1. 
   

   How can prosthetic joint infection (PJI) be distinguished from aseptic failure of the joint arthroplasty due to mechanical loosening or dislocation?

   
   
2. 
   

   Which pathogens are usually responsible for PJIs?

   
   
3. 
   

   What are the management strategies and treatment options for infected prosthetic joints?

   
   
4. 
   

   Can an infected prosthetic joint be effectively treated with retention of the prosthesis?

   
   
5. 
   

   What is the optimal antimicrobial regimen for a PJI?

   
   
6. 
   

   How should patients with PJI s on long-term parenteral or oral antibiotics be monitored and followed up?

Joint replacement surgery (arthroplasty) is one of the major accomplishments of modern medicine over the past four decades. It has provided excellent and cost-effective results in alleviating pain and improving mobility and quality of life in patients with debilitating joint disease. Most arthroplasties are performed on hips and knees, but shoulder, elbow, and other small joint replacements are not uncommon. It is estimated that over 750,000 prosthetic joints are implanted each year in the United States, with projections of over 4 million primary arthroplasties per year by 2030. Less than 10% of prosthesis recipients develop complications that require revision surgery. Fortunately, prosthetic joint infection (PJI) is less common than other forms of joint failure, such as aseptic loosening or dislocation. Nevertheless, given the exponential increase in total joint replacement surgeries, the cumulative numbers of PJI is increasing. Moreover, PJI represents the most dreaded and devastating complication of arthroplasty. Overall, about 1 to 2% of prosthetic joints become infected, with higher rates for total knee arthroplasties (TKA) compared with total hip arthroplasties (THA) (2.5% vs 1.5%). PJIs may occur at any time after arthroplasty. The risk of infection is higher after revision arthroplasty, with rates of 3.2% and 5.6% reported for revision THA and TKA, respectively. Consequences of PJI may include reoperation, potential for permanent removal of the prosthesis, large skeletal defects after surgery, prolonged courses of antimicrobial therapy, severe functional impairment, and persistent joint pain. Rarely, PJIs result in loss of limb or life. The cost of treatment of PJIs is in excess of $50,000 per episode, with an estimated total expenditure of over $250 million per year in the United States.

Conditions and events that predispose a prosthetic joint recipient to infection include host characteristics, index arthroplasty factors, and perioperative and postoperative factors. Host characteristics include advanced age, morbid obesity, diabetes mellitus, malignancy, HIV seropositivity, and systemic steroid use. Arthroplasties performed due to underlying rheumatoid arthritis are four times more likely to be infected compared to those for degenerative osteoarthritis. Other joint factors, such as primary arthroplasty undertaken due to trauma or fracture, revision arthroplasty, regardless of indication, and prior PJI, are independent risk factors for infection. Perioperative factors such as a National Nosocomial Infections Surveillance System surgical risk score greater than 2, American Society of Anesthesiologists greater than 2, duration of surgery greater than 75th percentile for the procedure or longer than three hours, postoperative superficial surgical site infection, wound drainage or wound hematoma, and distant nonarticular infections such as UTIs and soft tissue infection increase the risk for PJI. The use of systemic antimicrobial prophylaxis, Staphylococcus aureus decolonization in carriers, and the use of antimicrobial therapy in the fixation of impregnated bone cement reduce the risk of PJI in patients undergoing joint arthroplasty.

Depending on the onset of infection relative to the date of implantation, PJIs are classified as early (within three months), delayed (three months to two years), or late (after two years). There are three mechanisms by which prosthetic joints become infected: colonization of the device with microorganisms at the time of index surgery or immediately thereafter; direct spread from contiguous skin and soft tissue infections, penetrating trauma, or preexisting osteomyelitis; or hematogenous spread from distant infected foci. Generally, early and delayed infections are acquired at the time of implantation, and late infections occur predominantly from the hematogenous route, although there is significant overlap. Not surprisingly, in the very few well-designed cohort studies to elucidate the risk factors for PJIs, postoperative surgical site infections have been the most significantly and consistently identified risk factor for PJIs, with an associated odds ratio of over 35 in one study. Bacteremia from infections at distant sites such as the respiratory tract, urinary tract, or the skin can secondarily seed a well-fixed prosthesis and cause a PJI. Although this is relatively rare, it is particularly true for S aureus. It is estimated that one-third of joint recipients with ensuing S aureus bacteremia will develop S aureus PJI.

Two unique factors combine to make prosthetic joint infection an enormous challenge to eradicate: the presence of a foreign body and the production of bacterial biofilm.

The Foreign Body

The prosthetic implant is a foreign body devoid of microcirculation. This prevents adequate delivery of antibiotics to the site of infection. Furthermore, the implant is nonphagocytosable and renders granulocytes around the prostheses functionally defective. Local immunity is thus further impaired. It is estimated that as little as 100 colony-forming units of S aureus are sufficient to infect 95% of subcutaneous implants.

The Role of Biofilm

The first phase of PJI is adherence of microorganisms to the surface of the implant. These microorganisms then elaborate a glycocalyx of highly hydrated exopolysaccharides, better known as slime. Biofilm refers to slime and the bacteria embedded within. Bacteria in biofilm form organized and complex communities, with structural and functional heterogeneity, resembling multicellular organisms. Bacteria in biofilm even communicate by a signaling process known as quorum sensing. Sessile (attached) bacteria in biofilms are protected from antimicrobial agents and host immune responses, compared with free-living (planktonic) bacteria. Biofilm acts as a physical barrier to antibiotics and white blood cells (WBCs). As well, microorganisms in biofilms are in a low metabolic state (stationary phase) due to poor availability of nutrients such as glucose and oxygen. This renders them less susceptible to growth-dependent antibiotics, such as beta-lactams. Infections are therefore difficult to diagnose (low inoculum, few planktonic bacteria, mild inflammatory reaction), difficult to cure (poor access of antimicrobials to bacteria), and persist or recur after discontinuing antibiotics (revert to planktonic state after antibiotic pressure has been released).

The microbiology of PJIs may vary depending on time of onset, mechanism of infection, acuity of clinical presentation, and location of prosthetic joint. Gram-positive organisms, especially staphylococci, account for the vast majority of PJIs (Table 201-1). Infections in the early postoperative period, or those presenting with sudden onset joint pain and effusion, are commonly due to virulent organisms such as S aureus, gram-negative bacilli, and mixed infections. Delayed infections with subtle, chronic symptoms and signs may be caused by less virulent, more indolent bacteria such as coagulase-negative staphylococci, enterococci, and Propionibacterium acnes. Shoulder arthroplasties have a greater preponderance for P acnes infections, compared to lower extremity joints. Several Streptococcus species are responsible for some PJIs from contiguous pyogenic skin infections. Viridans streptococci, Peptococcus species, Peptostreptococcus species, and gram-negative rods may be recovered from PJIs following dentogingival processes or manipulations. More than one organism is recovered in about 10% to 12% of cases, and a similar proportion of patients have no organisms isolated at all (culture-negative PJI). The most common reason for the latter is prior use of antimicrobial therapy. Unusual bacteria, fungi, and mycobacteria are rarely isolated.

Aseptic Failure

Joint pain and malfunction may be septic (related to infection) or aseptic (biomechanical failure). In a recent review of over 50,000 hip and 60,000 knee revision procedures from the Nationwide Inpatient Sample database, 85.2% and 74.8% of surgeries, respectively, were due to aseptic failure. A major cause of aseptic implant failure is osteolysis from particulate wear debris. Relative motion between opposed surfaces leads to loss of prosthetic material, with generation of particles that get deposited in the space between implant and bone. These are phagocytosed by macrophages, and a granulomatous reaction ensues. Inflammatory mediators are released, stimulating osteoclastic bone resorption. Biopsy specimens from tissues surrounding the osteolytic lesions may demonstrate this process, commonly referred to as “arthroplasty effect.” Misalignment of the arthroplasty can also result in inappropriate (eccentric) mechanical loads, and cause implant loosening and mechanical damage to the implant material or the bone–implant interface.

Skin and Soft Tissue Infections

The skin and subcutaneous tissue overlying a prosthetic joint may become infected. The pain, tenderness, differential warmth, and erythema of cellulitis are clinically indistinguishable from a PJI. However, the pain in soft tissue infection may not be worse with movement, and a joint effusion (ballotable swelling) may be absent. Nonetheless, given the very low sensitivity and specificity of clinical findings, and high risk of subsequent spread of local infection to the joint, infectious diseases and orthopedic surgeon consultation should be sought for appropriate evaluation of the joint.

Sterile Joint Inflammation

Patients with prosthetic joints and inflammatory arthroplasties such as rheumatoid arthritis or systemic lupus erythematosus occasionally experience inflammation of their prosthesis from activation of the underlying disease. Gout and pseudogout may also result in joint pain that may be clinically indistinguishable from a PJI.

Does This Patient Have a PJI?

Distinguishing septic from aseptic implant failure preoperatively may be exceedingly difficult, but is of great importance. Failure to recognize a PJI at or before revision surgery will lead to implantation of a new prosthesis into an infected surgical site without appropriate debridement, and inadequate local or systemic antimicrobial treatment. This might lead to persistent infection and ultimately implant failure. On the other hand, suspicion of PJI where there is none can result in delayed reimplantation, extended length of hospital stay, increased number of surgeries and anesthetic exposures (with attendant risks), prolonged patient immobilization and rehabilitation, and increase in overall costs. No single clinical feature, laboratory test, or imaging study can unequivocally establish a diagnosis of PJI before surgery. It requires careful synthesis of information from multiple investigations. The tools available to clinicians and orthopedic surgeons in making a correct diagnosis, their accuracy and reliability, is the subject of discussion in the next section.

History and Physical Examination

The symptoms and signs of PJIs are variable, and depend to a large extent on the time of presentation and mechanism of infection. Pain at the implant site is the presenting symptom in over 90% of patients. It is of acute onset in early infections and may be associated with joint swelling, tenderness, erythema, and differential warmth. Systemic symptoms like fever, chills, rigors, and general malaise may occur, but are neither sensitive nor specific for PJI. Signs of surgical site infection, such as wound dehiscence and purulent drainage in the early postoperative period, should prompt suspicion for a PJI, as the implant is not hermetically separated from the superficial wound in the early postoperative period. Indeed, the presence of a sinus tract communicating with the device is diagnostic of PJI, although this is not a common finding. Late infections occur mostly from hematogenous seeding, and also tend to present with acute pain. Delayed infections are the most difficult to pick up, as the symptoms and local signs are subtle or absent. Joint pain, when present, is insidious in onset, is persistent and progressive, and may be associated with implant loosening and functional deterioration. Important historical information includes the type of prosthesis, date of implantation, past surgeries on the joint, comorbid conditions, current or prior antimicrobial therapies, and drug allergies or adverse effects, as these affect treatment decisions and prognosis. The elderly are disproportionately represented in total joint recipients and PJIs, and so a complete preanesthetic medical evaluation (PAME) is almost always indicated.

Laboratory Tests

Inflammatory Markers of Prosthetic Joint Infections

The peripheral WBC count, percentage of segmented neutrophils and band neutrophils, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) levels are routinely used to screen for PJIs. These tests, however, are neither sensitive nor specific enough to firmly rule out or establish the diagnosis. The WBC count may be normal in indolent infections due to lack of a systemic inflammatory response, and may be elevated in concurrent nonarticular infections. ESR, a nonspecific acute inflammatory marker, is usually elevated after joint surgery, and remains high for up to three months after surgery. Therefore, a single measurement is not useful particularly in early infections. Likewise, CRP is elevated after arthroplasty, and remains elevated for up to three weeks. Various studies have evaluated the performance characteristics and accuracy of these screening tests with varying results. In a recent meta-analysis from our group, CRP was found to have the best discriminative function of the three (Table 201-2). Newer cytokine markers of PJI, including IL-6, procalcitonin, and TNF-alpha are being studied with promising results, but there is currently not enough evidence to justify their routine use in the preoperative diagnosis of PJI.