Biomedical subjects
Publications and source records attributed to M A Pfaller.
Explore the source record for details and available documents.
A total of 63 clinical isolates of Proteus mirabilis collected over a 19-month period were typed by the Dienes test and ribotyping. Ribotyping was performed using the fully automated RiboPrinter Microbial Characterization System (Qualicon, Wilmington, Del.). Isolates that were indistinguishable by the Dienes test and/or ribotyping were characterized further by pulsed-field gel electrophoresis (PFGE). Most of the isolates represented unique strains as judged by the Dienes test and ribotyping. Forty isolates represented 40 different ribotypes and Dienes types. The remaining 23 isolates were grouped into 13 Dienes types, 12 ribotypes, and 14 PFGE types. The index of discrimination was 0.980 for the Dienes test, 0.979 for ribotyping, and 0.992 for PFGE. Both the Dienes test and ribotyping are useful methods for identifying individual strains of P. mirabilis. The Dienes test is simple, inexpensive, and easy to perform. It can be performed in virtually any laboratory and should be used in the initial epidemiologic characterization of P. mirabilis isolates.
The performance of the Etest for itraconazole susceptibility testing of 50 isolates of filamentous fungi was assessed in comparison with the National Committee for Clinical Laboratory Standards (NCCLS) proposed standard microdilution broth method. The NCCLS method employed RPMI 1640 broth medium, and MICs were read after incubation for 48 h at 35 degrees C. Etest MICs were determined with RPMI agar containing 2% glucose and with Casitone agar and were read after incubation for 24 h (Aspergillus spp. and Rhizopus spp.) and 48 h (all species except Rhizopus spp.) at 35 degrees C. The isolates included Aspergillus flavus, Aspergillus fumigatus, Aspergillus niger, Aspergillus terreus, Fusarium spp., Pseudallescheria boydii, Rhizopus spp., Paecilomyces variotii, and an Acremonium sp. Overall agreement between Etest and microdilution MICs was 96% with RPMI agar and 80% with Casitone agar. The agreement was 100% for all species except Rhizopus spp. (83%) and Paecilomyces varioti (0%) with RPMI agar. When Casitone agar was used, the agreement ranged from 50% with Rhizopus spp. to 100% with Fusarium spp., P. boydii, P. varioti, and an Acremonium sp. Notably, for Aspergillus spp., the agreement between itraconazole Etest MICs read at 24 h and reference microdilution MICs read at 48 h was 100% with both RPMI and Casitone agar. Both media supported the growth of all filamentous fungi tested. Where a discrepancy was observed between Etest and the reference method, the Etest MIC was generally higher. The Etest method using RPMI agar appears to be a useful method for determining itraconazole susceptibilities of Aspergillus spp. and other filamentous fungi.
Broth microdilution susceptibility tests of Candida species have now been standardized by the National Committee for Clinical Laboratory Standards (NCCLS). An eight-laboratory collaborative study was carried out in order to document reproducibility of tests of Candida parapsilosis ATCC 22019 and Candida krusei ATCC 6258 by the NCCLS method. Replicate broth microdilution tests were used to define control limits for 24- and 48-h MICs of amphotericin B, flucytosine, fluconazole, voriconazole, ketoconazole, itraconazole, caspofungin (MK 0991), ravuconazole (BMS 207147), posaconazole (SCH 56592), and LY 303366.
Modern epidemiology studies now require that nosocomial pathogens be characterized to the subspecies level whenever possible to better define infectious processes and modes of transmission. In general, if isolates are classified as different by at least one molecular typing method, they may be assumed to represent different strains and to reflect independent infections. If the isolates are the same, it may be assumed that cross infection has occurred or that the patients were infected by exposure to a common source. Typing methods may also be used to address clinical problems related to distinguishing reinfection versus relapse of an infection, and to examine the development of antifungal resistance among fungal pathogens during the course of antifungal therapy. Determining DNA fingerprints of sequential isolates from patients undergoing antifungal therapy has been useful in demonstrating the potential for the development of antifungal resistance in previously susceptible strains and for detecting the substitution of a more resistant strain for a more susceptible strain in the face of intense antimicrobial pressure. In order to be useful as an epidemiological typing method, a DNA fingerprinting system must effectively distinguish between genetically unrelated strains, identify the same strain in separate samples, and reflect genetic relatedness or unrelatedness (genetic distance) among strains.
The establishment of a standardized broth reference method for antifungal susceptibility testing of yeasts has opened the door to a number of interesting and useful developments. The adaptation of the reference macrodilution method to a microdilution method has significantly increased the clinical utility of antifungal susceptibility testing, and both methods are now included in the NCCLS document M27-A. The publication of quality control limits for five antifungal agents, coupled with the establishment of interpretive MIC breakpoints for three agents, provides useful parameters to survey clinical isolates of Candida and other yeast species. Adaptations of the M27 microdilution method for testing molds has also proved feasible. These developments have made it possible for a number of recent studies designed to expand the capabilities of laboratories to perform antifungal susceptibility testing and to enhance our understanding of trends in antifungal susceptibility. The availability of reference methods also provides a useful touchstone for the development of commercial products that promise to be more user friendly and to further improve test standardization. Products in varying stages of development include two colorimetric microdilution methods (Sensititre and KPI) and the Etest stable gradient MIC method. Preliminary data indicate that these methods are viable alternatives to the reference method for testing of yeasts. Furthermore, Etest may also prove useful for testing molds. Future expectations for antifungal susceptibility testing includes improved ability to detect amphotericin B resistance, development of an NCCLS document for susceptibility testing of molds, and application of these methods for testing dermatophytes. Incorporation of antifungal susceptibility testing methods into the clinical trials of new antifungal agents will facilitate the establishment of clinical correlates and further enhance the clinical utility of antifungal susceptibility testing.
One thousand seventy-three bacterial isolates were collected from patients with community acquired respiratory tract infections (CARTI) in 11 Latin American centers (7 countries) during 1997 and 1998. They were tested against numerous antimicrobial agents by the reference broth microdilution method as part of the ongoing multinational SENTRY Antimicrobial Surveillance Program. Among Streptococcus pneumoniae (553 isolates), approximately 61% were susceptible to penicillin. There was a great variation of the penicillin susceptibility rates among participating countries. The highest susceptibility rates were found in Argentina (76.7%) and Brazil (71.9%), while the lowest rate of penicillin susceptibility was detected in Mexico (33.3%). High level resistance to penicillin and resistance to cefotaxime were observed in nearly 10% of the isolates. The newer quinolones, levofloxacin (MIC(90) 2 microg/mL) and gatifloxacin (MIC90 0.5 microg/mL), were active against 100% of the isolates tested. Among the other non-beta-lactams drugs tested, the rank order of susceptibility against the pneumococci was: chloramphenicol (93.9%)>clindamycin (93.2%)> azithromycin (89.1%) > clarithromycin (88.7%)>tetracycline (78.5%)> trimethoprim/sulfamethoxazole (55.7%). The percentage of Haemophilus influenzae (361 isolates) isolates resistant to amoxicillin was 12. 7% (beta-lactamase positive). Among Moraxella catarrhalis (159 isolates) isolates, only 8.2% were susceptible. Clavulanic acid restored the activity of amoxicillin against both species. Trimethoprim/sulfamethoxazole was active against only 59.5% of H. influenzae, while susceptibility to this compound among M. catarrhalis was 96.1%. All other compounds tested were active against>95% of H. influenzae and M. catarrhalis isolates. These species were susceptible to levofloxacin (MIC90 < or = 0.5 microg/mL for both) and gatifloxacin (MIC90 < or = 0.03 microg/mL for both) with very low MICs. Our results indicate that penicillin resistance rates are particularly high among pneumococci in some countries. The newer fluoroquinolones show an excellent potency and spectrum against pathogens causing community acquired respiratory infections in Latin America.
The Meropenem Yearly Susceptibility Test Information Collection (MYSTIC) programme aims to provide in vitro surveillance data for geographically diverse institutions where meropenem is available for use. The in vitro activity of meropenem and eight comparator antimicrobial agents against 2340 significant pathogens obtained in 1999 was assessed and compared in 14 study centres in Brazil, Mexico and the USA. Isolates were further characterized for production of extended-spectrum beta-lactamases (ESBLs), AmpC beta-lactamases and carbapenemases. Carbapenems demonstrated their broad spectrum and potency, inhibiting > or = 95% of all isolates irrespective of the geographical region or centre type. The overall order of activity of the nine agents tested against all pathogens in 1999 was meropenem (96%) > imipenem (95%) > cefepime (92%) > gentamicin (89%) > piperacillin/tazobactam (88%) > ceftazidime = tobramycin (86%) > cefotaxime (84%) > ciprofloxacin (83%). Thus far, the results from the Americas indicate that meropenem has excellent potency and spectrum of activity despite being prescribed for the treatment of seriously ill patients. In contrast, other ESBLs, fluoroquinolones and aminoglycosides have lost activity in many institutions as a result of the selection of strains producing ESBLs or having AmpC and other resistance determinants. Carbapenem resistance was observed rarely and at a prevalence similar to those reported in earlier studies. Carbapenems appear to be a continuing reliable option for the treatment of serious nosocomial infection.
Gatifloxacin (formerly AM-1155 or CG 5501) is a new 8-methoxy fluoroquinolone with enhanced activity against Gram-positive cocci, especially Streptococcus pneumoniae and other streptococci. Recent clinical strains (599 isolates) were tested against gatifloxacin, three comparison fluoroquinolones, and penicillin by the reference broth microdilution, Etest (AB BIODISK, Solna, Sweden) and standardized disk diffusion methods (5 micrograms gatifloxacin disk). Gatifloxacin (MIC90, 0.5 microgram/ml) activity was generally comparable to that of trovafloxacin (MIC90, 0.25 microgram/ml), or sparfloxacin (MIC90, 0.5 microgram/ ml) and markedly superior to ofloxacin (MIC90, 2-4 micrograms/ml) against the streptococci. Rates of penicillin non-susceptibility were 41.9, 38.0, and 16.2% for S. pneumoniae (301 strains), viridans group streptococci (150 strains), and beta-haemolytic streptococci (148 strains). Etest results correlated well (95.7-100.0% +/- one log2 dilution) with the reference MIC results, but Etest tended to have elevated gatifloxacin MIC results compared to the broth microdilution method for the highly resistant isolates (MICs, > 2 micrograms/ml). Gatifloxacin disk zone diameters correlate well to reference MICs for all streptococci and proposed interpretive criteria (susceptible at < or = 1 microgram/ml or > or = 18 mm, and resistant at > or = 4 micrograms/ml or < or = 14 mm) did not produce discords between method results (absolute agreement). A nine laboratory quality control (QC) study conforming to the National Committee for Clinical Laboratory Standards (NCCLS) Guideline M23-T3 studied S. pneumoniae ATCC 49619 and gatifloxacin. Proposed ranges for QC of NCCLS tests were 0.12-0.5 microgram/ml for the broth microdilution test and 24-31 mm for the disk diffusion method. These reported results indicate that gatifloxacin was a potent fluoroquinolone with extensive activity against streptococcal isolates. In vitro test methods to measure this activity appear accurate and comparable; and QC guidelines have been established for routine clinical laboratory use pending approval by the NCCLS and the Food and Drug Administration (FDA).
Gatifloxacin (formerly AM-1155 and CG5501), a new 8-methoxy fluoroquinolone, has an expanded spectrum of activity against Gram-positive cocci and some anaerobic bacteria. This compound was tested against 600 recent clinical strains of rapidly growing aerobic species to establish susceptibility testing interpretive criteria for the reference broth microdilution and standardized disk (5-microgram) diffusion methods of the National Committee for Clinical Laboratory Standards (NCCLS). These strains included 285 Enterobacteriaceae (17 species), 165 staphylococci, 49 enterococci, and 101 nonfermentative Gram-negative bacilli. Based on achievable serum levels with projected gatifloxacin dosing regimens, MIC break points of < or = 2 micrograms/mL (> or = 18 mm) for susceptibility and > or = 8 micrograms/mL (< or = 14 mm) for resistance were selected. The absolute agreement between tests was 94.3% with no very major false-resistant errors. The quality control ranges (MIC and zone diameters) for the NCCLS recommended strains were determined in a nine-laboratory NCCLS protocol as follows: Escherichia coli ATCC 25922 = 0.008-0.03 microgram/mL and 31-37 mm; Enterococcus faecalis ATCC 29212 = 0.12-1 microgram/mL; Pseudomonas aeruginosa ATCC 27853 = 0.5-2 micrograms/ml and 21-27 mm; Staphylococcus aureus ATCC 25923 = 27-33 mm and S. aureus ATCC 29213 = 0.03-0.12 microgram/mL. Gatifloxacin appears to be a promising new fluoroquinolone with acceptable susceptibility testing methods for routine clinical laboratory use.
The Gram-positive cocci have clearly re-emerged as important pathogens world-wide in the past two decades. Staphylococci, including the coagulase-negative staphylococci and Staphylococcus aureus, and the enterococci account for approximately one-third of all blood stream infections and as much as 50% of nosocomial blood stream infections. Although Streptococcus pneumoniae is often considered a community-acquired pathogen, it is also an important cause of nosocomial infection. The hallmark of these Gram-positive pathogens is increasing resistance to available antimicrobial agents. Of particular note is resistance to glycopeptides (vancomycin and teicoplanin), aminoglycosides (high-level), and penicillins among the enterococci (especially E. faecium), resistance to penicillinase-resistant penicillins (oxacillin and methicillin) and fluoroquinolones (ciprofloxacin and ofloxacin) among staphylococci, and resistance to penicillin, other beta-lactams and macrolides among the pneumococci. The recent detection of decreased susceptibility to vancomycin among S. aureus is also quite ominous. In many instances the ability of the clinical laboratory to accurately characterize these resistant isolates is suboptimal, further compounding the problem. Increased understanding of resistance mechanisms and correlations of resistance genes with the phenotypic expression of resistance has allowed for modifications and improvements of reference susceptibility tests and interpretive breakpoints. New compounds for effective therapy of infection with multi-resistant Gram-positive species are clearly needed. To this end, the streptogramin combination, quinupristin/dalfopristin, has demonstrated significant activity against oxacillin-resistant staphylococci, penicillin-resistant streptococci, and vancomycin-resistant E. faecium. Other candidate drugs including Gram-positive active fluoroquinolones (clinafloxacin, grepafloxacin, moxifloxacin, gatifloxacin, and trovafloxacin) and novel compounds such as the everninomicin derivatives (SCH27899), ketolides, and oxazolidinones (linezolid) have been shown to be active against these organisms and are under rapid clinical development.
MK-0991 has demonstrated activity against a variety of fungal pathogens. We evaluated the MIC endpoint for MK-0991 by reading the endpoint using three methods and comparing these results with minimum fungicidal concentrations and electron micrographs. The concentration that resulted in 80% inhibition of fungal growth compared with control, similar to the endpoint for the azole antifungal agents, provided the most consistent results. Additionally, we investigated the time-kill properties of this agent against two isolates each of Candida albicans, Candida glabrata and Candida tropicalis at concentrations ranging from 0.125 x MIC to 16 x MIC. Kill curves were performed using RPMI buffered with morpholine propanesulfonic acid as growth media. Samples were obtained at predetermined time points over 24 h and plated for colony counting. Fungicidal activity was observed with one isolate of C. albicans, two isolates of C. glabrata, and one isolate of C. tropicalis. MK-0991 displayed concentration-dependent activity, which was fungicidal or fungistatic depending on the isolate tested.
DNA-based procedures have replaced earlier epidemiologic methodologies that relied on nonreproducible and insensitive measurements of phenotypic characteristics to identify a specific strain as the source of infection. The reliability (interlaboratory percent agreement for strain delineation) and sensitivity (recognition of subtle strain-to-strain variation) of similar DNA-based typing systems by different laboratories were evaluated. Ten isolates (five epidemiologic-related and five unrelated strains each) of Candida albicans, C. lusitaniae, C. parapsilosis, C. tropicalis, and Candida (Torulopsis) glabrata were characterized in a blinded fashion by three laboratories. All 50 isolates were subtyped in each laboratory by electrophoretic karyotyping (EK) analysis using contour-clamped homogenous electric field (CHEF) electrophoresis protocols. In addition, two laboratories also performed restriction endonuclease analysis of genomic DNA (REAG) using the restriction endonucleases SfiI and BssHII followed by CHEF electrophoresis separation of resulting fragments. DNA strain identification of the 50 isolates by the three different laboratories using similar CHEF methodologies demonstrated the following species-dependent, interlaboratory reproducibility: C. tropicalis (82%), C. parapsilosis (83%), C. albicans (90%), C. lusitaniae (93%), and C. glabrata (100%). In addition, agreement was higher by the CHEF method (83 to 100%), when compared with the strain types identified by the REAG (60 to 100%) method. Five to seven strains of each Candida species evaluated were detected by the different methodologies used for this study. This study indicates that these procedures are relatively discriminatory and reliable tools to study strain-to-strain variations in epidemiologic evaluations of these yeasts.
The SENTRY Antimicrobial Surveillance Program was established in January, 1997 to monitor the predominant pathogens and antimicrobial resistance patterns of nosocomial and community-acquired infections via a network of sentinel hospitals in the United States (30 sites), Canada (eight sites), Latin America (10 sites), and Europe (24 sites). During the first 12-month study period (January to December, 1997), a total of 9519 blood stream infections (BSI) were reported by SENTRY participants in the U.S. (6150), Canada (1727), and Latin America (1642). The Gram-positive cocci, Staphylococcus aureus, coagulase-negative staphylococci (CoNS), enterococci, and streptococci accounted for 53.9% (5131 infections) of all BSI (56.5% U.S., 55.7% Canada, and 42.9% Latin America). The staphylococci, Enterococcus spp., S. pneumoniae, beta-hemolytic streptococci, and viridans group streptococci accounted for 6 of the top 11 BSI pathogens in the U.S. and Canada, whereas only S. aureus (1st), CoNS (3rd), and Enterococcus spp. (9th) were among the top 11 pathogens in Latin American hospitals. The results of this survey affirm the importance of Gram-positive cocci as causes of BSI in both North America and Latin America and demonstrate that important antimicrobial resistance exists among isolates of staphylococci, streptococci, and enterococci from all three geographic regions. This includes oxacillin-resistance among S. aureus (26.9% U.S., 29.2% Latin America, and 4.0% Canada) and CoNS (71.5% U.S., 68.4% Latin America, and 65.6% Canada), penicillin resistance among viridans group streptococci (48.5% U.S., 45.1% Canada, and 33.3% Latin America) and pneumococci (36.1% U.S., 27.5% Canada, and 65.6% Latin America), high-level resistance (HLR) to aminoglycosides among enterococci (27.2 to 70.1% U.S., 33.3 to 75.7% Canada and 16.7 to 51.5% Latin America), and macrolide resistance among beta-hemolytic streptococci (12.4 to 14.2% U.S., 10.5 to 12.3% Canada, and 0.0 to 4.0% Latin America), viridans group streptococci (32.4 to 39.7% U.S., 22.5-35.2% Canada, and 20.0% Latin America), and pneumococci (10.0 to 10.6% U.S., 9.8-10.8% Canada, and 9.4-18.7% Latin America). BSI isolates of Gram-positive cocci from the U.S. and Latin America were considerably more resistant than those from Canada. New agents with Gram-positive activity will be essential for optimal treatment of BSI attributable to Gram-positive cocci in both North and Latin America.
Urinary tract infection (UTI) is common and involves pathogens with changing susceptibility patterns. The SENTRY Antimicrobial Surveillance Program evaluates international pathogen incidence patterns to detect and manage the emergence of resistant strains. We describe the antimicrobial resistance patterns among 1617 pathogens recovered from UTIs during the third-quarter of 1997 in North America (United States and Canada), as part of this worldwide program. The isolates were tested against more than 50 antimicrobial agents (20 reported) by reference broth microdilution methods, and selected isolates were characterized by pulsed-field gel electrophoresis (PFGE) and automated ribotyping. The five most frequently isolated species were Escherichia coli (48.6%), Enterococcus spp. (13.7%), Klebsiella spp. (12.0%), Pseudomonas aeruginosa (6.2%), and Enterobacter spp. or Proteus mirabilis (3.8% each). For each nation, imipenem and cefepime produced the widest spectrum of coverage among the beta-lactams and amikacin was best among the aminoglycosides. For Gram-negative species, high resistance among beta-lactam antimicrobial agents was noted especially for various penicillins against E. coli (37.9% to 42.8%) and for the cephalosporins tested against enterococci (99.4% and 100%). Approximately 7.0% of enterococci in the USA were vancomycin-resistant (88% with Van A). P. aeruginosa provided the most consistent levels of resistance, but the following agents were most active against these organisms: amikacin (96.6 to 97.4% susceptible), tobramycin (89.5 to 100.0%), piperacillin/tazobactam (89.5 to 100.0%), piperacillin (89.5 to 96.6%), imipenem (89.7 to 92.1%), cefepime (77.6 to 89.7%), and ceftazidime (82.9 to 86.2%). E. coli (2.2 to 2.7%), K. pneumoniae (6.2 to 6.4%), and a single Enterobacter cloacae strain produced extended-spectrum beta-lactamases; and five other Enterobacter spp. were likely to have expressed chromosomally mediated (Amp C) Stably derepressed cephalosporinases with associated resistance to ceftazidime (16.7 to 21.2% resistance). These data demonstrated that several UTI isolates in SENTRY hospitals have high levels of resistance to various classes of antimicrobial agents with little evidence of clonal dissemination.
National surveillance of blood stream infections (BSI) attributable to Candida spp. has been limited to date. Recent studies have suggested in increase in the proportion of BSI attributable to non-Candida albicans species and have also raised concerns regarding the emergence of antifungal resistance among Candida spp. The increased utilization of broad-spectrum antifungal agents and the recognition of Candida spp. as prominent pathogens with the potential for developing antifungal resistance, emphasize the need for ongoing surveillance of antifungal susceptibility patterns. In this investigation trends in species distribution and susceptibility to fluconazole among BSI isolates of Candida spp. referred to our laboratory by United States hospitals were evaluated over the 7-year period from 1992 to 1998. A total of 1579 BSI isolates from more than 50 medical centers were processed. Overall, C. albicans accounted for 52% of isolates followed by C. glabrata (18%), C. parapsilosis (15%), C. tropicalis (11%), and C. krusei (2%). The proportion of BSI isolates that were C. albicans ranged from 45% in 1992 to 60% in 1998. Among the non-C. albicans isolates, C. glabrata succeeded C. parapsilosis as the most common species beginning in 1995. Overall, the susceptibility of all Candida species (C. albicans plus all other species) to fluconazole remained stable (MIC90, 16 micrograms/mL). The fluconazole MIC90 for C. albicans was 0.5-2.0 micrograms/ml for all years studied except 1995 (8.0 micrograms/mL) and was 1.0 microgram/mL overall. The present study suggests a continued prominent role of C. albicans as a cause of BSI, and a constant level of susceptibility of Candida BSI isolates to fluconazole over 7 years. These data should serve as a baseline for future surveillance efforts for anti-fungal agents tested against yeast BSI isolates.
As part of the SENTRY Antimicrobial Surveillance Program, 1562 bacterial isolates were recovered from hospitalized patients with skin and soft tissue infections (SSTIs) in 30 United States (U.S.) and 8 Canadian medical centers between October and December, 1997. The overall rank order of recovery of the six most common pathogens was Staphylococcus aureus (42.6%) > Pseudomonas aeruginosa (11.3%) > Enterococcus spp. (8.1%) > Escherichia coli (7.2%) > Enterobacter spp. (5.2%) > beta-hemolytic streptocci (5.1%). With one exception, essentially the same order was observed in both the U.S. and Canada. The single exception was the Enterococcus group, which were the third most common isolate in the U.S. (9.6%), but the seventh most common isolate in Canada (3.7). Of note, 24.0% of S. aureus isolates were oxacillin resistant; vancomycin was uniformly active. Vancomycin resistance among Enterococcus spp. (16.5%) was observed only in the U.S. Several antimicrobial agents remained broadly active for SSTI isolates of P. aeruginosa, including meropenem, amikacin, tobramycin, and piperacillin with or without tazobactam. Imipenem resistance (MICs, > or = 8 micrograms/mL) was observed in 11.9% of isolates of P. aeruginosa and ceftazidime, and cefepime had equivalent activity (85.2% and 85.8% susceptible, respectively). Numerous beta-lactams, aminoglycosides and fluoroquinolones were broadly active against E. coli SSTI isolates (i.e. < 5% resistance). Extended-spectrum beta-lactamase production was uncommon both with E. coli and Klebsiella spp. in both nations. Cefepime, imipenem, and meropenem; the aminoglycosides; and fluoroquinolones were conspicuously more active against Enterobacter spp. than other agents tested. High-level, stably derepressed Amp C beta-lactamase production was commonly observed in this group (26.8%), but cefepime generally retained activity against these ceftazidime-resistant organisms. The results of this study serve to define the most common bacterial causes of SSTIs in North America, elucidate patterns of antimicrobial resistance and can be used as a basis for making initial empiric antimicrobial management decisions in hospitalized patients with such infections.
Twenty-three multi-resistant strains of Escherichia coli were isolated at a single hospital in Beijing, China between January 1997 and May 1998. All isolates produced extended spectrum beta-lactamases (ESBLs) as detected by the double disk synergy test and the Etest ESBL strip (AB BIODISK, Solna Sweden). Additional antimicrobial susceptibility testing showed that most isolates were resistant to gentamicin, tobramycin, tetracycline, trimethoprim/sulfamethoxazole, ciprofloxacin, and cefepime. All isolates remained susceptible to imipenem with MICs of < or = 0.5 microgram/ml. The isolates each produced several beta-lactamases (range 1-4 enzymes/strain) with pI values ranging from 5.2-8.4. Molecular epidemiologic typing revealed four ribotypes and eight pulsed field gel electrophoresis (PFGE) patterns with subgroups among the 23 isolates. Clusters of isolates with the same DNA type were observed as follows (ribotype/PFGE): Wards A (242-5/2, and 242-5/3a), B (242-5/4), and C (880-1/1a). Moreover, similar molecular types were observed in patients from two or more different wards. Further use of isoelectric focusing results and co-resistance patterns produced evidence of potential nosocomial dissemination of strains in only two instances (two identical strains on one ward and two identical strains on different wards). There were also strong similarities in beta-lactamase pIs and co-resistances among many of the strains throughout this medical center. These data document the wide genetic diversity among E. coli producing ESBLs, and a potential for nosocomial spread of these highly resistant organisms requiring increasingly more sophisticated molecular-based techniques and local interventions.