Prosthetic joint infection: Treatment

INTRODUCTION — Goals for management of prosthetic joint infection (PJI) include alleviation of symptoms and restoration of function [1-5].

Issues related to treatment of PJI will be reviewed here. Issues related to the epidemiology, clinical manifestations, diagnosis, and prevention of PJIs are discussed separately. (See "Prosthetic joint infection: Epidemiology, microbiology, clinical manifestations, and diagnosis" and "Prevention of prosthetic joint and other types of orthopedic hardware infection".)

Issues related to osteomyelitis are discussed separately. (See "Nonvertebral osteomyelitis in adults: Clinical manifestations and diagnosis" and "Nonvertebral osteomyelitis in adults: Treatment".)

SURGICAL MANAGEMENT

Overview — In general, management of PJIs consists of surgery and antimicrobial therapy. The approach depends on the timing and microbiology of infection, condition of the joint and implant, quality of the soft tissue envelope, and individual patient circumstances.

Whenever possible, initiation of antibiotic therapy should be delayed until specimens for culture are obtained (via joint aspiration, joint debridement, and/or hardware removal).

Surgical options include debridement and retention of prosthesis, resection arthroplasty with reimplantation (in one or two stages), resection arthroplasty with no reimplantation, or amputation. The choice of surgical approach is based on data from cohort studies [1].

Patients with a well-fixed prosthesis with no sinus tract within approximately 30 days of prosthesis implantation or <3 weeks of symptom onset may be candidates for debridement and retention of prosthesis, followed by systemic antibiotic therapy guided by microbiologic results [1]. (See 'Debridement and retention of prosthesis' below.)

Patients who do not meet criteria for debridement and retention of prosthesis may be considered for resection arthroplasty with reimplantation (in one or two stages) (see 'Resection arthroplasty with reimplantation' below):

●One-stage exchange arthroplasty consists of prosthesis resection with debridement of soft tissue and bone, patient redraping (as if performing a new surgical procedure, with different instruments), and implantation of a new prosthesis during the same surgery, followed by systemic antibiotic therapy guided by microbiologic results [1]. This procedure has been advocated in patients with an adequate soft tissue envelope, provided microbiology results are known preoperatively and the pathogen(s) are susceptible to antimicrobials with excellent oral bioavailability. One-stage exchange arthroplasty is less commonly performed in the United States as compared to that in Europe. (See 'One-stage exchange arthroplasty' below.)

●Two-stage exchange arthroplasty consists of prosthesis resection with debridement of soft tissue and bone and placement of a joint spacer with antimicrobial-impregnated cement, followed by systemic antibiotic therapy for four to six weeks, and subsequent implantation of a new prosthesis once the infection is considered controlled [1,6,7]. Two-stage exchange arthroplasty is the most common approach for management of PJI in the United States. (See 'Two-stage exchange arthroplasty' below.)

Long-term antibiotic suppression may be warranted in patients with residual hardware following surgery (eg, patients who undergo debridement with retention of prosthesis or patients who undergo one-stage exchange arthroplasty). (See 'Suppressive therapy' below.)

Patients who do not meet criteria for reimplantation may warrant permanent resection arthroplasty [1]. This includes nonambulatory patients, patients who are poor candidates for major surgery (including those with limited bone stock or soft tissue coverage), patients with infection due to a highly resistant organism, and patients who failed a previous exchange arthroplasty in whom the risk of subsequent PJI after another implantation is deemed unacceptably high. Patients with total knee arthroplasty infection may undergo permanent resection arthroplasty with arthrodesis [2]. (See 'Permanent resection arthroplasty' below.)

Amputation should be the last option; it may be appropriate for patients who do not meet criteria for the above approaches. (See 'Amputation' below.)

Debridement and retention of prosthesis — Debridement and retention of prosthesis (also termed "debridement, antibiotics, and implant retention" [DAIR], or "irrigation and debridement with polyethylene liner exchange") may be an appropriate strategy in the following circumstances [1]:

●Diagnosis of PJI within approximately 30 days of prosthesis implantation, or within approximately three weeks of symptom onset, in patients with a well-fixed prosthesis with no sinus tract [8]. In such cases, eradication of infection occurs in up to 70 percent of patients [9-12].

●Patients who are poor surgical candidates for resection arthroplasty; in such cases, the likelihood of relapsed infection is greater than for patients who undergo resection arthroplasty [1,13].

Debridement and retention of prosthesis is not an effective strategy if the prosthesis is loose or if a sinus tract is present; in such cases, resection arthroplasty is required for definitive management.

Debridement should be performed via open arthrotomy to facilitate polyethylene liner exchange; this approach has been associated with better outcomes than arthroscopic lavage (which precludes liner exchange) [9,12]. Extensive soft tissue debridement should be performed, including debridement of the implant surface and behind polyethylene surfaces. In the setting of significant soft tissue infection, more than one debridement may be warranted.

Following debridement with polyethylene liner exchange and retention of prosthesis, the initial approach to antibiotic therapy depends on the microbiology of infection:

●For patients with staphylococcal PJI who undergo debridement and retention of prosthesis, antibiotic therapy consists of pathogen-specific intravenous therapy in combination with rifampin for the first two to six weeks following debridement (table 1) [1,10,11,14].

•For patients with staphylococcal infection associated with hip, elbow, shoulder, or ankle PJI, subsequent therapy consists of pathogen-specific oral therapy in combination with rifampin to complete a total duration of three months [10,11,14-17].

•For patients with staphylococcal infection associated with knee PJI, subsequent therapy consists of pathogen-specific oral therapy in combination with rifampin to complete a total duration of six months [1,18].

Options for oral therapy that may be combined with rifampin are summarized in the table (table 2); fluoroquinolones are preferred. Clindamycin should be avoided due to its increased risk of Clostridioides difficile infection [9,10,19]. (See 'Use of adjunctive rifampin' below.)

For patients with staphylococcal PJI who cannot take rifampin because of drug resistance, allergy, toxicity, intolerance, or drug-drug interactions, antibiotic therapy consists of pathogen-specific intravenous therapy for four to six weeks (before transitioning to oral antibiotic suppression, if warranted) [1].

In carefully selected cases of staphylococcal PJI managed with debridement, retention of prosthesis, and rifampin-based therapy, eradication of infection can occur in nearly 90 percent of cases; among patients with staphylococcal PJI managed with debridement and retention in the absence of rifampin, eradication of infection occurs in 50 to 60 percent of cases [20-24]. Risk factors associated with treatment failure include presence of a sinus tract, symptom duration ≥21 days prior to debridement, and lack of a rifampin-based antibiotic regimen [11].

●For patients with PJI due to other pathogens who undergo debridement and retention of prosthesis, antibiotic therapy consists of pathogen-specific intravenous therapy or highly bioavailable oral therapy for four to six weeks (table 1) [1].

In a retrospective study including more than 460 patients with streptococcal PJI managed with implant retention, this approach was associated with treatment failure in 42 percent of cases [25]. Factors associated with failure included rheumatoid arthritis, late postoperative infection, and bacteremia; factors associated with success included exchange of removable components and treatment for at least 21 days with a beta-lactam agent.

Following administration of antibiotic therapy as described above, indefinite antibiotic suppression with an oral regimen may be warranted (table 3) [1]. The decision to use suppressive therapy must be made based on individual clinical circumstances, taking into consideration the nature of the organism and its antimicrobial susceptibility pattern, the potential for implant loosening and loss of bone stock, the nature of soft tissues, the hazards of prolonged antibiotic therapy, candidacy for further orthopedic surgery, and the age and immune status of the patient [1].

Resection arthroplasty with reimplantation — In general, patients with PJI presenting >30 days after prosthesis implantation require prosthesis removal for definitive management [26].

Options for surgical management include one-stage or two-stage arthroplasty; resection and reimplantation are performed in one or two procedures. The two-stage procedure is associated with the highest success rates to prevent PJI recurrence and is the preferred procedure in the United States; the one-stage procedure is more commonly performed outside the United States [1,2,15,27,28]. There are no randomized clinical trials comparing outcomes among patients who undergo one-stage or two-stage procedures [28]. Moreover, a relatively small number of patients have been included in case series and comparative investigations to define patients who are candidates for either procedure.

Candidates for one-stage exchange arthroplasty include patients with all of the following [1]:

●Total hip or knee arthroplasty

●Good soft tissue coverage (in the absence of a sinus tract)

●Good bone stock (no bone grafting anticipated)

●Microbiology known preoperatively

●Pathogen(s) are susceptible to antimicrobials with high oral bioavailability (to facilitate suppressive therapy)

●Immunocompetent

Candidates for two-stage exchange arthroplasty include patients with all of the following [1]:

●No prior two-stage exchange arthroplasty for infection (or prior two-stage exchange arthroplasty with other reason for failure); in selected circumstances, more than one two-stage exchange arthroplasty can be successful if the first one fails.

●Delayed reimplantation technically feasible.

●Good functional outcome anticipated.

●Patient able and willing to undergo two surgeries.

Patients with sinus tracts, poor soft tissue coverage, and/or difficult to treat pathogen(s) (such as methicillin-resistant staphylococci, enterococci, and Candida species) are typically managed with two-stage exchange arthroplasty.

One-stage exchange arthroplasty — One-stage exchange arthroplasty consists of prosthesis resection with debridement of the soft tissue and bone, patient redraping (as if performing a new surgical procedure, with different instruments), followed by implantation of a new prosthesis during the same surgery; many favor the use of antibiotic-impregnated cement to fix the new prosthesis [1,29]. One-stage exchange arthroplasty is less commonly performed in the United States compared with Europe.

In some circumstances, one-stage exchange arthroplasty is performed inadvertently, such as in the setting of revision for presumed aseptic loosening; in such cases, the diagnosis of PJI is made based on intraoperative cultures [30].

Following one-stage exchange arthroplasty, the initial approach to antibiotic therapy depends on the microbiology of infection:

●For patients with staphylococcal PJI who undergo one-stage exchange arthroplasty, antibiotic therapy consists of pathogen-specific intravenous therapy in combination with rifampin for two to six weeks (table 1), followed by pathogen-specific oral therapy in combination with rifampin to complete a total duration of three months [1]. Options for oral therapy that may be combined with rifampin include fluoroquinolones, trimethoprim-sulfamethoxazole, doxycycline, minocycline, dicloxacillin, and first-generation cephalosporins (table 2). (See 'Use of adjunctive rifampin' below.)

For patients with staphylococcal PJI who cannot take rifampin because of drug resistance, allergy, toxicity, or intolerance, antibiotic therapy consists of pathogen-specific intravenous therapy for four to six weeks [1].

●For patients with PJI due to other pathogens who undergo one-stage exchange arthroplasty, antibiotic therapy consists of pathogen-specific intravenous therapy or highly bioavailable oral therapy for four to six weeks (table 1) [1].

Thereafter, indefinite oral antibiotic suppression may be warranted (table 3) [1]. Determination regarding use of suppressive therapy must be made based on individual clinical circumstances, taking into consideration the nature of the organism and its antimicrobial susceptibility pattern, whether cultures are positive at the time of the one-stage revision, the potential for implant loosening and loss of bone stock, the nature of soft tissues, the hazards of prolonged antibiotic therapy, candidacy for further orthopedic surgery, and the immune status of the patient.

In selected case series, eradication of infection occurs in approximately 80 percent of patients who undergo one-stage exchange arthroplasty [24,31]. In one systematic review including 22 articles and more than 960 patients who underwent single-stage revision arthroplasty of an infected hip or knee joint, recurrent infection was observed in up to 18 percent of patients at ≤2 years follow-up [28].

Two-stage exchange arthroplasty — A two-stage exchange arthroplasty consists of:

●Stage 1: Prosthesis resection with debridement of soft tissue and bone and placement of an antibiotic-impregnated spacer, followed by systemic antibiotic therapy for four to six weeks.

●Stage 2: Implantation of new prosthesis; the need for additional antibiotic therapy is determined by cultures obtained at the time of reimplantation [1,6,7].

Eradication of infection occurs in more than 85 percent of patients who undergo two-stage exchange [24,32,33].

Stage one — Stage one consists of prosthesis resection with debridement of soft tissue and bone and placement of a spacer, followed by systemic antibiotic therapy for four to six weeks. We favor a two-week interval off of antibiotic therapy prior to hardware removal, to improve the diagnostic yield of operative cultures.

Antibiotic-impregnated cement spacers are commonly used in between stages; it is uncertain whether this practice confers additional benefit beyond the benefit associated with systemic antibiotic therapy [1,34-39]. Most commercially available antibiotic-impregnated cements are prepared with aminoglycosides, which have activity against gram-negative organisms. In the setting of infection due to gram-positive organisms, it may be appropriate to mix vancomycin with a commercial preparation containing an aminoglycoside [40].

A static or articulating spacer may be used; the choice does not appear to influence infection-related outcomes [41]. Use of an articulating spacer may permit earlier ambulation, increased range of motion, and may allow for a more extended antibiotic-free period prior to hardware reimplantation; use of a static spacer may limit range of motion and lead to muscle atrophy and arthrofibrosis [6,7].

Following removal of the infected prosthesis, antimicrobials with activity against the infecting organism(s) are typically administered for four to six weeks (table 1) [1].

It is can be difficult to know whether eradication of infection has been achieved following completion of systemic antimicrobial. The optimal approach to evaluating for signs of infection is uncertain [2]:

●We favor checking inflammatory markers (erythrocyte sedimentation rate or C-reactive protein) after completing antibiotics, prior to new prosthesis implantation [1]. Persistently elevated inflammatory markers without an alternative explanation should prompt workup for persistent infection (including repeat aspiration, frozen section, radiographic imaging, and/or open biopsy and culture) prior to reimplantation. However, the correlation between persistently elevated inflammatory markers and likelihood of residual infection at the time of new prosthesis implantation is not clear [5,42-44]. (See 'Monitoring during treatment' below.)

●Some favor obtaining synovial fluid culture to assess sterilization of the joint space following an observation period (at least two weeks) off all antimicrobials prior to new prosthesis implantation; others favor proceeding with new prosthesis implantation immediately following completion of antimicrobials if there are no clinical signs of infection [1]. In general, we favor obtaining synovial fluid cultures in selected cases when there is clinical concern regarding the presence of persistent infection. Molecular methods are promising but thus far, data are insufficient to support routine use of this tool [45]. Patients with known or suspected residual infection should undergo further debridement with spacer exchange and further antimicrobial therapy [2].

In one retrospective study including more than 250 patients with PJI who underwent two-stage exchange arthroplasty, the risk of failure was increased among patients with a positive culture at reimplantation (45 vs 21 percent) [46]. In another retrospective study including more than 150 patients with PJI, the presence of positive cultures obtained at the time of new prosthesis implantation was not associated with a worse outcome [47].

Stage two — Stage two consists of new prosthesis implantation. Arthrotomy is performed with operative inspection, collection of cultures and frozen section histology [48]. If frozen section is negative, hardware reimplantation may be performed. If frozen section is positive, then a determination of whether or not to proceed with implantation of new hardware must be made; intraoperative findings are important in such decision-making.

If there are no apparent signs of infection, the new prosthesis may be implanted; many favor the use of antibiotic-impregnated cement to fix the new prosthesis [1]. Empiric antibiotic therapy (guided by culture data obtained at the time of the original resection and at the time of reimplantation) should be administered pending final operative culture results.

Patients with positive cultures at the time of reimplantation should be managed with reinstitution of antimicrobial therapy to complete another treatment course, followed by chronic suppressive antimicrobial therapy. Patients with negative cultures at the time of reimplantation do not require further antimicrobial therapy.

Permanent resection arthroplasty — Permanent resection arthroplasty may be appropriate for patients who must undergo resection for management of infection, but who do not meet criteria for reimplantation or for whom reimplantation does not offer a significant functional improvement (such as patients who are nonambulatory for other reasons) [1]. Such patients may include nonambulatory patients, patients who are poor candidates for major surgery (including those with limited bone stock or soft tissue coverage), patients with infection due to highly resistant organisms, and patients who failed a previous exchange in whom the risk of recurrent infection after another implantation is deemed unacceptably high.

Patients with total hip arthroplasty infection may undergo permanent resection arthroplasty without arthrodesis (Girdlestone procedure); this is frequently associated with limb length discrepancy [49]. Retention of a prosthetic spacer may be associated with a reasonable outcome, based on limited data [50].

Patients with total knee arthroplasty infection may undergo permanent resection with arthrodesis; approaches include intramedullary nailing or external fixation (with or without placement of an antibiotic cement spacer) [2]. In such cases, adequate bone must be present to create a fusion, and the limb will be shorter than the contralateral limb. Patients will not have knee mobility, but the limb will be preserved.

Following permanent resection arthroplasty, a four- to six-week course of pathogen-directed parenteral antimicrobial therapy or highly oral bioavailable agent should be administered (table 1) [51].

Eradication of infection occurs in 60 to 100 percent of patients who undergo permanent resection arthroplasty, which is lower than the reported efficacy of two-stage exchange arthroplasty [1]. An increased risk of reinfection has been observed among patients with a preoperative sinus tract who undergo resection hip or knee arthroplasty with placement of a permanent joint spacer (either fully constructed of polymethylmethacrylate or with new prostheses); in one cohort study including 51 such patients, administration of antimicrobial suppression therapy did not prevent subsequent PJI [52].

Amputation — Amputation may be warranted in the following circumstances (any of the following) [1]:

●Necrotizing fasciitis

●Severe bone loss

●Inadequate soft tissue coverage

●Failed prior attempt of resection arthroplasty or arthrodesis to control infection

●Functional benefit to amputation over resection arthroplasty or arthrodesis

For patients in whom all infected bone and soft tissue was resected in the absence of bacteremia or sepsis, pathogen-specific antimicrobial therapy should administered for 24 to 48 hours following amputation (table 1).

For patients with bacteremia or sepsis, the duration of therapy should be guided by these conditions (see separate topics).

For patients with residual infected bone and/or soft tissue, a four- to six-week course of pathogen-directed parenteral antimicrobial therapy or highly oral bioavailable agent should be administered (table 1) [1]. (See "Nonvertebral osteomyelitis in adults: Treatment".)

Patients who are not surgical candidates — Antimicrobial therapy without accompanying surgery most often results in a delay in appropriate management and confusion regarding the microbiologic diagnosis. Therefore, this approach is warranted only for patients unable to undergo even a single surgical procedure [2].

In such cases, joint aspiration for microbiologic studies should be performed prior to initiation of antibiotics. Depending on culture results, patients should receive four to six weeks of pathogen-directed intravenous or highly bioavailable oral antimicrobial therapy (table 1). (See 'Definitive therapy' below.)

Thereafter, indefinite antibiotic suppression with an oral regimen is warranted (table 3). (See 'Suppressive therapy' below.)

ANTIBIOTIC THERAPY — Whenever possible, initiation of antibiotic therapy should be delayed until specimens for culture are obtained (via joint aspiration, joint debridement, and/or hardware removal).

In general, management of prosthetic joint infection usually includes a prolonged course of intravenous antibiotic therapy. Data regarding use of oral antibiotic therapy are emerging [53]. (See "Nonvertebral osteomyelitis in adults: Treatment", section on 'Systemic therapy'.)

Suggested antibiotic regimens are outlined in the table (table 1). Antibiotic therapy should be tailored to culture and susceptibility data when available.

Empiric therapy — For patients with PJI who present with sepsis or are otherwise too unstable to wait for culture data to guide therapy, empiric antibiotic therapy may be justified.

In such cases, empiric treatment of PJI should include activity against staphylococci (including methicillin-resistant strains) and aerobic gram-negative bacilli. Reasonable regimens include vancomycin in combination with a third- or fourth-generation cephalosporin (table 1).

Antibiotic therapy should be tailored to culture and susceptibility data when available.

Definitive therapy — The approach to antimicrobial therapy (including duration of therapy) depends on the surgical approach, as discussed above. (See 'Surgical management' above.)

Antimicrobial therapy should be tailored to culture data, as discussed in the following sections.

Staphylococci

Staphylococcus aureus — In general, definitive therapy for treatment of PJI due to S. aureus consists of parenteral antibiotic therapy; regimens are summarized in the table (table 1). In addition, use of adjunctive agents for treatment of S. aureus PJI may be warranted in some circumstances. (See 'Use of adjunctive rifampin' below.)

Parenteral agents with activity against methicillin-sensitive S. aureus (MSSA) include nafcillin, oxacillin, and cefazolin (table 1). Ceftriaxone may be as effective as other anti-staphylococcal beta-lactams for treatment of oxacillin-susceptible staphylococcal PJIs, based on retrospective data [47,54].

For outpatient antimicrobial therapy, treatment with nafcillin or oxacillin may be continued via infusion pump; alternatively, treatment with cefazolin may allow greater convenience. In one retrospective study, outpatient treatment of serious MSSA infections with cefazolin was associated with fewer drug emergent effects (12 versus 31 percent) and premature discontinuations (7 versus 34 percent) compared with nafcillin [55].

Patients with an MSSA infection and a history of type I hypersensitivity (anaphylaxis) to beta-lactams can be treated with either vancomycin or daptomycin [1]. Use of linezolid is limited by risks of myelosuppression, peripheral neuropathy, and optic neuritis.

The antibiotic of choice for treatment of PJI due to methicillin-resistant S. aureus (MRSA) is vancomycin; it is the antibiotic agent for which there is the greatest cumulative clinical experience, although there are no controlled trials. Acceptable alternative agents to vancomycin for treatment of PJI due to MRSA include daptomycin (and teicoplanin, where available).

Use of adjunctive agents for treatment staphylococcal PJI may be warranted in some circumstances. (See 'Use of adjunctive rifampin' below.)

Antibiotic agents that warrant further study for treatment of staphylococcal PJI include ceftaroline, telavancin, and dalbavancin [56]. We do not favor use of trimethoprim-sulfamethoxazole, linezolid, tedizolid, clindamycin, fluoroquinolones, quinupristin-dalfopristin, or tigecycline for definitive treatment of PJI due to S. aureus [1,14,57,58]. (See "Nonvertebral osteomyelitis in adults: Treatment" and "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Treatment of bacteremia".)

Coagulase-negative staphylococci — Most coagulase-negative staphylococci are methicillin resistant (an exception is Staphylococcus lugdunensis, which is almost universally methicillin susceptible). (See "Staphylococcus lugdunensis".)

In general, the approach to treatment of PJI due to coagulase-negative staphylococci is the same as the approach for treatment of PJI due to S. aureus. (See 'Staphylococcus aureus' above.)

Empiric treatment for PJI due to coagulase-negative staphylococci should consist of an agent with activity against methicillin-resistant staphylococci. If susceptibility testing demonstrates methicillin susceptibility, a beta-lactam with activity against methicillin-susceptible staphylococci should be used (table 1).

Use of adjunctive rifampin — For patients with staphylococcal PJI and residual hardware following surgery (eg, patients who undergo debridement with retention or patients who undergo one-stage exchange), we favor the use of rifampin (in combination with at least one other anti-staphylococcal agent) (table 2 and table 1). (See 'Debridement and retention of prosthesis' above and 'One-stage exchange arthroplasty' above.)

Rifampin has activity against microorganisms in biofilms, which are important in the pathophysiology of staphylococcal osteomyelitis (particularly in the setting of hardware) [10,14,59]. (See "Prosthetic joint infection: Epidemiology, microbiology, clinical manifestations, and diagnosis", section on 'Biofilm'.)

There are no data from large, randomized trials evaluating the efficacy of adjunctive rifampin for treatment of PJI [60]. In one systematic review and meta-analysis including 64 studies and more than 4000 patients with staphylococcal hip and/or knee PJI managed with debridement and prosthesis retention, the pooled risk ratio for efficacy of rifampin effectiveness was 1.10 (95% CI 1.00-1.22), suggesting rifampin may provide only modest benefit for prevention of treatment failure [60].

Rapid emergence of resistance in the setting of rifampin monotherapy is common. Therefore, rifampin should be initiated several days after surgical debridement and initiation of anti-staphylococcal therapy (eg, once the burden of organisms has been reduced). Rifampin should be coadministered with another active antibacterial agent.

Rifampin has significant drug-drug interactions and a high risk of toxicity; careful consideration of an individual's medications and comorbidities (particularly history of liver disease) is warranted prior to initiation of rifampin.

Gram-negative bacilli — For treatment of PJI due to gram-negative organisms, we favor fluoroquinolones (if susceptibility testing confirms sensitivity) since they have high bone penetration with oral administration. Other antibiotics for treatment of PJI due to gram-negative organisms include parenteral beta-lactams and carbapenems. Antibiotic dosing is summarized in the table (table 1).

PJIs due to Pseudomonas aeruginosa are difficult to cure, even with debridement and hardware removal. Agents with activity against P. aeruginosa include ciprofloxacin, levofloxacin, ceftazidime, cefepime, and meropenem (table 1) [1].

Enterococci — Treatment for enterococcal PJI should be tailored to susceptibility results. It is unclear whether combination therapy is superior to monotherapy [61]. We favor combination therapy in the setting of retained hardware; the regimen of ampicillin and ceftriaxone is generally well tolerated [62]. However, if antibiotic-impregnated material containing gentamicin is implanted at the time of debridement, monotherapy is likely sufficient; in such cases, we favor treatment with intravenous ampicillin or penicillin.

The treatment of choice for penicillin-resistant enterococcal PJI is vancomycin; acceptable alternative agents to include daptomycin (and teicoplanin, where available) [1].

Streptococci — The treatment of choice for streptococcal PJI is penicillin or ampicillin (table 1) [63]. Ceftriaxone is an equally acceptable agent and is favored for dosing convenience in the outpatient setting. Patients allergic to beta-lactams may be treated with vancomycin.

Cutibacterium acnes — Shoulder infection is the most common presentation of PJI due to Cutibacterium (formerly Propionibacterium) acnes. Options for treatment of infection due to C. acnes include penicillin or ceftriaxone (table 1); patients allergic to beta-lactams may be treated with vancomycin [1]. (See "Invasive Cutibacterium (formerly Propionibacterium) infections".)

Fungi — Management of fungal PJI is discussed separately. (See "Candida osteoarticular infections".)

Culture-negative — The most common cause of culture-negative PJI is administration of antibiotic therapy prior to collection of specimens for culture [64]. Other causes include infection with a pathogen that cannot be identified using routinely available methods and noninfectious mimic of PJI [2]. Techniques for identifying a pathogen in culture-negative PJI are discussed elsewhere. (See "Prosthetic joint infection: Epidemiology, microbiology, clinical manifestations, and diagnosis", section on 'Culture-negative infection'.)

Patients with culture-negative PJIs should receive antimicrobial agents with activity against gram-positive and gram-negative pathogens. (See 'Empiric therapy' above.)

Patients with culture-negative PJI have similar outcomes as patients with positive cultures following standard treatment regimens. In a retrospective study including 60 patients with PJI and negative cultures, five-year survival free of treatment failure was 94 percent in patients treated with two-stage exchange arthroplasties and 71 percent in patients who underwent debridement with retention of their prosthesis [64].

Suppressive therapy — Long-term oral antibiotic suppression may be warranted for patients with PJI and residual hardware following surgery (eg, patients who undergo debridement with retention or patients who undergo one-stage exchange). (See 'Debridement and retention of prosthesis' above and 'One-stage exchange arthroplasty' above.)

Decisions regarding the use and duration of suppressive therapy must be made based on individual clinical circumstances, taking into consideration the pathogen and its antimicrobial susceptibility pattern, the potential for implant loosening and loss of bone stock, the characteristics of soft tissues, the hazards of prolonged antibiotic therapy, candidacy for further orthopedic surgery, and age and immune status of the patient [1].

Options for suppressive therapy are summarized in the table (table 3). The optimal duration of suppressive therapy is uncertain. Many favor at least three to six months of therapy; some favor indefinite therapy [2]. Individual patient factors (such as clinical course, microbiology, patient preference) may impact the duration of suppressive therapy.

Suppressive oral antibiotic therapy typically postpones rather than prevents treatment failure [9,23,47]. Studies evaluating use of suppressive oral antibiotic therapy include:

●An observational study including 108 patients with knee implant infection managed with debridement, hardware retention, and initial intravenous antibiotic therapy followed by an extended course of oral antibiotic therapy; treatment failure rates were comparable among those treated less than 12 months and those treated longer than 12 months [65].

●A randomized trial including more than 100 patients with PJI managed with two-stage exchange arthroplasty, followed by randomization to three months of oral antibiotic suppressive therapy or no further antibiotic treatment; repeat infection during 24-month follow-up occurred less frequently among patients treated with suppressive antibiotics (5 versus 19 percent) [66].

●A cohort study including more than 130 patients with retained PJI; oral antibiotic suppressive therapy was effective and well-tolerated during two years of follow-up in 61 percent of cases [67].

●A cohort study including more than 80 patients with orthopedic infection in the setting of retained hardware; oral antibiotic suppressive therapy was successful among those treated for three months but not among those treated for six months [68].

●A retrospective review including more than 90 patients with PJI; the infection-free rate at five years was greater among those who received oral antibiotic suppressive therapy for at least six months than among those who did not (68 versus 41 percent) [69].

Duration of therapy — The duration of antimicrobial therapy for treatment of PJI is tailored to the surgical management, as discussed above. (See 'Surgical management' above.)

In general, management of complex orthopedic infections usually includes a prolonged course of intravenous antibiotic therapy; data regarding shortened antibiotic duration and use of oral antibiotic therapy are emerging [53]:

●In a randomized trial including 410 patients with PJI who underwent surgery and initial intravenous antibiotic treatment (mean duration 9 days; range 5 to 15 days) followed by a course of oral antibiotic therapy (in the absence of long-term suppressive therapy), persistent infection within 2 years after completing treatment was observed more frequently among patients who received 6 weeks of antibiotic therapy than among those who received 12 weeks of antibiotic therapy (18.1 versus 9.4 percent; risk difference 8.7 percentage points, 95% CI 1.8-15.6); thus, noninferiority was not shown [70]. The difference in risk was greatest among patients who underwent debridement with prosthesis retention (30.7 versus 14.5 percent; risk difference 16.2 percentage points, 95% CI 2.9-29.5). These data are important as we strive to define the optimal approach for management of PJI.

●In one meta-analysis including 10 studies (9 observational studies and 1 randomized trial) and more than 850 patients with PJI managed with debridement and retention or two-stage exchange arthroplasty, no significant difference between short-course and long-course antibiotics was observed (relative risk 0.87, 95% CI 0.62-1.22) [71]. The generalizability of these findings is limited; further study is needed.

MONITORING DURING TREATMENT — During treatment, patients should be monitored for clinical response and for adverse effects of therapy.

Clinical response — The clinical response to treatment is monitored via examination of the joint and wound.

We typically see patients for follow-up examination every two weeks initially and then every three months to monitor response. Most patients will have significant improvement in pain, edema, and erythema of the joint within 6 to 12 weeks of surgery. It may take 6 to 12 months for symptoms to return to baseline; in some cases, symptoms never return to baseline despite resolution of infection.

In the early post-operative period, monitoring the joint for clinical response to treatment can be challenging due to difficulty distinguishing normal post-operative pain and edema from persistent joint infection. Furthermore, after surgical debridement of the joint, signs of remaining infection are often delayed because residual microbes must multiply enough to elicit signs of infection.

Lack of wound healing may be due to infection or to other factors that affect wound healing. Inflammation of the wound may be a normal finding in the early post-operative setting. Alternatively, it may be indicative of surgical site infection, which is often unrelated to the underlying joint infection. Further discussion of factors that affect wound healing, including surgical site infection, are discussed elsewhere. (See "Risk factors for impaired wound healing and wound complications" and "Overview of the evaluation and management of surgical site infection".)

Serum inflammatory markers — We obtain serum inflammatory markers (erythrocyte sedimentation rate [ESR] and C-reactive protein [CRP]) at the beginning and end of parenteral therapy (eg, at the time of transition to oral suppressive therapy, if prescribed). If test results do not show improvement at the end of parenteral therapy, we repeat the tests again in two weeks and consider the possibility of refractory infection if results again show no improvement (see 'Refractory infection' below). We generally do not monitor serum inflammatory markers weekly during or after antibiotic therapy.

Inflammatory markers (especially the ESR) may remain elevated for prolonged periods of time following resolution of infection. Furthermore, many other conditions cause elevated inflammatory markers which can further complicate interpretation of these tests. Additional detail regarding interpretation of serum inflammatory markers is found elsewhere. (See "Prosthetic joint infection: Epidemiology, microbiology, clinical manifestations, and diagnosis", section on 'Serum inflammatory markers'.)

Adverse effects due to antimicrobial therapy — Laboratory monitoring is necessary during prolonged administration of antimicrobial therapy to monitor for adverse drug effects.

For patients on parenteral antimicrobial therapy, we obtain weekly complete blood count and chemistries. Some parenteral antibiotics require additional laboratory monitoring, as detailed in the table (table 4). Further details are found elsewhere. (See "Outpatient parenteral antimicrobial therapy", section on 'Monitoring'.)

For patients on oral suppressive antimicrobial therapy, we obtain a complete blood count, serum creatinine, and alanine aminotransferase at 2, 4, 8, and 12 weeks and then every 6 to 12 months thereafter.

REFRACTORY INFECTION — Clinical recognition of refractory infection can be challenging due to the lack of specificity of clinical findings and inflammatory markers, especially in the post-operative setting, as described above. (See 'Clinical response' above and 'Serum inflammatory markers' above.)

The management of refractory infections is complex. The clinical approach should involve discussions between the patient, orthopedic surgeon, infectious diseases specialist, and other providers involved in the patient's care.

If signs and symptoms of infection or elevation of serum inflammatory markers persist, we perform further workup for refractory infection, including joint aspiration and plain radiographs. Indium white blood count scanning or PET/CT scanning can be considered, if results of the workup are negative.

The diagnosis of refractory infection is confirmed by positive synovial fluid cultures that grow the same organism as recovered during the initial infection. (See "Prosthetic joint infection: Epidemiology, microbiology, clinical manifestations, and diagnosis", section on 'Diagnosis'.)

For patients who have a confirmed refractory infection, management depends on which surgical procedure was performed to initially treat the infection (ie, whether original hardware was removed, and whether a new joint was subsequently inserted). We suggest the following approach:

●Patients in whom the infected hardware was not removed – For these patients, we suggest removal of the hardware and a one- or two-stage exchange arthroplasty. (See 'Two-stage exchange arthroplasty' above.)

●Patients who initially underwent a one-stage exchange – We suggest removal of the hardware and a two-stage exchange arthroplasty for these patients. Less optimal alternatives include a repeat one-stage exchange arthroplasty or a debridement and retention of prosthesis (DAIR) procedure. (See 'Surgical management' above.)

●Patients who had a spacer placed for a two-stage exchange – For these patients, we remove the spacer and perform repeat debridement. Before placing a new joint, we perform additional microbiologic testing to detect residual infection (including repeat aspiration, frozen section, if available, radiographic imaging, and/or open biopsy and culture). We wait at least two weeks after the completion of parenteral antibiotics before performing microbiologic tests. (See 'Two-stage exchange arthroplasty' above.)

In all cases, a repeat course of antibiotic therapy is warranted after surgery (eg, for two-stage exchange arthroplasty, the antibiotics are given between the first stage and the second stage). Further details regarding surgical and antibiotic management are detailed above. (See 'Surgical management' above and 'Antibiotic therapy' above.)

If surgical intervention is not an option, chronic oral suppressive antibiotic therapy is indicated following completion of parenteral antibiotics. (See 'Suppressive therapy' above.)

For patients in whom refractory infection is suspected but is unable to be confirmed, clinical judgment is necessary to determine the next steps in management. We often treat patients who have clinical findings compatible with ongoing infection and persistently elevated inflammatory markers as though they have confirmed refractory infection. In the absence of clinical signs consistent with refractory infection, clinical observation may be reasonable.

RECURRENT INFECTION AND REINFECTION — Patients who have signs and symptoms of recurrent infection should always undergo a complete workup, including synovial fluid analysis and culture, as described elsewhere. (See "Prosthetic joint infection: Epidemiology, microbiology, clinical manifestations, and diagnosis", section on 'Diagnosis'.)

Recurrences may be caused by the same microorganism as the initial infection or by a different organism. When they are caused by a different organism, they are often termed "reinfections." In one study of 132 episodes of PJI, 21 of 36 nonfatal clinical failures (58 percent) were caused by a new pathogen [72]. Most failures occurred with two years of the initial infection.

The management of recurrent infections and reinfections follows the same approach as initial infections. (See 'Surgical management' above and 'Antibiotic therapy' above.)

OUTCOMES — Outcomes depend on the type of surgical procedure, the microbiology of infection, and individual patient circumstances. Definitions for treatment failure have been developed by expert panels [73,74].

Overall, outcome data suggest that treatment failure occurs in 9 to 44 percent of patients [10,70,75]. The wide range of outcome data is due to many factors, including variations in study design, patient populations, clinical management, definitions of treatment failure, and duration of follow-up.

Some data suggest that the type of surgical procedure is the main variable that affects outcome [22,72]. In one study that included 200 patients with PJI and rheumatoid arthritis, the type of surgical procedure predicted the rate of survival free of treatment failure at five years of follow-up; rates for two-stage exchange, permanent resection arthroplasty, and debridement with retention of prosthesis were 79, 61, and 32 percent, respectively [22]. Treatment failure was defined as refractory or recurrent PJI confirmed by culture, histopathology, or sinus tract, death due to PJI, or PJI diagnosed at an outside hospital.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Osteomyelitis and prosthetic joint infection in adults" and "Society guideline links: Outpatient parenteral antimicrobial therapy".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or email these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Deciding to have a knee replacement (The Basics)" and "Patient education: Deciding to have a hip replacement (The Basics)")

●Beyond the Basics topics (see "Patient education: Joint infection (Beyond the Basics)" and "Patient education: Total hip replacement (Beyond the Basics)" and "Patient education: Total knee replacement (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●General approach – In general, management of prosthetic joint infection (PJI) consists of surgery and antimicrobial therapy. The approach depends on a number of factors including the timing and microbiology of infection, condition of the joint and implant, and individual patient circumstances. (See 'Overview' above.)

●Surgical management – Surgical approaches for management of PJI include debridement and retention of prosthesis (ie, DAIR), resection arthroplasty with reimplantation (ie, one- or two-stage exchange arthroplasty), permanent resection arthroplasty, or amputation. (See 'Surgical management' above.)

Selection of the most appropriate surgical approach depends on the patient's comorbidities, timing of original prosthetic implantation, and timing of onset of symptoms.

•Patients presenting >30 days after prosthesis implantation – For these patients, definitive management consists of prosthesis removal. Options for replacement arthroplasty include one-stage or two-stage procedures. (See 'Resection arthroplasty with reimplantation' above.)

-One-stage exchange arthroplasty – This procedure consists of prosthesis resection with debridement of the soft tissue and bone followed by implantation of a new prosthesis during the same surgery. (See 'One-stage exchange arthroplasty' above.)

-Two-stage exchange arthroplasty – For this method, stage one consists of prosthesis resection, debridement of soft tissue and bone, and placement of a joint spacer, followed by systemic antibiotic therapy for four to six weeks; stage two consists of implantation of a new prosthesis (see 'Two-stage exchange arthroplasty' above). This is the most common approach used in the United States.

•Patients presenting within 30 days of prosthesis implantation or <3 weeks of symptom onset – These patients may be candidates for debridement, polyethylene liner exchange, and retention of prosthesis (ie, DAIR). (See 'Debridement and retention of prosthesis' above.)

•Patients who do not meet criteria for reimplantation but must undergo resection for management of infection – These patients may warrant permanent resection arthroplasty. Amputation should be the last option considered. (See 'Permanent resection arthroplasty' above and 'Amputation' above.)

•Patients who are not surgical candidates – This approach is warranted only for patients unable to undergo even a single surgical procedure. The patients are managed with a course of parenteral antibiotics followed by indefinite chronic oral antibiotic suppression. (See 'Patients who are not surgical candidates' above.)

●Antibiotic therapy – Antibiotic options for PJI caused by various organisms are outlined in the table (table 1). (See 'Antibiotic therapy' above.)

•For treatment of PJI due to methicillin-resistant Staphylococcus aureus (MRSA), we suggest vancomycin (Grade 2C); daptomycin is an acceptable alternative agent. (See 'Staphylococcus aureus' above.)

•For patients with staphylococcal PJI and residual hardware following surgery, we suggest adjunctive use of rifampin (in combination with at least one other anti-staphylococcal agent) (table 2 and table 1) (Grade 2C). (See 'Use of adjunctive rifampin' above.)

Antibiotic therapy should be tailored to culture and susceptibility data when available. The duration of therapy depends on the surgical management. (See 'Surgical management' above and 'Suppressive therapy' above.)

●Monitoring during treatment – Patients should undergo serial examinations in the outpatient setting to monitor for refractory or recurrent infection. Serum inflammatory markers may be helpful to detect clinical failure, but these tests are nonspecific. We suggest additional laboratory tests while patients are receiving antibiotic therapy to monitor for side effects. (See 'Monitoring during treatment' above and 'Refractory infection' above and 'Recurrent infection and reinfection' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Elie Berbari, MD, FIDSA, who contributed to earlier versions of this topic review.