• Summary
  Click here to print a Quick Reference Guide
• Background
• Clinical Diagnosis
• Investigation
• Treatment
• Non-Antimicrobial Treatment
• Empirical Antimicrobial Treatment
• Directed Antimicrobial Treatment (when microbiology results are known)
• Duration of Treatment
• Treatment Failure

Key Links

• Staphylococcus species
• Enterococcus species
• Streptococcus species

Criteria for use of guideline

• < 6 weeks from hip or knee replacement surgery and increasing or new joint pain,with any one of: swelling, erythema, tenderness, wound discharge, fever, systemic upset.

Investigations required

• Full blood count and erythrocyte sedimentation rate [ESR] or plasma viscosity [PV].
• C-reactive protein [CRP], urea and electrolytes, liver function tests, glucose.
• urine for glycosuria.
• Urinalysis and MSU for microbiology if positive.
• Plain X-ray, AP + lateral of affected joint +/- USS of hip [if deep collection suspected].
• If wound is discharging, send fluid or pus to Microbiology. If not possible then send wound swab.
• Blood cultures 2 sets if systemic signs of infection.

Non-Antimicrobial Management

• Refer urgently to surgeon who performed original operation, otherwise to on-call orthopaedic team.
• Surgeon will perform / arrange joint aspiration [link to protocol for prosthetic joint aspiration]
• If pre-operative aspirate or biopsies are culture positive or clinical evidence of prosthesis infection proceed to early surgical debridement with retention of prosthesis.
• Surgical debridement should be through the same surgical approach and involve dislocation of the hip or full eversion of the patella to allow a radical debridement.
• Send 4 tissue specimens and 1 specimen of joint fluid or pus to microbiology [not swabs].

Antimicrobial treatment

DO NOT START EMPIRICAL ANTIMICROBIAL THERAPY UNTIL AFTER JOINT ASPIRATION OR SURGICAL DEBRIDEMENT UNLESS THE PATIENT HAS SEVERE SEPSIS OR SEPTIC SHOCK.
Empirical therapy: [microbiological diagnosis not known] iv Vancomycin * see dosing guidelines + iv Gentamicin* 2mg/kg 8-12 hourly *adjust dose according to creatinine clearance and serum levels.

IF PRELIMINARY JOINT CULTURES GROW A RECOGNISED VIRULENT PATHOGEN (e.g. Staphylococcus aureus or beta-haemolytic Streptococcus) DISCUSS COMMENCING DIRECTED ANTIMICROBIAL THERAPY PRIOR TO SURGERY WITH THE RELEVANT CONSULTANT ORTHOPAEDIC SURGEON.

IF PRELIMINARY JOINT CULTURES GROW A LOW VIRULENCE ORGANISM (E.G. MEMBERS OF NORMAL SKIN FLORA) WITHOLD ANTIMICROBIAL THERAPY UNTIL AFTER SURGERY.

Directed therapy [usually commence post operatively [once further Microbiology samples have been collected], based on the positive pre-operative cultures], other pathogens or “case-by-case” – discuss with microbiology.

Antimicrobial allergy
See table or contact microbiology for advice

Duration of treatment
6 weeks in total assuming ESR/PV and CRP are normalizing.

Switch to oral agent[s]
When wound has healed and ESR/PV and CRP approaching normal, normally after 1 week IV. Early oral switch may not be appropriate for some pathogens e.g. Staphylococcus aureus, enterococci
Switch to oral after consultation with microbiology and orthopaedic consultant.

Referral criteria: All patients must be treated by either the original surgeon or a specialist lower limb joint replacement surgeon.

Back to top

Infection of a prosthetic hip or knee replacement is an uncommon occurrence but leads to pain, disability, prolonged hospitalisation and rarely loss of a limb. Such infections carry an increased risk of mortality and are associated with considerable cost. Mandatory surveillance of orthopaedic surgical site infections has been in place in England and Wales since 2004 and shows an average infection rate of 0.92% of total hip replacement operations, 0.51% total knee replacements and 3.58% of hemiarthroplasties (Health-Protection-Agency, 2007). Risk factors for prosthetic joint infection include: post operative wound infection, delayed wound healing, operative factors [including duration of operation and haematoma formation], concurrent malignancy, diabetes or rheumatoid arthritis, poor nutritional status and obesity (Berbari et al., 1998; Lentino, 2003). Staphylococcus aureus remains the most common cause of post operative surgical wound infections following prosthetic hip and knee implantation, accounting for 45%, more than half of which are meticillin-resistant (Health-Protection-Agency, 2007). Staphylococci, both S. aureus and coagulase negative staphylococci are the most common cause of joint prosthesis infection (Toms, 2006), (Lentino, 2003). The other causes largely comprise Enterobacteriaceae, enterococci, streptococci and Pseudomonas spp. Early infections more commonly involve S. aureus, streptococci and Gram negatives.

A practical definition of prosthetic joint infection includes at least one of (Osmon et al., 2005; Trampuz et al., 2003):

1. Growth of the same organism from at least two separate joint samples and local or systemic signs of infection.
2. Histopathological evidence of acute inflammation and organisms.
3. Pus [confirmed by white cell count on microscopy of intra-operative sample] around the prosthesis at operation or from an aspirate of joint fluid.
4. A sinus tract communicating with the joint.

An infected prosthesis can be managed in several different ways:

1. Debridement with retention of the prosthesis [this guideline]
2. A two stage revision, in which the prosthesis is removed and then replaced after a period of antimicrobial therapy.
3. A single stage procedure in which the prosthesis is removed and then replaced during the same operation followed by a period of systemic antimicrobial therapy.
4. Arthrodesis [Total Knee Replacement].
5. Amputation [Total Knee Replacement].
6. Excisional arthroplasty [Total Hip Replacement].

The highest rates of success have been obtained with two-stage revision procedures but this approach requires multiple operations and prolonged immobility with their associated risks (Meehan et al., 2003). Debridement with retention of the prosthesis can salvage some implants but success has been variable (Crockarell et al., 1998; Marculescu et al., 2006). This approach is more successful in patients who present early after the initial procedure. It has also been successful in those operated on rapidly after onset of symptoms and for patients with prosthetic knee joint infections caused by Gram positive organisms other than Staphylococcus aureus (Deirmengian et al., 2003a; Deirmengian et al., 2003b). Patients with “early” infections have a better outcome than late (Laffer et al., 2006). In a small series of early knee replacement infections 95% were cured with retention of the prosthesis (Laffer et al., 2006). Prolonged symptoms prior to presentation [>8 days] and presence of a sinus tract have been associated with treatment failure (Marculescu et al., 2006).

“Early” prosthetic joint or knee infection has been variably defined but <6 weeks from hip or knee replacement surgery is the definition used in this guideline.

This guideline relates to the management of an infected hip or knee prosthesis with debridement and retention of the prosthesis and is applicable to patients who have been operated on within the preceding 6 weeks, with a short duration of symptoms [1 week] and a stable prosthesis.

Back to top

Pain is almost universally present in early prosthetic joint infections (Lentino, 2003). Local warmth, erythema, tenderness, wound discharge or effusions, systemic upset and fever all support a diagnosis of possible infection (Lentino, 2003), but are not necessarily present.

Consider early prosthetic joint infection in patients < 6 weeks from hip or knee replacement surgery and increasing or new joint pain, with any one of: swelling, erythema, tenderness, wound discharge, fever or systemic upset.
Evidence level B

Back to top

A raised peripheral blood polymorphonuclear leucocye [neutrophil] count is insensitive for the diagnosis of prosthetic joint infection (Anguita-Alonso et al., 2005), but a full blood count [FBC] is indicated at baseline assessment because of the likely need for surgery and potential affects on bone marrow or blood cells of various antimicrobials.
Evidence level D

A raised C-reactive protein [CRP] has a reported sensitivity of 61-97% but is non-specific. A positive result gives support to a clinical diagnosis of prosthetic joint infection in the absence of other potential causes of a raised CRP, not may also be raised if infection is present at a site distant from the prosthetic joint (Anguita-Alonso et al., 2005). Serial [weekly] CRP measurements are recommended to monitor response to therapy
Evidence level D

Erythrocyte Sedimentation Rate [ESR] or Plasma Viscosity [PV] have similar sensitivity to CRP, but when combined with CRP the sensitivity increases (Sanzen & Sundberg, 1997). CRP returns to normal within two weeks of surgery in non infected cases, but ESR/PV may take more than 6 weeks (Aalto et al., 1984). Serial [weekly] ESR measurements are recommended to monitor response to therapy.
Evidence level B

Normal values of either CRP or ESR/PV can be found in infected joint replacements, but for both to be normal appears to uncommon (Sanzen & Sundberg, 1997)

Urea and electrolytes [U&E], liver function tests [LFT’s] are required to assess baseline renal and hepatic function because the antimicrobials commonly used in this situation may either affect these organs or dosing may be affected by renal or hepatic dysfunction. Serial [weekly] U&E and LFT measurements are recommended during therapy, with more frequent measurement in patients with renal impairment or on nephrotoxic drugs.
Evidence level B

Two sets of blood cultures, taken at different times, should be sent from patients with a fever, any symptoms of systemic upset [e.g. chills, myalgia, general malaise] or with a systemic inflammatory response syndrome or more severe inflammatory response [link to definitions].
Evidence level B

The sensitivity and specificity of plain x-rays, computed tomography scanning or magnetic resonance imaging for diagnosis of early prosthetic joint infection has not been assessed (Lentino, 2003) A plain X-ray is required to compare with previous X-rays in order to confirm implant stability. Any evidence of loosening / bone resorption indicates an unstable implant and excludes the patient from “early infection” and hence from these guidelines, [link to Guideline for late onset [>6 weeks] infected hip or knee replacement.]

Preoperative aspiration of joint fluid has a reasonably high sensitivity [82-94%] and specificity [94-97%] for prosthetic joint infection (Anguita-Alonso et al., 2005).
If CRP or ESR/PV or neutrophil count are raised in a patient with a painful prosthesis, or clinical suspicion of prosthesis involvement is strong [discharging wound sinus, wound inflammation, systemic inflammatory response and joint pain], joint aspiration is recommended. Two samples of pus or fluid and a biopsy, preferably in laminar flow theatre, should be sent for microbiology (Spangehl et al., 1999).

Intraoperative Sampling.
Multiple intraoperative samples are required to optimize the diagnosis of prosthetic joint infection. The LTHT protocol for intraoperative sampling of suspected hip or knee prosthesis infection should be followed.
Evidence level B

Gram-stain is unreliable for the intraoperative diagnosis of prosthetic joint infection because of the very low sensitivity and specificity (Della Valle et al., 1999; Spangehl et al., 1999). Evidence level B.

Back to top

A patient with suspected prosthetic joint infection should be referred urgently to the surgeon who performed original operation or to the on-call orthopaedic team if the primary consultant is unavailable. A bed-side review of the patient by the consultant or their deputy [not F1 or F2 trainee] should provided as a matter of urgency [within 6 hours] in a patient with suspected early prosthetic joint infection and signs of a systemic inflammatory response, severe sepsis or septic shock [link to definitions], otherwise, the patient should be seen early, ideally within 24 hours.

If initial aspirates or biopsies are positive or there is strong clinical evidence of prosthesis infection proceed to early surgical debridement with retention of prosthesis.

Back to top

DO NOT START EMPIRICAL ANTIMICROBIAL THERAPY UNTIL AFTER JOINT ASPIRATION OR SURGICAL DEBRIDEMENT UNLESS THE PATIENT HAS SEVERE SEPSIS OR SEPTIC SHOCK.

IF PRELIMINARY JOINT CULTURES GROW A RECOGNISED VIRULENT PATHOGEN (e.g. Staphylococcus aureus or beta-haemolytic Streptococcus) DISCUSS COMMENCING DIRECTED ANTIMICROBIAL THERAPY PRIOR TO SURGERY WITH THE RELEVANT CONSULTANT ORTHOPAEDIC SURGEON.

IF PRELIMINARY JOINT CULTURES GROW A LOW VIRULENCE ORGANISM (E.G. MEMBERS OF NORMAL SKIN FLORA) WITHOLD ANTIMICROBIAL THERAPY UNTIL AFTER SURGERY.

Empirical therapy
When preoperative joint aspiration is negative but debridement is still carried out because of ongoing clinical suspicion of prosthetic joint infection, empirical therapy will be required. Empirical therapy will also be required in the acutely unwell patient with severe sepsis or septic shock at presentation. In this setting broad-spectrum antimicrobial cover is necessary to provide activity against the most likely pathogens. NB In this setting 2 sets of Blood cultures should be taken prior to commencing antimicrobial therapy. Between Jan 2005 – July 2007 217 different Gram positive organisms had been isolated from tissue and fluid aspirate samples of LTHT orthopaedic patients and had all been susceptible to Vancomycin *.

Commence intravenous Vancomycin * see dosing guidelines + iv Gentamicin* 2mg/kg 8-12 hourly.
*Dose or frequency may need adjusting according to creatinine clearance.

Back to top

When preoperative joint aspiration is positive, directed therapy can be commenced [after operative samples have been collected] according to the previous culture results. In each case, susceptibility of the causative organisms will determine therapy. For some of the more common causative organisms specific recommendations can be made.

Staphylococcus species
meticillin susceptible:
Initial therapy: intravenous Flucloxacillin 2g 6-hourly plus oral Rifampicin 300-600mg 12-hourly.

Oral switch: oral Flucloxacillin 500mg 6-hourly plus oral Rifampicin 300-600mg 12-hourly.

meticillin resistant:
Initial therapy: intravenous Vancomycin * see dosing guidelines plus oral Rifampicin 300-600mg 12-hourly.

Reduce Rifampicin dose to 300mg 12-hourly in severe renal impairment, creatinine clearance <30ml/min

Oral switch: case-by-case depending on susceptibility.

In the absence of high quality evidence on which to base recommendations other antimicrobial agents can be considered based on susceptibility pattern of the organism[s] and cost. These may include combinations of Rifampicin with fusidic acid or Trimethoprim.

Enterococcus species
Amoxicillin-susceptible:
Amoxicillin 2g 4 hourly iv plus Gentamicin* 1mg/kg 8-12 hourly (Raymond et al., 1995)

Amoxicillin-resistant:
Vancomycin * see dosing guidelines iv plus Gentamicin* 1mg/kg 8-12 hourly (Raymond et al., 1995).
Switch to oral antimicrobials should be discussed with Microbiology on a case-by case basis.

Streptococcus species
Penicillin-susceptible:
Initial therapy: intravenous Benzyl penicillin 1.2g 4-hourly.

Oral switch: oral Amoxicillin 500mg tds

Other organisms should be discussed with Microbiology on a case-by case basis.

Back to top

6 weeks in total is recommended assuming PV and CRP are normal. If clinical signs and symptoms have not resolved within this time frame revision surgery should be considered, with further microbiological sampling and a review of antimicrobial therapy.

Switch to oral agent[s]
Evidence to guide appropriate timing from intravenous to oral antimicrobials is lacking. In practice this usually occurs when wounds have healed and CRP is approaching normal. Normally a minimum of one weeks intravenous therapy is used but longer may be appropriate for aggressive pathogens such as Staphylococcus aureus. If regimens are not given above, cases should be discussed on a case-by case basis.

Back to top

Back to top