Outcome of acute prosthetic joint infections due to gram-negative bacilli. treated with open débridement and retention of the prosthesis

Transcription

1 AAC Accepts, published online ahead of print on 17 August 2009 Antimicrob. Agents Chemother. doi: /aac Copyright 2009, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved. Outcome of acute prosthetic joint infections due to gram-negative bacilli treated with open débridement and retention of the prosthesis Nº de AACh AAC , Version 4 Juan C. Martínez-Pastor 1, Ernesto Muñoz-Mahamud 1, Félix Vilchez 1, Sebastián García- Ramiro 1, Guillem Bori 1, Josep Sierra 2, José A Martínez 2, Lluis Font 1, Josep Mensa 2, and Alex Soriano 2. 1 Department of Orthopaedics of Hospital Clinic of Barcelona. 2 Department of Infectious Diseases of Hospital Clínic of Barcelona. Hospital Clínic Universitari. IDIBAPS. Barcelona. Running title: Prosthetic infection by Gram-negative bacilli. Keywords: prosthetic joint infection, gram-negative bacilli, outcome, open débridement. Correspondence to: Dr. Alex Soriano Department of Infectious Diseases. Hospital Clínic of Barcelona. C/ Villarroel 170. Barcelona asoriano@clinic.ub.es Phone number: Fax:

2 Abstract Background: the aim of our study was to evaluate the outcome of acute prosthetic joint infections (PJI) due to Gram-negative bacilli (GNB) treated without implant removal. Methods: patients with an acute PJI due to GNB diagnosed from 2000 to 2007 were prospectively registered. Demographics, co-morbidity, type of implant, microbiology data, surgical treatment, antimicrobial therapy and outcome were recorded. CART analysis, Kaplan-Meier survival method and Cox regression model were applied. Results: 47 patients were included. Mean age was 70.7 years, there were 15 hip and 32 knee prostheses. The median number of days from arthroplasty was 20. The most frequent pathogens were Enterobacteriaceae in 41 cases and Pseudomonas in 20. Among the Enterobacteriaceae, 14 were resistant to ciprofloxacin, while all P. aeruginosa were susceptible. The median duration of intravenous and oral antibiotics was 14 and 64 days, respectively. A total of 35 (74.5%) patients were in remission after a median (IQR) followup of 463 ( ) days. Using a Kaplan-Meier survival curve, a C-reactive protein (CRP) 15 mg/dl (p=0.03) and to receipt of a fluoroquinolone, when all the isolated Gramnegatives were susceptible (p=0.0009), were associated with a better outcome. In a Cox regression model, CRP 15 mg/dl (OR:3.57, CI95%: , p=0.043) and receiving a fluoroquinolone (OR:9.09, CI95%: , p=0.005) were independently associated with a better outcome. Conclusion: open débridement without removing the implant had a success rate of 74.5% and factors associated with good prognosis were a CRP at the time of diagnosis 15 mg/dl and treatment with a fluoroquinolone.

3 Introduction Acute postoperative prosthetic joint infection is an uncommon but severe complication after joint arthroplasty. The infection rate is 1-3% and the most frequently isolated microorganisms are Gram positive cocci, including Staphylococcus aureus, coagulasenegative staphylococci and Streptococcus spp (12). However, it is of note that Gramnegative bacilli (GNB) are isolated in 10% of cases of PJI and frequently these infections are polymicrobial (9). The success rate in staphylococcal acute prosthetic joint infections treated with open débridement without implant removal and a prolonged course of antibiotics is higher than 75% (2,13,17). However, the experience using the same surgical and antibiotic treatment in infections due to Gram-negative bacilli is scarce (3,8). In addition, a major concern associated with prosthetic joint infections due to GNB is the emergence of resistant strains to many antibiotics and the lack of alternatives (15). The aims of the present study were to review our experience in acute prosthetic joint infections due to GNB treated with open débridement and retention of the implant followed by antibiotic treatment and to analyse those factors associated with outcome.

4 Patients and methods From January 2000 to December 2007 all patients with a prosthetic joint infection (hip hemiarthroplasty, total hip and knee arthroplasty) were prospectively registered in a database and retrospectively reviewed. All patients were treated in the same hospital in the bone and joint infection unit that includes orthopedic surgeons and infectious diseases specialists. Relevant information about demographics, co-morbidity, type of implant (hip or knee prosthesis), clinical manifestations, leukocyte count, value of C-reactive protein (CRP) at the time of admission for infection, surgical treatment, isolated microorganism, antimicrobial therapy and outcome were recorded. In the present study, only those cases with an acute, mono or polymicrobial prosthetic joint infection due to GNB were included. Acute prosthetic joint infection due to GNB in the present study was defined by the presence of local inflammation of acute onset (<15 days of symptoms duration), macroscopic evidence of extension of the infection through the capsule during open débridement and isolation of GNB in deep samples. Infections were classified as follows: 1) postoperative infections were those diagnosed within the 90 days after joint arthroplasty and 2) haematogenous infections those that represent haematogenous seeding of the joint from another primary site with no previous prosthesis dysfunction. In terms of débridement, pre-existing incisions were always used, necrotic tissue was excised and the joint was washed out with 6 litters of sterile water. The components were left in situ after confirming at the time of surgery that there were no signs of loosening. In knee arthroplasties the polyethylene component was removed and replaced by a new component, while in total hip arthroplasties the polyethylene, as well as the femoral head when possible, were substituted. Only when systemic or local signs of infection persisted after débridement, the patients were taken back to the operating room for repeating the

5 irrigation. When this was performed within the week after the first débridement, it was not considered as failure. At least three deep samples of synovial fluid or periprosthetic tissue were submitted to the microbiology laboratory. Synovial fluid was aspirated and 50% inoculated into aerobic and 50% into anaerobic blood culture flasks (BACTEC 9240 system, BD Diagnostic Systems). The volume inoculated in each flask was approximately 1-3 ml. Solid samples from periprosthetic tissue were taken and placed into sterile containers. Finally, swab cultures were obtained by passing a sterile swab over intracapsular area, bone or fluid and immediately keep in transport medium (AMIES transport medium). Each culture was transported, processed and analysed in the Microbiology laboratory. Blood culture flasks containing the aspirated synovial fluid were incubated in the BACTEC 9240 system up to 5 days (5). Positive cultures were Gram stained and the microorganisms were identified by conventional microbiological methods. The periprosthetic tissue and swabs were cultured in Thioglicolate broth, blood agar in aerobic conditions and Schaedler agar (in anaerobic conditions). All samples were incubated up to five days. Positive cultures were re-grown in an appropriated media. All isolated microorganisms were identified with standard biochemical procedures. In addition, blood cultures were performed to patients with fever at the moment of admission for infection. An antibiogram for all the isolates was performed by the microdilution method. After open débridement, a broad-spectrum intravenous antimicrobial regimen including vancomycin 1g/12h (monitored to attain a trough concentration of 15 mg/l) plus ceftazidime 2g/8h was started and maintained until obtaining definitive microbiological results. Intravenous antibiotics for GNB susceptible to third generation cephalosporins were ceftriaxone 1 g/24h or cefotaxime 2 g/6h, for ESBL-enterobacteriaceae imipenem 1g/8h or meropenem 1g/8h, for P. aeruginosa ceftazidime 2g/6h plus ciprofloxacin 400 mg/12-8h.

6 Oral antibiotics were ciprofloxacin 750 mg/12h or levofloxacin 500 mg/24h for fluoroquinolone-susceptible Gram-negatives, ceftibuten 400 mg/24h, cotrimoxazole 800 mg/12h or amoxiclavulanate 875/125 mg/8h for fluoroquinolone-resistant Gram-negatives. For staphylococci and streptococci, patients received levofloxacin, cotrimoxazole or linezolid 600 mg/12h according to the antibiogram, associated with rifampin 600 mg/24h. For enterococci patients received amoxicillin 1 g/8h or linezolid according to the antibiogram. The dosages were adjusted according to renal function. The duration of intravenous and oral antibiotics was not standardized and this was decided according to the clinical manifestations and the CRP values of each case. After being discharged, the patients were followed-up monthly while they were on treatment. Once the treatment was finished they were followed-up every 3 months where clinical response and secondary effects of the antimicrobial therapy were recorded. Outcome was evaluated according to the following definitions: 1) remission: when the patient showed no symptoms of infection, the prosthesis was retained and CRP values were lower than 1 mg/dl, and 2) failure: when inflammatory signs and high CRP values remained during the treatment or re-appeared after having completed it (relapse or re-infection depending on the isolated microorganism). Statistical analysis Continuous variables were expressed as mean and standard deviation (SD) or median and interquartile range (IQR). Continuous variables were: age, time from arthroplasty to diagnosis of infection (age of implant), leukocyte count, duration of intravenous and oral antibiotic. Categorical variables were: sex, co-morbidity (having or not having one or more of the following entities: diabetes mellitus, liver cirrhosis, chronic renal failure, rheumatoid

7 arthritis or chronic obstructive pulmonary disease), type of infection (post-surgical or haematogenous), type of prosthesis (hip or knee), positive blood cultures, the need of a second débridement, polymicrobial infection, infection due to Pseudomonas spp and antibiotic treatment. Continuous variables were compared using U-Mann-Whitney test and comparison of proportions was made using χ 2 test or Fisher exact test when necessary. Continuous and categorical variables were analysed using Classification and Regression Tree Analysis (CART) to identify predictors of outcome and useful breakpoints of continuous variables. The Kaplan-Meier survival method was used to estimate the cumulative probability of treatment failure from open débridement to the last visit. Log Rank test was applied to evaluate the influence of each variable. A Cox regression model was applied to identify the independent variables associated with failure after open débridement. Only those variables with significantly higher cumulative probability of failure in the Kaplan-Meier curve were included in the Cox regression model. Statistical significance was defined as a two-tailed p value < The analysis was done by the program SPSS (version 12.0; SPSS, Inc., Chicago, IL, U.S.A.) and CART software (Salford Systems, version 6.0, San Diego, CA, U.S.A.).

8 Results During the study period, a total of 47 patients met the inclusion criteria of the study. The main characteristics of the patients according to the outcome are summarized in table 1. Mean (SD) age was 70.7 (11.3) years, there were 15 hip prosthesis and 32 knee prosthesis. The median (IQR) number of days from arthroplasty to the diagnosis of the infection was 20 (16-28) days. There were 44 acute post-surgical deep infections with a median (IQR) number of days from arthroplasty of 19 ( ) and 3 haematogenous with 168, 673 and 682 days from arthroplasty, respectively. Gram-negative bacilli and their main traits of resistance are listed in table 2. The most common isolates were Enterobacteriaceae in 41 cases (Escherichia coli in 20) and Pseudomonas aeruginosa in 19. There were 28 polymicrobial infections (59.5%), 10 due to more than one Gram-negative bacillus and 18 due to a mixed infection including Gram-negative bacilli and Gram-positive cocci (table 3). In 21 out of 28 polymicrobial infections 2 different microorganisms were isolated, in 7 cases 2 different Gram-negatives and in 14 a Gram-negative and a Gram-positive. In the rest of polymicrobial infections (7 cases), 3 different microorganisms were identified, in 3 cases 3 different Gram-negatives, in 3 two different Gram-negatives and a Gram-positive and in 1 case one Gram-negative and two different Gram-positives. Among Enterobacteriaceae, 14 were resistant to ciprofloxacin, 8 extended spectrum beta-lactamase (ESBL) and 1 chromosomal beta-lactamase (AmpC) producers. All P. aeruginosa were susceptible to ciprofloxacin. Including B. fragilis, 13 patients had an infection due to a fluoroquinolone-resistant Gram-negative bacillus and they were associated with poor outcome (table 1). The median (IQR) duration of intravenous and oral antibiotics was 14 (8-24) and 64 ( ) days, respectively. Intravenous antibiotic regimens were a beta-lactam (n=24), a

9 beta-lactam with activity against P. aeruginosa plus ciprofloxacin (n=17) and ciprofloxacin alone (n=6). Fluoroquinolones (ciprofloxacin or levofloxacin) were the definitive antibiotic treatment in 35 cases. Among these patients, 28 had an infection due to fluorquinolonesusceptible Gram-negatives while the other 7 had susceptible and resistant Gram-negatives and they received other antibiotics (cotrimoxazole in 5 and amoxiclavulanate in 2 cases) in addition to fluoroquinolones. Among the 12 patients that did not receive fluorquinolones, all of them received an intravenous betalactam and 8 switched to oral therapy with cotrimoxazole in 6 cases, and ceftibuten and amoxiclavulanate in one case each. Those patients with an infection due to a fluorquinolone-susceptible Gram-negative who received a fluorquinolone (n=28), had a high success rate (26 out of 28, 92.8%) (table 1). A total of 35 (74.5%) patients were in remission after a median (IQR and range) follow-up of 463 ( and ) days after open débridement. There were 12 (24.5%) failures, 9 relapses, 2 reinfections and one case was not possible to classify because no microorganism was identified. Among relapses, in 8 cases the same Gram-negative bacilli of the index infection was isolated, E. coli in 6 cases (4 ESBL-producers), B. fragilis and E. cloacae in 1 case each. The other relapse was due to the Gram-positive component (coagulase-negative staphylococci) of the index infection. Reinfections were due to S. aureus and coagulase-negative staphylococci. Although P. aeruginosa was part of the flora identified in the index infection of 4 out of 12 failures, in no one case this microorganism was the cause of relapse. CART model only identified the C-reactive protein value as a predictor of outcome and the breakpoint to delineate the failure risk was 15 mg/dl. Using the Kaplan-Meier survival curve a C-reactive protein (CRP) 15 mg/dl (p=0.03) and receiving a fluoroquinolone, when all the isolated Gram-negatives were susceptible (p=0.0009), were associated with a

10 better outcome (figure 1 and 2). There were infections due to more than 1 GNB with different susceptibilities to fluoroquinolones and treatment with fluoroquinolones indeed showed a trend towards a better outcome but it was not significant (p=0.13, table 1). In the Cox regression model, including significant variables in univariate analysis (table 1), a CRP 15 mg/dl (OR:3.57, CI95%: , p=0.043) and receipt of a fluoroquinolone, when all the isolated Gram-negatives were susceptible (OR:9.09, CI95%: , p=0.005), were independently associated with a better outcome. Having an infection due to a fluoroquinolone-resistant strain was not an independent predictor of failure. Among patients with a CRP 15 mg/dl, those that received a fluoroquinolone when all the isolated strains were susceptible to ciprofloxacin had a remission rate of 95.4% (21 out of 22) while those with resistant strains or who did not receive a fluoroquinolone had a remission rate of 61.5% (8 out of 13). In addition, among patients with a CRP >15 mg/dl, those that received a fluoroquinolone when all the isolated strains were susceptible to ciprofloxacin had a remission rate of 83.3% (5 out of 6) while those with resistant strains or who did not receive a fluoroquinolone had a remission rate of only 16.7% (1 out of 6) (table 4).

11 Discussion The present study suggests that open débridement with retention of the implant followed by antibiotic treatment is a good option in acute prosthetic joint infections due to GNB. The success rate was 74.5%, similar to that reported by other authors in acute prosthetic joint infections due to staphylococci (2,4,7,13). A CRP 15 mg/dl and treatment with a fluoroquinolone when all the isolated GNB were susceptible to fluoroquinolones were factors independently associated with a better outcome. In fact, patients that received a fluoroquinolone when all the isolated strains were susceptible had a remission rate higher than 90% (26 out of 28). Previous published clinical experience with a similar therapeutic approach is scarce but supports the efficacy of fluoroquinolones. Brouqui et al (3), using a combination of ceftazidime and ciprofloxacin, cured nine osteosynthetic devices and four out of five prosthetic joint infections due to P. aeruginosa. Recently, Legout et al (8) reviewed their experience without removing the implant in 12 patients with an orthopaedic device infection (internal fixation or joint prosthesis) due to a Gram-negative bacilli. Antibiotic treatment consisted of intravenous cefepime for 4 weeks combined with oral ofloxacin or ciprofloxacin for 3 to 9 months and the cure rate was 67% (8 out of 12). The efficacy of fluoroquinolones in the treatment of implant infections and osteomyelitis due to Gram-negative bacilli is probably due to 2 facts: 1) their diffusion to synovial fluid and bone (11) and 2) their activity against biofilms. In an in vitro model of Pseudomonas biofilm, Tanaka et al (14) showed that the bactericidal action of beta-lactams against biofilm cells was affected by the low cell growth rate inside the biofilm, while that of fluoroquinolones was considerably greater and independent from the growth rate. Unfortunately, 14 out of 43 (32.5%) Enterobacteriaceae isolated in our study were resistant to ciprofloxacin. In addition, 8 prosthetic joint infections were due to ESBL-

12 Enterobacteriaceae producers, to our knowledge the first description of these pathogens as a cause of prosthetic joint infections, with a cure rate of only 50%. The activity of tigecycline against in vitro Gram-negative biofilms (Acinetobacter baumanii and Klebsiella pneumoniae) was evaluated in a previous study (1) and it was lower than in Gram-positive biofilms (6,10), however, there was an considerable synergism between N-acetylcisteine and tigecycline. In the future, it is necessary to evaluate new strategies to treat implantrelated infections due to Gram-negative bacilli. The duration of antimicrobial therapy in acute prosthetic joint infections has not been well established. Based on expert opinion, duration of 3-6 months is recommended (16). In our protocol, antibiotic therapy is maintained until clinical resolution of infectious symptoms and normalization of CRP. We have obtained good results using this protocol, most especially when a fluoroquinolone was administered. More than 60% of the infections in our study were polymicrobial and although it might seem a high rate, a recent article from Oxford (9) described 12 acute infections due to GNB and 8 (66.6%) were polymicrobial. In fact, in that study 47% of acute prosthetic joint infections were polymicrobial. The suggested duration of follow-up in prosthetic joint infections is at least 2 years based on experiences where the major pathogens were Gram-positives. A drawback of our study was that only 40% of our patients reach a 2-year follow-up, however, data about the outcome and natural history in prosthetic joint infections due to Gram-negatives is scarce and, in fact, this is the largest reported series of patients treated with conservative surgery and antimicrobial therapy. However, the small sample size implies limited statistical power and the conclusions should be interpreted with caution. Other drawback from our study was that the antimicrobial therapy was not randomized, but it was administered according to the

13 susceptibility pattern of the isolated microorganisms. The design of clinical trials in these infections is difficult and for this reason it is necessary to progress from observational cohort studies. In conclusion, open débridement without removing the implant in acute PJI due to Gramnegative bacilli had a success rate of 74.5%. Factors associated with good prognosis were a CRP at the moment of diagnosis 15 mg/dl and treatment with a fluoroquinolone when all the strains isolated were susceptible to ciprofloxacin. In the future, it will be necessary to evaluate new therapeutic strategies for those infections due to fluoroquinolone-resistant strains.

14 Acknowledgements: REIPI, Red Española para la Investigación en Patología Infecciosa. Potential conflict of interest: at the time of publication none of the authors disclosed any potential conflicts of interest.

15 REFERENCES 1. Aslam S, Trautner BW, Ramanathan V et al Combination of tigecycline and N-acetylcysteine reduces biofilm-embedded bacteria on vascular catheters. Antimicrob Agents Chemother. 51: Barberan J, Aguilar L, Carroquino G et al Conservative treatment of staphylococcal prosthetic joint infections in elderly patients. Am J Med. 119: Brouqui P, Rousseau MC, Stein A et al Treatment of Pseudomonas aeruginosa-infected orthopedic prostheses with ceftazidime-ciprofloxacin antibiotic combination. Antimicrob Agents Chemother. 39: Giulieri SG, Graber P, Ochsner PE et al Management of infection associated with total hip arthroplasty according to a treatment algorithm. Infection. 32: Hughes JG, Vetter EA, Patel R et al Culture with BACTEC Peds Plus/F bottle compared with conventional methods for detection of bacteria in synovial fluid. J Clin Microbiol. 39:

16 6. Labthavikul P, Petersen PJ, Bradford PA In vitro activity of tigecycline against Staphylococcus epidermidis growing in an adherent-cell biofilm model. Antimicrob Agents Chemother. 47: Laffer RR, Graber P, Ochsner PE et al Outcome of prosthetic kneeassociated infection: evaluation of 40 consecutive episodes at a single centre. Clin Microbiol Infect. 12: Legout L, Senneville E, Stern R et al Treatment of bone and joint infections caused by Gram-negative bacilli with a cefepime-fluoroquinolone combination. Clin Microbiol Infect. 12: Moran E, Masters S, Berendt AR et al Guiding empirical antibiotic therapy in orthopaedics: The microbiology of prosthetic joint infection managed by debridement, irrigation and prosthesis retention. J Infect. 55: Raad I, Hanna H, Jiang Y et al Comparative Activities of Daptomycin, Linezolid, and Tigecycline against Catheter-Related Methicillin-Resistant Staphylococcus Bacteremic Isolates Embedded in Biofilm. Antimicrob Agents Chemother. 51: Rimmele T, Boselli E, Breilh D et al Diffusion of levofloxacin into bone and synovial tissues. J Antimicrob Chemother. 53:

17 12. Soriano A, Bori G, Garcia-Ramiro S et al Timing of antibiotic prophylaxis for primary total knee arthroplasty performed during ischemia. Clin Infect Dis. 46: Soriano A, Garcia S, Bori G et al Treatment of acute post-surgical infection of joint arthroplasty. Clin Microbiol Infect. 12: Tanaka G, Shigeta M, Komatsuzawa H et al Effect of the growth rate of Pseudomonas aeruginosa biofilms on the susceptibility to antimicrobial agents: betalactams and fluoroquinolones. Chemotherapy. 45: Wenzel RP The antibiotic pipeline--challenges, costs, and values. N Engl J Med. 351: Zimmerli W, Trampuz A, Ochsner PE Prosthetic-joint infections. N Engl J Med. 351: Zimmerli W, Widmer AF, Blatter M et al Role of rifampin for treatment of orthopedic implant-related staphylococcal infections: a randomized controlled trial. JAMA. 279:

18 Table 1. Characteristics of patients according to the outcome. Characteristics Failure Remission (n=12)* (n=35) Median (IQR) of age (years) 71 ( ) 75 (63-79) 0.96 Sex (female) 7 (58.3) 24 (68.3) 0.37 Comorbidity # 6 (50) 15 (42.9) 0.46 Type of prosthesis Hip Knee 4 (33.5) 8 (66.7) 11 (31.4) 24 (68.6) 0.58 Median (IQR) of age of prosthesis 19 ( ) 21 (13-31) 0.71 (days) Type of infection Post-surgical Haematogeneous (91.4) 3 (8.6) 0.40 Median (IQR) of Leukocyte count 8750 ( ( (cells/mm 3 ) 11122) 10417) Median (IQR) of C-reactive protein 11.4 ( ( (mg/dl) 21.4) 12.9) C-reactive protein > 15 mg/dl 6 (50) 6 (17.1) 0.03 Bacteremia 0 4 (11.4) 0.29 Polymicrobial infection 7 (58.3) 21 (60) 0.59

19 Mixed infection with Gram positives 3 (25) 15 (42.9) 0.32 Infection due to Pseudomonas spp & 4 (33.5) 16 (45.7) 0.34 Infection due to fluoroquinoloneresistant strain 8 (66.7) 5 (14.3) Infection due to an ESBL- 4 (33.3) 4 (11.4) 0.17 Enterobacteriaceae Need of a 2 on débridement 3 (25) 4 (11.4) 0.24 Median (IQR) of antibiotic therapy (oral and intravenous) duration (days) 84.5 ( (47-135) ) Median (IQR) duration of oral 95 ( ( antibiotic therapy (days) $ 165.2) 101.5) Treatment with fluoroquinolones 7 (58.3) 28 (80) 0.13 Treatment with fluoroquinolones when all isolated GNB were susceptible 2 (16.7) 26 (74.3) GNB: Gram-negative bacilli. ESBL, extended spectrum beta-lactamase producer. * Including relapse and U Mann-Whitney test, Chi-square test or Fisher exact test when necessary. # Diabetes mellitus, liver cirrhosis, chronic renal failure, rheumatoid arthritis or chronic obstructive pulmonary disease. & P. aeruginosa in 19 cases and P. stutzeri in 1. $ The information is referred to those patients that received oral antibiotics (n=41).

20 Table 2. Gram-negative bacilli isolated from deep peri-prosthetic samples of either monomicrobial and polymicrobial infections. Microorganism Number (%) Gram-negative bacilli 63 Enterobacteriaceae 41 (65) Escherichia coli Resistant to ciprofloxacin ESBL-producer* AmpC-producer* Proteus mirabilis Resistant to ciprofloxacin Enterobacter cloacae Resistant to ciprofloxacin Klebsiella pneumonia Resistant to ciprofloxacin and ESBL-producer Other Enterobacteriaceae # Resistant to ciprofloxacin 20 (48.8) (19.5) 2 7 (17) 0 2 (4.9) 1 4 (9.5) 1 Non-fermenter Gram-negative bacilli 21 (33.3) Pseudomonas aeruginosa Resistant to ciprofloxacin Other non-fermenters $ Resistant to ciprofloxacin 19 (90.4) 0 2 (9.6) 0 Anaerobes 1 (1.6)

21 Bacteroides fragilis 1 GNB, Gram-negative bacilli. The number of microorganisms isolated was higher than the number of patients due to polymicrobial infections. * 5 ESBL-producers and the AMPc-producer were resistant to ciprofloxacin. # Citrobacter diversus, Serratia marcescens, Providencia rettgeri, Morganella morgagnii. $ Pseudomonas stutzeri, Acinetobacter baumanii.

22 Table 3. Other pathogens isolated in polymicrobial infections. Microorganism Number (%) Gram-positive cocci * 19 Coagulase-negative staphylococci 8 (42.1) Staphylococcus aureus 3 (15.8) Enterococcus faecalis 6 (31.6) Streptococcus viridans 1 (5.2) Streptococcus agalactiae 1 (5.2) * The number of patients with a polymicrobial infection including Gram-negatives and Gram-positives was 18 (see text), but the total number of Gram-positives was 19 because in one patient 2 different Gram-positives were isolated.

23 Table 4. Outcome according to the value of C-reactive protein and the antimicrobial therapy received. Value of CRP Failure, Remission, P* n=12 (%) n=35 (%) 15 mg/dl - no fluoroquinolone or resistant strain (n=13) 5 (83.3) 8 (27.6) - fluoroquinolone (n=22) > 15 mg/dl - no fluoroquinolone or resistant strain (n=11) - fluoroquinolone (n=12) 1 (16.7) 5 (83.3) 1 (16.7) 21 (72.4) (16.7) 5 (83.3) 0.04 * Fisher exact test.

24 Figure 1. Cumulative probability of survival (free of failure) according to the C-reactive protein value at the time of admission for infection. % free of failure acum CRP 15 mg/dl CRP > 15 mg/dl Log Rank test, p= Follow-up from open débridement (days)

25 Figure 2. Cumulative probability of survival (free of failure) according to receipt of a fluoroquinolone when isolates were susceptible or non-receipt a fluoroquinolone or receipt of a fluoroquinolone it when at least 1 isolate was resistant to ciprofloxacin. % free of failure Fluoroquinolone Non-fluoroquinolone or resistant isolate Log Rank test, p= Follow-up from open débridement (days)