Complications in Total Knee Arthroplasty

Donald T. Reilly, MD. PhD

Introduction

At the recent State of the Art Update in Orthopaedics 2000 in Whistler, British Columbia, Lawrence G. Morawa, MD, moderated a group of presentations centered around the diagnosis and treatment of complications in total knee arthroplasty (TKA).

Diagnosing and Treating Infection
Although infection in total knee arthroplasty (TKA) is a relatively infrequent complication, it can be devastating in terms of morbidity and cost. Lester S. Borden, MD,[1] reviewed the diagnosis and treatment of infection in TKA. He began by stressing that this is a surgical disease. Long-term antibiotic depression is rarely indicated and is generally used only in patients for whom surgery is contraindicated.

Risk factors for knee infection include:
• multiple previous infections
• history of previous infection
• inflammatory arthritis (delayed infection)
• insulin dependent diabetes
• postthrombotic syndrome
• malnutrition.


Borden reviewed the importance of perioperative antibiotics. The institutional incidence of primary TKA infection should be around 1 %. Revision total knee replacement, however, carries at least a 2-fold increased risk for infection. In addition, late infections occur in approximately 2 per 1,000 TKAs annually.


Waldman and colleagues[2] retrospectively analyzed 290 patients with TKAs performed between 1982 and 1993 to define the risk for infection associated with dental surgery.

They identified 62 TKAs with late infections (occurring more than 6 months after procedure). Seven of these late infections were temporally and bacteriologically associated with dental procedures. Eight of 9 patients received no antibiotic prophylaxis. Fifty-six percent of the patients with late infections had positive risk factors, including diabetes and rheumatoid arthritis. These authors suggested prophylaxis for extensive dental work.


Diagnosing TKA infections is challenging, and prompt diagnosis and treatment are essential for a successful outcome. Windsor and colleagues[3] found that 96% of patients with infected TKAs presented with pain and 77 % had swelling. Only 27% had fever and drainage.


Treatment options for an infected TKA include:
• antibiotic suppression alone
• aggressive wound debridement, drainage, and antibiotic suppression therapy
• resection arthroplasty
• arthrodesis
• 2-stage reimplantation
• amputation.


The definitive diagnosis of infection is recovery of neutrophils by aspiration with greater than 30,000 leukocytes, 75% of which are polymorphonuclear. Five percent of infected TKAs in the Windsor study had negative cultures.
With an established infection, suppression is suggested only for patients in which surgery is not feasible. A low-virulence-organism, secure implant components, and well-tolerated oral antibiotics are required for cases in which suppression is chosen. Debridement with retention of the implant has a greater rate of success when there is meticulous synovectomy, copious irrigation, and parenteral antibiotics for 4 to 6 weeks in cases of acute infection.


According to Borden, management of infection with arthroscopy and multiple irrigations has yielded a lower rate of success. One-stage reimplantation, a more popular approach in Europe, may be possible for patients with acutely infected cementless TKAs and allows for debridement of the prosthesis-bone interface. Reimplantation usually requires antibiotic-impregnated cement. The gold standard for treatment of TKA infection remains the 2-stage reimplantation. This is especially successful in patients with chronically infected TKAs (ie, patients in whom symptoms persist for longer than 3 weeks.


Simmons and colleagues[4] performed a meta-analysis of 77 studies of 2-stage reimplantation and found an average success rate of 80%. This technique is more successful for cases of osteoarthritis with low-virulence bacteria and less successful for cases of rheumatoid arthritics and when high-virulence bacteria or multiple-organism infections are present. According to the Backe and colleagues,[5] a second 2-stage reimplantation following a failed 2-stage reimplantation has a success rate of approximately 80%.


There are few attractive surgical options for failed treatment of infected TKA. Arthrodesis should be considered when multiple surgical attempts fail to eradicate infection. Adequate bone stock, however, must be present for arthrodesis to be successful. Knee arthrodesis is challenging surgically and can be complicated by nonunion, malunion, or recurrent infection. A modular titanium intramedullary nail has been used in an attempt to reduce the incidence of nonunion and the rate of complications.


Waldman[6] reviewed the results of knee arthrodesis after infected TKA in 21 patients with a mean age of 64 years. Patients were followed for a mean of 2.4 years, and the mean number of previous operations was 4. Solid arthrodesis was achieved in 20 of 21 patients at approximately 6 months by using an intramedullary nail.


Borden discussed his personal approach to 2-stage reimplantation. His technique includes meticulous debridement with preservation of all noninvolved bone, preservation of collateral ligaments with resection of the posterior cruciate ligament, and the use of spacers (Prostalac). Prostalac remains interesting but controversial. He stressed the importance of skin closure. In the interval between removal and reimplantation, a patient should receive 4 to 6 weeks of intravenous antibiotics and no antibiotics for 1 to 2 weeks. An erythrocyte sedimentation rate and C-reactive protein test can be used to decide whether to reimplant. Borden emphasized the importance of antibiotics in cement used for reimplantation.


Looks Good, Feels Bad
Although the rate of dissatisfaction in patients with TKA is reported to be approximately less than 1%, evaluating a patient with pain or limited function is of utmost importance. James A. D'Antonio, MD,[7] reviewed the work-up of a patient with a painful or dysfunctional TKA.


D'Antonio presented a case of a 62-year-old active man who presented with pain. Plain radiographs and an examination appeared normal. However, slight rotation in several views other than ideal positioning showed loosening of an uncemented component. It is important to obtain fluoroscopic views if perfectly tangential radiographs of the prosthesis-bone interface are not available.


The second case was a 65-year-old man who reported that his knees were not supporting him. The examination revealed a very stable, well-aligned knee with no effusion. Radiographs showed no abnormalities. The patient had a history of alcoholism, which led to D'Antonio to the conclusion that the problem was with the patient not the arthroplasty. Evaluation of a patient who is unsatisfied with their

TKA should include:

• detailed history

• detailed physical examination, including chief complaints; pain; function; and neurovascular, psychosocial, and hip examinations

• radiographs and laboratory tests.


This work-up usually yields a diagnosis before surgery is necessary. D'Antonio stressed the importance of delaying surgery until the surgeon has reached a definitive diagnosis. Patients is key for surgeons encountering the small percentage of patients with TKA who are unhappy with unexplained pain or dysfunction.


Looks Bad, Feels Good
There are many failure mechanisms in TKA. These include:

• polyethylene wear

• instability

• aseptic Loosening

• extensor mechanism dysfunction

• unexplained pain

• infection.


George D. Markovich, MD,[8] examined the issue of "silent" osteolysis and the management of bone loss in TKA.

Femoral defects distally and posteriorly can be prevented with metal blocks on the femoral component to maintain the joint line.

Tibial wedges, half block, full block, and oblique full blocks are also useful.

Stems can be used enhance fixation when augments are used or bone quality is poor.

Markovich's presented his experience with 50 revision TKAs with metal augmentation followed for a maximum of 10 years.

To date, none of his patients treated with this technique have had failure of fixation of the components.

Osteolysis due to polyethylene wear may be present in patients with few symptoms. As such, patients often do not present with symptoms until after extensive structural damage has already occurred.

Early treatment is central to the prevention of widespread bone loss.

There are few data to guide the decision to intervene.
Regardless of whether a patient has osteolysis, the surgeon should continue to focus on the goals of restoration of bone stock, reestablishment of the joint line, and recreation of stability in the revision setting.


Lonner and colleagues[9] evaluated a total of 102 revision TKAs to determine the prodromal symptoms and radiographic findings associated with failure. The most important indicator for failure was pain, occurring in 84% of patients at an average of 13 months. Radiographs underestimated the diagnosis of osteolysis to be 4%. Osteolysis was confirmed during surgery in 22% of patients. The authors recommended an annual questionnaire and weight-bearing radiographs to ensure adequate surveillance of TKA patients.


Radiographs often underestimate the extent of bone loss. This inadequacy has prompted the development of more exact techniques, such as dual-energy x-ray absorptiometry and microradiographic evaluation, into the clinical setting. These techniques, however, remain experimental. Because of the shortcomings of radiographs in diagnosing osteolysis, the surgeon needs to be prepared for more than is preoperatively visible.


Markovich briefly discussed pharmaceutical intervention to prevent osteolysis. Shanbaga and colleagues[10] evaluated oral bisphosphonate therapy in a canine total hip replacement model. The dogs were randomized into 3 groups of 8, and a right uncemented total hip replacement was done on each animal. The control group (group 1) received no particulate debris. In groups 2 and 3, a mixture of fabricated ultra-high molecular weight polyethylene, titanium alloy, and cobalt chrome alloy was introduced into the proximal femoral gap. Group 3 also received a once-daily dose of 5 mg of alendronate sodium begun on day 7 and continued until the time of sacrifice.


Radiographically, 1 of 8 control dogs and 6 of 7 dogs from group 2 had periprosthetic radiolucencies with development of endosteal scalloping. In contrast, only 1 of 8 animals from group 3 had periprosthetic radiolucencies. Tissues from both experimental groups had significant macrophage infiltration. Levels of prostaglandin E2 and interleukin-1 were also significantly higher in the experimental groups than in controls. Continuous administration of alendronate effectively inhibited bone lysis for the 24-week duration of the study. Markovich stressed, however, that the clinical usefulness of this treatment is still in question.


Unstable Total Knee Replacement
Instability is the leading cause of failure in TKA.

A patient with preoperative ligament dyslaxity requires prosthetic substitution.

In patients with a primary TKA, a knee that is unbalanced after surgery was not balanced properly during surgery.

Secondary instability after surgery may result from after delayed rupture, wear, or loosening.


Patient history and clinical findings are important in the diagnosis of instability.

Radiographs are not usually helpful. Presentation usually takes the form of dissatisfaction with the knee, multiple falls, instability, pain, and effusion.

Patients are not usually cognizant of instability and usually describe situations in which they descend ramps and their knee tends to give way. Instability must be looked for in flexion, extension, and both positions (global instability).



References
• Borden LS. Diagnosing and treating infection. State of the Art Update in Orthopaedics 2000. Whistler, BC: February 12-16.
• Waldman BJ, Mont MA, Hungerford DS. Total knee arthroplasty infections associated with dental procedures. Clin Orthop. 1997;343:164-72.
• Windsor RE, Bono JV. Infected total knee replacements. J Am Acad Orthop Surg. 1994 Jan;2:44-53.
• Simmons TD, Stern SH. Diagnosis and management of the infected total knee arthroplasty. Am J Knee Surg. 1996;9:99-106.
• Backe HA Jr, Wolff DA, Windsor RE. Total knee replacement infection after 2-stage reimplantation: results of subsequent 2-stage reimplantation. Clin Orthop. 1996; 331:125-31.
• Waldman BJ, Mont MA, Payman KR, et al. Infected total knee arthroplasty treated with arthrodesis using a modular nail. Clin Orthop. 1999;367:230-7.
• D'Antonio J. Looks good, feels bad. State of the Art Update in Orthopaedics 2000. Whistler, BC: February 12-16.
• Markovich GD. Looks bad, feels good. State of the Art Update in Orthopaedics 2000. Whistler, BC: February 12-16.
• Lonner JH, Siliski JM, Scott RD. Prodromes of failure in total knee arthroplasty. J Arthroplasty. 1999;14:488.
• Shanbhag AS, Hasselman CT, Rubash HE. Inhibition of wear debris mediated osteolysis in a canine total hip arthroplasty model. Clin Orthop. 1997;344:33-43.