There is no uniformly accepted definition for prosthetic joint infection. The presentation is often varied, and the classical features of infection, such as fever, leukocytosis, and other typical signs of sepsis, are often absent. A prosthesis is a foreign body, and its presence alone is a risk factor for infection. Also, the microorganism load that will produce an established infection in a prosthetic joint is often less compared to a native joint.
The strongest factor that defines prosthetic joint infection is perhaps the recovery of a microorganism from a joint aspirate or surgically obtained peri-prosthetic material. Other factors that support prosthetic joint infection include the presence of a sinus tract communicating with the joint space, wound dehiscence, purulence around the joint without an alternative explanation, loosening of the prosthesis, elevated synovial fluid leukocyte count with neutrophil predominance, and elevated inflammatory markers. However, failure to identify a pathogen does not rule out infection.
The implicated pathogens of prosthetic joint infection are best categorized into three groups based on when they are likely to be isolated from culture materials in relation to the duration of the implant.
Early infection, which is less than three months of implant placement), is typically caused by highly virulent organisms including Staphylococcus aureus, aerobic gram-negative Bacilli, and beta-hemolytic Streptococcus.
Delayed infection, from 3 months to 1 year, is typically caused by low virulent organisms including coagulase-negative Staphylococcus, Propionibacterium acnes, and Enterococci. To a lesser extent, S. aureus may also be implicated.
Late infection (over one year; some authors use over 2 years) are typically hematogenous and often caused by Staphylococcus aureus, coagulase-negative Staphylococcus, viridans streptococcus, Enterococci, and occasionally gram-negative bacilli. See figure.
In summary, Staphylococci account for more than 50% of total prosthetic joint infections. Approximately 20% may be polymicrobial, 15% caused by gram-negative, and about 10% are culture negative.
Currently, in the United States, physicians perform about one million total hip and knee arthroplasties, but this number is expected to quadruple over the next one to two decades. With increasing number of surgeries, it is expected there will be more infections. The current rate of prosthetic joint infection varies from one center to another but typically range between 0.5% to 1.0% for hip and shoulder replacements, and 0.5% to 2% for knee replacements.
Infection rates are higher in the first 2 years compared to after two years following surgery.
The risk factors for infection include:
Underlying the pathogenesis of prosthetic joint infection is the simple fact that prostheses are foreign bodies and they can promote the formation of biofilm by several notable organisms. Biofilm is a thin film of microorganism-embedded glycocalyx (exopolysaccharides) which often coats the surface of the prosthesis and create a barrier against antimicrobial agents. Also, certain forms of Staphylococcus aureus, a notable player in prosthetic joint infections may form small colony variants which often grow slowly and are associated with recurrent and difficult-to-treat infections.
The presentation can be early, delayed, or late as outlined previously. Virulent pathogens typically cause early infections, most of which are acquired around the time of surgery and usually are accompanied with erythema, induration at the incision site, and wound drainage. Delayed infections may also be acquired around the time of surgery, but low virulent organisms typically cause them. Late infections, however, are mostly hematogenous and often without fever, draining wounds, or local signs of infection. Overall, symptoms are very non-specific, and most patients present with joint swelling or pain. Other common clinical presentations include wound dehiscence, purulence around the prosthesis, sinus tract communicating with the joint cavity, joint instability and ambulation difficulties.
The accurate diagnosis of prosthetic joint infection often involves the combination of multiple factors including symptoms, signs, synovial fluid cell count, serum inflammatory markers, and culture.
The threshold for diagnosing a prosthetic joint infection using some of these parameters is much lower compared to septic arthritis of a native joint. For example, unlike septic arthritis, when synovial fluid cell counts are typically tens of thousands, a synovial fluid white blood cell count (WBC) of more than 4200 cells/microliter supports hip prosthetic joint infection, while a WBC count of more than 1700 cells/microliter supports knee prosthetic joint infection. Neutrophil predominance is typically a common feature.
A culture is an important tool for the diagnosis of prosthetic joint infection. Aspirated joint fluid culture should be sent in multiple sets of culture media. Also, obtaining multiple intraoperative culture samples is encouraged. This is especially important for the recovery and accurate diagnosis of prosthetic joint infection caused by low virulence organisms because having just one positive culture is not enough to make a diagnosis. The sensitivity of synovial fluid culture is only 85%, so a negative culture does not rule out infection. However, the specificity of synovial fluid culture is approximately 95%, and positive cultures often imply the presence of prosthetic joint infection.
Imaging may have some role in diagnosis, but it is mostly complimentary. Perhaps the most important study is a plain film of the affected joint. Even though overall sensitivity and specificity are low, plain imaging may provide useful clues about joint effusion, joint alignment, bone-cement interface lucencies, and periosteal reactions.
Physicians seldom order CT and MRI scans, but when requested, the scans may show bony erosion, abscess, sinus tract, or loosening of the prosthesis. Technetium bone scans have poor specificity and typically are not helpful in early and delayed infections. FDG-PET scan and WBC scans have high specificities and sensitivities and may be a good adjunct to diagnosis.
Current Classification of Prosthetic Joint Infection
Treatment of prosthetic joint infection often includes a combination of medical (antibiotics) and surgical methods. Treatment principles include:
Antibiotic therapy is often necessary but should be delayed until after culture materials (at least joint aspiration culture) are obtained. The exception is in the rare situation of sepsis or overwhelming infection. Empirical antibiotic therapy should be tailored according to the timing of infection. As a general guide, the antibiotic choice should at least provide adequate coverage against Staphylococcus aureus (including methicillin-resistant Staphylococcus aureus (MRSA)), coagulase negative staphylococcus and aerobic gram-negative bacilli. Antibiotics with excellent activity against biofilms such as Rifampin and fluoroquinolones are often included in treatment. Other possible oral antibiotic choices for prosthetic joint infection include minocycline, linezolid, and trimethoprim-sulfamethoxazole.
Surgical options can be further categorized into five options as follow:
Two novel methods recently added to the diagnostic armamentarium of prosthetic joint infection include:
The management of a prosthetic joint infection is best done with a multidisciplinary team consisting of an orthopedic surgeon, infectious disease expert, rehabilitation specialist, nurse and pharmacist. These patients are often left bedridden for long periods, and hence a physical therapy consult is recommended for joint movement and muscle training. A dietary consult may help prevent muscle wasting, and the nurse should ensure that the patient has prophylaxis against deep vein thrombosis and pressure ulcers. The pharmacist should check up on cultures and make sure that the patient is on the right antibiotics. After discharge, most patients need active physical therapy for months to regain joint motion and muscle strength. (Level V)
There are no large randomized clinical trials on prosthetic joint infections and outcomes are based on small retrospective case series and case reports. The success of the different procedures varies from 0-100%, depending on the severity of the infection, time of treatment, patient co-morbidity and duration of infection. The difficulty in dealing with prosthetic joint infections is whether to remove all the hardware but this also leaves the patient with a disability. Use of spacers has been useful, but it is not the answer to all type of prosthetic joint infections. Overall, good results have been obtained with delayed exchange arthroplasty with success rates ranging from 40-80%. Unfortunately, the majority of patients do have some degree of pain, joint instability and limited use of the joint after an infection. (Level V)