Fluoroquinolone resistance in clinical isolates of Klebsiella oxytoca.

BACKGROUND: Prevalence of fluoroquinolone-resistant Klebsiella oxytoca has been reported worldwide. METHODS: We recovered ten clinical K. oxytoca isolates from patients with acute cystitis, asymptomatic bacteriuria or acute bacillary diarrhea in Japan. Out of ten isolates, one fluoroquinolone-susceptible isolate was included as a control. Fluoroquinolone resistance was characterized genetically by PCR and DNA sequencing methods. Outer membrane protein (OMP) profiles were determined by SDS-PAGE. RESULTS: In nine clinical isolates of levofloxacin-resistant K. oxytoca, nucleotide sequences in the quinolone-resistance-determining regions showed amino acid mutations such as Thr83Ile and Asp87Gly in GyrA and Ser80Ile in ParC. Combined effects of reduced 36-kDa OMP production and amino acid mutations in GyrA and ParC were shown by two K. oxytoca isolates exhibiting higher minimum inhibitory concentrations for fluoroquinolones than other fluoroquinolone-resistant isolates. CONCLUSIONS: In clinical K. oxytoca isolates, the various mechanisms of fluoroquinolone resistance may include reduced 36-kDa OMP production as well as GyrA and ParC mutations.

[Comparison of susceptibilities for fosfomycin determined by various methods in clinical isolates in 2003 -2004].

Antimicrobial susceptibilities for fosfomycin (FOM), cephalexin, cefpodoxime, cefdinir, cefditren, ampicillin, sulbactam/ampicillin, imipenem (IPM), panipenem, meropenem (MEPM), biapenem, levofloxacin (LVFX), gatifloxacin, pazufloxacin, prulifloxacin and sulfamethoxazole/trimethoprim were determined by an agar dilution method using Mueller-Hinton agar (MHA) in Escherichia coli, Klebsiella spp., Serratia marcescens, Citrobacter spp., Enterobacter spp. and Proteus mirabilis, which were isolated from patients in 2003-2004. Those for FOM were determined by the agar dilution methods using MHA containing glucose-6-phosphate (G6P) under aerobic conditions, MHA under anaerobic conditions and nutrient agar under aerobic conditions. Those for FOM, LVFX, IPM and MEPM were also determined by an Etest method. The results by the agar dilution method showed that carbapenems had good antibacterial activities in all isolates, whereas MIC ranges for other antimicrobials were broad. Our results showed that the agar dilution method for FOM using MHA containing G6P under aerobic conditions provided reliable MICs in E. coli, which agreed with data previously reported. The results by the agar dilution method for LVFX, IPM and MEPM showed the high rate of agreement compared with those by the Etest method. In E. coli, the results for FOM by the agar dilution method using MHA containing G6P showed the high rate of agreement compared with the Etest results, although the rate was affected by bacterial species and culture conditions in various ways.

Recommended initial loading dose of teicoplanin, established by therapeutic drug monitoring, and outcome in terms of optimal trough level.

Teicoplanin has a long serum half-life, and therefore it takes time to reach a steady-state concentration. An initial loading procedure has been recommended for teicoplanin to enable prompt reaching of the optimal serum trough level (10-15 microg/ml). However, the dose of teicoplanin that should be administered to patients with varying renal function levels remains inconclusive. In this study, we monitored the serum concentrations of teicoplanin in patients with methicillin-resistant Staphylococcus aureus (MRSA) pneumonia and compared different teicoplanin serum concentrations and their clinical efficacy, investigating the significance of the mean dose administered during the initial 3 days. The study included 48 patients with MRSA pneumonia. The peak and trough concentrations of teicoplanin were determined utilizing a fluorescence polarization immunoassay and a two-compartment Bayesian population model. Teicoplanin was given at a loading dose of 400 or 800 mg on the first day, followed by maintenance doses of 200 or 400 mg. The mean initial dose (MID) over the first 3 days was calculated as: (loading dose + dose on 2nd day + dose on 3rd day)/3. Patients with an MID of 266.7 mg or less (400 mg for loading, 200 mg over the 2nd and 3rd days) did not have a trough level that exceeded 10 microg/ml at the point before the injection on the 4th day. Even in patients with hemodialysis (HD), an MID of 266.7 mg was not enough to provide a trough level of 10 microg/ml. Patients with an MID than 533.3 mg had significantly elevated trough levels, showing better outcomes. A multiple regression formula for predicting trough level before the fourth day of administration is given as: 0.034 + 0.030 x (MID; mg) - 0.057 x creatinine clearance (Ccr; ml/min). These findings suggest that 800 mg as an initial dose, followed by 400 mg maintenance doses over the following 2 days, makes it possible to safely attain an optimal trough level, even in the patients with HD.

Release of exotoxin A, peptidoglycan and endotoxin after exposure of clinical Pseudomonas aeruginosa isolates to carbapenems in vitro.

BACKGROUND: Production of several cell-associated components and extracellular enzymes can play important roles in the pathogenesis of Pseudomonas aeruginosa infections. METHODS: We characterized the time course of morphological changes, production of exotoxin A (ETA) and release of peptidoglycan (PG) and endotoxin (ET) in clinical P. aeruginosa isolates after exposure to carbapenems including imipenem, panipenem, meropenem and biapenem at 0.5, 2, 8 and 32 microg/ml. RESULTS: The amount of ETA in the supernatant of bacterial cultures exposed to carbapenems correlated with the number of viable cells, independently of morphological changes. Formation of ovoid cells and rapid cell lysis induced by carbapenems above the minimal inhibitory concentrations (MICs) decreased ETA production and ET release, while filamentation and prolonged cell lysis increased ETA production and/or ET release. Neither the number of viable cells nor bacterial morphology was related to the amount of PG released throughout the 6-hour observation period. CONCLUSION: Exposure to concentrations of carbapenems above the MIC resulted in rapid cell lysis of P. aeruginosa and decreased ETA levels and ET release, while filamentation and prolonged cell lysis induced by exposure to sub-MICs of carbapenems were associated with increased ETA production and/or greater ET release.

Mechanisms of resistance to fluoroquinolones and carbapenems in Pseudomonas putida.

OBJECTIVES: Pseudomonas putida is an uncommon opportunistic pathogen, usually susceptible to antimicrobial agents. Data concerning resistance to antimicrobial agents in clinical P. putida isolates are limited. PATIENTS AND METHODS: Susceptibilities to fluoroquinolones, carbapenems and other antibiotics were characterized in five clinical isolates of P. putida recovered from different patients with urinary tract infections as causative pathogens. Fluoroquinolone and carbapenem resistance were characterized genetically by the methods of PCR and DNA sequencing. Outer membrane protein (OMP) profiles were characterized by SDS-PAGE. RESULTS: Four of five isolates were resistant or intermediate to both fluoroquinolones and carbapenems. Nucleotide sequences in the quinolone resistance-determining regions suggested that amino acid mutations such as Thr-83-->Ile in GyrA and Glu-469-->Asp in GyrB may contribute to high resistance to fluoroquinolones. Four metallo-beta-lactamase-producing isolates that showed resistance to carbapenems carried the IMP-type metallo-beta-lactamase genes. A combined effect of reduced production of 46 kDa OMP and metallo-beta-lactamase production was shown by a P. putida isolate exhibiting the highest MICs of carbapenems. CONCLUSIONS: This study identified mechanisms of resistance to fluoroquinolones and carbapenems in clinical P. putida isolates.

Characterization of fluoroquinolone and carbapenem susceptibilities in clinical isolates of levofloxacin-resistant Pseudomonas aeruginosa.

BACKGROUND: Pseudomonas aeruginosa can rapidly acquire resistance to antibiotics, including fluoroquinolones and carbapenems. METHODS: We characterized fluoroquinolone, carbapenem and other beta-lactam susceptibilities and analyzed fluoroquinolone and carbapenem resistance in 16 clinical isolates of levofloxacin-resistant P. aeruginosa. RESULTS: All levofloxacin-resistant isolates showed high MICs (> or =32 microg/ml) for fluoroquinolones including norfloxacin, levofloxacin, sparfloxacin, gatifloxacin and pazufloxacin, whereas the MICs for sitafloxacin were between 2 and 16 microg/ml. These isolates had both a Thr83Ile mutation in GyrA and a Ser87Leu mutation in ParC. An additional mutation, Glu469Asp in GyrB, was detected in 3 isolates. Three of 16 isolates found during antibiotic therapy showed resistance to carbapenems (MIC, 16-32 microg/ml) because of a reduced production of OprD. Fluoroquinolones, beta-lactams and sulfamethoxazole-trimethoprim were used for 3 months before the isolation of levofloxacin-resistant P.aeruginosa. CONCLUSIONS: Emergence of resistant isolates to both fluoroquinolones and carbapenems during antibiotic therapy is a serious clinical problem. Our results suggest that susceptibilities to fluoroquinolones as well as carbapenems should be monitored during a prolonged course of antibiotic therapy against P. aeruginosa infection.

Control of pulmonary absorption of water-soluble compounds by various viscous vehicles.

Effects of various viscous vehicles on the pulmonary absorption of water-soluble drugs were examined by an in situ pulmonary absorption experiment. Gelatin, polyvinylacohol (PVA), hydroxypropylcellose (HPC), chondroitin sulfate A sodium salt (CS), polyacrylic acid (PAA), methylcellulose #400 (MC400) and hyaluronic acid sodium salt (HA) were used as models of viscous vehicles. 5(6)-Carboxyfluorescein (CF) and fluorescein isothiocayanate-labeled dextran with an average molecular weight of 4000 (FD4) were used as water-soluble drugs. The plasma concentration of CF was controlled and regulated in the presence of these viscous vehicles, especially gelatin (1-5%) and polyvinyl alcohol (PVA) 1%. In the pharmacokinetic analysis, the Cmax values of CF significantly decreased, and its Tmax values increased in the presence of these viscous vehicles compared with the control. The MRT and MAT values of CF with these vehicles were significantly higher than those without these vehicles. Therefore, these findings indicated that the viscous vehicles were effective to regulate the absorption rate of CF. On the other hand, the pulmonary absorption of FD4 was not so much affected even in the presence of gelatin and PVA, although PVA slightly decreased MRT value, and significantly decreased Tmax value. Furthermore, we examined the release rate of CF from the cellulose tube containing various concentrations of gelatin. The release rate of CF from the cellulose tube with gelatin was inversely related to the viscosity of gelatin. In addition, the release rate of CF was inversely related to DeltaMAT (DeltaMAT = MATgel(MAT with gelatin)-MATsol(MAT without gelatin)) in the presence of varying concentrations of gelatin. These findings indicated that these viscous vehicles were effective to control the pulmonary absorption of CF, a water-soluble drug with low molecular weight and they might be useful to increase the local concentration of drugs in the lung.

Effect of basic amino acids on susceptibility to carbapenems in clinical Pseudomonas aeruginosa isolates.

We evaluated effects of medium composition, including basic amino acid content and pH, on susceptibility to carbapenems such as imipenem, panipenem and meropenem, in clinical isolates of Pseudomonas aeruginosa. Susceptibility to carbapenems was reduced by basic amino acids in the medium, while susceptibilities to ceftazidime and aztreonam were not. Among carbapenems, susceptibility to panipenem was most sharply reduced by addition of basic amino acids to 1:16 Mueller-Hinton agar (MHA). In 174 of 175 clinical isolates, MICs for carbapenems were affected to different degrees by medium composition. One isolate, in which MICs for carbapenems did not differ between MHA and 1:16 MHA, showed reduced production of porin (OprD). Our results suggest that susceptibility to individual carbapenems, especially panipenem, is difficult to evaluate based on MICs for other carbapenems determined on MHA. For a better prediction of antibiotic efficacy, it may be important to evaluate the susceptibility for each carbapenem individually.

Characterization of Pseudomonas aeruginosa isolates from patients with urinary tract infections during antibiotic therapy.

We characterized susceptibilities and genotypes in a series of Pseudomonas aeruginosa isolates from five cases of urinary tract infections (UTIs) to evaluate clonal shifts of carbapenem resistance. In one case, a series of isolates showed different susceptibility patterns for carbapenems but an identical genotype. In another case, genotypes varied among 4 P. aeruginosa isolates from recurrent UTIs over 9 months. Although the patient had been treated with no antibiotic immediately before isolation, the susceptibility patterns for carbapenems and ceftazidime varied. Further analysis in these two cases of outer membrane protein profiles showed that loss of OprD production resulted in reduced susceptibilities to carbapenems in all of the carbapenem-resistant isolates. Loss of OprD production was likely due to oprD gene inactivation in both of cases, since the carbapenem-resistant isolates showed no cross resistance to levofloxacin and chloramphenicol compared with the carbapenem-susceptible isolates. There was another case in which all isolates showed similar susceptibility patterns for carbapenems and ceftazidime, and an identical genotype during the intermittent use of antibiotics over 5 months. In two cases, a single course of antibiotic therapy resulted in eradication of P. aeruginosa. Our results suggest that clonal shifts of carbapenem resistance in P. aeruginosa may result from loss of OprD during antibiotic treatment. Therefore, it is important for clinicians to monitor susceptibilities to antibiotics, especially carbapenems, in P. aeruginosa isolated during therapy.

Detection of mutations in quinolone resistance-determining regions in levofloxacin- and methicillin-resistant Staphylococcus aureus: effects of the mutations on fluoroquinolone MICs.

The minimum inhibitory concentrations (MICs) of 18 antibiotics were determined for 66 clinical isolates of staphylococci. Genotypes, mutations in the quinolone resistance-determining regions (QRDRs), and effect of efflux were determined in the 18 levofloxacin-resistant isolates, for which the MICs of levofloxacin were high (> or =8 microg/ml). The increased levofloxacin resistance mainly resulted from some combinations of mutations in the QRDRs, although NorA-mediated efflux may play a minor role in resistance. A combination of mutations in GrlA (Ser80Phe), GrlB (Pro451Ser), and GyrA (Ser84Leu) was found in 4 methicillin-resistant Staphylococcus aureus (MRSA) isolates that were unrelated genotypically. The mutations in grlA QRDR varied in the isolates classified as being in an identical pulsed-field gel electrophoresis (PFGE) group, although the grlB, gyrA, and gyrB QRDRs were the same. These results suggest that the patterns of amino acid mutations in the QRDRs can provide distinct epidemiologic information from PFGE genotypes in fluoroquinolone-resistant MRSA. A combination of at least three mutations in GrlA, GrlB, and/or GyrA is required to increase the MICs of fluoroquinolones, although all of the levofloxacin-resistant MRSA retained the MICs of sitafloxacin in the range of 1 to 2 microg/ml.

Detection of macrolide resistance in Streptococcus pneumoniae.

We determined the susceptibilities of recent clinical isolates of Streptococcus pneumoniae to 19 antibiotics. The frequency of erythromycin nonsusceptibility was high, i.e. 8/13 (61.5%), 10/14 (71.4%) and 11/11 isolates (100%) from 13 penicillin-susceptible, 14 penicillin-intermediate and 11 penicillin-resistant S. pneumoniae, respectively. Macrolide resistance was detected by polymerase chain reaction (PCR) and disk diffusion methods. Of these erythromycin-nonsusceptible pneumococcal isolates, 13/29 (44.8%) isolates contained genomic copies of MEFA and showed non-'D'-shaped zones of inhibition observed around rokitamycin and/or clindamycin disks. Sixteen out of 29 isolates (55.2%) contained copies of ERMB and showed typical 'D'-shaped zones of inhibition, except one isolate. Although the macrolide resistance determinants, MEFA and ERMB, could be identified by PCR and disk diffusion methods, PCR methods were more reliable in elucidating these determinants. The susceptibility pattern to 14-, 15- and 16-membered macrolides and clindamycin differed between the MEFA+ and ERMB+ isolates. All isolates were susceptible to levofloxacin, sparfloxacin and vancomycin. The MICs of sitafloxacin were lowest among the fluoroquinolones examined for 38 isolates.

Effects of mupirocin at subinhibitory concentrations on flagella formation in Pseudomonas aeruginosa and Proteus mirabilis.

Colonization of Pseudomonas aeruginosa at trachea, nares and oropharynx can cause ventilator-acquired pneumonia. To identify beneficial effects of antibiotics on expression of virulence factors related to colonization by such pathogens, we evaluated the effect of mupirocin on flagella formation in P. aeruginosa and on motility and flagella formation in Proteus mirabilis. In P. aeruginosa, subinhibitory concentrations of mupirocin inhibited flagella formation, which was associated with reduced flagellin expression. In P. mirabilis, subinhibitory concentrations of mupirocin dose-dependently suppressed bacterial motility and flagella formation, again with reduced flagellin expression. Our results indicate that subinhibitory concentrations of mupirocin can suppress expression of virulence factors in P. aeruginosa and P. mirabilis.