Antimicrobial therapy in prosthetic joint infectionan approach to the most relevant and current clinical problems

Resumen

Prosthetic joint infections (PJI) are defined as difficult-to-treat, mainly due to the presence of bacterial biofilm and other bacterial adaptive forms. An aggressive treatment including surgery and long antimicrobial therapy must be provided, the economical burden of this complication being very important. Thus, PJI constitute a first-order health care problem, and it is expected that the number of infected devices will increase in the future. Our knowledge on the efficacy of antimicrobial therapy in the setting of PJI mainly derives from experimental models. Clinical studies are very difficult to perform, due to the specialized nature of this infection, the need for a long follow up and the difficulties for collecting patients. Previous case series usually present with small and quite heterogeneous samples, so the conclusions drawn from these studies are not always very reliable. The clinical studies presented in this thesis have been performed in the setting of the multidisciplinary Bone and Joint Unit of the Hospital Universitario de Bellvitge, which is integrated in the Spanish Network for the Research in Infectious Diseases (REIPI), thus allowing an appropriate clinical approach to PJI. These studies try to answer clinical questions on PJI regarding the efficacy of the antimicrobial treatment. The specific aims have been classified according to the specific surgical approach with which the patients were managed. A. Antimicrobial therapy in PJI managed with implant retention A.1. Infection by staphylococci Aim 1 – To measure the impact of rifampin in the outcome of a large cohort of PJI by S. aureus Multicenter observational retrospective study of 345 cases of PJI by S. aureus managed with implant retention, including 81 cases by methicillin-resistant S. aureus (MRSA). Overall failure was 45%. Treatment with rifampin was an independent predictor of a favourable outcome, the specific type of staphylococci having a similar overall prognosis. Aim 2 – To assess the efficacy of a short schedule of levofloxacin plus rifampin in staphylococcal PJI Open, comparative, multicenter clinical trial where patients with acute staphylococcal PJI undergoing debridement plus implant retention were randomized to receive either a short treatment of 8 weeks of levofloxacin plus rifampin, or a standard long treatment of 3 months (hip prosthesis) or 6 months (knee prosthesis). Overall success rate was 93% and 65% in the per-protocol and intention-to-treat analysis, respectively, with no significant differences between the two-arms. Aim 3 – To evaluate daptomycin plus rifampin for fluoroquinolone resistant staphylococcal PJI Retrospective observational multicenter study of 20 patients with acute staphylococcal PJI undergoing debridement plus implant retention and treated with daptomycin (10 mg/kg/d) plus rifampin for 6 weeks. Two (10%) patients were withdrawn due to toxicity. In the other 18 patients, clinical and microbiological cure were observed in 50% and 73% of cases, respectively. There were no differences as compared with a historical cohort of cases treated with alternative rifampin-based combinations, except for a lesser rate of failure while patients were still under treatment with daptomycin plus rifampin. A.2. Infection by Gram-negative bacilli Aim 4 – To assess the impact of fluoroquinolones in the outcome of a large cohort of PJI by Gram-negative bacilli Observational retrospective multicenter study of 172 cases of PJI by Gram-negative bacilli undergoing debridement plus implant retention. Overall success was 68% after a median follow-up of 25 months. Treatment with ciprofloxacin was an independent predictor of success (79% vs 41%). A.3. Infection in the elderly Aim 5 – Comparative evaluation of the antibiotic efficacy in patients carrying total hip prosthesis or hip hemiarthroplasties Observational retrospective study of 210 patients with hip-PJI, comparing patients carrying either hip-hemiarthroplasties (HHA) or total hip arthroplasties (THA). Patients with HHA were older, had more underlying conditions, and infection by Gram-negative bacilli was more frequent. Overall failure of 123 patients undergoing debridement plus implant retention was 63% after a median follow up of 347 days. While the specific device (HHA or THA) was not associated with a higher likelihood of failure, crude and related mortality were significantly higher in the HHA group. B. Antimicrobial therapy in PJI managed with implant removal Aim 6 – To evaluate linezolid in PJI by Gram-positive microorganisms managed with a two-step exchange procedure. Prospective non-comparative multicenter clinical trial involving 25 patients treated with linezolid for 6 weeks after prosthesis removal. Three (12%) were withdrawn due to toxicity. Among the other 22 (88%) patients, 20 (91%) were considered to be clinically cured. Among them, cultures taken at surgical site at the time of reimplantation were positive in 1 (5%) patient. Thus, clinical and microbiological cure among these 22 patients was 86%. C. Antimicrobial activity on biofilms of multi-resistant Gram-negative bacilli Aim 7 – To study the activity of colistin against multi-resistant P. aeruginosa biofilm in an in vitro experimental model. Based on the CDC Biofilm reactor, the in vitro experiments were conducted during 72 hours, after an initial conditioning phase of 28 hours. Three different strains of P. aeruginosa were used: the referral strain PAO-1 (colistin- and carbapenem-susceptible) and two clinical strains (HUB-1 and HUB-2, both colistin-susceptible and carbapenemresistant). Two clinically relevant concentrations of colistin were used at constant infusion (1.25 mg/L and 3.50 mg/L), as well as doripenem (as bolus every 8 hours, Cmax 25 mg/L). Monotherapies of colistin produced initial killing followed by regrowth and emergence of colistin-resistance. The combination with doripenem gave place to a more sustained killing and lesser rate of regrowth. Additivity and synergy were observed at different times with the combination therapy. Emergence of colistin resistance was avoided, too. These effects were also observed in the two carbapenem-resistant clinical strains.