problems with postoperative pain and loosening limited success. The success with hip arthroplasty was encouraging but the complexities of the knee joint hindered similar progress. The simple hinge-like prostheses of the 1950s did not take into account the complexities of knee mechanics, and high rates of failure with aseptic loosening resulted. Infection also contributed to an unacceptable failure rate. In the 1970s a joint which took into account the more complex movement of femoral condyles over the tibia was developed by Frank Gunston (a Canadian orthopaedic surgeon from Sir John Charnley's Hip Centre) but failed through inadequate fixation of the prosthesis. By 1974 replacing the patellofemoral joint (either preserving or sacrificing the cruciate ligaments) had become the standard operation.[17] More natural kinematics can be aided by retaining cruciate ligaments.

By 1994 Ranawat reported the best results so far with 95.6% success at 14-year follow-up.[18] There was excellent durability reported for total condylar knee replacement in 1999 at 20-year follow-up.[19] In 2002 there were 19 different companies in the USA distributing knee implants used in about 130,000 knee replacements. These comprise three types (cruciate-preserving, cruciate-substituting and total condylar III (TCIII)). The success of knee replacement surgery has now equalled, and perhaps even surpassed, that of hip replacement.

Further reading and references

  • Richmond JC; Surgery for osteoarthritis of the knee. Rheum Dis Clin North Am. 2008 Aug34(3):815-25.

  1. Joint replacement (primary): hip, knee and shoulder; NICE Quality standard, March 2022

  2. Pandit H, Aslam N, Pirpiris M, et al; Total knee arthroplasty: the future. J Surg Orthop Adv. 2006 Summer15(2):79-85.

  3. Clinical and cost effectiveness of total and partial knee replacement in patients with medial compartment osteoarthritis

  4. Coggon D, Reading I, Croft P, et al; Knee osteoarthritis and obesity. Int J Obes Relat Metab Disord. 2001 May25(5):622-7.

  5. Memtsoudis SG, Hargett M, Russell LA, et al; Consensus statement from the consensus conference on bilateral total knee arthroplasty group. Clin Orthop Relat Res. 2013 Aug471(8):2649-57. doi: 10.1007/s11999-013-2976-9. Epub 2013 Apr 6.

  6. Osteoarthritis: care and management in adults; NICE Clinical Guideline (February 2014 - last updated December 2020)

  7. Woolhead GM, Donovan JL, Dieppe PA; Outcomes of total knee replacement: a qualitative study. Rheumatology (Oxford). 2005 Aug44(8):1032-7. Epub 2005 May 3.

  8. A community for measuring health outcomes using SF tools;

  9. The Health Assessment Questionnaire; Aramis

  10. Kornaat PR, Bloem JL, Ceulemans RY, et al; Osteoarthritis of the knee: association between clinical features and MR imaging findings. Radiology. 2006 Jun239(3):811-7.

  11. Total Knee Arthroplasty Roentgenographic Evaluation and Scoring System; Knee Society

  12. Love C, Tomas MB, Marwin SE, et al; Role of nuclear medicine in diagnosis of the infected joint replacement. Radiographics. 2001 Sep-Oct21(5):1229-38.

  13. Esser S, Bailey A; Effects of exercise and physical activity on knee osteoarthritis. Curr Pain Headache Rep. 2011 Dec15(6):423-30. doi: 10.1007/s11916-011-0225-z.

  14. Mini-incision surgery for total knee replacement; NICE Interventional Procedures Guidance, May 2010

  15. Activities after a Knee Replacement: OrthoInfo; American Academy of Orthopaedic Surgeons

  16. Knee replacement duration - a pooled registry data study

  17. Ranawat CS; History of total knee replacement. J South Orthop Assoc. 2002 Winter11(4):218-26.

  18. Ranawat CS, Flynn WF Jr, Deshmukh RG; Impact of modern technique on long-term results of total condylar knee arthroplasty. Clin Orthop Relat Res. 1994 Dec(309):131-5.

  19. Rodriguez JA, Bhende H, Ranawat CS; Total condylar knee replacement: a 20-year followup study. Clin Orthop Relat Res. 2001 Jul(388):10-7.