Abstract
The agar dilution, broth microdilution, and disk diffusion methods were compared to determine the in vitro susceptibility of 428 extended-spectrum-β-lactamase (ESBL)-producing Escherichia coli and Klebsiella pneumoniae to fosfomycin. Fosfomycin showed very high activity against all ESBL-producing strains. Excellent agreement between the three susceptibility methods was found for E. coli, whereas marked discrepancies were observed for K. pneumoniae.
Fosfomycin tromethamine is a stable salt of fosfomycin which is licensed for the single-dose treatment of acute uncomplicated urinary tract infections (UTIs) caused by susceptible organisms (8, 12). After many years of fosfomycin use, both in Spain and worldwide, fosfomycin continues to be active against the most common uropathogens, and there is a very low incidence of resistant strains in Escherichia coli (about 2%) (2, 12, 14).
Isolation of extended-spectrum-β-lactamase (ESBL)-producing E. coli and Klebsiella pneumoniae strains is increasing in Spain, and a high proportion of these isolates are recovered from outpatients with uncomplicated UTIs (4, 7, 11). It is common to find that the same plasmid coding for ESBL also contains genes conferring resistance to several groups of antimicrobial agents, such as aminoglycosides and cotrimoxazole (7, 10). The concurrence of quinolone resistance, particularly in ESBL-producing K. pneumoniae, is frequent (6, 10, 13), there being few alternatives for the appropriate oral treatment of uncomplicated UTIs caused by ESBL-producing microorganisms.
The present study was designed to assess the in vitro activity of fosfomycin by three different methods against 428 ESBL-producing strains, made up of 290 (68%) E. coli and 138 (32%) K. pneumoniae isolates.
The strains were collected in two Spanish studies: the first, developed in 2000, included 40 hospitals representing different regions of Spain, and the second was carried out in Seville, Spain, at the University Hospital of Virgen Macarena (UHVM), using strains recovered between 1995 and 2001 (7, 11). In both studies ESBL production was determined by broth microdilution according to CLSI guidelines (5).
The strains collected from the multicenter study included 170 E. coli and 70 K. pneumoniae strains, while those obtained from the UHVM included 120 E. coli and 68 K. pneumoniae strains. One-hundred thirty-two out of 290 (45.5%) E. coli and 16 out of 138 (11.6%) K. pneumoniae strains were isolated from outpatients with community-acquired infections, and among these isolates found in the community, 66 (22.7%) E. coli and 9 (6.5%) K. pneumoniae strains were isolated from urine samples of women with uncomplicated UTIs.
The MICs of fosfomycin (Zambon, Milan, Italy) were determined by the agar dilution and broth microdilution methods in cation adjusted Mueller-Hinton medium, supplemented with 25 mg/liter of G-6-P (glucose-6-phosphate; Sigma Chemical Co.). A twofold dilution across a range of 0.25 to 512 mg of fosfomycin per liter was used. The inocula were prepared to achieve 1 × 104 CFU per spot (agar dilution) or 5 × 105 CFU/ml (broth microdilution) according to CLSI recommendations (5). At the same time, disks containing 200 μg fosfomycin and 50 μg G-6-P (Oxoid) were tested following the method outlined by CLSI (9).
Fosfomycin breakpoints for the interpretative criteria for E. coli were used according to CLSI guidelines (5, 9). Because of the absence of accepted breakpoints for K. pneumoniae, those proposed by CLSI for E. coli have been used (5, 9).
Susceptibility results of the broth microdilution and disk diffusion methods were compared with those obtained from the agar dilution method used as the reference method. Agreement and discrepancies between the evaluated and reference method were classified as complete agreements, very major errors, major errors, and minor errors. Differences in proportions among paired categorical data were calculated by the McNemar test and for independent categorical data by the chi-square test. The Wilcoxon test was used for comparing MICs. Statistical significance was established as P < 0.05, and very significant was established as P < 0.01.
By the agar dilution method, 417 (97.4%) strains were susceptible to fosfomycin, with MICs of ≤64 μg/ml. The fosfomycin 50% minimum inhibitory concentration (MIC50) was 2 μg/ml, and the MIC90 was 32 μg/ml (range, 0.5 to 512 μg/ml). Most isolates with intermediate susceptibility or resistance to fosfomycin (2.6%) were K. pneumoniae isolates: four resistant strains (MIC, ≥256 μg/ml) and six intermediate strains (MIC, 128 μg/ml). Only one strain of E. coli of all 290 tested showed intermediate susceptibility to fosfomycin (MIC, 128 μg/ml). K. pneumoniae isolates have the highest MICs for fosfomycin (MIC50 and MIC90 of 16 to 64 μg/ml) but are still within the susceptible range, whereas more than 90% of E. coli isolates yielded very low MICs (≤4 μg/ml) (Fig. 1). These results are similar to those described in previous reports of non-ESBL-producing isolates (2, 8, 14), confirming that fosfomycin retains its activity against ESBL-producing isolates, and cross-resistance with other classes of antimicrobial agents is not a problem at present.
FIG. 1.
Distributions of fosfomycin MICs obtained with the agar dilution method for ESBL-producing E. coli and K. pneumoniae isolates.
Important discrepancies have been reported between broth and agar dilution MICs for fosfomycin, and, so far, agar dilution is the only approved fosfomycin MIC susceptibility testing method (5). The use of disks containing 200 μg of fosfomycin and 50 μg of G-6-P in Mueller-Hinton medium is only indicated for testing with E. coli (9).
However, in common with others (1, 3, 6) who have used Mueller-Hinton broth supplemented with G-6-P, we observed only minor differences (average, ±1.93 dilutions; range, 0 to 7). In 241 (56.4%) strains the MIC broth values were higher than those obtained with agar (P < 0.001). The complete percentage agreement between the broth microdilution method and agar dilution results was higher for E. coli (98.6%) than for K. pneumoniae strains (72.5%) (P < 0.001). Table 1 compares the categories obtained by the microdilution broth and disk diffusion methods with those obtained by the agar dilution method used as reference. For both evaluated methods, the error rate observed for K. pneumoniae was significantly higher than that for E. coli (P < 0.001).
TABLE 1.
Correlation of broth microdilution and disk diffusion methods with the reference agar dilution method
| ESBL-producing strain | No. | Method | Error rate, % (no. of strains)
|
|||
|---|---|---|---|---|---|---|
| Very major | Major | Minor | Total | |||
| E. coli | 290 | Broth microdilution | 0 | 0.3 (1) | 1.03 (3) | 1.4 (4) |
| Disk diffusion | 0 | 0 | 0.3 (1) | 0.3 (1) | ||
| K. pneumoniae | 138 | Broth microdilution | 1.4 (2) | 12.3 (17) | 13.7 (19) | 27.5 (38) |
| Disk diffusion | 0.7 (1) | 10.9 (15) | 3.6 (5) | 15.2 (21) | ||
| Total | 428 | Broth microdilution | 0.46 (2) | 4.2 (18) | 5.1 (22) | 10.2 (44) |
| Disk diffusion | 0.23 (1) | 3.5 (15) | 1.4 (6) | 5.1 (22) | ||
Our study confirms that the result of fosfomycin susceptibility testing is dependent on the method used and the microorganisms tested. Significant differences were not found between the two evaluated methods for clinical categories for E. coli (broth, P = 0.625; disk, P = 1.00). However, the results of susceptibility testing of K. pneumoniae showed poor correlation with the broth microdilution (P < 0.001) and disk diffusion methods (P = 0.001) reporting greater resistance to fosfomycin. Besides, for K. pneumoniae, the error rate obtained with the broth microdilution was significantly higher than that with the disk method (P = 0.01).
When broth microdilution or disk diffusion methods are used to determine the susceptibility to fosfomycin of K. pneumoniae, the results for resistance and intermediate susceptibility should be confirmed by the reference method. Because most automated systems for antimicrobial susceptibility testing are microdilution-based methods, resistance to fosfomycin may be overestimated in laboratories employing such systems, especially for isolates other than E. coli.
Our data suggest an excellent in vitro activity of fosfomycin against ESBL-producing E. coli and K. pneumoniae strains. Since fosfomycin is only approved in the United States for E. coli UTI treatment, further clinical studies are required to assess the clinical efficacy of fosfomycin for the treatment of UTIs caused by ESBL-producing K. pneumoniae.
Acknowledgments
This study was partially supported by a grant from Pharmazan and was developed under the auspices of the Spanish Network for Research on Infectious Diseases (REIPI CO3/14; Instituto de Salud Carlos III, Madrid, Spain).
REFERENCES
- 1.Allenberger, F., and I. Klare. 1999. In vitro activity of fosfomycin against vancomycin resistant enterococci. J. Antimicrob. Chemother. 43:211-217. [DOI] [PubMed] [Google Scholar]
- 2.Andreu, A., J. I. Alos, M. Gobernado, F. Marco, M. de la Rosa, and J. A. Garcia-Rodriguez. 2005. Etiology and antimicrobial susceptibility among uropathogens causing community acquired lower urinary tract infections: a nationwide surveillance study. Enferm. Infecc. Microbiol. Clin. 23:1-3. [DOI] [PubMed] [Google Scholar]
- 3.Barry, A. L., and P. C. Fuchs. 1991. In vitro susceptibility testing procedures for fosfomycin tromethamine. Antimicrob. Agents Chemother. 35:1235-1238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bou, G., M. Cartelle, M. Tomás, D. Canle, F. Molina, R. Moure, J. M. Eiros, and A. Guerrero. 2002. Identification and broad dissemination of the CTX-M-14 beta-lactamase in different Escherichia coli strains in the northwest area of Spain. J. Clin. Microbiol. 40:4030-4036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Clinical and Laboratory Standards Institute. 2005. Performance standards for antimicrobial susceptibility testing 15th informational supplement. Approved standard M100-S15. Clinical and Laboratory Standards Institute, Wayne, Pa.
- 6.Fuchs, P. C., A. L. Barry, and S. D. Brown. 1999. Fosfomycin tromethamine susceptibility of outpatient urine isolates of Escherichia coli and Enterococcus faecalis from ten North American medical centres by three methods. J. Antimicrob. Chemother. 43:137-140. [DOI] [PubMed] [Google Scholar]
- 7.Hernandez, J. R., A. Pascual, R. Canton, and L. Martinez-Martinez. 2003. Extended-spectrum beta lactamase producing Escherichia coli and Klebsiella pneumoniae in Spanish hospitals (GEIH-BLEE project 2002). Enferm. Infecc. Microbiol. Clin. 21:77-82. [PubMed] [Google Scholar]
- 8.Marchese, A., L. Gualco, E. A. Debbia, G. C. Schito, and A. M. Schito. 2003. In vitro activity of fosfomycin against gram-negative urinary pathogens and the biological cost of fosfomycin resistance. Int. J. Antimicrob. Agents 22:853-859. [DOI] [PubMed] [Google Scholar]
- 9.National Committee for Clinical Laboratory Standards. 2003. Performance standards for antimicrobial disk susceptibility tests, 8th ed. Approved standard M3-A8. National Committee for Clinical Laboratory Standards, Wayne, Pa.
- 10.Paterson, D. L., L. Mulazimoglu, J. M. Casellas, W. C. Ko, H. Goosens, A. Von Gottberg, S. Mohapatra, G. M. Trenholme, K. P. Klugman, J. G. McCormack, and V. L. Yu. 2000. Epidemiology of ciprofloxacin resistance and its relationship to extended spectrum betalactamase production in Klebsiella pneumoniae isolates causing bacteremia. Clin. Infect. Dis. 30:473-478. [DOI] [PubMed] [Google Scholar]
- 11.Rodriguez-Baño, J., M. D. Navarro, L. Romero, L. Martinez-Martinez, M. A. Muniain, E. J. Perea, R. Perez-Cano, and A. Pascual. 2004. Epidemiology and clinical features of infections caused by extended-spectrum β-lactamase-producing Escherichia coli in nonhospitalized patients. J. Clin. Microbiol. 42:1089-1094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Schito, G. C. 2003. Why fosfomycin trometamol as first line therapy for uncomplicated UTI? Int. J. Antimicrob. Agents 22(Suppl. 2):79-83. [DOI] [PubMed] [Google Scholar]
- 13.Schwaber, M. J., S. Navon-Venezia, and Y. Carmelli. 2005. High level of antimicrobial coresistance among extended-spectrum β-lactamase-producing Enterobacteriaceae. Antimicrob. Agents Chemother. 49:2137-2139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ungheri, D., E. Albini, and G. Belluco. 2002. In vitro susceptibility of quinolone resistant clinical isolates of Escherichia coli to fosfomycin trometamol. J. Chemother. 14:237-240. [DOI] [PubMed] [Google Scholar]