I. Epidemiology and Overview
A. Prevalence
1. The risk of periprosthetic infection following primary knee arthroplasty (1% to 2%) is slightly higher than that following hip arthroplasty (0.3% to 1.3%).
2. The risk of infection is higher after revision procedures: 3% for hips and 6% for knees.
B. Risk factors
1. Postoperative surgical site infection or hematoma formation
2. Wound healing complications
3. Malignant disease
4. Prior joint arthroplasty
5. Prior surgery or infection of the joint or adjacent bone
6. Perioperative nonarticular infection
7. Rheumatoid arthritis
*Javad Parvizi, MD, or the department with which he is affiliated has received research or institutional support from Stryker.
8. Psoriasis
9. Diabetes
C. Prophylaxis—Administration of antibiotics within 60 minutes before surgery is the most effective method for prevention of periprosthetic infection.
D. Classification—A classification of infected total joint arthroplasty is outlined in
Table 1.
II. Presentation and Etiology
A. General symptoms
1. Pain at the implant site is a consistent symptom.
2. Pain at the implant site is associated with infection in >90% of patients.
B. Typical patient presentation and etiology for several types of periprosthetic infections are listed in
Table 2.
III. Diagnosis
A. History and physical examination
1. A detailed history and physical examination can diagnose periprosthetic infection with reasonable
[Table 1. Classification of Periprosthetic Joint Infections]
[Table 2. Etiologies, Signs, and Symptoms of Periprosthetic Joint Infections]
[
Figure 1. Radiograph of the hip of a patient with periprosthetic infection demonstrating cortical resorption, progressive radiolucencies, and localized osteolysis. Various voids in the cement can be seen crossing onto the cortex and resulting in resorption of the bone.]
certainty; laboratory tests simply confirm the diagnosis.
2. Many times, signs and symptoms of periprosthetic infection overlap those of hematoma formation, aseptic loosening, and instability; thus, additional diagnostic tests are often required.
B. Imaging studies
1. Radiographic signs (Figure 1)
a. Periosteal reaction
b. Scattered foci of osteolysis
c. Generalized bone resorption in absence of implant wear
2. Radionuclide studies
a. When infection is clinically suspected but cannot be confirmed by aspiration or serology, a bone scan is indicated.
b. Radionuclide studies are the imaging modality of choice for periprosthetic infections, having 99% sensitivity and 30% to 40% specificity.
c. Technetium Tc 99m detects inflammation, and indium In-111 detects leukocytes in periprosthetic tissue.
d. A triple scan can distinguish infection from conditions with high metabolic activity, such as fracture or bone remodeling, improving the specificity to 95%.
3. Positron emission tomography (PET)
a. PET has recently been shown to have a role in the diagnosis of periprosthetic infection, with a sensitivity of 98% and specificity of 98%.
b. PET scanning uses fluorinated glucose, which travels to areas of high metabolic activity.
C. Serologic tests
1. Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are nonspecific markers of inflammation that are very useful in diagnosis when combined.
a. Combined, ESR and CRP have a sensitivity of 99% and a specificity of 89%.
b. The CRP level returns to normal within 21 days of surgery, whereas ESR may take up to 90 days to return to normal.
c. Continued elevation of levels of CRP and ESR is concerning for infection.
2. Interleukin-6 may be useful for diagnosis and for monitoring the progress of infection.
D. Joint aspiration
1. Joint aspiration is performed when there is a strong suspicion of infection.
2. Aspiration has a sensitivity of 57% to 93% and a specificity of 88% to 100%. Sensitivity can be improved by repeat aspiration.
3. False-negative results can be caused by improper technique or by the patient taking antibiotics at the time of aspiration.
4. False-positive results are due to contamination, which can be avoided by using more sterile technique.
E. Microbiology
1. A definitive diagnosis can be made when the same organism is recovered either from repeated joint aspirations or from at least three of five periprosthetic specimens obtained at surgery.
2. The false-positive rate is not known with certainty but may be approximately 8%.
F. Histopathology
1. Gram stain
a. Gram stain has an unacceptable level of sensitivity (0% to 23%) and hence is considered unreliable for determining the presence of infection.
b. In other words, a negative Gram stain means little, whereas a positive Gram stain is almost definitive for infection.
2. Frozen section
a. Frozen section has become another valuable tool for the diagnosis of infection.
b. It is most useful in equivocal cases, when pre-operative investigations are confounded by false elevations in the ESR and CRP or when the intraoperative appearance of the joint is worrisome for infection.
c. Most studies report favorable results, with a sensitivity approaching 85% and a specificity of approximately 90% to 95%.
d. Mirra and associates first reported the value of intraoperative frozen section in 1976, stating that more than five polymorphonuclear leukocytes (PMNs) per high-power field is a probable sign of infection.
G. Molecular techniques
1. Polymerase chain reaction of the aspirate
a. This technique relies on amplification of bacterial DNA.
b. The test is so sensitive that contamination with a few bacteria usually leads to a false-positive result.
2. Microarray and proteomic technology
a. These techniques attempt to target specific bacterial genes.
b. They produce a profile of the genes (microarray) or the proteins (proteomics) of the joint aspirate or the periarticular tissues and rely on identifying signature genes/proteins that indicate infection.
IV. General Treatment Principles
A. Overview
1. Treatment of periprosthetic infection usually involves multiple surgeries.
2. The treatment also usually involves an extensive course of antimicrobial therapy.
B. Characteristics of periprosthetic infections
1. Biofilm formation
a. Biofilm formation is characteristic of periprosthetic infections; the biofilm is believed to form and establish within 4 weeks.
b. Bacteria produce an extracelluar matrix, such as glycocalyx, that allows adherence to the implant surface.
c. Eradication of the infection is difficult once the biofilm has formed.
d. Removal of the prosthesis for established infections (>4 weeks) is recommended.
e. Early postoperative and acute hematogenous infections are less likely to be associated with the development of biofilm or prosthetic loosening; the chance of eradication of infection without prosthesis removal is higher in these cases compared with cases of indolent disease.
2. Microbial colonization
a. Microbial colonization can occur at the time of implantation.
b. It can be a result of direct spread from a contiguous focus or can result from hematogenous seeding.
C. Treatment options
1. Surgical
a. Debridement with retention of prosthesis
b. Resection arthroplasty with reimplantation
c. Definitive resection arthroplasty with or without arthrodesis
d. Amputation
2. Nonsurgical—Suppressive antimicrobial therapy
V. Surgical Treatment
A. Two-stage replacement arthroplasty—Two-stage resection and replacement arthroplasty is the gold standard for treatment of infection beyond 4 weeks.
1. Procedure
a. Removal of prosthesis
b. Surgical debridement of the joint
c. Administration of antimicrobials with subsequent delayed reimplantation
2. Prerequisites
a. Adequate bone stock
b. Medical fitness for multiple surgical procedures
c. Confirmation of eradication of infection (normal serology and negative aspiration)
3. Antimicrobials should be stopped for at least 2 weeks before aspiration and serology.
4. Antimicrobials are administered for 4 to 6 weeks following resection to eradicate osteomyelitis.
5. The time interval between the two surgical procedures is highly variable.
a. No specific protocol is in place, but early reimplantation has less success.
b. In infected total knee arthroplasty, reimplantation within 2 weeks has a success rate of approximately 35% compared to success rates of 70% to 90% with delayed reimplantation (>6 weeks) and more extensive antimicrobial therapy.
c. The proper timing of reimplantation after parenteral antibiotic therapy should be based on the clinical appearance of the wound and improvement in serologic markers of infection, such as the CRP level and the ESR.
i. At the time of reimplantation, tissue specimens should be sent for culture and histopathologic evaluation.
ii. If all preoperative and intraoperative indices are acceptable, then reimplantation is indicated.
iii. If the frozen section suggests continued acute inflammation, the wound should be debrided again, a cement spacer reapplied, and the wound closed.
6. Antibiotic-impregnated spacers (static or dynamic) are often used in the interim between the two stages.
a. Dynamic spacers allow joint motion and have the advantage of better patient satisfaction and ease of revision.
b. Static spacers allow delivery of higher doses of antibiotics (since these are usually mixed in the operating room as opposed to be being pre-made) and, because of lack of motion, provide a more favorable environment for wound healing.
7. Antibiotic-impregnated cement should be used during cemented reimplantation. Cementless reimplantation in the hip has better outcome.
B. One-stage replacement arthroplasty
1. Procedure
a. All prosthetic components, infected bone, and soft tissue are excised.
b. The new prosthesis is implanted during the same surgery.
c. Intravenous antibiotics are administered for a variable period of time after the revision
2. Indications
a. Mainly for managing infected hip prostheses
b. Used more commonly in Europe
3. Advantages
a. Single procedure
b. Lower cost
c. Earlier mobility
d. Patient convenience
4. Disadvantage—Risk of reinfection from residual microorganisms
5. Outcomes
a. Overall success rates for both hip and knee have been variable (75% to 100%) but not as successful as two-stage reimplantation.
b. Successful outcomes are more likely in the following circumstances:
i. Infection caused by a low-virulence organism with a good sensitivity profile
ii. Patient not immunocompromised
iii. No sinus tract formation
iv. Healthy soft tissue
v. Full debridement
vi. Prolonged course of postoperative antibiotics
vii. No bone graft used
6. Antibiotic-impregnated bone cement is often used for fixation of the new prosthesis in either a oneor two-stage revision.
C. Debridement with retention of prosthesis
1. Procedure
a. Debridement of infected tissue and exchange of polyethylene insert with large-volume irrigation
b. Prolonged postoperative antibiotic therapy
2. Indications—Considered when infection developed within 4 weeks of surgery or after an inciting event such as dental extraction.
3. Advantages—Limited surgery with preservation of prosthesis and bone stock.
4. Disadvantages
a. Risk of leaving infected foreign body in place
b. Failure rate is generally approximately 20%, with occasional reports of success up to 85%, depending on the outcome factors listed above.
D. Resection arthroplasty
1. Procedure
a. Definitive removal of all infected components and tissue
b. No subsequent implantation
2. Indications—Currently are limited, but include the following:
a. Poor quality of bone and soft tissue
b. Recurrent infections
c. Infection with multi-drug-resistant organism
d. Medical conditions that preclude major surgery (reimplantation)
e. Failure of multiple previous exchange arthroplasties
f. Also an acceptable alternative in elderly non-ambulatory patients
3. Disadvantages
a. Shortened limbs
b. Poor function
c. Patient dissatisfaction
4. Outcomes—Overall success rate for eradication of infection depends on whether the hip or the knee is involved.
a. For total hip arthroplasty, the success rate is between 60% and 100%.
b. For total knee arthroplasty, the success rate is 50% to 89%.
E. Arthrodesis
1. Procedure—Arthrodesis provides bony ankylosis of a joint.
2. Indications—Arthrodesis is used when subsequent joint reimplantation is not feasible due to poor bone stock or recurrent infection with virulent organisms.
3. Outcomes—Overall success with both eradication of infection and bony fusion can be achieved in 71% to 95% of cases.
F. Amputation
1. Procedure—Above-the-knee (transfemoral) amputation.
2. Indications
a. Recalcitrant total knee arthroplasty infections, when all other options have been exhausted
b. Intractable infection with severe pain, soft-tissue compromise, bone loss of a severe nature that precludes the use of a prosthesis, and vascular compromise
G. Antimicrobial therapy
1. May be curative when the infected joint is removed and periprosthetic tissue is debrided
2. Antibiotics should be withheld until aspiration or intraoperative cultures are obtained, unless overwhelming sepsis is present.
3. Initial empiric therapy for most common pathogens—First-generation cephalosporin or vancomycin when any of the following factors are present:
a. True sensitivity to penicillins
b. Prior history of methicillin-resistant S aureus (MRSA) infection
c. Exposure to MRSA (in institutionalized patients)
d. Infection by an unidentified organism
4. Following identification of the microorganism and susceptibility testing, the antibiotic regimen should be tailored to the findings.
5. Antimicrobials should be given systemically for a minimum of 4 weeks. Antimicrobial-impregnated cement beads/spacers also should be used.
VI. Nonsurgical Treatment
A. Treatment—Suppressive antimicrobial therapy.
B. Indications—Considered for frail, elderly, and ill patients in whom surgery is not possible or is refused by the patient.
C. Goals
1. Provide symptomatic relief
2. Maintain joint function
3. Prevent systemic spread of infection rather than eradication of infection
D. Outcomes
1. Success rates of only 10% to 25%
2. Complications occur in 8% to 22% of patients.
VII. Antimicrobial-Impregnated Devices
A. Antimicrobial-impregnated devices take the form of solid spacers, beads, or dynamic spacers.
B. Doses of antibiotics in spacers
1. For each bag of cement (40 g), 3 g of vancomycin and 4 g of tobramycin may be added.
2. Gentamicin is also available in powder form in some institutions and may be used in place of tobramycin.
C. Advantages of spacers
1. Reduce dead space
2. Provide joint stability
3. Deliver high local doses of antimicrobials
D. Disadvantages of spacers
1. May result in lower mechanical properties of bone cement (when used for definitive fixation)
2. Potential for allergic reactions or local or systemic toxicity
3. Potential for emergence of antibiotic-resistant organisms
4. Heat-labile antimicrobials cannot be added to cement. (Tobramycin, vancomycin, gentamicin powder, and amphotericin-B are heat-stable antimicrobials.)
E. Elution technology
1. In recent years, some efforts have been made to coat the implant surface with antibiotics or a material with antibacterial properties such as silver.
2. The antibiotic in the coating or the material on the surface of the implant is believed to prevent bacterial attachment (biofilm formation) and subsequent infection. So far, all the described technologies involve elution of the antibiotics. None of these technologies is available for clinical use.
3. One advantage of antibiotic-impregnated devices is that the antibiotic is localized to the site of infection.
4. Disadvantages of elution technologies
a. Acidification of the milieu due to the degradation of the polymer
b. The fragility of the polymers
c. The discharge of antibiotic independent of the infection state of the site
d. Unstable and unpredictable release kinetics—The release kinetics of elution materials, including methylmethacrylate cement, is often unstable and unpredictable, resulting in large fluctuations in the local concentrations. Hence, biodegradable polymers such as polylactic acid (PLA), polyglycolic acid (PGA), or a combination thereof (PLAGA); collagen; and other carriers have been used to deliver antibiotics. The release of antibiotic from these materials usually depends directly on the rate of degradation of the surface coating and can be modified by changes in composition.
e. A major limitation of the elution technology relates to the potential for emergence of resistant organisms. The drop in the level of the eluded antibiotic to below minimal inhibitory concentration, which occurs at some point in the elution cycle, raises concern for potential emergence of resistant organisms.
F. Covalent bonding of antimicrobials
1. Another technology under evaluation is the covalent bonding of antimicrobials to the surface of the implant.
2. This technology overcomes most of the limitations of elution technology.
Top Testing Facts
1. Predisposing factors for periprosthetic infection are postoperative surgical site infection or hematoma formation, wound healing complications, malignant disease, prior joint arthroplasty, prior surgery or infection of the joint or adjacent bone, perioperative nonarticular infection, rheumatoid arthritis, psoriasis, and diabetes.
2. Pain at the implant site is a consistent symptom, associated with infection in more than 90% of cases.
3. Periosteal reaction, scattered foci of osteolysis, and generalized bone resorption in the absence of implant wear are radiographic signs of periprosthetic infection. When infection is clinically suspected but cannot be confirmed by aspiration or serology, a bone scan is indicated.
4. ESR and CRP are nonspecific markers of inflammation that are very useful in diagnosis when combined. Combined, ESR and CRP have a sensitivity of 99% and specificity of 89%. The level of CRP returns to normal within 21 days of surgery, whereas ESR may take up to 90 days to return to normal. Continued elevated level of CRP and ESR is concerning for infection.
5. Joint aspiration is performed when there is a strong suspicion of infection. Aspiration has a sensitivity of 57% to 93% and a specificity of 88% to 100%. Sensitivity can be improved by repeat aspiration.
6. Molecular techniques such as polymerase chain reaction, although valuable, are extremely sensitive and hence carry a high incidence of false-positive results.
7. Two-stage resection and replacement arthroplasty is the gold standard for treatment of infection beyond 4 weeks.
8. Incision and debridement are potentially effective for infections within 4 weeks of index arthroplasty or after an inciting event such as dental extraction.
9. Antibiotics should be withheld until aspiration or intraoperative cultures are obtained, unless overwhelming sepsis is present.
10. Spacers reduce dead space, provide joint stability, and deliver high local doses of antimicrobials.
Bibliography
Bauer TW, Parvizi J, Kobayashi N, Krebs VJ: Diagnosis of periprosthetic infection. J Bone Joint Surg Am 2006;88:869-882.
Greidanus NV, Masri BA, Garbuz DS, et al: Use of erythrocyte sedimentation rate and C-reactive protein level to diagnose infection before revision total knee arthroplasty: A prospective evaluation. J Bone Joint Surg Am 2007;89:1409-1416.
Hanssen AD, Osmon DR: Evaluation of a staging system for infected hip arthroplasty. Clin Orthop Relat Res 2002;403: 16-22.
Langlais F: Can we improve the results of revision arthroplasty for infected total hip replacement? J Bone Joint Surg Br 2003;85:637-640.
Marculescu CE, Berbari EF, Cockerill FR III, Osmon DR: Fungi, mycobacteria, zoonotic and other organisms in prosthetic joint infection. Clin Orthop Relat Res 2006;451:64-72.
Mirra JM, Amstutz HC, Matos M, Gold R: The pathology of the joint tissues and its clinical relevance in prosthesis failure. Clin Orthop Relat Res 1976;117:221-240.
Parvizi J, Ghanem E, Menashe S, Barrack RL, Bauer TW: Periprosthetic infection: What are the diagnostic challenges? J Bone Joint Surg Am 2006;88(Suppl 4):138-147.
Segawa H, Tsukayama DT, Kyle RF, Becker DA, Gustilo RB: Infection after total knee arthroplasty: A retrospective study of the treatment of eighty-one infections. J Bone Joint Surg Am 1999;81:1434-1445.
Spangehl MJ, Duncan CP, O'Connoll JX, Masri BA: Prospective analysis of preoperative and intraoperative investigations for the diagnosis of infection at the sites of two hundred and two revision total hip arthroplasties. J Bone Joint Surg Am 1999;81:672-683.
Zimmerli W, Trampuz A, Ochsner PE: Prosthetic-joint infections. N Engl J Med 2004;351:1645-1654.