Trends in Production of Extended-Spectrum -Lactamases among Enterobacteria of Medical Interest: Report of the Second Italian Nationwide Surv
http://www.100md.com
微生物临床杂志 2006年第5期
Laboratory of Microbiology, Ospedale di Circolo and University of Insubria, Varese
Department of Microbiological Sciences, University of Catania, Catania
Department of Molecular Biology, University of Siena, Siena
Department of Sciences and Biomedical Technologies, University of L'Aquila, L'Aquila, Italy
ABSTRACT
Results of a 2003 survey carried out in Italy to evaluate the prevalence of extended-spectrum -lactamase (ESBL)-producing enterobacteria are presented. Eleven Italian Microbiology Laboratories investigated 9,076 consecutive nonreplicate isolates (inpatients, 6,850; outpatients, 2,226). ESBL screening was performed by MIC data analysis. Confirmation was obtained using the double-disk synergy test and the combination disk test based on CLSI methodology. ESBL determinants were investigated by colony blot hybridization and confirmed by sequencing. Results were compared to those of the 1999 Italian survey (8,015 isolates). The prevalence of ESBL producers was 7.4% among isolates from inpatients (in 1999, 6.3%) and 3.5% among outpatients (no data were available for 1999). Among hospitalized patients, the most prevalent ESBL-positive species was Escherichia coli (Klebsiella pneumoniae in 1999). Proteus mirabilis was the most prevalent ESBL-positive species among outpatients. In both groups, most ESBL-positive pathogens were obtained from urinary tract infections. TEM-type ESBLs were the most prevalent enzymes (45.4%). Non-TEM, non-SHV determinants emerged: CTX-M-type in E. coli and K. pneumoniae, and PER-type in P. mirabilis, Providencia spp., and E. coli. With the exception of 3/163 P. mirabilis isolates and 1/44 Providencia stuartii isolate (all of which were intermediate for imipenem), carbapenems were active against all ESBL-positive enterobacteria. Susceptibility to other drugs was as follows: 84.7% for amikacin, 84.4% for piperacillin-tazobactam, 48.0% for gentamicin, and 32.8% for ciprofloxacin. Carbapenems appear to be the drug of choice. Amikacin and -lactam/-lactamase inhibitor combinations represent an alternative in non-life-threatening infections. The appearance of ESBL-positive enterobacteria in the community makes it mandatory that family physicians learn how to treat these pathogens.
INTRODUCTION
Extended-spectrum -lactamases (ESBLs) are enzymes capable of hydrolyzing a wide range of expanded-spectrum -lactams, including most recent cephalosporins, but which are inactive against cephamycins and carbapenems (15). To date, more than 150 different natural ESBL variants (which are detected most frequently in enterobacteria) are known, and they represent a worldwide problem in hospitalized patients (29). Infections caused by ESBL-positive organisms often involve compromised patients, making it difficult to eradicate these pathogens in high-risk wards such as intensive care units (1).
The overall prevalence of ESBL-positive enterobacteria varies greatly among different geographical areas, with the highest reported value (44.9%) in Latin America (32). According to published reports, in Europe, ESBLs appeared to be increasing among enterobacteria in the periods 1997 through 1999 to 2001 and 2002 (3, 21). However, the prevalence of ESBLs differs from country to country, with the highest percentages in Greece (27.4%) and Portugal (15.5%) and the lowest in The Netherlands and Germany (2.0 and 2.6%, respectively) (3). Klebsiella pneumoniae and Escherichia coli are the most common ESBL-positive species, but all enterobacteria can harbor plasmid-mediated ESBL genes (4). Recently, ESBL-positive isolates of Proteus mirabilis have been reported from France (7), different European countries (21), and New York City (26). In enterobacteria, classical ESBLs evolved from the TEM and SHV families (4). In recent years, several new ESBLs of non-TEM, non-SHV types emerged, such as enzymes of the CTX-M, PER, VEB, and GES lineages (13). Particularly, CTX-M-type enzymes increased in E. coli and K. pneumoniae isolates from Spain, the United Kingdom, and Russia (2, 8, 16).
In Italy, a nationwide survey carried out in 1999 among hospitalized patients (28) showed the following: (i) 6.3% of enterobacterial isolates were ESBL positive; (ii) ESBL determinants were more often detected in K. pneumoniae and P. mirabilis; and (iii) in the large majority of isolates (92.4%), enzymes of the TEM and/or the SHV family accounted for the ESBL phenotype.
Monitoring the ESBL prevalence and type in enterobacteria of clinical interest may contribute to delineating the breadth of the problem and to defining appropriate therapeutic options (29). The present nationwide survey aimed at evaluating the prevalences of different ESBL types and the associated resistance determinants in enterobacterial isolates obtained from both inpatients and outpatients. The goal was to analyze changes in ESBL types and antimicrobial susceptibilities as compared to the 1999 survey.
MATERIALS AND METHODS
Design of the study. Eleven clinical microbiology laboratories distributed throughout Italy were enrolled in the study. Hospitals of the following cities were represented: Bergamo, Milano, Novara, Varese, Verona, Firenze, Ancona, Napoli, Sassari, Catania, and Palermo. From September to December 2003 (4 months), up to 1,000 consecutive, nonreplicate enterobacterial isolates were collected at each laboratory (750 from inpatients and 250 from outpatients). Each isolate was identified to the species level. Susceptibilities to different -lactams were evaluated by automated methods using either the Vitek system (bioMerieux, Marcy l'Etoile, France) or the Phoenix system (Becton Dickinson Diagnostic Systems, Sparks, MD). Ampicillin-susceptible strains of E. coli, P. mirabilis, Salmonella enterica spp., and Shigella spp. were regarded as non-ESBL producing and were not further investigated. Ampicillin-resistant isolates of the above strains, as well as any other enterobacterial species, were suspected to produce ESBLs when showing a MIC of 2 μg/ml for ceftazidime and/or cefotaxime and/or ceftriaxone and/or aztreonam. The isolates were stored at –70°C at each laboratory and shipped to reference centers (Varese and Catania) for controlling species identification and defining the resistance phenotype. A case report form containing the laboratory code, patient's data, ward of admission, specimen source, and species identification was shipped with each suspected ESBL-positive isolate.
Identification and phenotypic characterization of ESBL-positive isolates. At reference centers, presumptive ESBL-positive isolates were reidentified to the species level by the API system (bioMerieux). Bacteriological media and susceptibility disks were obtained from Oxoid Limited, Basingstoke, United Kingdom. Confirmatory tests for ESBL production were performed according to species groups.
E. coli, Klebsiella spp., and P. mirabilis. Unless specified otherwise, 10-μg clavulanate disks were used throughout. (i) For quantitative assay, based on the methodology of the Clinical and Laboratory Standards Institute (CLSI) (formerly NCCLS) (20), susceptibilities to cefotaxime (30 μg) and ceftazidime (30 μg) alone and plus clavulanate were determined. As recommended, results were positive when the drug tested in combination with clavulanate produced a 5-mm increase in the zone diameter over its zone when tested alone. (ii) For qualitative assay, medium and inoculum were as described above. Disks containing aztreonam (30 μg), ceftriaxone (30 μg), cefotaxime (30 μg), and ceftazidime (30 μg) were placed around a disk containing amoxicillin (20 μg) plus clavulanate. Center-to-center distance for E. coli and Klebsiella spp. was 25 mm and for P. mirabilis, 30 mm. Results were interpreted according to the method of Jarlier (14).
Enterobacteria other than E. coli, Klebsiella spp., and P. mirabilis. (i) Quantitative assays were performed and interpreted as above. (ii) For qualitative assay, medium, inoculum, and disks were as described above. The only exception was that ceftriaxone was replaced with cefepime (30 μg), since this drug is more stable than other extended-spectrum cephalosporins to hydrolysis by AmpC -lactamases (15). Thus, cefepime was used for detecting synergistic activity in enterobacteria overproducing AmpC. The center-to-center distance was 25 mm. Enterobacter spp. isolates were tested at both 20 and 25 mm. Results were interpreted according to the method of Jarlier (14).
Two additional combination tests were used for both groups: cefpodoxime (10 μg) alone or with clavulanate (1 μg) and cefpirome (30 μg) alone or with clavulanate (10 μg). Cefpodoxime is regarded as a useful tool for analyzing suspected ESBL production (20), whereas cefpirome (as cefepime) is stable in the presence of AmpC enzymes (15).
Resistance phenotype of ESBL-positive isolates. Susceptibility of ESBL-positive isolates was evaluated by the disk diffusion technique according to CLSI guidelines (20). The following drugs were tested: amoxicillin plus clavulanate (20 plus 10 μg), ampicillin plus sulbactam (10 plus 10 μg), piperacillin plus tazobactam (100 plus 10 μg), cefoxitin (30 μg), imipenem (10 μg), meropenem (10 μg), amikacin (30 μg), gentamicin (10 μg), and ciprofloxacin (5 μg). Results were interpreted on the basis of CLSI criteria (20).
Molecular methods. Isolates confirmed as ESBL producing were investigated by colony blot hybridization, using random-primed 32P-labeled DNA probes for blaTEM-, blaSHV-, blaPER-, and blaCTX-M-type genes essentially as described previously (17, 22, 24). The ESBL nature of TEM- and SHV-type determinants was confirmed by sequencing as described previously (24).
RESULTS
Prevalence of ESBL-producing organisms. A total of 9,076 nonreplicate isolates (6,850 from hospitalized patients and 2,226 from outpatients) were studied during a 4-month period (September to December 2003). Overall, 504/6,850 (7.4%) and 79/2,226 (3.5%) isolates showed synergy between clavulanate and at least one -lactam. The prevalence of ESBL-positive isolates was different among participating centers (inpatients, range, 3.7 to 12.7%; outpatients, range, 0.9 to 7.3%).
Among inpatients (Table 1), E. coli (n = 161; 31.9%) was the most common ESBL-positive species, followed by P. mirabilis (n = 132; 26.2%), K. pneumoniae (n = 76; 15.1%), Enterobacter aerogenes (n = 38; 7.5%), and Providencia stuartii (n = 36; 7.1%). When data were expressed as ESBL prevalence within each species, P. stuartii and P. mirabilis appeared to harbor these enzymes at the highest frequency (36.7% and 25.7%, respectively). ESBL prevalence in E. coli was relatively low (4.4%). Among outpatients (Table 2), P. mirabilis (n = 31; 39.2%) and E. coli (n = 27; 34.2%) were the most frequent ESBL producers, followed by P. stuartii (n = 8; 10.1%) and K. pneumoniae (n = 5; 6.3%).
Patient population and source of specimens. Overall, 8,926 patients were studied (6,720 inpatients and 2,206 outpatients). The mean age was 58 years (median, 66; standard deviation, ±26). ESBL-positive enterobacteria were isolated from 558 patients. The mean age of patients infected by ESBL producers was 62 years (median, 71; standard deviation, ±28).
Isolates carrying ESBL determinants were obtained most frequently from urine samples in both hospital and community patients (65.8% and 86.1%, respectively). The remaining samples were obtained from respiratory tract infections (10.4%, inpatients; 8.9%, outpatients), surgical wounds (8.8%, inpatients; none, outpatients), bloodstream infections (6.5%, inpatients; none, outpatients). Among the hospitalized patients, ESBL-positive isolates were most frequently detected in medical (52.4%) and surgical (24.0%) wards but also from ICUs (16.5%), pediatrics (5.3%), and oncohematology wards (1.8%).
Distribution of -lactamase gene types in enterobacterial species. The presence of -lactamase genes of the TEM, SHV, CTX-M, or PER type was investigated in ESBL-producing isolates by colony blot hybridization. The ESBL nature of TEM and SHV determinants was confirmed by sequencing. Table 3 shows the distribution of different types of -lactamase genes in enterobacterial species. TEM-type enzymes were more prevalent than SHV-type enzymes (45.4% versus 21.3%, respectively). Eleven percent of ESBL-positive enterobacteria hybridized with both TEM and SHV probes. In these isolates, at least one of the two gene types accounted for the ESBL phenotype. Hybridization with the CTX-M- or PER-type probes occurred in a remarkable number of isolates (19.7% and 2.6%, respectively). Of interest is that CTX-M-type genes were frequently associated (89%) with TEM-type determinants in E. coli. This association was not detected in K. pneumoniae. PER-type genes were frequently associated with TEM-type determinants in P. mirabilis (100%) and Providencia spp. (78%).
As shown in Table 3, TEM-type ESBLs were particularly prevalent in P. mirabilis, P. stuartii, Morganella morganii, and Citrobacter koseri, whereas SHV-type enzymes (either alone or in association with TEM-type determinants) were widely distributed in Klebsiella and Enterobacter species. CTX-M-type enzymes were found most frequently in E. coli and K. pneumoniae. Notably, in E. coli these enzymes accounted for more than 50% of ESBL determinants. Finally, PER-type determinants were detected (though at a low rate) in P. mirabilis, Providencia spp., and E. coli.
In vitro susceptibility to clinically relevant drugs. Drugs potentially active against ESBL-positive Enterobacteriaceae include -lactam/-lactamase inhibitor combinations, cephamycins, carbapenems, aminoglycosides, and fluoroquinolones. As shown in Table 4, meropenem (100%) and imipenem (99.3%) were the most active drugs against ESBL-positive isolates. The only exception was represented by three P. mirabilis isolates and one P. stuartii isolate classified as intermediate to imipenem. Cefoxitin, which is not hydrolyzed by ESBLs, was active against 83.9% of isolates. High cefoxitin resistance was found in Enterobacter spp. (>85%), Serratia marcescens (100%), and Citrobacter freundii (100%), likely due to the expression of the chromosomal AmpC -lactamase. Among -lactam/-lactamase inhibitor combinations, ampicillin-sulbactam was the least effective (49.1%), whereas amoxicillin-clavulanate and piperacillin-tazobactam were more active (64.2% and 84.4%, respectively). P. mirabilis and P. stuartii isolates were highly susceptible to piperacillin-tazobactam (98.8 and 100%, respectively). Susceptibility to gentamicin was fairly low (48.0%), whereas 84.7% of ESBL-producing enterobacteria were susceptible to amikacin. Most of the amikacin-resistant isolates belonged to K. pneumoniae, E. aerogenes, or S. marcescens species. Resistance to fluoroquinolones was widely spread among ESBL-positive enterobacteria: overall, two-thirds of isolates were resistant to ciprofloxacin. It is noted that E. coli, P. mirabilis, and P. stuartii isolates were resistant to this drug in more than 70% of cases.
Association of drug resistance with -lactamase gene types. When isolates were grouped by gene type, interesting resistance patterns emerged. As shown in Table 5, TEM-positive enterobacteria were widely susceptible to ampicillin-sulbactam, amoxicillin-clavulanate, and piperacillin-tazobactam. In contrast, SHV-positive bacteria (carrying either SHV only or TEM plus SHV determinants) showed high rates of resistance to these combinations (up to 78.1%). Even in the latter group, piperacillin-tazobactam was the most effective drug. Enterobacteria carrying CTX-M- or PER-type determinants had similar phenotypes characterized by resistance to ampicillin-sulbactam and amoxicillin-clavulanate but remarkable susceptibility to piperacillin-tazobactam.
Consistent with the notion that ESBL determinants are often associated with other resistance traits (23), TEM- and PER-type determinants were associated in over 70% of cases with both fluoroquinolone and aminoglycoside resistance (data not shown). With regard to isolates carrying other ESBLs, those carrying SHV-type determinants alone or associated with TEM-type genes were frequently resistant to amikacin (Table 5). CTX-M-positive strains were frequently resistant to fluoroquinolones (67.8%) but susceptible to both amikacin and gentamicin in at least two-thirds of cases.
DISCUSSION
The increasing frequency of ESBL-producing enterobacteria among hospitalized patients is an important problem for both microbiologists and clinicians, because of the difficulty in correctly detecting, reporting, and treating infections caused by these organisms (23). In recent years, true community-acquired infections caused by ESBL-producing enterobacteria have also been described (25). This survey defines the current epidemiology of ESBLs among hospitalized and community patients in Italy. In addition, compared to the 1999 survey, the study provides evolutionary data regarding prevalence and distribution of ESBL-producing species, types of produced enzymes, and susceptibility to potentially active drugs.
In 2003, the overall prevalences of ESBL-producing enterobacteria among inpatients and outpatients were 7.4% and 3.5%, respectively. Thus, compared to the 1999 survey, data show a moderate increase of ESBL-producing organisms among hospitalized patients (28). The spread of ESBL-positive enterobacteria in the community presumably started after the 1999 survey, since at that time these pathogens were not detected.
Compared to data for 1999, the prevalence of ESBL-producing E. coli was increased among hospitalized patients. This is currently the most frequent ESBL-positive species, followed by P. mirabilis. On the other hand, ESBL-positive isolates of K. pneumoniae (the leading species in 1999) showed a significant decrease. Taken together, these findings agree with the current wide diffusion of ESBLs in medical wards and may reflect the positive impact of infection control interventions in high-risk wards. Concerning the community, E. coli and P. mirabilis appeared to be the most common ESBL-producing strains. This was not unexpected, since most isolates were obtained from urinary tract infections. Notably, P. stuartii isolates accounted for approximately 10% of ESBL producers in the community. ESBL-positive P. stuartii has been previously detected at high frequency in a large Italian university hospital (30). This species is also frequent among hospitalized patients enrolled in this survey (7.1%). Thus, ESBL-positive P. stuartii, an emerging problem in Italy, needs increased attention by clinical microbiologists.
Compared to data from the 1999 survey, the prevalence of isolates carrying TEM-type alone or both TEM- and SHV-type determinants was unchanged (45.4% versus 46.8% and 11.0% versus 11.0%, respectively). In contrast, a decrease of isolates carrying SHV-type determinants was observed (21.3% in 2003 versus 34.6% in 1999). This was likely due to the reduced isolation rate of K. pneumoniae. The most notable variation in 2003 was the marked increase of isolates harboring non-TEM, non-SHV determinants (22.3% versus 7.6% in 1999). This was mostly due to the emergence and spread of CTX-M-type enzymes in E. coli isolates. This finding is in agreement with recent European reports (5, 16) and confirms the current relevance of CTX-M enzymes in pathogenic enterobacteria. Clinical microbiology laboratories should take into account this changing epidemiology and adjust accordingly the screening and confirmatory procedures for ESBL production. The experience of this survey indicates that testing both ceftazidime and cefotaxime is an effective choice for ESBL screening. In fact, CTX-M-type enzymes may be undetected by screening with ceftazidime alone (2), whereas a few TEM-type ESBL enzymes can be lost by using cefotaxime alone (15). Use of both ceftazidime and cefotaxime (alone and in association with clavulanate) is suggested by CLSI for confirmatory tests (6), at least in the case of E. coli, Klebsiella spp., and P. mirabilis.
This survey also demonstrates that PER-type determinants, which previously were restricted to nonfermenting species such as Pseudomonas aeruginosa and Acinetobacter baumanii (31), are now harbored by P. mirabilis, Providencia spp., and E. coli. PER-positive isolates were found in 5/11 hospitals, thus revealing the spread of this determinant in Italy.
Therapy for infections caused by ESBL-producing enterobacteria is usually difficult, since these organisms not only are resistant to penicillins, cephalosporins, and the monobactam aztreonam but often are characterized by associated resistance to other classes of antimicrobials (27). ESBL-positive enterobacteria isolated in the 2003 survey remained fully susceptible to carbapenems, with only 3/163 P. mirabilis and 1/44 P. stuartii isolates showing an imipenem MIC of 8 μg/ml. On the contrary, our survey showed that susceptibility to ciprofloxacin had dramatically decreased, from 58% to 32% (1999 and 2003 data, respectively). A marked association between ESBL production and resistance to ciprofloxacin was observed, especially in E. coli, P. mirabilis, P. stuartii, and E. aerogenes. Notably, species that frequently harbored TEM-type determinants (i.e., P. mirabilis and P. stuartii) were characterized by resistance to both ciprofloxacin and gentamicin in more than 70% of cases. Overall, these data support the choice of carbapenems as empirical therapy in the case of life-threatening infections or nosocomial outbreaks (9, 10, 11). Use of fluoroquinolones may be justified only in selected cases of ESBL-related infections.
Excluding carbapenems, amikacin and piperacillin-tazobactam were the most effective drugs in vitro (overall susceptibility, 84.7% and 84.4%, respectively). In both cases, resistance was found in isolates of species carrying SHV-type determinants. Species harboring TEM-, CTX-M-, and PER-type determinants (e.g., P. mirabilis, P. stuartii, and E. coli) were consistently susceptible to amikacin and piperacillin-tazobactam. The latter three species represent two-thirds of ESBL producers and are frequently associated with urinary tract infections. Based on these data, piperacillin-tazobactam alone or together with amikacin would be a useful option for urinary tract infections caused by P. mirabilis, E. coli, or P. stuartii. As suggested previously (33), in the case of non-life-threatening infections and in nonoutbreak situations, it is not necessary to administer carbapenems. This approach is intended to preserve the therapeutic value of these precious drugs. The heavy use of carbapenems, in fact, may favor the selection of Stenotrophomonas maltophilia (a species naturally resistant to these drugs) and other nonfermenting gram-negative organisms resistant to carbapenems (17). It is notable that K. pneumoniae isolates producing carbapenem-hydrolyzing enzymes are emerging in the United States (34) and that different enterobacteria encoding metallo--lactamases have been recently detected in the Mediterranean area (12, 18, 19).
ACKNOWLEDGMENTS
The excellent contribution of many friends and colleagues is gratefully acknowledged: A. Goglio and F. Vailati, Ospedali Riuniti, Bergamo; E. Magliano and G. Ortisi, Ospedale Niguarda Ca' Granda, Milano; G. Fortina and G. L. Molinari, Ospedale Maggiore, Novara; R. Fontana and G. Lo Cascio, Universita di Verona; P. Varaldo and E. Manso, Universita di Ancona; P. Nicoletti and P. Pecile, Ospedale Careggi, Firenze; S. Esposito and S. Noviello, Universita di Napoli; E. Muresu, Universita di Sassari; G. Nicoletti, Universita di Catania; and A. Giammanco, Universita di Palermo, Italy.
This work was supported by grants from the Italian Ministry of Education, University and Scientific Research (MIUR) (Rome) and the Italian Ministry of Health (ISS) (Rome). We thank Wyeth-Lederle Italia for support.
REFERENCES
Babini, G., and D. M. Livermore. 2000. Antimicrobial resistance amongst Klebsiella spp. from intensive care units in southern and western Europe in 1997-1998. J. Antimicrob. Chemother. 45:183-189.
Bonnet, R. 2004. Growing group of extended-spectrum -lactamases: the CTX-M enzymes. Antimicrob. Agents Chemother. 48:1-14.
Bouchillon, S. K., B. M. Johnson, D. J. Hoban, J. L. Johnson, M. J. Dowzicky, D. H. Wu, M. A. Visalli, and P. A. Bradford. 2004. Determining incidence of extended-spectrum -lactamase producing Enterobacteriaceae, vancomycin-resistant Enterococcus faecium and methicillin-resistant Staphylococcus aureus in 38 centres from 17 countries: the PEARLS study 2001-2002. Int. J. Antimicrob. Agents 24:119-124.
Bradford, P. A. 2001. Extended-spectrum -lactamases in the 21st century: characterization, epidemiology, and detection of this important resistance threat. Clin. Microbiol. Rev. 14:933-951.
Brigante, G., F. Luzzaro, M. Perilli, G. Lombardi, A. Colì, G. M. Rossolini, G. Amicosante, and A. Toniolo. 2005. Evolution of CTX-M-type -lactamases in isolates of Escherichia coli infecting hospital and community patients. Int. J. Antimicrob. Agents 25:157-162.
Clinical and Laboratory Standards Institute. 2005. Performance standards for antimicrobial susceptibility testing. CLSI document M100-S15. Clinical and Laboratory Standards Institute, Wayne, Pa.
De Champs, C., R. Bonnet, D. Sirot, C. Chanal, and J. Sirot. 2000. Clinical relevance of Proteus mirabilis in hospital patients: a two-year survey. J. Antimicrob. Chemother. 45:537-539.
Edelstein, M., M. Pimkin, I. Palagin, I. Edelstein, and L. Stratchounski. 2003. Prevalence and molecular epidemiology of CTX-M extended-spectrum -lactamase-producing Escherichia coli and Klebsiella pneumoniae in Russian hospitals. Antimicrob. Agents Chemother. 47:3724-3732.
Endimiani, A., F. Luzzaro, M. Perilli, G. Lombardi, A. Colì, A. Tamborini, G. Amicosante, and A. Toniolo. 2004. Bacteremia due to Klebsiella pneumoniae isolates producing the TEM-52 extended-spectrum -lactamase: treatment outcome of patients receiving imipenem or ciprofloxacin. Clin. Infect. Dis. 38:243-251.
Endimiani, A., F. Luzzaro, G. Brigante, M. Perilli, G. Lombardi, G. Amicosante, G. M. Rossolini, and A. Toniolo. 2005. Proteus mirabilis bloodstream infections: risk factors and treatment outcome related to the expression of extended-spectrum -lactamases. Antimicrob. Agents Chemother. 49:2598-2605.
Giamarellou, H. 2005. Multidrug resistance in Gram-negative bacteria that produce extended-spectrum -lactamases (ESBLs). Clin. Microbiol. Infect. 11(Suppl. 4):1-16.
Ikonomidis, A., D. Tokatlidou, I. Kristo, D. Sofianou, A. Tsakris, P. Mantzana, S. Pournaras, and A. N. Maniatis. 2005. Outbreaks in distinct regions due to a single Klebsiella pneumoniae clone carrying a blaVIM-1 metallo--lactamase gene. J. Clin. Microbiol. 43:5344-5347.
Jacoby, G. A., and L. S. Munoz-Price. 2005. The new -lactamases. N. Engl. J. Med. 352:380-391.
Jarlier, V., M.-H. Nicolas, G. Fournier, and A. Philippon. 1988. Extended broad-spectrum -lactamases conferring transferable resistance to newer -lactam agents in Enterobacteriaceae: hospital prevalence and susceptibility patterns. Rev. Infect. Dis. 10:867-878.
Livermore, D. M. 1995. -Lactamases in laboratory and clinical resistance. Clin. Microbiol. Rev. 8:557-584.
Livermore, D. M., and P. M. Hawkey. 2005. CTX-M: changing the face of ESBLs in the UK. J. Antimicrob. Chemother. 56:451-454.
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 -lactamase. J. Clin. Microbiol. 39:1865-1870.
Luzzaro, F., J.-D. Docquier, C. Colinon, A. Endimiani, G. Lombardi, G. Amicosante, G. M. Rossolini, and A. Toniolo. 2004. Emergence in Klebsiella pneumoniae and Enterobacter cloacae clinical isolates of the VIM-4 metallo--lactamase encoded by a conjugative plasmid. Antimicrob. Agents Chemother. 48:648-650.
Miriagou, V., E. Tzelepi, D. Giannelli, and L. S. Tzouvelekis. 2003. Escherichia coli with a self-transferable, multiresistant plasmid coding for metallo--lactamase VIM-1. Antimicrob. Agents Chemother. 47:395-397.
National Committee for Clinical Laboratory Standards. 2003. Performance standards for antimicrobial susceptibility testing. NCCLS document M100-S13. National Committee for Clinical Laboratory Standards, Wayne, Pa.
Nijssen, S., A. Florijn, M. J. M. Bonten, F. J. Schmitz, J. Verhoef, and A. C. Fluit. 2004. Beta-lactam susceptibilities and prevalence of ESBL-producing isolates among more than 5000 European Enterobacteriaceae isolates. Int. J. Antimicrob. Agents. 24:585-591.
Pagani, L., E. Dell'Amico, R. Migliavacca, M. M. D'Andrea, E. Giacobone, G. Amicosante, E. Romero, and G. M. Rossolini. 2003. Multiple CTX-M-type extended-spectrum -lactamases in nosocomial isolates of Enterobacteriaceae from a hospital in northern Italy. J. Clin. Microbiol. 41:4264-4269.
Paterson, D. L., and R. A. Bonomo. 2005. Extended-spectrum -lactamases: a clinical update. Clin. Microbiol. Rev. 18:657-686.
Perilli, M., E. Dell'Amico, B. Segatore, M. R. de Massis, C. Bianchi, F. Luzzaro, G. M. Rossolini, A. Toniolo, G. Nicoletti, and G. Amicosante. 2002. Molecular characterization of extended-spectrum -lactamases produced by nosocomial isolates of Enterobacteriaceae from an Italian nationwide survey. J. Clin. Microbiol. 40:611-614.
Pitout, J. D. D., P. Nordmann, K. B. Laupland, and L. Poirel. 2005. Emergence of Enterobacteriaceae producing extended-spectrum -lactamases (ESBLs) in the community. J. Antimicrob. Chemother. 56:52-59.
Saurina, G., J. M. Quale, V. M. Manikal, E. Oydna, and D. Landman. 2000. Antimicrobial resistance in Enterobacteriaceae in Brooklyn, N.Y.: epidemiology and relation to antibiotic usage patterns. J. Antimicrob. Chemother. 45:895-898.
Schwaber, M. J., S. Navon-Venezia, D. Schwartz, and Y. Carmeli. 2005. High levels of antimicrobial coresistance among extended-spectrum--lactamase-producing Enterobacteriaceae. Antimicrob. Agents Chemother. 49:2137-2139.
Spanu, T., F. Luzzaro, M. Perilli, G. Amicosante, A. Toniolo, G. Fadda, and the Italian ESBL Study Group. 2002. Occurrence of extended-spectrum -lactamases in members of the family Enterobacteriaceae in Italy: implications for resistance to -lactams and other antimicrobial drugs. Antimicrob. Agents Chemother. 46:196-202.
Stürenburg, E., and D. Mack. 2003. Extended-spectrum -lactamases: implications for the clinical microbiology laboratory, therapy, and infection control. J. Infect. 47:273-295.
Tumbarello, M., R. Citton, T. Spanu, M. Sanguinetti, L. Romano, G. Fadda, and R. Cauda. 2004. ESBL-producing multidrug-resistant Providencia stuartii infections in a university hospital. J. Antimicrob. Chemother. 53:277-282.
Vahaboglü, H., R. Ozturk, G. Aygun, F. Coskunkan, A. Yaman, A. Kaygusuz, H. Leblebicioglu, I. Balik, K. Aydin, and M. Otkun. 1997. Widespread detection of PER-1-type extended-spectrum -lactamases among nosocomial Acinetobacter and Pseudomonas aeruginosa isolates in Turkey; a nationwide multicenter study. Antimicrob. Agents Chemother. 41:2265-2269.
Winokur, P. L., R. Canton, J. M. Casellas, and N. Legakis. 2001. Variations in the prevalence of strains expressing beta-lactamase phenotype and characterization of isolates from Europe, the Americas, and the Western Pacific region. Clin. Infect. Dis. 15(Suppl. 2):94-103.
Wong-Beringer, A., J. Hindler, M. Loeloff, A. M. Queenan, N. Lee, D. A. Pegues, J. P. Quinn, and K. Bush. 2002. Molecular correlation for the treatment outcomes in bloodstream infections caused by Escherichia coli and Klebsiella pneumoniae with reduced susceptibility to ceftazidime. Clin. Infect. Dis. 34:135-146.
Yigit, H., A. M. Queenan, G. J. Anderson, A. Domenech-Sanchez, J. W. Biddle, C. D. Steward, S. Alberti, K. Bush, and F. C. Tenover. 2001. Novel carbapenem-hydrolyzing -lactamase, KPC-1, from a carbapenem-resistant strain of Klebsiella pneumoniae. Antimicrob. Agents Chemother. 45:1151-1161.(Francesco Luzzaro, Marili)
Department of Microbiological Sciences, University of Catania, Catania
Department of Molecular Biology, University of Siena, Siena
Department of Sciences and Biomedical Technologies, University of L'Aquila, L'Aquila, Italy
ABSTRACT
Results of a 2003 survey carried out in Italy to evaluate the prevalence of extended-spectrum -lactamase (ESBL)-producing enterobacteria are presented. Eleven Italian Microbiology Laboratories investigated 9,076 consecutive nonreplicate isolates (inpatients, 6,850; outpatients, 2,226). ESBL screening was performed by MIC data analysis. Confirmation was obtained using the double-disk synergy test and the combination disk test based on CLSI methodology. ESBL determinants were investigated by colony blot hybridization and confirmed by sequencing. Results were compared to those of the 1999 Italian survey (8,015 isolates). The prevalence of ESBL producers was 7.4% among isolates from inpatients (in 1999, 6.3%) and 3.5% among outpatients (no data were available for 1999). Among hospitalized patients, the most prevalent ESBL-positive species was Escherichia coli (Klebsiella pneumoniae in 1999). Proteus mirabilis was the most prevalent ESBL-positive species among outpatients. In both groups, most ESBL-positive pathogens were obtained from urinary tract infections. TEM-type ESBLs were the most prevalent enzymes (45.4%). Non-TEM, non-SHV determinants emerged: CTX-M-type in E. coli and K. pneumoniae, and PER-type in P. mirabilis, Providencia spp., and E. coli. With the exception of 3/163 P. mirabilis isolates and 1/44 Providencia stuartii isolate (all of which were intermediate for imipenem), carbapenems were active against all ESBL-positive enterobacteria. Susceptibility to other drugs was as follows: 84.7% for amikacin, 84.4% for piperacillin-tazobactam, 48.0% for gentamicin, and 32.8% for ciprofloxacin. Carbapenems appear to be the drug of choice. Amikacin and -lactam/-lactamase inhibitor combinations represent an alternative in non-life-threatening infections. The appearance of ESBL-positive enterobacteria in the community makes it mandatory that family physicians learn how to treat these pathogens.
INTRODUCTION
Extended-spectrum -lactamases (ESBLs) are enzymes capable of hydrolyzing a wide range of expanded-spectrum -lactams, including most recent cephalosporins, but which are inactive against cephamycins and carbapenems (15). To date, more than 150 different natural ESBL variants (which are detected most frequently in enterobacteria) are known, and they represent a worldwide problem in hospitalized patients (29). Infections caused by ESBL-positive organisms often involve compromised patients, making it difficult to eradicate these pathogens in high-risk wards such as intensive care units (1).
The overall prevalence of ESBL-positive enterobacteria varies greatly among different geographical areas, with the highest reported value (44.9%) in Latin America (32). According to published reports, in Europe, ESBLs appeared to be increasing among enterobacteria in the periods 1997 through 1999 to 2001 and 2002 (3, 21). However, the prevalence of ESBLs differs from country to country, with the highest percentages in Greece (27.4%) and Portugal (15.5%) and the lowest in The Netherlands and Germany (2.0 and 2.6%, respectively) (3). Klebsiella pneumoniae and Escherichia coli are the most common ESBL-positive species, but all enterobacteria can harbor plasmid-mediated ESBL genes (4). Recently, ESBL-positive isolates of Proteus mirabilis have been reported from France (7), different European countries (21), and New York City (26). In enterobacteria, classical ESBLs evolved from the TEM and SHV families (4). In recent years, several new ESBLs of non-TEM, non-SHV types emerged, such as enzymes of the CTX-M, PER, VEB, and GES lineages (13). Particularly, CTX-M-type enzymes increased in E. coli and K. pneumoniae isolates from Spain, the United Kingdom, and Russia (2, 8, 16).
In Italy, a nationwide survey carried out in 1999 among hospitalized patients (28) showed the following: (i) 6.3% of enterobacterial isolates were ESBL positive; (ii) ESBL determinants were more often detected in K. pneumoniae and P. mirabilis; and (iii) in the large majority of isolates (92.4%), enzymes of the TEM and/or the SHV family accounted for the ESBL phenotype.
Monitoring the ESBL prevalence and type in enterobacteria of clinical interest may contribute to delineating the breadth of the problem and to defining appropriate therapeutic options (29). The present nationwide survey aimed at evaluating the prevalences of different ESBL types and the associated resistance determinants in enterobacterial isolates obtained from both inpatients and outpatients. The goal was to analyze changes in ESBL types and antimicrobial susceptibilities as compared to the 1999 survey.
MATERIALS AND METHODS
Design of the study. Eleven clinical microbiology laboratories distributed throughout Italy were enrolled in the study. Hospitals of the following cities were represented: Bergamo, Milano, Novara, Varese, Verona, Firenze, Ancona, Napoli, Sassari, Catania, and Palermo. From September to December 2003 (4 months), up to 1,000 consecutive, nonreplicate enterobacterial isolates were collected at each laboratory (750 from inpatients and 250 from outpatients). Each isolate was identified to the species level. Susceptibilities to different -lactams were evaluated by automated methods using either the Vitek system (bioMerieux, Marcy l'Etoile, France) or the Phoenix system (Becton Dickinson Diagnostic Systems, Sparks, MD). Ampicillin-susceptible strains of E. coli, P. mirabilis, Salmonella enterica spp., and Shigella spp. were regarded as non-ESBL producing and were not further investigated. Ampicillin-resistant isolates of the above strains, as well as any other enterobacterial species, were suspected to produce ESBLs when showing a MIC of 2 μg/ml for ceftazidime and/or cefotaxime and/or ceftriaxone and/or aztreonam. The isolates were stored at –70°C at each laboratory and shipped to reference centers (Varese and Catania) for controlling species identification and defining the resistance phenotype. A case report form containing the laboratory code, patient's data, ward of admission, specimen source, and species identification was shipped with each suspected ESBL-positive isolate.
Identification and phenotypic characterization of ESBL-positive isolates. At reference centers, presumptive ESBL-positive isolates were reidentified to the species level by the API system (bioMerieux). Bacteriological media and susceptibility disks were obtained from Oxoid Limited, Basingstoke, United Kingdom. Confirmatory tests for ESBL production were performed according to species groups.
E. coli, Klebsiella spp., and P. mirabilis. Unless specified otherwise, 10-μg clavulanate disks were used throughout. (i) For quantitative assay, based on the methodology of the Clinical and Laboratory Standards Institute (CLSI) (formerly NCCLS) (20), susceptibilities to cefotaxime (30 μg) and ceftazidime (30 μg) alone and plus clavulanate were determined. As recommended, results were positive when the drug tested in combination with clavulanate produced a 5-mm increase in the zone diameter over its zone when tested alone. (ii) For qualitative assay, medium and inoculum were as described above. Disks containing aztreonam (30 μg), ceftriaxone (30 μg), cefotaxime (30 μg), and ceftazidime (30 μg) were placed around a disk containing amoxicillin (20 μg) plus clavulanate. Center-to-center distance for E. coli and Klebsiella spp. was 25 mm and for P. mirabilis, 30 mm. Results were interpreted according to the method of Jarlier (14).
Enterobacteria other than E. coli, Klebsiella spp., and P. mirabilis. (i) Quantitative assays were performed and interpreted as above. (ii) For qualitative assay, medium, inoculum, and disks were as described above. The only exception was that ceftriaxone was replaced with cefepime (30 μg), since this drug is more stable than other extended-spectrum cephalosporins to hydrolysis by AmpC -lactamases (15). Thus, cefepime was used for detecting synergistic activity in enterobacteria overproducing AmpC. The center-to-center distance was 25 mm. Enterobacter spp. isolates were tested at both 20 and 25 mm. Results were interpreted according to the method of Jarlier (14).
Two additional combination tests were used for both groups: cefpodoxime (10 μg) alone or with clavulanate (1 μg) and cefpirome (30 μg) alone or with clavulanate (10 μg). Cefpodoxime is regarded as a useful tool for analyzing suspected ESBL production (20), whereas cefpirome (as cefepime) is stable in the presence of AmpC enzymes (15).
Resistance phenotype of ESBL-positive isolates. Susceptibility of ESBL-positive isolates was evaluated by the disk diffusion technique according to CLSI guidelines (20). The following drugs were tested: amoxicillin plus clavulanate (20 plus 10 μg), ampicillin plus sulbactam (10 plus 10 μg), piperacillin plus tazobactam (100 plus 10 μg), cefoxitin (30 μg), imipenem (10 μg), meropenem (10 μg), amikacin (30 μg), gentamicin (10 μg), and ciprofloxacin (5 μg). Results were interpreted on the basis of CLSI criteria (20).
Molecular methods. Isolates confirmed as ESBL producing were investigated by colony blot hybridization, using random-primed 32P-labeled DNA probes for blaTEM-, blaSHV-, blaPER-, and blaCTX-M-type genes essentially as described previously (17, 22, 24). The ESBL nature of TEM- and SHV-type determinants was confirmed by sequencing as described previously (24).
RESULTS
Prevalence of ESBL-producing organisms. A total of 9,076 nonreplicate isolates (6,850 from hospitalized patients and 2,226 from outpatients) were studied during a 4-month period (September to December 2003). Overall, 504/6,850 (7.4%) and 79/2,226 (3.5%) isolates showed synergy between clavulanate and at least one -lactam. The prevalence of ESBL-positive isolates was different among participating centers (inpatients, range, 3.7 to 12.7%; outpatients, range, 0.9 to 7.3%).
Among inpatients (Table 1), E. coli (n = 161; 31.9%) was the most common ESBL-positive species, followed by P. mirabilis (n = 132; 26.2%), K. pneumoniae (n = 76; 15.1%), Enterobacter aerogenes (n = 38; 7.5%), and Providencia stuartii (n = 36; 7.1%). When data were expressed as ESBL prevalence within each species, P. stuartii and P. mirabilis appeared to harbor these enzymes at the highest frequency (36.7% and 25.7%, respectively). ESBL prevalence in E. coli was relatively low (4.4%). Among outpatients (Table 2), P. mirabilis (n = 31; 39.2%) and E. coli (n = 27; 34.2%) were the most frequent ESBL producers, followed by P. stuartii (n = 8; 10.1%) and K. pneumoniae (n = 5; 6.3%).
Patient population and source of specimens. Overall, 8,926 patients were studied (6,720 inpatients and 2,206 outpatients). The mean age was 58 years (median, 66; standard deviation, ±26). ESBL-positive enterobacteria were isolated from 558 patients. The mean age of patients infected by ESBL producers was 62 years (median, 71; standard deviation, ±28).
Isolates carrying ESBL determinants were obtained most frequently from urine samples in both hospital and community patients (65.8% and 86.1%, respectively). The remaining samples were obtained from respiratory tract infections (10.4%, inpatients; 8.9%, outpatients), surgical wounds (8.8%, inpatients; none, outpatients), bloodstream infections (6.5%, inpatients; none, outpatients). Among the hospitalized patients, ESBL-positive isolates were most frequently detected in medical (52.4%) and surgical (24.0%) wards but also from ICUs (16.5%), pediatrics (5.3%), and oncohematology wards (1.8%).
Distribution of -lactamase gene types in enterobacterial species. The presence of -lactamase genes of the TEM, SHV, CTX-M, or PER type was investigated in ESBL-producing isolates by colony blot hybridization. The ESBL nature of TEM and SHV determinants was confirmed by sequencing. Table 3 shows the distribution of different types of -lactamase genes in enterobacterial species. TEM-type enzymes were more prevalent than SHV-type enzymes (45.4% versus 21.3%, respectively). Eleven percent of ESBL-positive enterobacteria hybridized with both TEM and SHV probes. In these isolates, at least one of the two gene types accounted for the ESBL phenotype. Hybridization with the CTX-M- or PER-type probes occurred in a remarkable number of isolates (19.7% and 2.6%, respectively). Of interest is that CTX-M-type genes were frequently associated (89%) with TEM-type determinants in E. coli. This association was not detected in K. pneumoniae. PER-type genes were frequently associated with TEM-type determinants in P. mirabilis (100%) and Providencia spp. (78%).
As shown in Table 3, TEM-type ESBLs were particularly prevalent in P. mirabilis, P. stuartii, Morganella morganii, and Citrobacter koseri, whereas SHV-type enzymes (either alone or in association with TEM-type determinants) were widely distributed in Klebsiella and Enterobacter species. CTX-M-type enzymes were found most frequently in E. coli and K. pneumoniae. Notably, in E. coli these enzymes accounted for more than 50% of ESBL determinants. Finally, PER-type determinants were detected (though at a low rate) in P. mirabilis, Providencia spp., and E. coli.
In vitro susceptibility to clinically relevant drugs. Drugs potentially active against ESBL-positive Enterobacteriaceae include -lactam/-lactamase inhibitor combinations, cephamycins, carbapenems, aminoglycosides, and fluoroquinolones. As shown in Table 4, meropenem (100%) and imipenem (99.3%) were the most active drugs against ESBL-positive isolates. The only exception was represented by three P. mirabilis isolates and one P. stuartii isolate classified as intermediate to imipenem. Cefoxitin, which is not hydrolyzed by ESBLs, was active against 83.9% of isolates. High cefoxitin resistance was found in Enterobacter spp. (>85%), Serratia marcescens (100%), and Citrobacter freundii (100%), likely due to the expression of the chromosomal AmpC -lactamase. Among -lactam/-lactamase inhibitor combinations, ampicillin-sulbactam was the least effective (49.1%), whereas amoxicillin-clavulanate and piperacillin-tazobactam were more active (64.2% and 84.4%, respectively). P. mirabilis and P. stuartii isolates were highly susceptible to piperacillin-tazobactam (98.8 and 100%, respectively). Susceptibility to gentamicin was fairly low (48.0%), whereas 84.7% of ESBL-producing enterobacteria were susceptible to amikacin. Most of the amikacin-resistant isolates belonged to K. pneumoniae, E. aerogenes, or S. marcescens species. Resistance to fluoroquinolones was widely spread among ESBL-positive enterobacteria: overall, two-thirds of isolates were resistant to ciprofloxacin. It is noted that E. coli, P. mirabilis, and P. stuartii isolates were resistant to this drug in more than 70% of cases.
Association of drug resistance with -lactamase gene types. When isolates were grouped by gene type, interesting resistance patterns emerged. As shown in Table 5, TEM-positive enterobacteria were widely susceptible to ampicillin-sulbactam, amoxicillin-clavulanate, and piperacillin-tazobactam. In contrast, SHV-positive bacteria (carrying either SHV only or TEM plus SHV determinants) showed high rates of resistance to these combinations (up to 78.1%). Even in the latter group, piperacillin-tazobactam was the most effective drug. Enterobacteria carrying CTX-M- or PER-type determinants had similar phenotypes characterized by resistance to ampicillin-sulbactam and amoxicillin-clavulanate but remarkable susceptibility to piperacillin-tazobactam.
Consistent with the notion that ESBL determinants are often associated with other resistance traits (23), TEM- and PER-type determinants were associated in over 70% of cases with both fluoroquinolone and aminoglycoside resistance (data not shown). With regard to isolates carrying other ESBLs, those carrying SHV-type determinants alone or associated with TEM-type genes were frequently resistant to amikacin (Table 5). CTX-M-positive strains were frequently resistant to fluoroquinolones (67.8%) but susceptible to both amikacin and gentamicin in at least two-thirds of cases.
DISCUSSION
The increasing frequency of ESBL-producing enterobacteria among hospitalized patients is an important problem for both microbiologists and clinicians, because of the difficulty in correctly detecting, reporting, and treating infections caused by these organisms (23). In recent years, true community-acquired infections caused by ESBL-producing enterobacteria have also been described (25). This survey defines the current epidemiology of ESBLs among hospitalized and community patients in Italy. In addition, compared to the 1999 survey, the study provides evolutionary data regarding prevalence and distribution of ESBL-producing species, types of produced enzymes, and susceptibility to potentially active drugs.
In 2003, the overall prevalences of ESBL-producing enterobacteria among inpatients and outpatients were 7.4% and 3.5%, respectively. Thus, compared to the 1999 survey, data show a moderate increase of ESBL-producing organisms among hospitalized patients (28). The spread of ESBL-positive enterobacteria in the community presumably started after the 1999 survey, since at that time these pathogens were not detected.
Compared to data for 1999, the prevalence of ESBL-producing E. coli was increased among hospitalized patients. This is currently the most frequent ESBL-positive species, followed by P. mirabilis. On the other hand, ESBL-positive isolates of K. pneumoniae (the leading species in 1999) showed a significant decrease. Taken together, these findings agree with the current wide diffusion of ESBLs in medical wards and may reflect the positive impact of infection control interventions in high-risk wards. Concerning the community, E. coli and P. mirabilis appeared to be the most common ESBL-producing strains. This was not unexpected, since most isolates were obtained from urinary tract infections. Notably, P. stuartii isolates accounted for approximately 10% of ESBL producers in the community. ESBL-positive P. stuartii has been previously detected at high frequency in a large Italian university hospital (30). This species is also frequent among hospitalized patients enrolled in this survey (7.1%). Thus, ESBL-positive P. stuartii, an emerging problem in Italy, needs increased attention by clinical microbiologists.
Compared to data from the 1999 survey, the prevalence of isolates carrying TEM-type alone or both TEM- and SHV-type determinants was unchanged (45.4% versus 46.8% and 11.0% versus 11.0%, respectively). In contrast, a decrease of isolates carrying SHV-type determinants was observed (21.3% in 2003 versus 34.6% in 1999). This was likely due to the reduced isolation rate of K. pneumoniae. The most notable variation in 2003 was the marked increase of isolates harboring non-TEM, non-SHV determinants (22.3% versus 7.6% in 1999). This was mostly due to the emergence and spread of CTX-M-type enzymes in E. coli isolates. This finding is in agreement with recent European reports (5, 16) and confirms the current relevance of CTX-M enzymes in pathogenic enterobacteria. Clinical microbiology laboratories should take into account this changing epidemiology and adjust accordingly the screening and confirmatory procedures for ESBL production. The experience of this survey indicates that testing both ceftazidime and cefotaxime is an effective choice for ESBL screening. In fact, CTX-M-type enzymes may be undetected by screening with ceftazidime alone (2), whereas a few TEM-type ESBL enzymes can be lost by using cefotaxime alone (15). Use of both ceftazidime and cefotaxime (alone and in association with clavulanate) is suggested by CLSI for confirmatory tests (6), at least in the case of E. coli, Klebsiella spp., and P. mirabilis.
This survey also demonstrates that PER-type determinants, which previously were restricted to nonfermenting species such as Pseudomonas aeruginosa and Acinetobacter baumanii (31), are now harbored by P. mirabilis, Providencia spp., and E. coli. PER-positive isolates were found in 5/11 hospitals, thus revealing the spread of this determinant in Italy.
Therapy for infections caused by ESBL-producing enterobacteria is usually difficult, since these organisms not only are resistant to penicillins, cephalosporins, and the monobactam aztreonam but often are characterized by associated resistance to other classes of antimicrobials (27). ESBL-positive enterobacteria isolated in the 2003 survey remained fully susceptible to carbapenems, with only 3/163 P. mirabilis and 1/44 P. stuartii isolates showing an imipenem MIC of 8 μg/ml. On the contrary, our survey showed that susceptibility to ciprofloxacin had dramatically decreased, from 58% to 32% (1999 and 2003 data, respectively). A marked association between ESBL production and resistance to ciprofloxacin was observed, especially in E. coli, P. mirabilis, P. stuartii, and E. aerogenes. Notably, species that frequently harbored TEM-type determinants (i.e., P. mirabilis and P. stuartii) were characterized by resistance to both ciprofloxacin and gentamicin in more than 70% of cases. Overall, these data support the choice of carbapenems as empirical therapy in the case of life-threatening infections or nosocomial outbreaks (9, 10, 11). Use of fluoroquinolones may be justified only in selected cases of ESBL-related infections.
Excluding carbapenems, amikacin and piperacillin-tazobactam were the most effective drugs in vitro (overall susceptibility, 84.7% and 84.4%, respectively). In both cases, resistance was found in isolates of species carrying SHV-type determinants. Species harboring TEM-, CTX-M-, and PER-type determinants (e.g., P. mirabilis, P. stuartii, and E. coli) were consistently susceptible to amikacin and piperacillin-tazobactam. The latter three species represent two-thirds of ESBL producers and are frequently associated with urinary tract infections. Based on these data, piperacillin-tazobactam alone or together with amikacin would be a useful option for urinary tract infections caused by P. mirabilis, E. coli, or P. stuartii. As suggested previously (33), in the case of non-life-threatening infections and in nonoutbreak situations, it is not necessary to administer carbapenems. This approach is intended to preserve the therapeutic value of these precious drugs. The heavy use of carbapenems, in fact, may favor the selection of Stenotrophomonas maltophilia (a species naturally resistant to these drugs) and other nonfermenting gram-negative organisms resistant to carbapenems (17). It is notable that K. pneumoniae isolates producing carbapenem-hydrolyzing enzymes are emerging in the United States (34) and that different enterobacteria encoding metallo--lactamases have been recently detected in the Mediterranean area (12, 18, 19).
ACKNOWLEDGMENTS
The excellent contribution of many friends and colleagues is gratefully acknowledged: A. Goglio and F. Vailati, Ospedali Riuniti, Bergamo; E. Magliano and G. Ortisi, Ospedale Niguarda Ca' Granda, Milano; G. Fortina and G. L. Molinari, Ospedale Maggiore, Novara; R. Fontana and G. Lo Cascio, Universita di Verona; P. Varaldo and E. Manso, Universita di Ancona; P. Nicoletti and P. Pecile, Ospedale Careggi, Firenze; S. Esposito and S. Noviello, Universita di Napoli; E. Muresu, Universita di Sassari; G. Nicoletti, Universita di Catania; and A. Giammanco, Universita di Palermo, Italy.
This work was supported by grants from the Italian Ministry of Education, University and Scientific Research (MIUR) (Rome) and the Italian Ministry of Health (ISS) (Rome). We thank Wyeth-Lederle Italia for support.
REFERENCES
Babini, G., and D. M. Livermore. 2000. Antimicrobial resistance amongst Klebsiella spp. from intensive care units in southern and western Europe in 1997-1998. J. Antimicrob. Chemother. 45:183-189.
Bonnet, R. 2004. Growing group of extended-spectrum -lactamases: the CTX-M enzymes. Antimicrob. Agents Chemother. 48:1-14.
Bouchillon, S. K., B. M. Johnson, D. J. Hoban, J. L. Johnson, M. J. Dowzicky, D. H. Wu, M. A. Visalli, and P. A. Bradford. 2004. Determining incidence of extended-spectrum -lactamase producing Enterobacteriaceae, vancomycin-resistant Enterococcus faecium and methicillin-resistant Staphylococcus aureus in 38 centres from 17 countries: the PEARLS study 2001-2002. Int. J. Antimicrob. Agents 24:119-124.
Bradford, P. A. 2001. Extended-spectrum -lactamases in the 21st century: characterization, epidemiology, and detection of this important resistance threat. Clin. Microbiol. Rev. 14:933-951.
Brigante, G., F. Luzzaro, M. Perilli, G. Lombardi, A. Colì, G. M. Rossolini, G. Amicosante, and A. Toniolo. 2005. Evolution of CTX-M-type -lactamases in isolates of Escherichia coli infecting hospital and community patients. Int. J. Antimicrob. Agents 25:157-162.
Clinical and Laboratory Standards Institute. 2005. Performance standards for antimicrobial susceptibility testing. CLSI document M100-S15. Clinical and Laboratory Standards Institute, Wayne, Pa.
De Champs, C., R. Bonnet, D. Sirot, C. Chanal, and J. Sirot. 2000. Clinical relevance of Proteus mirabilis in hospital patients: a two-year survey. J. Antimicrob. Chemother. 45:537-539.
Edelstein, M., M. Pimkin, I. Palagin, I. Edelstein, and L. Stratchounski. 2003. Prevalence and molecular epidemiology of CTX-M extended-spectrum -lactamase-producing Escherichia coli and Klebsiella pneumoniae in Russian hospitals. Antimicrob. Agents Chemother. 47:3724-3732.
Endimiani, A., F. Luzzaro, M. Perilli, G. Lombardi, A. Colì, A. Tamborini, G. Amicosante, and A. Toniolo. 2004. Bacteremia due to Klebsiella pneumoniae isolates producing the TEM-52 extended-spectrum -lactamase: treatment outcome of patients receiving imipenem or ciprofloxacin. Clin. Infect. Dis. 38:243-251.
Endimiani, A., F. Luzzaro, G. Brigante, M. Perilli, G. Lombardi, G. Amicosante, G. M. Rossolini, and A. Toniolo. 2005. Proteus mirabilis bloodstream infections: risk factors and treatment outcome related to the expression of extended-spectrum -lactamases. Antimicrob. Agents Chemother. 49:2598-2605.
Giamarellou, H. 2005. Multidrug resistance in Gram-negative bacteria that produce extended-spectrum -lactamases (ESBLs). Clin. Microbiol. Infect. 11(Suppl. 4):1-16.
Ikonomidis, A., D. Tokatlidou, I. Kristo, D. Sofianou, A. Tsakris, P. Mantzana, S. Pournaras, and A. N. Maniatis. 2005. Outbreaks in distinct regions due to a single Klebsiella pneumoniae clone carrying a blaVIM-1 metallo--lactamase gene. J. Clin. Microbiol. 43:5344-5347.
Jacoby, G. A., and L. S. Munoz-Price. 2005. The new -lactamases. N. Engl. J. Med. 352:380-391.
Jarlier, V., M.-H. Nicolas, G. Fournier, and A. Philippon. 1988. Extended broad-spectrum -lactamases conferring transferable resistance to newer -lactam agents in Enterobacteriaceae: hospital prevalence and susceptibility patterns. Rev. Infect. Dis. 10:867-878.
Livermore, D. M. 1995. -Lactamases in laboratory and clinical resistance. Clin. Microbiol. Rev. 8:557-584.
Livermore, D. M., and P. M. Hawkey. 2005. CTX-M: changing the face of ESBLs in the UK. J. Antimicrob. Chemother. 56:451-454.
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 -lactamase. J. Clin. Microbiol. 39:1865-1870.
Luzzaro, F., J.-D. Docquier, C. Colinon, A. Endimiani, G. Lombardi, G. Amicosante, G. M. Rossolini, and A. Toniolo. 2004. Emergence in Klebsiella pneumoniae and Enterobacter cloacae clinical isolates of the VIM-4 metallo--lactamase encoded by a conjugative plasmid. Antimicrob. Agents Chemother. 48:648-650.
Miriagou, V., E. Tzelepi, D. Giannelli, and L. S. Tzouvelekis. 2003. Escherichia coli with a self-transferable, multiresistant plasmid coding for metallo--lactamase VIM-1. Antimicrob. Agents Chemother. 47:395-397.
National Committee for Clinical Laboratory Standards. 2003. Performance standards for antimicrobial susceptibility testing. NCCLS document M100-S13. National Committee for Clinical Laboratory Standards, Wayne, Pa.
Nijssen, S., A. Florijn, M. J. M. Bonten, F. J. Schmitz, J. Verhoef, and A. C. Fluit. 2004. Beta-lactam susceptibilities and prevalence of ESBL-producing isolates among more than 5000 European Enterobacteriaceae isolates. Int. J. Antimicrob. Agents. 24:585-591.
Pagani, L., E. Dell'Amico, R. Migliavacca, M. M. D'Andrea, E. Giacobone, G. Amicosante, E. Romero, and G. M. Rossolini. 2003. Multiple CTX-M-type extended-spectrum -lactamases in nosocomial isolates of Enterobacteriaceae from a hospital in northern Italy. J. Clin. Microbiol. 41:4264-4269.
Paterson, D. L., and R. A. Bonomo. 2005. Extended-spectrum -lactamases: a clinical update. Clin. Microbiol. Rev. 18:657-686.
Perilli, M., E. Dell'Amico, B. Segatore, M. R. de Massis, C. Bianchi, F. Luzzaro, G. M. Rossolini, A. Toniolo, G. Nicoletti, and G. Amicosante. 2002. Molecular characterization of extended-spectrum -lactamases produced by nosocomial isolates of Enterobacteriaceae from an Italian nationwide survey. J. Clin. Microbiol. 40:611-614.
Pitout, J. D. D., P. Nordmann, K. B. Laupland, and L. Poirel. 2005. Emergence of Enterobacteriaceae producing extended-spectrum -lactamases (ESBLs) in the community. J. Antimicrob. Chemother. 56:52-59.
Saurina, G., J. M. Quale, V. M. Manikal, E. Oydna, and D. Landman. 2000. Antimicrobial resistance in Enterobacteriaceae in Brooklyn, N.Y.: epidemiology and relation to antibiotic usage patterns. J. Antimicrob. Chemother. 45:895-898.
Schwaber, M. J., S. Navon-Venezia, D. Schwartz, and Y. Carmeli. 2005. High levels of antimicrobial coresistance among extended-spectrum--lactamase-producing Enterobacteriaceae. Antimicrob. Agents Chemother. 49:2137-2139.
Spanu, T., F. Luzzaro, M. Perilli, G. Amicosante, A. Toniolo, G. Fadda, and the Italian ESBL Study Group. 2002. Occurrence of extended-spectrum -lactamases in members of the family Enterobacteriaceae in Italy: implications for resistance to -lactams and other antimicrobial drugs. Antimicrob. Agents Chemother. 46:196-202.
Stürenburg, E., and D. Mack. 2003. Extended-spectrum -lactamases: implications for the clinical microbiology laboratory, therapy, and infection control. J. Infect. 47:273-295.
Tumbarello, M., R. Citton, T. Spanu, M. Sanguinetti, L. Romano, G. Fadda, and R. Cauda. 2004. ESBL-producing multidrug-resistant Providencia stuartii infections in a university hospital. J. Antimicrob. Chemother. 53:277-282.
Vahaboglü, H., R. Ozturk, G. Aygun, F. Coskunkan, A. Yaman, A. Kaygusuz, H. Leblebicioglu, I. Balik, K. Aydin, and M. Otkun. 1997. Widespread detection of PER-1-type extended-spectrum -lactamases among nosocomial Acinetobacter and Pseudomonas aeruginosa isolates in Turkey; a nationwide multicenter study. Antimicrob. Agents Chemother. 41:2265-2269.
Winokur, P. L., R. Canton, J. M. Casellas, and N. Legakis. 2001. Variations in the prevalence of strains expressing beta-lactamase phenotype and characterization of isolates from Europe, the Americas, and the Western Pacific region. Clin. Infect. Dis. 15(Suppl. 2):94-103.
Wong-Beringer, A., J. Hindler, M. Loeloff, A. M. Queenan, N. Lee, D. A. Pegues, J. P. Quinn, and K. Bush. 2002. Molecular correlation for the treatment outcomes in bloodstream infections caused by Escherichia coli and Klebsiella pneumoniae with reduced susceptibility to ceftazidime. Clin. Infect. Dis. 34:135-146.
Yigit, H., A. M. Queenan, G. J. Anderson, A. Domenech-Sanchez, J. W. Biddle, C. D. Steward, S. Alberti, K. Bush, and F. C. Tenover. 2001. Novel carbapenem-hydrolyzing -lactamase, KPC-1, from a carbapenem-resistant strain of Klebsiella pneumoniae. Antimicrob. Agents Chemother. 45:1151-1161.(Francesco Luzzaro, Marili)