Total knee arthroplasty (TKA) infection is a complication requiring multiple hospitalizations, operations, and outpatient visits placing a significant burden on both patient and treating surgeons. When revision surgery is needed, two kinds of treatment can be applied----one-stage and two-stage reimplantation procedure. The two-stage reimplantation procedure is considered the gold standard for treatment of subacute and chronic deep periprosthetic infections. Placement of antibiotic-loaded spacer is a standard procedure for eradication of peri-prosthetic joint infection. For better patient outcome, the purpose of this study was to determine the success rate of total knee infection patients treating with two-stage reimplantation procedure.

*Background and study aims: Total knee arthroplasty (TKA) infection is a devastating complication requiring multiple hospitalizations, operations, and outpatient visits placing a significant burden on both patients and treating surgeons. The incidence of prosthetic joint infection after primary hip or knee arthroplasty is about 2% among the Medicare population. When revision surgery is needed, two kinds of treatment include one-stage and two-stage re-implantation procedure can be applied. Bengtsonet al., in an earlier report, evaluated 107 patients with a 75.7% success rate after one-stage revision. In the case of two-stage TKA revision success rates of 89% to 100% are noted with follow-up periods of 2 to 7.5 years. Delayed re-implantation after administration of intravenous antibiotics appears to offer better success rates than direct-exchange techniques. Therefore, the two-stage re-implantation procedure is considered the gold standard for treatment of subacute and chronic deep prosthetic joint infections (PJI). However, the protocols of the antibiotic therapy after resection arthroplasty have varied in different reports. The purpose of this study will be to evaluate the clinical outcomes of two-stage re-implantation for infected knee arthroplasty using the standardized protocol of combined parenteral and oral antibiotic therapy and the criterion for re-implantation.

*Study participants

Study period: from March 2011 to June 2019.

*Procedures and operations

-The first stage The first stage consisted of the removal of the prosthesis and all hardware, debridement of all infected and devitalized tissues, removal of all biofilms, and aqueous povidone iodine irrigation. For primary joint arthroplasty with cement, all the bone cement was removed. Samples for culture and histology were performed at the time of removal of prosthesis (ROP). The Long-leg splint was applied to the limb until the next replacement procedure. The first stage was using antibiotic-impregnated bone cement as a static spacer block. The types of antibiotic regimens were determined according to the results of cultures form the preoperative joint aspirations. In this study, 1 to 2 g of vancomycin (for gram-positive bacteria) and/or 1 to 2 g of ceftazidime (for gram-negative bacteria) per 40g package of cement would be hand-mixed according to the previous culture. Any past history of adverse events of the antibiotics used in the cement would be clarified to avoid any possible allergic reactions to the antibiotic-loaded cement. If the patient had allergic reactions to vancomycin or ceftazidime, daptomycin or gentamicin 1 to 2 g will be used for alternative antibiotic cement.

All patients received at least two weeks of intravenous antibiotics after first stage operation, and then additional 2 to 4 weeks of oral antibiotic therapy according to the laboratory and culture results performed preoperatively or at the time of the first-stage procedure. If the cultures revealed certain drug-resistant pathogens, including methicillin-resistant Staphylococcus aureus, methicillin-resistant Staphylococcus epidermidis (especially when the vancomycin MIC≧1 mg/L) or vancomycin-resistant enterococci, daptomycin or linezolid would be selected as the therapy option. Intravenous antibiotic therapy would be continued until the serum CRP decreased to normal range (less than 1 mg/dL). When the CRP levels remained normal (< 1 mg/dL), no pathogens were identified from aspirations, and there were no signs of active infection clinically, the re-implantation surgery would be scheduled.

• The second stage (reimplantation) In the second stage of reimplantation, four sets of bacterial cultures from different adjacent knee joint tissue (medial and lateral condyles, patella, and femur-tibia joint space) will be obtained. Prophylactic antibiotic will be used in the second stage. All patients will be examined clinically at 1st week, 2nd month, 6th month and one year postoperatively at the orthopedic department. Laboratory tests including complete blood counts (CBC) and differential counts (D/C), liver alanine aminotransferase (ALT), creatinine (Cre) and CRP will be performed at each visit. Free of recurrent infection will be defined when the clinical evaluation of the joint did not show any sign of infection and CRP will be less than 1 mg/dL. Antibiotic therapy, either intravenous or oral form, will be discontinued if cultures of debridement are sterile. If persistent infection is documented in the second-stage reimplantation procedure, the patients will be transferred to the medical ward for an additional 4 to 6 weeks of antibiotics (including intravenous form treatment at least for 2 weeks) until CRP sustained below 1 mg/dL.
• Antibiotic cement In this study, 1-2 g of vancomycin (for gram-positive bacteria) and/or 1-2 g of ceftazidime (for gram-negative bacteria) per 40-g package of cement will be hand-mixed according to the previous culture identified. Any past history of adverse events of the antibiotics used in the cement will be clarified to avoid any possible allergic reactions to the antibiotic-loaded cement. If patient had allergic reactions to vancomycin or ceftazidime, daptomycin or gentamicin 1-2 g will be used for alternative antibiotic cement.
• Follow-up Patients will be arranged to follow-up mainly at Orthopedic clinic regularly. Infectious Disease specialist will be consulted immediately during follow-up if necessary.
• Definition Recurrent and persisted infection Recurrent or persisted infection will be defined as swelling and pain of the joint with drainage, elevated CRP, persistent positive culture after two-stage of implantation during follow-up at clinic.
• Case record A standard case record form will be used to collect patients' information, including age, sex, underlying diseases, primary and secondary diagnoses, surgery, cultures, and the timing of replacement. Complete blood counts, differential counts, ALT, Creatinine and C-Reactive Protein will be recorded.

Inclusion criteria:

1. Patients received primary total knee arthroplasty with acute prosthetic joint infection who were failed to antibiotics treatment or chronic prosthetic joint infection of knee AND
2. Documented prosthetic joint infection by pathology findings or cultures from either bone tissue, joint effusion, synovial fluid, or purulent discharge from aseptic operation technique
3. Patient age ≧ 20 years old

Exclusion criteria:

1. Culture result shown Candida, mycobacterium or slow growing pathogens (e.g. nocardia)
2. Patients with severe systemic diseases can not received second stage operation
3. Pregnancy, terminal stage patients, inmates, aboriginal patients, student, mental disorder patients and other vulnerable patients will be excluded.

The investigators collected cases of infected TKAs treated with two-stage reimplantation procedure since 2011 in National Taiwan University Hospital.