Phenotypic detection of plasmid-acquired AmpC in Escherichia coli--evaluation of screening criteria and performance of two commercial methods for the phenotypic confirmation of AmpC production.

The phenotypic detection of plasmid-acquired AmpC (pAmpC) in Escherichia coli is challenging, and molecular methods are required for confirmation. In addition to cefoxitin resistance, multiresistance and high-level resistance to cephalosporins have both been suggested as criteria for targeting isolates with pAmpC, but data to support these proposed criteria are lacking. A Swedish collection of 378 isolates with either putative chromosomal hyperproduction of AmpC (cAmpC) or pAmpC were subjected to disk diffusion and minimum inhibitory concentration (MIC) determination with the Etest. The frequency of resistance to gentamicin, ciprofloxacin, and trimethoprim among cAmpC and pAmpC was compared to elucidate the issue of multidrug resistance. Lastly, methods for the phenotypic confirmation of pAmpC were compared. One in-house disk diffusion method, one method employing NeoSensitabs (Rosco), and one Etest method (bioMérieux) were compared. The analysis of histograms based on both disk diffusion and the Etest showed that resistance [according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST)] to cefotaxime and/or ceftazidime occurred in almost all isolates. By coining resistance instead of non-susceptibility, the number of isolates required to subject to phenotypic testing/genotypic confirmation dropped by more than 40 %, without compromising the sensitivity substantially. Further, almost 70 % of isolates with pAmpC were non-multidrug resistant, clearly indicating that this is an inappropriate criterion for further investigation. The phenotypic tests all had more than 90 % sensitivity, and the best sensitivities were obtained with the in-house method and with the ceftazidime ± cloxacillin NeoSensitab. In conclusion, clinical resistance to cefotaxime and/or ceftazidime seems to be an appropriate criterion for pAmpC screening, and several phenotypic methods perform well for the phenotypic confirmation of AmpC production prior to genotypic confirmation.

Dynamics of penicillin-susceptible clones in invasive pneumococcal disease.

In a 10-year period, 1987-1997, there was a >4-fold increase in the rate of pneumococcal bacteremia in Sweden. Invasive pneumococcal isolates (n=1136), which were obtained from 18 Swedish clinical microbiology laboratories from 1987 through 1997, and other national and international isolates were serotyped, and their clonal relationships were determined by molecular typing. The increase in invasive pneumococcal disease in Sweden during this period was associated particularly with an increase in isolates of serotypes 1 and 14. A 3-fold increase of type 14 was seen from 1987 through 1992, and a 10-fold increase of type 1 occurred from 1992 through 1997. One dominating penicillin-susceptible clone of type 14 was responsible for the increase of type 14 during the first 5 years. This clone also was found in Canada and the United States and was shown by multilocus sequence typing to correspond to a previously identified hyper-virulent clone. A novel penicillin-susceptible clone of type 1, which was not found among invasive isolates from 1987 or 1992, was responsible for the increase of serotype 1 during the last 5 years. These results illustrate the ability of virulent penicillin-susceptible pneumococcal clones to emerge and spread rapidly within a country.

Activity of trovafloxacin against blood isolates of Streptococcus pneumoniae in Sweden.

The in vitro activity of the fluoroquinolone trovafloxacin (CP-99,219) against 257 blood isolates of Streptococcus pneumoniae obtained from Swedish hospitals was compared with those of commonly prescribed oral antibiotics and also with those of ciprofloxacin and ofloxacin against a collection of strains resistant (n = 6) or intermediately resistant (n = 22) to penicillin (Pc-R). The MICs of trovafloxacin for Pc-R strains of pneumococci ranged from 0.032 to 0.25 mg/l. No difference was seen between the clinical isolates and the Pc-R strains (MIC50 = 0.064 mg/l and MIC90 = 0.125 mg/l for both collections). For the Pc-R strains, the MIC50 and MIC90 values of ciprofloxacin were 0.5 and 1 mg/l, and those of ofloxacin 2 and 4 mg/l. The incidence of resistance in the two collections (clinical isolates/Pc-R strains) was 3%/39% for tetracycline, 1%/18% for macrolides, and 3%/57% for trimethoprim/sulfamethoxazole. The results of the current study suggest that the clinical efficacy of trovafloxacin in the treatment of pneumococcal infections should be investigated.

In vitro activity of amoxicillin plus clavulanic acid against Haemophilus influenzae and Branhamella catarrhalis.

The in vitro activity of amoxicillin in the presence of clavulanic acid against clinical isolates of Haemophilus influenzae and Branhamella catarrhalis was assessed in comparison with ampicillin, amoxicillin, cefaclor and erythromycin. The isolates were selected so as to yield equal numbers of beta-lactamase producing and non-beta-lactamase producing strains of the two species. MICs obtained by agar dilution indicated that amoxicillin in the presence of clavulanic acid was the most active of the drugs tested. Clavulanic acid potentiated the activity of amoxicillin against beta-lactamase-producing strains of both Haemophilus influenzae and Branhamella catarrhalis. Further studies on a few strains of each species revealed that the beta-lactamase of Haemophilus influenzae (TEM-1) rapidly inactivated ampicillin and slowly inactivated cefaclor but not cefuroxime. The Branhamella catarrhalis enzyme rapidly inactivated cefaclor, ampicillin and to some extent cefuroxime. Clavulanic acid afforded protection against the beta-lactamase action of both species when beta-lactam antibiotics were added to bacterial cultures.