Show Summary Details
Page of

Total knee replacement 

Total knee replacement
Total knee replacement

William J. Long

and Giles R. Scuderi

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2020. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 10 August 2020

Summary points

  • Modern total knee replacement is a safe, effective, durable, and reproduceable solution to address symptomatic degenerative joint disease of the knee.


The era of modern total knee arthroplasty (TKA) has undergone numerous design changes over the last three decades. It is an effective procedure that reduces pain and improves function. Despite multiple advances, surgical management and successful clinical outcomes continue to rely on the basic principles of accurate soft tissue balancing, well positioned components, and a neutrally aligned mechanical axis.


A complete history and physical examination are important in localizing pain and disability in the degenerative knee. Initial management may include non-steroidal anti-inflammatory drugs (NSAIDs), intra-articular injections, and possibly surgical arthroscopy. Other important modalities often overlooked in early management include an exercise programme and weight loss, both of which have been shown to be of benefit to patients with symptomatic arthritis.

Cemented TKA remains the gold standard for surgical management of symptomatic unicompartmental, bicompartmental, and tricompartmental arthritis of the knee that fails non-operative management. Predictable and durable results have been achieved in older and younger patient populations. Gender differences in preoperative functional decline and pre-existing quadriceps weakness suggest that a more aggressive approach and earlier intervention may be indicated when addressing females with progressive knee arthritis.

Athletic activity and high demand following TKA is a growing area of interest as little is known about the impact of these activities on long-term survivorship. Surgeon surveys have demonstrated that for the most part, moderate non-impact activities were recommended, although whether these guidelines are followed by younger, healthy, more active individuals is unknown. Implantable monitors may provide further insight into the forces in a total knee replacement following surgery, allowing surgeons to develop more scientific basis for these recommendations.

Design considerations


Concern arose in total hip fixation when the well-designed cemented total hip system popularized by Charnley was applied to younger patients. Multiple centres reported higher rates of failures in this younger cohort which prompted the development of uncemented surfaces. Forces at the cement bone interface in the hip include a combination of shear, tension, and compression. This is distinctly different from the total knee setting, where the interface is primarily subjected to compressive forces, for which a cement interface is much better suited.

Long-term comparative studies of cemented versus uncemented total knees have shown superior survivorship in the cemented group with some designs. Other studies have demonstrated equivalent results. No studies have demonstrated superior survivorship with uncemented designs.

We continue to use a cemented, posterior stabilized design and have had excellent results in both our large outcome series, survivorship study, and in those selected patients less than 50 years old (Figure 8.6.1). A similar result, with greater than 95% survivorship at 10 years has also been noted elsewhere in younger patients.

Fig. 8.6.1 A) Anteroposterior and B) lateral radiographs of a cemented IB II TKA functioning well at 15-year follow-up.

Fig. 8.6.1
A) Anteroposterior and B) lateral radiographs of a cemented IB II TKA functioning well at 15-year follow-up.

Antibiotic cement

There are currently a number of commercially available, pre-mixed antibiotic cements for prophylactic use in total joint arthroplasty. Results from the Norwegian Arthroplasty Registry revealed at 1.8 times higher rate of infection when only systemic antibiotic therapy was used in cemented total hip arthroplasty. One randomized study in cemented primary total knees, demonstrated no infections with antibiotic cement, versus a 3.1% rate in total knees with standard cement. All infections occurred in patients with diabetes mellitus. These results provide evidence for the use of antibiotics in high-risk groups and have created debate as to the benefit of routine use of low-dose antibiotic cement in primary TKA.

Posterior cruciate ligament management

Excellent clinical results have been reported with both modern posterior cruciate ligament-retaining and posterior stabilized designs. A meta-analysis of eight randomized trials comparing the two techniques failed to demonstrate a difference in outcomes.

Some studies have shown superior outcomes with posterior stabilized components in specific subsets of patients. Rheumatoid knees have been shown to have a higher failure rate with one posterior cruciate-retaining design, though more recent studies have presented more favourable results with other posterior cruciate ligament-retaining designs. Patients with a planovalgus foot deformity are also at increased risk of failure when a posterior cruciate ligament-retaining component is used.

Fixed versus mobile bearing

Multiple studies have examined the clinical outcomes of mobile versus fixed bearing designs and have failed to demonstrate a benefit to the mobile bearing design. Bearing dislocation remains a risk with this design if the knee is not appropriately balanced. One of the strongest arguments for mobile bearings is that they are more forgiving of component malrotation, and thus decrease patellar based complications, but again this has not been shown clinically. A second theoretical benefit of mobile bearings is decreased wear. An early retrieval study has demonstrated increased polyethylene wear with this type of bearing, but this has not yet been confirmed in clinical outcome studies.

Articular conformity

We currently employ a conforming, high flexion design, with highly cross-linked polyethylene, for the vast majority of our primary TKAs. Our early outcome studies with this new design have been promising, as have in vivo biomechanical studies, but long-term follow-up does not yet exist.

Designs with flat-on-round articulations created point and edge loading resulting in increased failure rates. The current high flexion fixed bearing designs employed today involve a conforming articular tibial geometry. The tibial component allows a higher degree of flexion without posterior edge loading, and also has an anterior recess for the patellar component as the knee goes into deep flexion.

Patellar resurfacing

Improvements in implant design have resulted in a significant decrease in patellar component-related modes of failure in TKA. Modifications include more anatomical trochlear geometry, tibial polyethylene recession, and the redesign of metal-backed patellae. It must be emphasized that although metal backing has been associated with high rates of failure with some designs, this is not always the case and is likely to be design specific.

A number of randomized clinical trials exist comparing patellar resurfacing versus non-resurfacing in modern designs with no significant difference in outcomes shown. The two longest demonstrate no clinical difference at a minimum of 10-year follow-up.


Tibial modularity was introduced for a number of reasons: biomechanical studies showed improved stress distribution with a metal-backed design; lower rates of loosening were observed in some early metal-backed designs versus a similar monoblock polyethylene component; and the option for isolated liner exchange existed if symmetric laxity or polyethylene wear occurred. However, recent long-term follow-up results have begun to question the benefit of modularity, as clinical results have demonstrated significant wear rates with some modular implants, although this is likely design, preparation (sterilization), and technique specific as other modular designs have not demonstrated increased wear.

Recent reconsideration of the monoblock design has been prompted by a number of long-term studies demonstrating excellent survivorship. A 98% survival at 14 years in patients over 75 years of age and no failures in a younger active population with monoblock designs were reported in recent studies. Other authors have not only noted excellent survival, but also a significant cost savings of over 700 USD per case.

A new design consisting of compression-moulded polyethylene on a porous tantalum base plate has introduced the possibility for long-term biological fixation into the porous metal surface. Basic science research involving porous metal has shown a number of benefits including its bone ingrowth qualities, the low modulus of elasticity, and initial friction fit. Early clinical results have been presented, but no long-term follow-up studies have been published.

Gender-specific designs

Review of the National Joint Registry for England and Wales demonstrated a lower level of satisfaction in females following TKA. Other authors have confirmed this observation with increased rates of pain and stiffness in radiographically well fixed and well aligned components. Anatomical studies as far back as 1996 have demonstrated important size and alignment discrepancies between male and female knees. The combined results of these anatomical and clinical studies have lead to gender-specific femoral component designs (Figure 8.6.2). Our early surgical results were reviewed, and demonstrated a statistically significant reduction in lateral patellar release rates in women with a modified femoral design. Clinical results following the introduction of these modifications are not yet available.

Fig. 8.6.2 A) Anteroposterior and B) lateral radiographs of a modern cemented Gender High Flex LPS TKA.

Fig. 8.6.2
A) Anteroposterior and B) lateral radiographs of a modern cemented Gender High Flex LPS TKA.

Surgical technique


A significant body of literature has recently been published regarding less invasive surgical techniques in TKA. Advances in perioperative management associated with these new protocols include preoperative education, multimodal pain management, rapid mobilization, and less invasive surgical techniques. Most studies have shown a steep learning curve, with limited early, or no benefits and more complications with mini-approaches. We favour a mini medial parapatellar approach as it is familiar to surgeons, can be gradually introduced, and can be extended to a formal medial parapatellar approach if necessary.

Ligament balancing

When alignment is not sufficiently corrected, due to poor bony cuts, or ligament imbalance, the components will be unequally loaded and subjected to excessive stress, resulting in eventual loosening of the prosthesis. Despite significant advances in component design and instrumentation, TKA remains a soft tissue balancing procedure. When a fixed angular deformity exists, the ligament on the concave side is contracted, and the other on the convex side is stretched. It is very rare for complete incompetence to exist. The principle of balancing of the collateral ligaments is to release the contracted side to the length of the stretched side.

Varus knees may require release of contracted medial structures including the superficial medial collateral ligament and the pes anserinus. Following standard bone cuts including a conservative tibial cut, a titrated subperiosteal release is preformed until gap balancing is achieved. When a varus–valgus imbalance exists, it is important to achieve soft tissue balancing prior to recutting bone.

Valgus knees are addressed by similarly releasing the contracted lateral structures. In this case our preferred technique is to ‘pie crust’ lateral structures, again in a titrated manner. Structures released include the arcuate ligament, lateral collateral ligament, and the iliotibial band. The popliteus tendon is preserved. Care must be taken when performing this procedure in the posterolateral region of the knee due to the proximity of the peroneal nerve. Good clinical success has been demonstrated with this technique in a number of studies. In selected elderly patients with low physical demands, a constrained condylar knee prosthesis without ligamentous release is another option with no failures reported at an average 7.8-year follow-up.


Rotational malalignment in TKA is associated with an increased rate of pain, wear, and patellar instability together resulting in mechanical failure. The most reliable reference for femoral component rotation is the epicondylar axis, which averages three degrees of external rotation with respect to the posterior condylar axis. Special consideration must be taken when simply referencing the posterior condylar axis when a hypoplastic lateral femoral condyle, or varus joint line obliquity exists.

On the tibial side, two popularly used techniques exist for establishing rotation. The first is to apply the tibial template and then flex and extend the knee allowing the tibial component to find the appropriate rotation, and then fix it in that position. The second method employs external landmarks including the medial third of the tubercle, the shaft of the tibia, and the second metatarsal. We employ the latter technique, and its superiority has been demonstrated in a recent comparison study.


Significant advances have been achieved in surgical navigational tools. Studies have shown improvements in femoral component alignment alone; both tibial and femoral component alignment; and component rotation. Though this improved accuracy should be reflected in improved outcomes, this has not been demonstrated clinically. The benefit of computer-aided techniques in decreasing outliers (excess alignment errors) is particularly useful with unusual bony deformities; periarticular hardware precluding intramedullary femoral guidance; and in combination with minimally invasive techniques, where exposure and landmark recognition may be compromised.

In future, computer-aided guidance may also provide a benefit in surgical education and for the low-volume orthopaedic surgeon. Despite these benefits, the cost, increased surgical time, learning curve, and unfamiliar new technique, have been barriers to widespread adoption of computer navigation into total joint surgery.

Pain management

Current trends in pain management include a multimodal approach from the preoperative period through the outpatient rehabilitation phase. This includes NSAIDs, long acting narcotics, local infiltration, peripheral nerve blocks, and epidural pain management. In fact this may be the most important advance in patient care management as a result of the introduction of minimally invasive and rapid rehabilitation protocols. Early mobilization in turn, decreases associated postoperative morbidity including rates of deep vein thrombosis.


The success of cemented total knee replacement supports its continued use. Early posterior ligament-retaining designs, such as the Kinematic Total Knee Prosthesis (Howmedica, Rutherford, NJ) have had long-term success. In a 10–14-year follow-up study of this prosthetic design, the investigators reported a 6.5% revision rate. Results with a modular fixed-bearing posterior cruciate design at a minimum 15 years demonstrated a 92.6% survival rate. Of 139 patients there were five reoperations, four due to polyethylene wear, and one due to a loose cemented femoral component. Similarly, in the subset of patients under the age of 55, a 95% survivorship at 12 year follow-up has been published.

Results with the total condylar prosthesis, which sacrificed the posterior cruciate ligament, were excellent, supporting the belief that cemented knee replacement is a durable and predictable procedure. Despite this success, the posterior stabilized prosthesis was introduced. The intent was to design a prosthesis that improved stair-climbing ability, increased range of motion, and prevented tibial subluxation. The 9–12-year results with the posterior stabilized prosthesis and an all polyethylene tibial component demonstrated 87% good to excellent results. An analysis of failures revealed a 3% rate of tibial loosening. This prompted a modification to a metal-backed tibial component. In a subsequent 10–12-year follow-up study on this design, a 96% good to excellent result was reported. Despite the occurrence of two loose femoral components, there were no loose tibial components. Further modifications of patellar geometry, posterior femoral condylar offset, tibial shape, and conformity have resulted in the modern modular, cemented, third-generation posterior-stabilized prosthesis. Long-term studies do not yet exist for this design, but at a mean 48 months, no cases of aseptic loosening, no patellar complications, no evidence of osteolysis or wear, and an average Knee Society Knee Score of 96 were noted.

Speculating that the level of activity would influence the longevity of cemented TKA, the long-term results and functional outcome in patients who were 55 years or younger at the time of index procedure have been evaluated. All patients were rated good to excellent at an average follow-up of 8 years. The 18-year cumulative survivorship was 94%. This represented an active group of patients who regularly participated in exercises which placed high stresses on the cement interface. Although there was one case of polyethylene wear, there were no cases of component loosening.

Registry data

Registry data can provide some insight into both demographic as well as survivorship figures in large groups of patients followed over many years. Survivorship method analysis was performed on 2629 consecutive cemented primary TKAs performed by one surgeon (Dr J. Insall). The Total Condylar series had an average annual failure rate of 0.46% and a 21-year success rate of 90.77%. The Posterior Stabilized prosthesis with an all polyethylene tibia had an average annual rate of failure of 0.38% and a 16-year success rate of 94.10%, and this prosthesis with a metal-backed tibial component had an annual failure rate of 0.14% and a 14-year success rate of 98.10%. The Posterior Stabilized series with modular components had an average annual rate of failure of 0.59% and a 10-year success rate of 93.63%. The Constrained Condylar knee series had an average annual failure rate of 0.26% and a 7-year success rate of 98.12%.

Results from the Swedish knee registry of 41 223 primary arthroplasties, demonstrate decreased revision rates as the age at surgery increased. The overall cumulative revision rate (7.8% overall) decreased in the period 1976–1997 (7% decrease) primarily due to reductions in loosening (11%) and infection (7%). A 1.4 times increased risk of revision was noted with an uncemented tibial design. Overall loosening accounted for 44% of revisions. A satisfaction study in 27 372 patients operated on from 1981–1995 found that only 8% were dissatisfied at 2–17 years postarthroplasty, and that this was constant, regardless of when the procedure was performed over this 15-year period. A close review of population trends revealed a growing rate of total knee replacements, which is estimated to increase by at least one-third by 2030.

Review of the Norwegian Arthroplasty Registry demonstrated an 80.1% survival for unicompartmental knee replacements (n=2288), compared with 92% for total knee replacements (n=3032) and this value was lower in all age categories. Early failures in 7174 primary total knees implanted from 1994–2000 were examined. Five-year survival of tricompartmental total knees was 95–99% for different brands, and the differences were not statistically significant. When comparing bicompartmental to TKA, there was a higher rate of revision for patellar resurfacing following bicompartmental replacement, but a 2.5 times decrease in revision rates due to infection.

The Mayo Clinic Joint Registry, containing 11 606 primary TKAs, was analysed from 1978–2000. Higher survivorships in older patients, inflammatory arthritis, cruciate retention, and cemented all polyethylene tibial designs were noted. Survivorship was 91% at 10 years and 78% at 20 years.

An examination of 18 530 TKAs performed in Ontario, Canada from 1984–1991 was performed: 17,229 of these were primary total knees. Revision rates were estimated to be 4.3–8.0% at 7 years. Risk factors associated with significantly increased rates of revision included a younger age; an increased number of comorbid diagnoses; bilateral total knees; osteoarthritis; urban inhabitants; and having the primary knee surgery performed at a teaching hospital. The authors felt that the increased revision rate at teaching centres was likely due to the increased complexities of surgeries performed there, on patients with an increased number of comorbidities, but were not able to prove this.

Further reading

Diduch, D.R., Insall, J.N., Scott, W.N., Scuderi, G.R., and Font-Rodriguez, D. (1997). Total knee replacement in young, active patients. Long-term follow-up and functional outcome. Journal of Bone and Joint Surgery, 79A, 575–82.Find this resource:

Font-Rodriguez, D.E., Scuderi, G.R., and Insall, J.N. (1997). Survivorship of cemented total knee arthroplasty. Clinical Orthopaedics and Related Research, 345, 79–86.Find this resource:

Fuchs, R., Mills, E.L., Clarke, H.D., Scuderi, G.R., Scott, W.N., and Insall, J.N. (2006). A third-generation, posterior-stabilized knee prosthesis: early results after follow-up of 2 to 6 years. Journal of Arthroplasty, 21(6), 821–5.Find this resource:

Long, W.J. and Scuderi, G.R. (2008). Fixed varus and valgus deformities. In: Lotke, P.A. and Lonner, J.H. (eds) Master Techniques in Orthoaedic Surgery. Knee Arthroplasty, 3rd edn, pp.111–126. Philadelphia, PA: Lippincott Williams & Wilkins.Find this resource:

    Robertsson, O., Knutson, K., Lewold, S., Lidgren, L. (2001). The Swedish Knee Arthroplasty Register 1975-1997: an update with special emphasis on 41,223 knees operated on in 1988-1997. Acta Orthopaedica Scandinavica, 72(5), 503–13.Find this resource: