Aim. Debridement, antibiotic, and implant retention (DAIR) is an accepted treatment of early and late acute Total Knee Arthroplasty (TKA) infections. DAIR failure may adversely affect the outcome of a subsequent two-stage exchange arthroplasty. Controversy exists on risk factors that can affect DAIR's results. The aim of the study is to review presurgical, intrasurgical and postsurgical variables that could affect DAIR's result. Method. A retrospective study of 27 DAIRs performed between 2015–2019 to treat late acute TKA infections was carried out. Patients were divided into two groups depending on DAIR's outcome [Healing (H) vs non-healing group (NH)] according on the Delphi-based multidisciplinary consensus criteria on success after treatment of periprosthetic joint infection. We reviewed presurgical variables, including epidemiological variables (Age, Sex, comorbidities, ASA, Charlson, BMI, alcohol dependency), prosthesis variables (prosthesis type, primary cause of operation, primary TKA surgery center), infection variables (concomitant infection, previous antibiotic treatment, c-reactive protein, synovial WBC count, synovial % PMN, pathogen), KLIC score and CRIME 80 score. Surgical variables such as surgery duration and type of surgery (elective vs urgent). Post-surgical variables like antibiotic treatment duration and destination at discharge. Normal distribution was assessed by Shapiro-Wilk test. Mann Whitney U test was used to compare the two independent sample variables. Chi-squared test was used for qualitative variables. P-value was established at 0.05 and statistical power at 80%. Results. Infection Healing was achieved in 63% of patients. In presurgical variables, alcohol dependency, hypertension, liver disease, previous surgery performed in another institution were more frequent in NH group (p< 0.05). KLIC score value equal or greater than 4 had a higher risk of surgical failure (p < 0.05). Regarding surgical variables, the healing group had more negative cultures than de non-healing one (p<0.05). Regarding post-surgical variables, long term antibiotic treatment (six months) achieved more healing after DAIR (p<0.05). Conclusions. Alcohol dependency, hypertension, liver disease and KLIC score values equal or greater than 4, may increases the risk of DAIR failure. Finally, we observed that the long-term antibiotic treatment (6 months) favors healing after DAIR

Aim. We report on the performance of a simple algorithm using a combination of synovial fluid White blood cell count(WBC), C-reactive protein(CRP) and α-Defensin(AD) tests to aid in the diagnosis of prosthetic joint infections. Methods. Sixty-six synovial fluid samples were collected prospectively in patients with suspected PJI (hip and knee). All samples were tested by: WBC counts (read manually) and CRP test (Alere-Afinion™ validated in-house); and on 37 of these with AD test. Synovial fluid samples were collected in 5 ml ethylenediaminetetraacetic acid (EDTA) tubes. Samples that were very viscous were pre-processed by the addition of 100µl of hyaluronidase solution. Grossly blood stained and clotted samples were excluded. A clinical diagnosis of infection was based IDSA definitions. 1. Cut offs of >3000 × 10. 6. cells/L for total synovial WBC count and >12mg/L for CRP were used to define infection. 2,3. . Results. Of 66 samples tested, 20 samples were categorised as clinically infected. Combination of WBC count and CRP yielded a sensitivity of 95% (95% CI: 75.13% to 99.87%) and specificity of 100% (95% CI: 92.29% to 100.00%). Only one patient, who had a chronic infection with S.epidermidis and S.warneri, had a CRP and WBC count that was falsely negative (<5mg/L and 93 × 10. 6. cells/L respectively). AD test was used on 37 samples (of which 20 were infected). Sensitivity of this test alone was 85.71% (95% CI: 63.66% to 96.95%) and specificity 87.5% (95% CI: 61.65% to 98.45%). There were 2 falsely positive AD test results (one of whom had a metal on metal prosthesis) and 3 false negative results (2 E.coli infections and one patient with chronic infection with S.epidermidis and S.warneri). Conclusion. Use of a combination of synovial fluid WBC count and CRP (both of which can be performed using simple and inexpensive laboratory tests), has a sensitivity of 95% and 100% specificity in the diagnosis of PJI. AD test may be useful on some occasions when near patient testing result may affect patient management