Service de Microbiologie, Hôpital Beaujon, 92110 Clichy,1 Service de Microbiologie, Hôpital Tenon, 75020 Paris,2 Bio-Rad, 92430 Marnes-la-Coquette,3 Service de Microbiologie, Faculté de Pharmacie, Université de Bordeaux 2, 33076 Bordeaux cedex, France4
Received 25 April 2002/ Returned for modification 25 September 2002/ Accepted 17 March 2003
Osiris is a video zone size reader for disk diffusion tests
featuring a built-in extended expert system (EES). The efficacy
of the EES for the identification of the ß-lactam
susceptibility phenotypes of
Pseudomonas aeruginosa isolates
was evaluated. Thirteen ß-lactams were tested in four
laboratories by the disk diffusion test with 53 strains with
well-characterized resistance mechanisms, including the production
of 12 extended-spectrum ß-lactamases (ESBLs). The
plates were read with the Osiris system and the results were
interpreted with the ESS, and then the phenotype identified
by the EES was compared to the resistance mechanism. The strains
were also screened for the presence of ESBL production by a
double-disk synergy test by placing the strains between an extended-spectrum
cephalosporin-containing disk and a clavulanic acid-containing
disk at distances of 30, 20, 15, and 10 mm from each other.
Overall, the EES accurately identified the phenotypes of 88.2%
of the strains and indicated an association with several mechanisms
for 3.8% of the strains. No phenotype was identified in four
strains with low levels of penicillinase production. Misidentifications
were observed for two penicillinase-producing strains: one strain
with partially derepressed cephalosporinase production and one
strain overexpressing the MexA-MexB-OprM efflux system. The
production of only four ESBLs was detected by the standard synergy
test with a 30-mm distance between the disks. The production
of five further ESBLs was identified by reducing the distance
to 20 mm, and the production of the last three ESBLs was detected
only at a distance of 15 or 10 mm. Our results indicate that
the Osiris EES is an effective tool for the identification of
P. aeruginosa ß-lactam phenotypes. A specific double-disk
synergy test with reduced disk distances is necessary for the
detection of ESBL production by this organism.
Pseudomonas aeruginosa is characterized by its natural resistance
to ß-lactam agents and its ability to acquire additional
mechanisms of resistance to these drugs. The most frequent mechanisms
of acquired resistance are production of transferable ß-lactamases,
overproduction of the derepressed AmpC cephalosporinase, overexpression
of the MexA-MexB-OprM efflux system, and porin D2 deficiency
(
16). Each ß-lactam phenotype, defined as the expression
of a given mechanism of resistance to these agents, is characterized
by a specific susceptibility pattern. Thus, classical ß-lactamases
belonging to the CARB, OXA, or TEM group are associated with
a penicillinase production phenotype (resistance to carboxy-
and ureidopenicillins and susceptibility to ceftazidime), whereas
extended-spectrum ß-lactamases (ESBLs), such as PER-1,
VEB-1, or OXA, TEM, and SHV derivatives, are capable of hydrolyzing
extended-spectrum cephalosporins (
3,
14). Derepressed cephalosporinase
confers resistance to all ß-lactams with the exception
of carbapenems (
3,
4). Overexpression of the MexA-MexB-OprM
efflux system mainly affects the activities of carboxypenicillins
and monobactams (
24). Lastly, carbapenem resistance is mostly
due to porin D2 deficiency and is independent of the susceptibility
to other ß-lactam agents (
15). Osiris is a video zone
size reader system for the reading and interpretation of susceptibility
tests performed by the disk diffusion method. The inhibition
zone diameter around each antibiotic disk is measured by the
camera and compared with predefined breakpoints in order to
classify the tested strains as susceptible, intermediate, or
resistant. The software includes a built-in extended expert
system (EES) which detects inconsistent results, identifies
the susceptibility patterns associated with specific phenotypes,
and interprets the susceptibility results according to the phenotype
recognized. For
P. aeruginosa, eight ß-lactam phenotypes
corresponding to the most frequently encountered resistance
mechanisms can be identified by the EES (Table
1). The goal
of the present study was to evaluate the efficacy of the Osiris
EES for identifying the ß-lactam phenotypes of
P. aeruginosa by using strains with well-characterized resistance
mechanisms. We also analyzed the susceptibility patterns associated
with the different mechanisms in order to improve the EES and,
more particularly, to determine indicators for ESBL production.
| fig.ommitted |
TABLE 1. Identification of P. aeruginosa ß-lactam phenotypes by the Osiris EES
| |
Strains. A total of 53
P. aeruginosa strains were collected for the study,
including 7 ticarcillin-susceptible (wild-type) strains and
46 strains with well-characterized ß-lactam resistance
mechanisms. The ß-lactam phenotypes of the strains
tested are listed in Table
2. Some of these were reference strains
previously reported by others (
7,
10-
13,
17,
18,
20,
21,
23).
The other strains were clinical isolates whose resistance mechanisms
had been identified by biochemical and molecular techniques.
ß-Lactamases were characterized by their isoelectric
points, their inhibitor-substrate profiles with clavulanic acid
and cloxacillin (
4), and PCR-fragment length polymorphism analysis
(
2). Strains overexpressing the MexA-MexB-OprM system were kindly
provided by P. Plesiat. Five to seven strains with different
levels of phenotypic expression of each of the three most frequent
resistance mechanisms (production of PSE-1, cephalosporinase
derepression, and efflux) were included in the study. Most strains
had a single resistance mechanism. However, a few ß-lactamase-producing
strains were also resistant to carbapenems, which was presumably
due to an associated impermeability. Strains were collected
by Bio-Rad Laboratories (Marnes-la-Coquette, France), kept frozen
at -70°C prior to testing, subcultured onto tryptic soy
agar, and sent to each of the participating laboratories.
| fig.ommitted |
TABLE 2. ß-Lactam phenotypes of the 53 strains tested
| |
Antimicrobial agents. Disks containing ticarcillin (75 µg), ticarcillin-clavulanic
acid (75/10 µg), piperacillin (75 µg), piperacillin-tazobactam
(75/10 µg), cefotaxime (30 µg), cefoperazone (30
µg), ceftazidime (30 µg), cefsulodin (30 µg),
cefepime (30 µg), cefpirome (30 µg), aztreonam (30
µg), imipenem (10 µg), and meropenem (10 µg)
were supplied by Bio-Rad.
Susceptibility tests. The 13 aforementioned ß-lactam agents were tested in each laboratory with the 53 strains by the disk diffusion method according to the guidelines of the Comité de l'Antibiogramme de la Société Française de Microbiologie (CASFM) (6). Briefly, the strains were subcultured onto tryptic soy agar. A bacterial cell suspension in Mueller-Hinton broth adjusted to 106 CFU/ml was used to inoculate a square petri plate containing Mueller-Hinton agar (Bio-Rad) by flooding. Disks containing the antibiotics were distributed onto the plates according to a predefined scheme by using a disk dispenser. After incubation at 37°C for 18 h, the plates were read with the Osiris video instrument and the results were interpreted with the EES according to the instructions of the manufacturer. The phenotype identified by the EES was compared to the ß-lactam resistance mechanism determined by biochemical and molecular techniques.
Detection of ESBLs. Strains were screened for the presence of ESBLs by the double-disk synergy test, as described by Jarlier et al. (8). On the square petri plate inoculated for susceptibility testing, the ticarcillin-clavulanic acid disk was placed next to the ceftazidime, cefepime, and aztreonam disks at a distance of 30 mm (center to center). Moreover, an additional test was done in order to detect OXA-derived ESBLs, which are poorly susceptible to clavulanic acid (14). A round Mueller-Hinton agar plate was inoculated with the bacterial suspension in the same way described above for the disk diffusion procedure. Three ceftazidime disks were placed at distances of 20, 15, and 10 mm, respectively, from a central amoxicillin-clavulanic acid disk. The test result was considered positive when an enhancement of the inhibition zone around at least one of the ceftazidime, cefepime, or aztreonam disks toward the clavulanic acid disk was observed.
Susceptibility patterns. The means and ranges of the inhibition zone diameters for each
phenotype are shown in Table
3. As expected, penicillinase-producing
strains showed decreased susceptibilities to all antibiotics
tested except ceftazidime and the carbapenems. The observed
susceptibility patterns were highly variable according to the
type of ß-lactamase produced. Compared to the levels
of resistance in strains producing enzymes of the CARB, OXA-10,
and OXA-1 groups, the production of OXA-2 and TEM-1 resulted
in lower levels of resistance to ticarcillin and piperacillin;
and strains producing these enzymes were more susceptible to
the combination of ticarcillin and piperacillin with ß-lactamase
inhibitors. The activity of cefoperazone was mostly affected
by CARB and OXA-10, and that of cefsulodin was mostly affected
by CARB and TEM-1. OXA-1-producing strains remained susceptible
to cefsulodin and aztreonam but displayed the greatest level
of resistance to cefepime and cefpirome.
| fig.ommitted |
TABLE 3. Inhibition zone diameters with the data analyzed by phenotype
| |
Cephalosporinase derepression decreased the activities of all
drugs except carbapenems, with a wide range of ceftazidime inhibition
zone diameters detected (Table
3). ESBLs also conferred resistance
to ceftazidime in all strains with the exception of the two
TEM-4-producing strains, for which the ceftazidime inhibition
zone was above the 21-mm breakpoint, despite a visible synergy
with the ticarcillin-clavulanic acid disk. The susceptibility
patterns varied according to the type of ß-lactamase
produced. Strains producing OXA-2- and TEM-derived ESBLs showed
the greatest susceptibilities to ß-lactamase inhibitors,
whereas strains producing OXA-14 were poorly susceptible to
the ß-lactamase inhibitors. The activities of cefepime
and cefpirome were most affected by VEB-1 and SHV-2a, and the
activity of aztreonam was most affected by OXA-18 and VEB-1.
We compared the susceptibility patterns of cephalosporinase-overproducing strains and ESBL-producing strains in order to determine which antibiotics were predictive of ESBL production in ceftazidime-resistant strains. When compared with the ceftazidime inhibition zone diameter, the piperacillin-tazobactam and cefotaxime inhibition zone diameters were the best indicators of ESBL production. Thus, 70% of ESBL-producing strains remained susceptible to piperacillin-tazobactam, whereas only 10.7% of the cephalosporinase-overproducing strains remained susceptible to piperacillin-tazobactam (positive predictive value , 90.3%; negative predictive value , 67.6%). The inhibition zone of piperacillin-tazobactam was at least 6 mm greater than that of ceftazidime for 90% of the ESBL-producing strains but for only 10.7% of the cephalosporinase-overproducing strains (PPV, 90%; NPV, 86.2%). The cefotaxime inhibition zone was greater than the ceftazidime inhibition zone for 65% of the ESBL-producing strains and 0% of the cephalosporinase-overproducing strains (PPV, 100%; NPV, 66.6%).
As reported previously (24), strains overexpressing the MexA-MexB-OprM efflux system showed decreased susceptibilities to all antibiotics with the exception of imipenem, and the loss of activity was most emphasized with ticarcillin and aztreonam. Porin D2 deficiency conferred isolated resistance to carbapenems, with the activity of imipenem being compromised more than that of meropenem.
Evaluation of EES. The phenotypes recognized in each of the four laboratories were identical for 44 (83%) of the isolates tested, whereas discrepancies were found for the remaining 9 (17%) isolates. Overall, the EES was able to correctly and accurately identify the ß-lactam phenotype in 187 (88.2%) cases. In an additional 8 (3.8%) cases, the EES indicated an association of several resistance mechanisms but did not specify the mechanisms involved. The phenotype detected was incorrect in 10 (4.7%) other cases, and no phenotype could be identified in the remaining 7 (3.3%) cases.
The results analyzed by phenotype are shown in Table 4. The correct phenotype was identified for all wild-type strains and the strain with porin D2 deficiency. No phenotype could be identified in one or several centers for the four strains displaying a low-level penicillinase production phenotype (production of an OXA-2, TEM-1, or PSE-1 derivative) because of a ticarcillin inhibition zone greater then 14 mm or intermediate susceptibility to ticarcillin-clavulanic acid or piperacillin (Table 5). The two PSE-4-producing strains were misidentified as ESBL producers in two centers because synergy was observed between the cefepime and ticarcillin-clavulanic acid disks. The mechanism for one strain with low-level derepressed cephalosporinase production was misidentified as efflux in two centers because the piperacillin inhibition zone was slightly greater than the 18-mm breakpoint (Table 4). Conversely, the mechanism for one strain with high-level efflux was misidentified as cephalosporinase production in one center because of a piperacillin inhibition zone slightly below the breakpoint. The accuracy of ESBL detection varied considerably according to the distance used in the double-disk synergy test. Production of only 25% of the ESBLs was detected by the standard test with the 30-mm distance recommended for use when testing members of the family Enterobacteriaceae (8). Most of the strains for which no synergy was observed by the double-disk test were misclassified by the EES as cephalosporinase overproducers (Table 4). The percentage of ESBL producers detected was increased to nearly 80% by using an additional screening test with reduced distances of 20, 15, and 10 mm. Only PER-1 production by one strain and the production of TEM-related enzymes were identified at 30 mm (Table 6). The production of five additional ESBLs (PER-1, VEB-1, OXA-18, an OXA-2 derivative, and SHV-2a) was detected by reducing the distance to 20 mm (Fig. 1). OXA-14 and OXA-15 production was detected only by use of a 15- or a 10-mm distance. In contrast, TEM-4 production could not be identified with a reduced distance because of a large ceftazidime inhibition zone. For ESBL-producing strains that were also resistant to carbapenems (by production of VEB-1 and an OXA-2 derivative), the EES classified the phenotype as "association of several resistance mechanisms."
| fig.ommitted |
TABLE 4. Identification of ß-lactam phenotypes by the EES
| |
| fig.ommitted |
TABLE 5. Phenotypes incorrectly identified or not identified by the EES
| |
| fig.ommitted |
TABLE 6. Detection of ESBL production by the double-disk synergy test between disks of ceftazidime and clavulanic acid placed at various distances
| |
| fig.ommitted |
FIG. 1. Detection of ESBL production by the double-disk diffusion test. TIC, ticarcillin; PIP, piperacillin; CFP, cefoperazone; CTX, cefotaxime; ATM, aztreonam; TCC, ticarcillin-clavulanic acid; CAZ, ceftazidime; TZP, piperacillin-tazobactam; CPO, cefpirome; FEP, cefepime; IPM, imipenem; CFS, cefsulodin; AN, amikacin; MEM, meropenem; TM, tobramycin; CIP, ciprofloxacin; AMC, amoxicillin-clavulanic acid. (A) PER-1 production detected by a synergy between a ceftazidime disk and a ticarcillin-clavulanic acid disk placed 30 mm apart; (B) PER-1 production detected by a synergy between a ceftazidime disk and an amoxicillin-clavulanic acid disk placed 20 mm apart; (C) OXA-15 production detected by synergy between a ceftazidime disk and an amoxicillin-clavulanic acid disk placed 15 mm apart.
| |
The routine detection of the ß-lactam phenotypes of
clinical isolates of
P. aeruginosa is particularly important
because of therapeutic problems related to acquired resistance
in this species. It can help clinicians prescribe empiric antibiotic
therapy and can help microbiologists monitor trends in antimicrobial
resistance, such as the dissemination of ESBLs.
Overall, the Osiris EES was able to correctly identify the ß-lactam phenotypes of 88.2% of the strains tested. The panel of strains that we used in the study included a high proportion with unusual phenotypes, such as the production of ESBL, TEM, or OXA enzymes, which did not reflect the average frequency of ß-lactam phenotypes among clinical P. aeruginosa isolates. Thus, the EES would probably perform even better in the average clinical laboratory. The performance was variable according to the mechanism involved. The wild-type pattern and porin D2 deficiency were correctly identified. Among the penicillinase producers, the high-level penicillinase production phenotype, observed with producers of PSE-1, CARB-3, the OXA-10 group, and the OXA-1 group, was recognized in all strains. In contrast, no phenotype was identified in one or several laboratories in strains associated with a low-level penicillinase production pattern due to the production of a TEM-1, OXA-2, or PSE-1 derivative. These errors were due to discrepancies between the inhibition zone diameters of ticarcillin, ticarcillin-clavulanic acid, or piperacillin and the expected phenotype. Moreover, two PSE-4 producers were identified as ESBL producers because of a positive synergy test result between cefepime and clavulanic acid. In laboratories where this synergy was not detected, the phenotype was correctly identified as high-level penicillinase production. Others have reported (22) similar false-positive results of the double-disk test with cefepime for species of the family Enterobacteriaceae. This phenomenon may be more frequent in P. aeruginosa, since the PSE and OXA enzymes more readily hydrolyze cefepime than ceftazidime (1). Thus, cefepime should not be used for the detection of ESBLs by the double-disk test in this particular species.
As for the penicillinase production phenotype, the susceptibility pattern was highly variable according to the type of ß-lactamase produced. TEM and OXA-2 producers were susceptible to ß-lactamase inhibitors, whereas OXA-10 and OXA-1 producers were poorly susceptible to these inhibitors. Producers of enzymes of the OXA-1 group displayed a particular pattern characterized by susceptibility to ceftazidime, cefsulodin, and aztreonam and resistance to piperacillin-tazobactam, cefepime, and cefpirome. However, the pattern associated with the producers of CARB enzymes was highly variable and not easily differentiable from that for producers of enzymes of the OXA-10 group. Thus, it seemed very difficult to infer the ß-lactamase involved from the susceptibility pattern observed for penicillinase-producing strains of P. aeruginosa.
The EES was able to ascertain the phenotype in most strains with derepressed cephalosporinase production or active efflux. These phenotypes can usually be differentiated by susceptibility to ureidopenicillins and extended-spectrum cephalosporins. High-level cephalosporinase production confers resistance to piperacillin and ceftazidime, whereas most strains with active efflux remain susceptible to these drugs (3, 4, 24). However, the susceptibility pattern varies according to whether the cephalosporinase production is partially or totally derepressed and according to the level of expression of the MexA-MexB-OprM efflux system. In our study, the piperacillin inhibition zone diameter was slightly above the 18-mm breakpoint in two laboratories for one strain with partially derepressed cephalosporinase production, and the resistance mechanism was therefore misidentified as active efflux. Conversely, one strain with high-level efflux displayed piperacillin and ceftazidime inhibition zone diameters slightly below their respective breakpoints in one laboratory, and its mechanism of resistance was misidentified as derepressed cephalosporinase production. The use of other antibiotics such as aztreonam or cefepime should be considered to improve the differentiation between these two phenotypes. Indeed, the activities of these drugs were affected significantly more than the activity of ceftazidime against strains with efflux but remained superior or equal to that of ceftazidime against cephalosporinase-overproducing strains (Table 4).
ESBL production in P. aeruginosa isolates, previously limited to a few geographical areas such as Turkey (20, 21), has recently been increasingly reported in other parts of the world (5, 9-11, 18, 19). It is therefore important to identify this phenotype in order to monitor the spread of such resistance genes in this species. According to our results, the presence of an ESBL should be suspected in ceftazidime-resistant, piperacillin-tazobactam-susceptible strains or when the inhibition zone of cefotaxime is greater than that of ceftazidime. However, these indicators were insufficient to differentiate cephalosporinase derepression and ESBL production in all strains. The detection of ESBLs in members of the family Enterobacteriaceae is based on a synergy test between an extended-spectrum cephalosporin disk and a clavulunic acid disk placed 30 mm apart (8, 22). It has been suggested that the incidence of ESBLs in P. aeruginosa may be underestimated since the double-disk synergy test may fail to detect strains that produce OXA-derived enzymes, which are poorly susceptible to clavulanic acid (14). Our study confirms that the standard test with a 30-mm distance is insufficient to identify most ESBLs produced by this species. Strains that produced only TEM-derived enzymes and one PER-1 enzyme were detected by use of a 30-mm distance. It was necessary to reduce the distance to 20 mm to detect strains that produce VEB-1, SHV-2a, and OXA-18, whereas strains that produce the other OXA-derived ESBLs were identified only at a distance of 10 or 15 mm. In addition, the EES did not indicate the presence of an ESBL, despite a positive synergy test, in strains which were also resistant to carbapenems; and the mechanisms of resistance were therefore classified as an "association of several mechanisms." Resistance to imipenem in P. aeruginosa is mostly due to porin D2 deficiency and is independent of susceptibility to other ß-lactams (15). Thus, the database definition of phenotypes other than porin D2 deficiency should not take into account susceptibility to carbapenems.
In this study, disk diffusion tests were performed and interpreted according to the guidelines of the CASFM. However, there are differences between the CASFM and the NCCLS in terms of the ß-lactam breakpoints for P. aeruginosa. Therefore, our results may not have been similar if inhibition zone diameters had been interpreted according to the NCCLS breakpoints. Thus, the validity of our results is limited to countries where the CASFM guidelines are used.
In conclusion, the Osiris EES is an effective tool for the identification of P. aeruginosa ß-lactam phenotypes. An upgrade of the EES could improve the identification of tricky phenotypes such as low-level penicillinase production, partially derepressed cephalosporinase production, and high-level active efflux. Ceftazidime-resistant strains suspected of producing an ESBL on the basis of their susceptibility patterns should be tested by a specific double-disk synergy test with reduced disk distances.
We thank C. Bizet, H. Marchandin, P. Nordmann, and P. Plesiat
for providing strains.
- Aubert, D., L. Poirel, J. Chevalier, S. Leotard, J.-M. Pages, and P. Nordmann. 2001. Oxacillinase-mediated resistance to cefepime and susceptibility to ceftazidime in Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 45:1615-1620.
- Bert, F., C. Branger, and N. Lambert-Zechovsky. 2002. Identification of PSE and OXA ß-lactamase genes in Pseudomonas aeruginosa using PCR-restriction fragment length polymorphism. J. Antimicrob. Chemother. 50:11-18.
- Cavallo, J. D., R. Fabre, F. Leblanc, M. H. Nicolas-Chanoine, A. Thabaut, and the GERPB. 2000. Antibiotic susceptibility and mechanisms of ß-lactam resistance in 1310 strains of ß-lactam: a French multicentre study (1996). J. Antimicrob. Chemother. 46:133-136.
- Chen, H. Y., M. Yuan, and D. M. Livermore. 1995. Mechanisms of resistance to ß-lactam antibiotics amongst Pseudomonas aeruginosa isolates collected in the UK in 1993. J. Med. Microbiol. 43:300-309.
- Claeys, G., G. Verschraegen, T. de Baere, and M. Vaneechoutte. 2001. PER-1 ß-lactamase-producing Pseudomonas aeruginosa in an intensive care unit. J. Antimicrob. Chemother. 45:924-925.
- Comité de l'Antibiogramme de la Société Française de Microbiologie. 2002. Communiqué 2001-2002. Société Française de Microbiologie, Paris, France. [Online.] http://www.sfm.asso.fr/.
- Danel, F., L. M. C. Hall, D. Gur, and D. M. Livermore. 1997. OXA-15, an extended-spectrum variant of OXA-2 ß-lactamase, isolated from a Pseudomonas aeruginosa strain. Antimicrob. Agents Chemother. 41:785-790.
- Jarlier, V., M.-H. Nicolas, G. Fournier, and A. Philippon. 1988. Extended broad-spectrum ß-lactamases conferring transferable resistance to newer ß-lactam agents in Enterobacteriaceae: hospital prevalence and susceptibility pattern. Rev. Infect. Dis. 10:867-878.
- Luzzaro, F., E. Mantengoli, M. Perilli, G. Lombardi, V. Orlandi, A. Orsatti, G. Amicosante, G. M. Rossolini, and A. Toniolo. 2001. Dynamics of a nosocomial outbreak of multidrug-resistant Pseudomonas aeruginosa producing the PER-1 extended-spectrum ß-lactamase. J. Clin. Microbiol. 39:1865-1870.
- Marchandin, H., H. Jean-Pierre, C. De Champs, D. Sirot, H. Darbas, P. F. Perigault, and C. Carriere. 2000. Production of a TEM-24 plasmid-mediated extended-spectrum ß-lactamase by a clinical isolate of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 44:213-216.
- Naas, T., L. Philippon, L. Poirel, E. Ronco, and P. Nordmann. 1999. An SHV-derived extended-spectrum ß-lactamase in Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 43:1281-1284.
- Naas, T., L. Poirel, A. Karim, and P. Nordmann. 1999. Molecular characterization of In50, a class 1 integron encoding the gene for the extended-spectrum beta-lactamase VEB-1 in Pseudomonas aeruginosa. FEMS Microbiol. Lett. 176:411-419.
- Nordmann, P., E. Ronco, T. Naas, C. Duport, Y. Michel-Briand, and R. Labia. 1993. Characterization of a novel extended-spectrum ß-lactamase from Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 37:962-969.
- Nordmann, P., and M. Guibert. 1998. Extended-spectrum ß-lactamases in Pseudomonas aeruginosa. J. Antimicrob. Chemother. 42:128-131.
- Pai, H., J. Kim, J. Kim, J. H. Lee, K. W. Choe, and N. Gotoh. 2001. Carbapenem resistance mechanisms in Pseudomonas aeruginosa clinical isolates. Antimicrob. Agents Chemother. 45:480-484.
- Péchère, J.-C., and T. Köhler. 1999. Patterns and modes of ß-lactam resistance in Pseudomonas aeruginosa. Clin. Microbiol. Infect. 5:S15-S18.
- Philippon, L. N., T. Naas, A.-T. Bouthors, V. Barakett, and P. Nordmann. 1997. OXA-18, a class D clavulanic acid-inhibited extended-spectrum ß-lactamase from Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 41:2188-2195.
- Poirel, L., E. Ronco, T. Naas, and P. Nordmann. 1999. Extended-spectrum ß-lactamase TEM-4 in Pseudomonas aeruginosa. Clin. Microbiol. Infect. 5:651-652.
- Poirel, L., D. Girlich, T. Naas, and P. Nordmann. 2001. OXA-28, an extended-spectrum variant of OXA-10 ß-lactamase from Pseudomonas aeruginosa and its plasmid- and integron-located gene. Antimicrob. Agents Chemother. 45:447-453.
- Vahaboglu, H., R. Öztürk, G. Aygün, F. Coskunkan, A. Yaman, A. Kaygusuz, H. Leblebicioglu, I. Balik, K. Aydin, and M. Otkun. 1997. Widespread detection of PER-1-type extended-spectrum ß-lactamases among nosocomial Acinetobacter and Pseudomonas aeruginosa isolates in Turkey: a nationwide multicenter study. Antimicrob. Agents Chemother. 41:2265-2269.
- Vahaboglu, H., R. Öztürk, H. Akbal, S. Saribas, O. Tansel, and F. Coskunkan. 1998. Practical approach for detection and identification of OXA-10-derived ceftazidime-hydrolyzing extended-spectrum ß-lactamases. Antimicrob. Agents Chemother. 36:827-829.
- Vercauteren, E., P. Descheemaeker, M. Leven, C. C. Sanders, and H. Goossens. 1997. Comparison of screening methods for detection of extended-spectrum ß-lactamases and their prevalence among blood isolates of Escherichia coli and Klebsiella spp. in a Belgian teaching hospital. J. Clin. Microbiol. 35:2191-2197.
- Yoneyama, H., and T. Nakae. 1993. Mechanism of efficient elimination of protein D2 in outer membrane of imipenem-resistant Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 37:2385-2390.
- Ziha-Zarifi, I., C. Llanes, T. Köhler, J.-C. Pechere, and P. Plesiat. 1999. In vivo emergence of multidrug-resistant mutants of Pseudomonas aeruginosa overexpressing the active efflux system MexA-MexB-OprM. Antimicrob. Agents Chemother. 43:287-291.
作者：
Frédéric Bert Zahia Ould-Hocine Manette Juvin 2007-5-10