Study design and population. This retrospective cohort study was conducted in two large orthopedic centers, having a specialized septic surgery unit. PJI episodes caused by Enterococcus species, alone or in combination with another microorganism(s), treated at one of the participating institutions from January 2010 to December 2017 were included. Data on PJI were collected using the same definition criteria, diagnostic procedure and outcome evaluation. The study was approved by the lead ethical committee (approval No. EA04/040/14) and was conducted in accordance with the Declaration of Helsinki. The study was registered with the public clinical trial identification NCT0253022 at https://www.clinicaltrials.gov.
Study definitions. PJI was defined according to the following definition criteria, as previously used [20-22] . According to these criteria, at least one of the following criteria are needed for the diagnosis of PJI: (i) macroscopic purulence surrounding the prosthesis, (ii) presence of communicating sinus tract, (iii) increased synovial fluid leukocyte count and differential (>2000 leukocytes/µl or >70% granulocytes), (iv) isolation of enterococci from synovial fluid, periprosthetic tissue or sonication culture, (v) positive histopathology, defined as >23 granulocytes per 10 high-power fields, corresponding to type II or type III periprosthetic membrane [23]. If enterococci grew in only one microbiological specimen, the microbiological finding was sufficient for the diagnosis of PJI only if additional (non-microbiological) criterion was present, as defined above.
Infections were classified according to their temporal appearance after surgery into early (<3 months), delayed (3-24 months) and late infection (>24 months) [24]. In addition, infections were defined as acute PJI (new onset symptoms for ≤4 weeks), or chronic PJI (symptoms duration >4 weeks). The hematogenous route of PJI was defined when (i) the onset of the symptoms was >3 months after implantation and occurred after an initial uneventful initial course and (ii) the infection presented with acute onset or the same Enterococcus species grew in blood cultures or from a distant infectious focus. Each case was evaluated and classified by an interdisciplinary team of orthopedic surgeons (DA, RT, CP) and infectious disease specialists (AT, NR).
Infection-free interval describes the interval from primary implantation or last septic surgery of the prosthesis to the diagnosis of an enterococcal infection.
Treatment success was defined by absence of relapse or persistence of PJI due to enterococci or death related to enterococcal PJI.
Clinical success was defined by the presence of all following criteria at last follow-up: (i) infection-free status, characterized by a healed wound without fistula, drainage, and no recurrence of the infection, (ii) no subsequent surgical intervention for persistent or perioperative infection after re-implantation, and (iii) no PJI-related death (within 3 months).
Microbiological testing. An automated broth microdilution assay was used to determine the antimicrobial susceptibility of all antibiotics except for fosfomycin, and the results were interpreted according to European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria. For fosfomycin, Etest (bioMérieux, Marcy-l’Étoile, France) was performed in Müller Hinton agar (BD, Heidelberg, Germany) according to manufacturer’s instructions. After incubation at 37°C for 24 h, the minimal inhibitory concentration (MIC) was recorded as the concentration value where the inhibition ellipse intersected the scale of the strip.
Surgical treatment. All patients underwent revision surgery. Patients with acute (early postoperative or late haematogenous) infection with symptoms lasting <4 weeks were treated with retention of the prosthesis, change of the mobile parts and meticulous debridement. In contrast, patients with chronic PJI, with signs of infection lasting ≥4 weeks or with a loose prosthesis were treated with one-stage or two-stage revision, depending on the local soft tissue and bone conditions and the revision history.
Antimicrobial treatment. Empiric antibiotic treatment was started intravenously after surgery and was subsequently adapted according to the susceptibility of the isolated organism. The intravenous treatment was typically continued for at least 2 weeks, followed by oral antibiotic treatment, as previously described ([25, 26]. In case of a two-stage revision, antibiotics were administered without interruption until re-implantation. After re-implantation, antibiotics were continued to complete a total duration of at least 12 weeks (or longer, if the prosthesis-free interval was >6 weeks).
Adequate antimicrobial therapy was considered when the antibiotic was appropriate against enterococcal infection according to its activity, dose, oral bioavailability and bone penetration. The antibiotics were chosen according to institutional treatment guidelines (“Pocket Guide on Diagnosis and Treatment of PJI”, see Supplementary Material) and patient history of antibiotic allergies. The following intravenous doses were used in patients with normal renal function: vancomycin 15-20 mg/kg every 12 h, daptomycin 8-10 mg/kg once daily, fosfomycin 5 g every 8 h, penicillin G 5 million IU every 6 h, ampicillin 2 g every 6 h, gentamicin 3 mg/kg once daily. In case of concomitant infectious endocarditis, higher doses were administered according to guidelines on infective endocarditis.
Follow-up evaluation. Patients were scheduled for follow-up in the outpatient clinic at 3, 6,12 and 24 months after revision surgery. Clinical, laboratory and radiological evaluation was performed and interpreted interdisciplinary by an orthopedic surgeon and an infectious disease specialist. Further follow-up evaluation was performed by phone contact using a standardized case-report form.
Statistical analysis. Categorical variables were compared using the Fisher's exact test, for comparison of continuous variables the Mann-Whitney U test was applied. A two-sided p-value of <0.05 was considered significant. The probability of infection-free survival and 95% confidence interval (CI) was estimated using the Kaplan-Meier survival method. Outcomes between groups were compared using Fishers exact test. An alpha level of 0.05 was considered significant. For statistical analysis and graphics, Prism software (version 8.2; GraphPad, La Jolla, CA, USA) was used.