The number of patients who undergo joint replacement surgery continues to increase. It has been projected that almost 3.5 million total knee arthroplasties will have been performed by 2030.

Prosthetic joint infections (PJIs) have been associated with significant morbidity and may lead to poor functional outcomes. For primary joint replacement surgeries, infection rates are <1% for hip and shoulder prostheses, <2% for knee prostheses, and <9% for elbow prostheses; however, infection rates associated with revision procedures are as high as 20%.

Risk factors for PJI include a history of rheumatoid arthritis, psoriasis, immunosuppression (either through medication or illness), poor nutritional status, obesity, diabetes mellitus, age of 70 years or older, presence of a malignancy and history of previous joint arthroplasty.

Inoculation of microorganisms into the surgical wound most commonly occurs during the perioperative period (during surgery or immediately thereafter); however, hematogenous spread from a distant site of infection or contiguous spread from an adjacent site of infection may occur.

Infections may present as a superficial cellulitis, an abscess, or more deep seated involving the joint space. PJI are classified as early (<3 months after surgery), usually caused by highly virulent organisms (Staphylococcus aureus or gram-negative bacilli); delayed (3 to 24 months after surgery), usually caused by less virulent organisms (coagulase-negative staphylococci or Propionibacterium acnes); and late (more than 24 months after surgery), predominantly caused by hematogenous seeding.

Early infection may present with acute onset of fever, joint pain, effusion, erythema, and warmth at the implant site. Clinically significant cellulitis and formation of a sinus tract with purulent discharge may also occur.

Delayed infection usually presents with more subtle signs and symptoms such as implant loosening and persistent joint pain. Infection is often difficult to distinguish from aseptic failure and often a combination of preoperative and intraoperative tests are necessary for accurate diagnosis.

Peripheral leukocyte count with differential may be normal or slightly elevated in the presence of infection.

Low sensitivity C-reactive protein is almost always elevated surrounding surgery and typically returns to normal within weeks. Repetitive measurements establishing a pattern with clinical correlation are often more informative than a single value.

Synovial fluid analysis is encouraged as a rapid, accurate test for differentiating infection from aseptic failure. Leukocyte count and differential values for diagnosing PJI are considerably lower than those for septic arthritis in native joints. In patients without underlying inflammatory joint conditions, a synovial fluid leukocyte count of >1,700 per cubic millimeter or >65% neutrophils is 94% to 97% sensitive and 88% to 98% specific for diagnosing PJI.

Histopathologic diagnosis requires the presence of 1 to 10 or more neutrophils per high-power field (magnification of 400) and has a sensitivity of more than 80% and a specificity of more than 90% for diagnosis of PJI.

Gram staining of synovial fluid and periprosthetic tissue for microbiologic diagnosis when positive has a specificity of >97%; however, if negative, a sensitivity of only 25%. Periprosthetic tissue cultures remain the most reliable means of diagnosing PJI and identifying the associated organism(s). At least three intraoperative tissue specimens should be sampled for culture. Superficial wound or sinus tract cultures should be avoided as they are often positive for colonization of surrounding skin. Periprosthetic cultures may be negative due to a low number of microorganisms, fastidious organisms, improper culturing technique or laboratory methods, or because of prior antimicrobial exposure. When possible, antimicrobial therapy should be discontinued for 2 weeks before tissue culturing to better detect low-grade infection (typically presents only with early loosening of the prosthesis and persistent pain, often without systemic or local clinical signs of infection).

Plain radiographs may help diagnose infection, especially when done serially over time after implantation. Prosthesis loosening will appear as new subperiosteal bone growth. Transcortical sinus tracts are specific for infection. Implant migration and periprosthetic osteolysis can also occur without infection and thus expert opinion is often necessary for diagnosis. Arthrography is useful for detecting implant loosening, pseudobursae, and abscesses.

Criteria for diagnosis of PJI include (1) isolation of the same organism in at least two cultures of synovial fluid or periprosthetic tissue, (2) purulence of synovial fluid at the implant site, (3) acute inflammation on histopathologic examination of periprosthetic tissue, or (4) presence of a sinus tract communicating with the prosthesis.

Management of PJI requires a multidisciplinary approach that includes adequate surgical debridement combined with long-term antimicrobial therapy. Eradicating the infection and improving the functional status of the joint is the ultimate goal of therapy. Often, a patient’s comorbidities prohibit surgical intervention and thus medical treatment alone may be justified. Often in this clinical scenario, lifelong antibiotic suppression is utilized.

Several surgical modalities exist for management of PJI and include debridement with retention of the prosthesis; a one-stage exchange where the infected prosthesis is removed and a new one implanted during the same surgery; a two-stage exchange where a resection arthroplasty is performed with delayed reimplantation during a second surgery; resection arthroplasty with or without arthrodesis; and amputation.

Zimmerli and colleagues developed a validated treatment algorithm that has been associated with an overall success rate of more than 80%.

Debridement with retention is a reasonable option for patients with early or acute hematogenous infection and the duration of symptoms is <3 weeks, the implant is stable, the soft tissue is in good condition, and an agent with activity against the organisms in the biofilm is available.

If the duration of symptoms exceeds 3 weeks, retention of the implant is not advisable. A one-stage exchange is possible if the soft tissues are in good condition, there are no severe coexisting medical illnesses, and the organism(s) is readily treatable. In patients with compromised soft tissue, a two-stage exchange is preferred. After removal of the hardware, a spacer or external-fixation device is inserted to maintain the length of the extremity. Antibiotics are then administered until the new hardware is inserted. If possible, it is preferable to discontinue antibiotics 2 weeks prior to reinsertion so that reliable cultures may be obtained for determining duration of therapy.