Argentinean collaborative multicenter study on the in vitro comparative activity of piperacillin-tazobactam against selected bacterial isolates recovered from hospitalized patients

Argentinean collaborative multicenter study on the in vitro comparative activity of piperacillin-tazobactam against selected bacterial isolates recovered from hospitalized patients

Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537 www.elsevier.com/locate/diagmicrobio Antimicrobial susceptibility studies Argentin...

104KB Sizes 0 Downloads 1 Views

Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

www.elsevier.com/locate/diagmicrobio

Antimicrobial susceptibility studies

Argentinean collaborative multicenter study on the in vitro comparative activity of piperacillin-tazobactam against selected bacterial isolates recovered from hospitalized patients Jose´ Marı´a Casellas1,a,*, Gabriella Tome´a,b,c, Carlos Bantard, Pamela Bertolinie, Ne´stor Bla´zquezf, Noemı´ Bordag, Elsa Coutoh, Norma Cudmanii, Josefina Guerrerae, Marı´a Josefina Jua´rezj, Teresa Lo´pezk, Ana Littvikk, Emilce Me´ndezl, Rodolfo Notariom, Graciela Poncen, Mirta Quinterosh, Francisco Salamonen, Mo´nica Sparoo, Emma Sutichp, Susana Vayletq, Lidia Wolffr a

Centro de Estudios en Antimicrobianos (CEA), Buenos Aires City (Reference Center), Buenos Aires, Argentina b Instituto Maternidad Santa Rosa, Vicente Lo´pez, BA Province, Argentina c Instituto Cardiovascular Infantil, Buenos Aires City, Argentina d Laboratorio Nanni, Parana´, Entre Rı´os Province, Argentina e Hospital Municipal de San Isidro, San Isidro, BA Province, Argentina f Hospital Zonal de Bariloche, Bariloche, Rı´o Negro Province, Argentina g Instituto de Diagno´stico ABC, Rosario, Santa Fe Province, Argentina h Hospital Francisco J. Mun˜iz, Buenos Aires City, Argentina i Hospital de Clı´nicas Nicola´s Avellaneda, S.M. de Tucuma´n, Tucuma´n Province, Argentina j Hospital Centro Gallego, Buenos Aires City, Argentina k Hospital Rawson, Co´rdoba, Co´rdoba Province, Argentina l Hospital Cullen, Santa Fe, Santa Fe Province, Argentina m Hospital Espan˜ol, Rosario, Santa Fe Province, Argentina n Hospital General San Martı´n, Parana´, Entre Rı´os Province, Argentina o Hospital Municipal de Tandil, Tandil, BA Province, Argentina p Hospital IPAM, Rosario, Santa Fe Province, Argentina q Hospital Penna, Bahı´a Blanca, BA Province, Argentina r Clı´nica Privada Ve´lez Sarsfield, Co´rdoba, Co´rdoba Province, Argentina Received 22 January 2003

Abstract The in vitro activity of piperacillin-tazobactam and several antibacterial drugs commonly used in Argentinean hospitals for the treatment of severe infections was determined against selected but consecutively isolated strains from clinical specimens recovered from hospitalized patients at 17 different hospitals from 9 Argentinean cities from different geographic areas during the period November 2001-March 2002. Out of 418 Enterobacteriaceae included in the Study 84% were susceptible to piperacillin-tazobactam. ESBLs putative producers were isolated at an extremely high rate since among those isolates obtained from patients with hospital acquired infections 56% of Klebsiella pneumoniae, 32% of Proteus mirabilis and 25% Escherichia coli were phenotypically considered as ESBLs producers Notably P.mirabilis is not considered by NCCLS, 2002 for screening for ESBL producers. ESBLs producers were 100% susceptible to imipenem and 70% were susceptible to piperacillin-tazobactam whereas more than 50% were resistant to levofloxacin. The isolates considered as amp C beta lactamase putative producers showed 99% susceptibility to carbapenems while 26.7% were resistant to piperacillin-tazobactam and 38.4% to levofloxacin. Noteworthy only 4% of the Enterobacteriaceae isolates were resistant to amikacin. Piperacillin-tazobactam was the most

1 Current affiliation for J.M. Casellas is CIBIC/Sanatorio Parque, Presidente Roca 740, Rosario CP 2000, Provincia de Santa Fe, Argentina. Part of the content of the manuscript has been presented at the following congresses: 13th Argentinean Congress for Intensive Care Medicine,

0732-8893/03/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0732-8893(03)00131-7

Buenos Aires City, Argentina, 14-18 Sep 2002, Abs. 119; 9th Meeting of the Argentinean Association for Microbiology, Cordoba City, Argentina, 28 –30 Nov 2002, Abs. P130. *Corresponding author. Fax: ⫹54-11-4745-2097. E-mail address: [email protected] (J. M. Casellas).

528

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

active agent against Pseudomonas aeruginosa isolates (MIC90: 128 ␮g/ml; 78% susceptibility) but showed poor activity against Acinetobacter spp (MIC90:⬎256 ␮g/ml; 21.7% susceptibility). Only 41.7% Acinetobacter spp isolates were susceptible to ampicillin-sulbactam. Piperacillin-tazobactam inhibited 100% of Haemophilus influenzae isolates (MIC90 ⬍ 0.25 ␮g/ml) but only 16.6% of them were ampicillin resistant. The activity of piperacillin-tazobactam against oxacillin susceptible Staphylococcus aureus or coagulase negative staphylococci was excellent (MIC90 2 ␮g/ml; 100% susceptibility). Out of 150 enterococci 12 isolates (8%) were identified as E.faecium and only three isolates (2%), 2 E.faecium and 1 E.faecalis were vancomycin resistant. All the enterococci isolates were susceptible to linezolid. Piperacillin-tazobactam showed excellent activity (MIC90 2 ␮g/ml; 92% susceptibility). Regarding pneumococci all the isolates showed MICs of 16 ␮g/ml for piperacillin-tazobactam. Among 34 viridans group streptococci only 67% were penicillin susceptible and 85.2% ceftriaxone susceptible whereas piperacillin-tazobactam was very active (MIC90 4 ␮g/ml). Piperacillin-tazobactam is therefore a very interesting antibacterial drug to be used, preferably in combination (ie: amikacin-vancomycin) for the empiric treatment of severe infections occurring in hospitalized patients in Argentina. Caution must be taken for infections due to ESBL producers considering that the inoculum effect MICs can affect MIC values. © 2003 Elsevier Inc. All rights reserved.

1. Introduction Resistance has emerged with alarming incidence in Latin American countries particularly that due to Ambler class A ␤-lactamases, both broad (BSBLs) and extended spectrum (ESBLs). Isolation rates of ESBL producers in some Argentinean hospitals is as high as 50% and includes isolates of Klebsiella pneumoniae, Escherichia coli, Proteus mirabilis, Salmonella spp, Shigella spp and other species. Most of these ESBLs are of the CTX-M type (Casellas, 2001; Casellas, 1999). E.coli, Shigella spp and Salmonella spp resistance to cotrimoxazole and chloramphenicol, Group C meningococci resistance to penicillin, as well as gonococci resistance to penicillin and tetracyclines are important in Latin American countries (Casellas et al., 1994; Guzman et al., 2000). Resistance to ciprofloxacin also has been observed (Me´ ndez et al., 1999). In addition Latin American countries also participate of those resistance problems which are of global concern such as: methicillin resistance in staphylococci, diminished susceptibility to ␤-lactams in pneumococci and streptococci of the viridans group; Group A streptococci and pneumococci resistance to the so called MALSKO antibacterial agents (macrolides, azalides, lincosamides, streptogramins and ketolides); resistance to cephalosporins due to amp C chromosomal enzymes and Pseudomonas aeruginosa and Acinetobacter spp. resistance to multiple antibacterial drugs. On the other hand, enterococci resistance to glycopeptides is not yet an important problem in Latin America (Guzma´ n et al., 2000; Rossi et al., 2000). A VISA isolate has recently been reported in Brazil (Rossi et al., 2000). ␤-lactamase production by Haemophilus influenzae is particularly lower in South America as compared to other geographic areas (Casellas et al., 1994; Guzma´ n et al., 2000). Epidemiologic surveillance of in vitro resistance in Latin America is carried out in some countries by different sentinel programs, such as the WHONET program, but this program includes few results obtained by the dilution method in our country. Other surveillance programs are supported by grants of pharmaceutical companies such as the SENTRY program. Unfortunately, the SENTRY program includes few laboratories in each country (i.e., there are only two laboratories in Argentina, both in Buenos

Aires, representing the whole country). Other international collaborative programs from pharmaceutical companies such as the Alexander, MYSTIC, and RESISTNET do not include Argentinean laboratory results. Piperacillin-tazobactam is a combination of piperacillin with tazobactam, a ␤-lactamase inhibitor. Piperacillin-tazobactam combines the intrinsic activity of piperacillin against P.aeruginosa, Enterococcus spp, pneumococci and other streptococci with the ␤-lactamase inhibitor activity of tazobactam, which restores the activity of piperacillin against ␤-lactamase Ambler class A producers such as E. coli, K. pneumoniae, P. mirabilis, etc., which frequently rendered piperacillin inactive due to BSBLs, such as TEM-1, TEM-2 or SHV-1. Tazobactam is also a strong inhibitor of ␤-lactamases produced by anaerobes. Consequently, anaerobes show nearly 100% susceptibility to piperacillin-tazobactam. It is also a good inhibitor of the penicillinase produced by staphylococci. Therefore piperacillin-tazobactam is active against methicillin-susceptible staphylococci. Due to its broad spectrum of activity and few adverse effects, piperacillin-tazobactam has been considered a drug for empiric initial therapy in ICU infections avoiding intensive use of carbapenems and selection of resistance to these drugs (Peterson, 2001; Joshi et al., 1999). A paper related to the activity of piperacillin-tazobactam has been recently published after the end of this study (Johnson et al., 2002) and includes isolates from Latin America. Unfortunately, it makes no reference regarding differences among isolates from different continents or other countries. A collaborative study has been carried out including selected isolates obtained from hospitalized patients during the period November 2001-March 2002 performed in centers of eleven different Argentinean cities. The centers were selected from cities from the 5 most populous Argentinean geographic areas: Northeastern-Mesopotamia (Rosario, Santa Fe-Parana´ ); Northwestern area (Tucuma´ n); Central area (Co´ rdoba); Pampa (Buenos Aires Province as Tandil, Bahia Blanca). Since Buenos Aires City and surroundings is the most populous city including 1/3 of Argentinean inhabitants, four centers were included from that area. The study was designed in order to compare the activity of piperacillin-tazobactam with that of other antibacterial agents com-

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

monly used in our country for the treatment of severe bacterial infections using a broth microdilution method.

2. Materials and methods 2.1. Centers and strains Seventeen centers participated in this study. Each center was required to send 30 to 120 consecutive isolates to the Reference Center (CEA) according to the following percentage of species distribution: Streptococcus pneumoniae 8%, Viridans group streptococci 2%, Enterococcus faecalis 13%, Enterococcus faecium 1%, Staphylococcus aureus 13%, coagulase negative staphylococci 13%, Haemophilus influenzae 3%, E.coli 13%, Klebsiella spp 8%, P.mirabilis 4%, Enterobacter spp 5%, other Enterobacteriaceae 5%, P.aeruginosa 8% and Acinetobacter spp 5%. Every isolate had to be obtained from hospitalized patients and only one isolate of the same species was permitted from each patient. Strains were identified at each center in accordance to ASM recommendations (Murray, 1998) 2.2. Susceptibility testings Susceptibility to antibacterial drugs was determined in each center by the microdilution method in accordance with the NCCLS recommendations (NCCLS, 2002). Special panels prepared by Sensititre (Trek Diagnostic Systems. West Sussex. UK) were used. Mu¨ eller Hinton broth was used for testing, 3-5% lysed horse blood was added for testing Streptococcus spp and Haemophilus Test Medium was used for testing Haemophilus spp. Inoculation and interpretation were performed following instructions of the manufacturer. Different panels were used for a) Gram-positive normal growers; b) Gram-negative normal growers; c) fastidious Gram-positive organisms and d) fastidious Gram-negative organisms. P.aeruginosa and Acinetobacter spp multiresistant (carbapenem resistant isolates) were tested by agar dilution for their susceptibility to colistin (colistin methansulphonate, Bristol Myers & Squibb) in Mu¨ eller Hinton agar 2.2.1. Quality control strains In the accordance with the NCCLS recommendations, the following strains were used for quality control (NCCLS, 2002): Pseudomonas aeruginosa ATCC 27853; Enterococcus faecalis ATCC 29212; Staphylococcus aureus ATCC 29213; Escherichia coli ATCC 25922; Escherichia coli ATCC 35218; Streptococcus pneumoniae ATCC 49619; and Haemophilus influenzae ATCC 49247. 2.2.2. Interpretation of results Isolates were considered susceptible, intermediate or resistant according to the NCCLS 2002 recommendations (NCCLS, 2002). Since no break points are provided by

529

Table 1 Distribution of the isolates included in the study A. GRAM NEGATIVE Enterobacteriaceae Non fermenter gram negative bacilli Haemophilus influenzae TOTAL B. GRAM POSITIVE Staphylococcus aureus Coagulase negative staphylococci Streptococcus pneumoniae Enterococcus spp Viridans group streptococci TOTAL TOTAL

n 418 173 30 621 154 141 100 150 34 579 1200 isolates

NCCLS for susceptibility to colistin we considered an arbitrary break point of MIC ⱖ4 ␮g/ml for resistance as previously suggested (Catchpole et al., 1997) (Montero et al., 2002) 2.3. ESBL putative producers The screening was performed: a) by means of the Jarlier test (Jarlier et al., 1988), using Neosensitab tablets (Rosco, Denmark) of cefotaxime, ceftazidime and cefepime distributed around a central amoxicillin-clavulanate tablet; b) by using disks containing cefotaxime or ceftazidime and lithium clavulanate as previously described (Casellas et al., 1989); c) by using E-test (AB Biodisk, Sweden) strips containing both ceftazidime in one end and ceftazidimeclavulanate in the opposite end, and cefotaxime in one end and cefotaxime-clavulanate in the opposite end. A test was considered positive when the MIC for the combinations resulted in two dilutions lower, as compared to that of the cephalosporins alone. 2.4. Submission of isolates, demographic data and reporting of results The isolates were sent to the Reference Center (CEACentro de Estudios en Antimicrobianos, in Buenos Aires City) by each investigator. The identification and susceptibility test with strains selected at random were there rechecked.

3. Results A total of 1200 isolates (1000 from adults and 200 from pediatrics) were included in the study according to the distribution presented in Table 1. The median age for the adult patients was 58.6 years. One third of the patients were in ICU at the moment of the collection of the specimens.

530

Table 2 Activity of piperacillin-tazobactam against selected Enterobacteriaceae groups Organisms (n tested)

Escherichia coli (ESBL negative, 108)

Klebsiella pneumoniae (ESBL negative, 44)

Klebsiella pneumoniae (ESBL phenotype, 56)

Proteus mirabilis (ESBL phenotype, 16)

Piperacillin-tazobactam(2) Ampicillin-sulbactam(3) Cefuroxime Ceftriaxone Ceftazidime Cefepime Imipenem Amikacin Levofloxacin Piperacillin-tazobactam Ampicillin-sulbactam Cefuroxime Ceftriaxone Ceftazidime Cefepime Imipenem Amikacin Levofloxacin Piperacillin-tazobactam Ampicillin-sulbactam Cefuroxime Ceftriaxone Ceftazidime Cefepime Imipenem Amikacin Levofloxacin Piperacillin-tazobactam Ampicillin-sulbactam Cefuroxime Ceftriaxone Ceftazidime Cefepime Imipenem Amikacin Levofloxacin Piperacillin-tazobactam Ampicillin-sulbactam Cefuroxime Ceftriaxone Ceftazidime Cefepime Imipenem Amikacin Levofloxacin

MIC (␮g/ml)

Percentage by category

Range

50%

90%

S(1)

I(1)

R(1)

ⱕ0.25–128 ⱕ0.5–64 ⱕ0.25–16 ⱕ0.25–1 ⱕ0.25–1 ⱕ0.25–1 ⱕ0.12–2 ⱕ0.5–16 ⱕ0.25–32 1–⬎256 4–⬎256 8–⬎128 2–⬎128 ⱕ0.25–⬎128 0.5–128 ⱕ0.12–4 4–64 ⱕ0.25–64 ⱕ0.25–16 2–128 ⱕ0.25–8 ⱕ0.25–1 ⱕ0.25–0.5 ⱕ0.25–0.5 ⱕ0.12–0.25 0.5–32 ⱕ0.25–8 1–⬎256 4–⬎256 8–⬎128 2–⬎128 ⱕ0.25–⬎128 1–⬎128 ⱕ0.12–1 2–⬎256 ⱕ0.25–64 ⱕ0.25–4 4–64 8–⬎128 4–⬎128 ⱕ0.25–16 4–⬎128 ⱕ0.12–4 ⱕ0.5–⬎256 1–⬎128

2 8 4 ⱕ0.25 ⱕ0.25 ⱕ0.25 ⱕ0.12 2 ⱕ0.25 8 32 32 32 4 4 ⱕ0.12 8 8 1 4 4 ⱕ0.25 ⱕ0.25 ⱕ0.25 ⱕ0.12 2 ⱕ0.25 8 64 32 8 2 8 ⱕ0.12 8 ⱕ0.25 1 32 ⬎128 32 1 16 0.5 4 32

8 16 8 ⱕ0.25 ⱕ0.25 ⱕ0.25 ⱕ0.12 4 2 64 64 ⬎128 ⬎128 64 16 0.5 32 32 4 8 4 ⱕ0.25 ⱕ0.25 ⱕ0.25 ⱕ0.12 4 ⱕ0.25 ⬎256 ⬎256 ⬎128 ⬎128 32 64 0.5 32 32 2 32 ⬎128 ⬎128 4 ⬎128 4 16 ⬎128

97.2 21.3 97.2 100 100 100 100 100 87.8 75.6 5.2 16.0 — 7.4 24.3 100 86.4 37.8 100 86.4 100 100 100 100 100 95.5 95.4 58.9 7.1 5.3 — 7.1 30.0 100 84.0 81.5 100 6.2 0 0 62.5 43.7 100 87.5 6.2

0.9 11.1 0.9 0 0 0 0 0 7.4 13.5 5.2 32.4 — — 54.1 0 10.8 5.4 0 4.5 0 0 0 0 0 4.5 2.3 8.9 14.2 0 — — 0 0 10.7 5.3 0 31.2 0 0 0 6.2 0 0 0

1.9 67.6 1.9 0 0 0 0 0 4.8 10.9 89.6 51.6 100 92.6 21.6 0 2.8 56.8 0 9.1 0 0 0 0 0 0 2.3 32.2 78.7 94.7 100 92.9 70.0 0 5.3 13.2 0 62.6 100 100 37.5 50.1 0 12.5 93.8

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

Escherichia coli (ESBL phenotype, 37)

Antibacterial agent

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

(2)

(1)

Other Enterobacteriaceae (11) Salmonella spp (4) Enterobacter agglomerans (4) Citrobacter koseri (2) Enterobacter sakazaki (1)

S: susceptible; I: intermediate; R: resistant. Results for piperacillin-tazobactam (1/8) expressed in piperacillin values (3) Results for ampicillin-sulbactam (1/2) expressed in ampicillin values

ⱕ0.25–⬎256 2–⬎256 1–⬎128 0.5–⬎128 ⱕ0.25–128 ⱕ0.25–64 ⱕ0.12–16 ⱕ0.5–32 ⱕ0.25–64 ⱕ0.25–64 2–128 ⱕ0.25–8 ⱕ0.25–2 ⱕ0.25–1 ⱕ0.25–0.5 ⱕ0.12–2 1–4 ⱕ0.25–2

32 32 32 8 2 1 0.5 2 ⱕ0.25 8 8 4 ⱕ0.25 ⱕ0.25 ⱕ0.25 0.25 1 ⱕ0.25

128 ⬎256 ⬎128 ⬎128 64 8 2 8 8 32 16 8 0.25 0.25 0.25 0.25 4 1

35.7 17.8 21.4 51.7 75.8 90.1 99.1 98.2 61.6

37.6 17.0 4.5 18.7 4.5 0 0 0.9 7.1

26.7 65.2 74.1 29.6 19.7 9.9 0.9 0.9 31.3

3.1. Activity of piperacillin-tazobactam against Enterobacteriaceae

Piperacillin-tazobactam Ampicillin-sulbactam Cefuroxime Ceftriaxone Ceftazidime Cefepime Imipenem Amikacin Levofloxacin Piperacillin-tazobactam Ampicillin-sulbactam Cefuroxime Ceftriaxone Ceftazidime Cefepime Imipenem Amikacin Levofloxacin amp C producers group: Enterobacter spp (61) Citrobacter freundii (14) Serratia spp (18) Morganella spp (11) Providencia spp (8) (total 112)

531

The activity of piperacillin-tazobactam against the 418 Enterobacteriaceae isolates is shown in Table 2. Imipenem was the most active drug in terms of susceptibility (meropenem was not tested) followed by amikacin (92.2% susceptibility). Levofloxacin, as representative of the fluoroquinolones group, notably showed slight activity, as only 63.2% isolates were susceptible. Piperacillin-tazobactam activity against E.coli, K.pneumoniae, P.mirabilis and amp C putative producers is shown in Table 2. Regarding class A beta lactamase producers, comparative results for phenotypical ESBLs producers and non producers are presented. The percentage of ESBL producers was very high: 25.5% of E.coli; 44% of K.pneumoniae and 56% of P.mirabilis. ESBLs isolates were similarly distributed with no significant differences between centers from different Argentinean provinces. The most frequent specimen sources from which ESBLs isolates were obtained were: abscesses, intra abdominal infections and surgical wounds. Notably ESBLs producers (particularly P.mirabilis and E.coli isolates) showed high resistance to levofloxacin. Genotypic identification of ESBLs is still in process, but according to the initial results available it remains clear that no TEM-derivative enzyme was present and most isolates contained the CTX-M-2 ESBL enzyme, which may account for the higher activity of ceftazidime as compared to cefotaxime. As expected imipenem showed full susceptibility (4 ␮g/ml was the highest MIC). Strikingly, P.mirabils isolates producing ESBLs were 100% susceptible against piperacillin-tazobactam while its activity against K.pneumoniae and E.coli ESBL producers was lower. When each isolate is considered for susceptibility against amikacin and piperacillin-tazobactam, the activity of the combination amikacin-piperacillin-tazobactam was found to be higher than 90%. Results for amp C producers isolates showed excellent susceptibility results for imipenem (99.1%), amikacin (98.2%) and cefepime (90.1%) and only a modest activity for piperacillin-tazobactam. 3.2. Activity of piperacillin-tazobactam against nonfermenter Gram-negative bacilli The results obtained from non-fermenter Gram-negative bacilli are shown in Table 3 for 100 P.aeruginosa and 60 Acinetobacter spp. Piperacillin-tazobactam showed the highest susceptibility values (78%) against P.aeruginosa, followed by imipenem (70.0%) and amikacin (68.0%). Twenty-two percent of these isolates were multiresistant (data not shown) and 20/22 were inhibited by colistin at a concentration of ⱕ4 ␮g/ml (data not shown). The susceptibility to levofloxacin was poor (47.0% susceptibility; MIC90 of 32 ␮g/ml). Imipenem was the most active drug against Acinetobacter spp isolates (MIC90: 16 ␮g/ml and 85.0% susceptibility). Eight out of the 60 isolates were

532

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

Table 3 Activity of piperacillin-tazobactam against non fermenter gram negative bacilli Organisms (n tested)

MIC (␮g/ml)

Antibacterial agent

Range Piperacillin-tazobactam(2)

Pseudomonas aeruginosa (100)

Ceftazidime Cefepime Imipenem Amikacin Levofloxacin Piperacillin-tazobactam Ampicillin-sulbactam(3) Ceftriaxone Ceftazidime Cefepime Imipenem Amikacin Levofloxacin

Acinetobacter spp (60)

Percentage by category 50%

90%

S(1)

I(1)

R(1)

0.5–⬎256

16

128

78.0

0

22.0

ⱕ0.25–⬎128 ⱕ0.25–⬎128 ⱕ0.12–⬎64 1–⬎256 ⱕ0.25–⬎128 ⱕ0.25–⬎256 ⱕ0.5–⬎256 1–⬎128 1–⬎128 ⱕ0.25–128 ⱕ0.12–⬎64 2–⬎256 ⱕ0.25–⬎128

8 8 2 4 8 64 32 ⬎128 32 16 1 32 8

32 32 16 64 32 ⬎256 64 ⬎128 ⬎128 64 16 ⬎256 32

62.0 61.0 70.0 68.0 47.0 21.7 31.7 7.1 23.3 36.7 85.0 35.0 16.7

14.0 20.0 12.0 5.0 5.0 40.0 10.0 11.7 40.0 51.8 5.0 15.0 28.3

24.0 19.0 18.0 27.0 48.0 38.3 58.3 81.2 36.7 11.5 10.0 50.0 55.0

(1)

S: susceptible; I: intermediate; R: resistant. Results for piperacillin-tazobactam (1/8) expressed in piperacillin values (3) Results for ampicillin-sulbactam (1/2) expressed in ampicillin values (2)

resistant to all the antibacterial tested except to colistin at 4 ␮g/ml (data not shown). Amikacin showed poor activity against Acinetobacter spp (35.0%) as well as ampicillinsulbactam (31.7% susceptibility) the latter being recommended for the treatment of infections due to Acinetobacter spp.

3.4. Activity of piperacillin-tazobactam against Grampositive cocci Table 5 lists the activity of piperacillin-tazobactam against Gram-positive cocci. Out of 113 S.aureus isolates, 38 (33.6%) were oxacillin resistant and 75 (66.4%) oxacillin susceptible. Only the data coresponding to the latter group is shown, since piperacillin-tazobactam, as other beta lactams is considered clinically inactive against oxacillin resistant strains independently of the MIC results. The MIC90 for piperacillin-tazobactam against oxacillin susceptible isolates was 2 ␮g/ml. This value is fourfold lower than that for ceftriaxone. Resistance to erythromycin was 21.6% with a MIC90 value of 8 ␮g/ml. Imipenem showed the highest activity en terms of MIC values. Linezolid resulted 100%

3.3. Activity of piperacillin-tazobactam against H.influenzae The susceptibility of 30 H.influenzae isolates for piperacillin-tazobactam was 100% (Table 4). Notably only 16.6% of our isolates were ampicillin resistant. Piperacillintazobactam showed an excellent activity (MIC90 ⱕ0.25 ␮g/ml).

Table 4 Activity of piperacillin-tazobactam against Haemophilus influenzae isolates Organisms (n tested)

(4)

Haemophilus influenzae (30)

(1)

MIC (␮g/ml)

Antibacterial agent

(2)

Piperacillin-tazobactam Ampicillin-sulbactam(3) Cefuroxime Ceftriaxone Ceftazidime Cefepime Imipenem Levofloxacin

Range

50%

90%

ⱕ0.25–0.5 ⱕ0.5–2 ⱕ0.25–4 ⱕ0.25–2 ⱕ0.25–0.5 ⱕ0.25–2 ⱕ0.12–0.5 ⱕ0.25–2

ⱕ0.25 ⱕ0.5 0.5 ⱕ0.25 ⱕ0.25 ⱕ0.25 ⱕ0.12 ⱕ0.25

ⱕ0.25 1 4 ⱕ0.25 ⱕ0.25 ⱕ0.25 0.25 ⱕ0.25

S: susceptible; I: intermediate; R: resistant. Results for piperacillin-tazobactam (1/8) expressed in piperacillin values (3) Results for ampicillin-sulbactam (1/2) expressed in ampicillin values. (4) 16.6% were ampicillin resistant all of then beta lactamase producers (2)

Percentage by category S(1)

100 100 100 100 100 100 100 100

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

533

Table 5 Activity of piperacillin-tazobactam against gram positive cocci Organisms (n tested)

Antibacterial agent

MIC (␮g/ml) Range

Staphylococcus aureus, methicillin susceptible (75)

Coagulase negative Staphylococci, methicillin susceptible (38)

Enterococci(3) (150)

Streptococcus pneumoniae (100)(4)

Viridans group streptococci (34)

Piperacillin-tazobactam(2)

Percentage by category 50%

ⱕ0.25–2

1

90%

S(1)

I(1)

R(1)

2

100

0

0

Penicillin Ceftriaxone Imipenem Vancomycin Linezolid Erythromycin Clindamycin Levofloxacin Piperacillin-tazobactam

ⱕ0.015–⬎32 2–16 ⱕ0.03–1 0.25–2 ⱕ0.12–4 ⱕ0.12–⬎64 ⱕ0.12–16 ⱕ0.25–2 ⱕ0.25–2

4 4 ⱕ0.03 1 2 ⱕ0.12 ⱕ0.12 ⱕ0.25 ⱕ0.25

32 8 0.26 1 4 8 0.25 0.5 2

9.3 94.7 100 100 100 88.0 97.4 100 100

0 5.3 0 0 0 5.3 1.3 0 0

90.7 0 0 0 0 6.7 1.3 0 0

Penicillin Ceftriaxone Imipenem Vancomycin Linezolid Erythromycin Clindamycin Levofloxacin Piperacillin-tazobactam Penicillin Imipenem Vancomycin Linezolid Erythromycin Levofloxacin Piperacillin-tazobactam

ⱕ0.015–8 ⱕ0.06–16 ⱕ0.03–0.5 ⱕ0.12–2 ⱕ0.12–2 ⱕ0.12–⬎64 ⱕ0.12–⬎64 ⱕ0.25–4 0.5–⬎128 0.12–⬎32 ⱕ0.03–⬎64 0.25–⬎64 0.5–8 ⱕ0.12–⬎64 ⱕ0.25–⬎64 ⱕ0.25–8

0.25 2 ⱕ0.03 1 1 ⱕ0.12 ⱕ0.12 ⱕ0.25 4 4 2 1 1 8 2 ⱕ0.25

2 8 0.12 2 2 8 0.25 0.5 16 16 8 4 2 ⬎64 32 0.25

9.1 97.3 100 100 100 73.0 94.6 97.3 — 88.0 — 98.0 100 18.0 56.0 —

0 2.7 0 0 0 5.4 0 2.7 — — — — — 23.0 8.0 —

90.9 0 0 0 0 21.6 5.4 0 — 12.0 — 2.0 0 59.0 36.0 —

Penicillin Imipenem Vancomycin Linezolid Ceftriaxone Erythromycin Clindamycin Levofloxacin Piperacillin-tazobactam

ⱕ0.015–4 ⱕ0.03–0.5 ⱕ0.12–1 0.5–8 ⱕ0.06–2 ⱕ0.12–⬎64 ⱕ0.12–8 ⱕ0.25–16 ⱕ0.25–16

ⱕ0.015 ⱕ0.03 0.12 1 ⱕ0.06 ⱕ0.12 ⱕ0.12 0.5 ⱕ0.25

0.12 ⱕ0.03 0.5 2 0.25 2 ⱕ0.12 2 4

77.0 100 100 100 91.0 85.0 95.0 91.0 —

15.0 0 0 0 9.0 0 3.0 3.0 —

8.0 0 0 0 0 15.0 2.0 6.0 —

ⱕ0.015–⬎32 ⱕ0.06–32 ⱕ0.12–2 ⱕ0.12–4 ⱕ0.03–1 ⱕ0.12–4 ⱕ0.12–4 ⱕ0.25–4

0.03 ⱕ0.06 0.5 1 ⱕ0.03 ⱕ0.12 ⱕ0.12 1

1 2 1 2 0.25 0.5 0.5 2

67.6 85.2 97.1 91.2 — 76.4 79.4 97.1

17.6 7.4 — — — 11.7 5.8 2.9

14.8 7.4 2.9 8.8 — 11.9 14.8 0

Penicillin Ceftriaxone Vancomycin Linezolid Imipenem Erythromycin Clindamycin Levofloxacin (1)

S: susceptible; I: intermediate; R: resistant. Results for piperacillin-tazobactam (1/8) expressed in piperacillin values. (3) E.faecalis (137), E.faecium (12), E.avium (1). (4) Results based on respiratory tract break points when pertinent. (2)

active but the MIC90 (4 ␮g/ml) was in the upper limit of the susceptibility break point. Against 141 coagulase negative staphylococci, 73% were oxacillin resistant. Table 5 shows the results obtained for

oxacillin susceptible isolates, which do not exhibit a significant difference with the data presented for S.aureus, except higher resistance against erythromycin. Only 12 out of 150 enterococci tested were classified as

534

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

E.faecium. Notably, only 3 isolates, 2 E.faecium and 1 E.faecalis were resistant to vancomycin. The highest susceptibility values were shown by linezolid (100%) and vancomycin (98%). Piperacillin-tazobactam and imipenem showed remarkable activity (MIC90 16 and 8 ug/ml respectively). On the other hand, poor activity was found for levofloxacin (MIC90 32 ␮g/ml; 56% susceptibility). The activity of piperacillin-tazobactam against pneumococci is controversial, since no break points are presented in the 2002 NCCLS documment (NCCLS, 2002). Considering recent arguments presented by Johnson et al. (Johnson et al., 2002) stating that an 8-16 ␮g/ml break-point could be considered for piperacillin-tazobactam against pneumococci, then only one strain (8 ug/ml) could have been considered as resistant and the activity was excellent (MIC90: 0.25 ug/ml). Thirty-four viridans group streptococci were included in this study. Our isolates showed lower penicillin resistance values than those presented by Johnson et al. (Johnson et al., 2002), 27 of our isolates were identified as S.mitis. Vancomycin, linezolid, imipenem and levofloxacin were the most active drugs. The susceptibility of the viridans group isolates were ⬍80% for erythromycin and clindamycin. Again no NCCLS break points exists for viridans streptococci. However piperacillin-tazobactam showed good activity (MIC90: 4 ug/ml)

4. Discussion As far as we are aware, this is the first study performed in Argentina including as much as 1200 isolates from hospitalized patients in 18 centers located in 11 different Argentinean cities from 5 different geographic areas. This study was designed to determine the susceptibility of piperacillin-tazobactam as compared to other antimicrobials, against a selected number of bacterial strains, supposed to be the more frequent isolates recovered in Argentina from hospitalized patients, particularly in ICU. Therefore results cannot be accounted for incidence, but susceptibility percentages and MIC values can be considered. As expected E. coli were prevalently recovered from urine specimens (42%) followed by blood cultures (20%) (data non shown). The activity of piperacillin-tazobactam against all E. coli isolates was excellent, only 3.4% of the strains were resistant in contrast to more than 20% resistant isolates observed for third and fourth generation cephalosporins. E.coli high rate of resistance to levofloxacin (35.7%) was an unexpected finding, probably due to the overuse of quinolones in Argentina (Casellas et al., 1994; Guzma´ n et al., 2000). Of concern is the finding that 25.5% of the E. coli isolates were putative ESBL producers. This figure is considerably higher when compared to similar findings in U.S.A (Peterson, 2001). The isolates were kept to determine the ESBL genotype but as MICs were much lower for ceftazidime than for cefotaxime, most isolates can

be presumed CTX-M2 producers, which is prevalent in Argentina (Rossi et al., 1999; Casellas, 2001; Casellas, 1999). polymerase chain reaction (PCR) studies performed by now have confirmed that most of our strains are genotypically CTX-M-2 producers (Radice M, Quinteros M, personal communication, 2002). However, MIC90 for ceftazidime was 64 ␮g/ml suggesting participation of other enzymes (SHV2, SHV5-PER2) (Casellas, 1999; Casellas, 2001). Unexpectedly putative ESBL producer E. coli isolates were equally distributed among patients in ICU than among those in other hospital units (data non shown), but the isolation of such strains was related to previous antibacterial administration. K. pneumoniae isolates were also collected prevalently from urine and blood (29% each), but a significant number (17%) was obtained from lower respiratory tract and most (80%) corresponded to patients staying in ICU (data non shown). Putative ESBL producers were detected in 56% of the isolates, a high figure if compared to that from Argentinean surveillance programs carried out during 1999-2000 (SADEBAC, 2002). MICs values suggest that most isolates are cefotaximase producers (probably of the CTX-M-2 type). Piperacillin-tazobactam was active against 100% of ␤-lactamase non-producer isolates, however 33% of putative ESBL producers were resistant. An important inoculum effect was observed for cefepime as we previously showed for cefpirome (Casellas et al., 1989). This inoculum effect was much lower for piperacillin-tazobactam than for fourth generation cephalosporins as recently published in USA (Smith Moland et al., 2002). Oppositely to the results that we obtained with E. coli isolates, K. pneumoniae strains showed only a 10% resistance level to levofloxacin. Surprisingly as much as 32% out of the 50 P. mirabilis included in the study were phenotypicaly putative ESBL producers. Cefotaximase producers were prevalent (MIC90 for ceftazidime 4 ␮g/ml and for cefotaxime/ceftriaxone ⬎128 ␮g/ml). Previous results (Radice M, Quinteros M, personal communication, 2002) have shown that most of these isolates are CTX-M-2 producers. However these strains were susceptible to piperacillin-tazobactam suggesting a lower production of the enzyme in this species or a stronger inhibition by tazobactam as compared to K. pneumoniae isolates. A striking resistance to levofloxacin (93.8%) among ESBL producers was observed. Unfortunately NCCLS does not recommend the investigation of ESBLs in P. mirabilis isolates. Our feeling is that all the Ambler class A ␤-lactamases can develop ESBL producer strains by point mutations and should be investigated as well as E. coli and K. pneumoniae. Blood cultures were the most important source for E.cloacae (47.5%) (data non shown). Considering a MIC value for ceftazidime ⬎64 ␮g/ml as criteria for amp C derepression (Livermore, 1995), 17.5% of the E. cloacae isolates, (the most prevalent species among Ambler class C ␤-lactamase producers) were probably derepressed for its production. Interestingly the data are coincident with the

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

resistance found in piperacillin-tazobactam (17.5%), showing that piperacillin-tazobactam is active against basal and induced amp C producers but not against derepressed strains. Cefepime had a slightly higher activity (10% resistance), but it is less affected by derepression. Levofloxacin showed only a 12.5% resistance rate. Among the 418 strains of Enterobacteriaceae included in the study the following conclusions can be drawn: even levofloxacin is slightly less active than ciprofloxacin against Enterobacteriaceae, our results probably indicate that a resistance mechanism to fluorquinolonesis is present in those isolates. Therefore fluoroquinolones might not be recommended for the empiric treatment of severe hospital infectious in our country, considering that intermediate and resistant isolates account for as much as 36.8% of the isolates. Moreover ciprofloxacin and probably levofloxacin are related to the selection of ESBL producers (Paterson et al., 2000; Ko et al., 2002). As a consequence of either ESBL or amp C beta lactamase production, resistance to ceftriaxone/cefotaxime was very high (34.6%). These drugs have also shown to be responsible for the selection of ESBL producers (Casellas, 1999; Paterson et al., 2000; Ko et al., 2002). It is recommended to avoid their use in ICU whenever possible (Casellas, 2001). The activity of amikacin was remarkable. Only 4% of the isolates were resistant. As amikacin cannot be used alone due to its inactivity in purulent collections, acid pH and anaerobic medium, the drug is frequently used in Latin American countries in severe infections in combination with other antibacterial drugs. Piperacillin-tazobactam appears to be a good drug to be associated with amikacin. No imipenem resistant Enterobacteriaceae isolates were detected in our investigation. The same results were observed with meropenem, which was assayed at the Reference Center (data not shown). The activity of piperacillin-tazobactam and imipenem against P.aeruginosa was similar. Meropenem was not included in the study, but determinations at the Reference Center showed 84% susceptibility rate for this carbapenem. Colistin (not included in the study) was tested for 54 isolates at the Reference Center and 97% of the isolates showed an MIC of ⱕ4 ␮g/ml. About 40% isolates were resistant to both ceftazidime and cefepime. Levofloxacin was poorly active (47% susceptibility; MIC90: 32ug/ml) as well as amikacin (58% susceptibility; MIC90: 64 ␮g/ml) Resistance to Acinetobacter spp was outstanding as only 85% of the isolates were susceptible to imipenem and 15% were resistant to all the antibacterials tested. All other antibacterials showed less than 40% activity including ampicillin-sulbactam. Forty-five of the 60 isolates were tested for susceptibility to colistin at the Reference Center and 43 (95%) were inhibited by ⱕ4 ␮g/ml (data not shown). Sulbactam did not result effective in our study, as it was previously suggested for Argentinean isolates (Casellas et al., 1997) Imipenem was absolutely ineffective against S.maltophilia isolates probably due to the carbapenemase produced

535

by this species (Livermore, 1995). Other beta lactams were also inactive. Only levofloxacin and cotrimoxazole (data non shown) were active, even though 15% of the strains were resistant to levofloxacin. Our results are consistent with those previously found in South America for H.influenzae (Guzma´ n et al., 2000). Production of beta lactamases in this species is infrequent in South America, as compared to other countries. The isolates were 100% susceptible to piperacillin-tazobactam. Regarding Gram-positive cocci oxacillin susceptible staphylococci were 100% susceptible to piperacillin-tazobactam. Among S. aureus MRSA represented the 51.3% of the isolates and MRCNS accounted for 73% of coagulase negative isolates. This high percentage of oxacillin resistance has been continuously increasing over the last five years in our country probably due to a similar increase in the use of invasive and prosthetic devices (Johnson et al., 2001). Figures are in accordance to those found in the surveillance program of the Argentinean Society for Clinical Bacteriology, SADEBAC (SADEBAC, 2002). MRSA and MRCNS are considered clinically resistant to all beta lactams by international consensus. The rates found for penicillin resistance in S.pneumoniae (15% intermediate and 8% resistant) are lower than those found in other continents and similar to those found in other Latin American countries (Rossi et al., 2000). It must be considered that only 17% of the patients included in the study were pediatrics and that no otitis media isolates were included which is the main source of pneumococci with diminished susceptibility to beta lactams. At present there is agreement that resistance to penicillin is only related to meningitis and does not predict failures in respiratory acquired infections (Heffelfinger, 2000). Considering the NCCLS (NCCLS, 2002) readjusted break points for ampicillin in respiratory tract infections, a strain with a 2 ␮g/ml MIC is interpreted as susceptible; those with a 4 ␮g/ml MIC, intermediate, and those with ⱖ8 ␮g/ml as resistant. Ampicillin susceptibility was tested by E test for 78 of the 100 isolates included in the study and all of them were found to be susceptible when considering break points for respiratory tract infections, except one strain which was intermediate (4 ␮g/ml). That strain had a MIC of 8 ␮g/ml for piperacillintazobactam. NCCLS do not give any criteria for piperacillin susceptibility for pneumococci. Piperacillin is an aminopenicillin derivate and reaches much higher concentrations than ampicillin in serum and tissues and moreover piperacillin-tazobactam is recommended for the treatment of pneumonia in ICU (Peterson, 2001; Johnson et al., 2002). If we consider a conservative break point of 8 ␮g/ml for piperacillin-tazobactam, as suggested by Johnson et al. then 100% of the strains included in our study could have been considered as susceptible. Moreover the MIC90 found for piperacillin-tazobactam was 0.25 ␮g/ml. We did not found resistant isolates to levofloxacin in our study and considering “respiratory” break points all the isolates should have been considered susceptible to ceftriaxone.

536

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537

Our results confirm the results of the SADEBAC surveillance program (SADEBAC, 2002) showing that glycopeptide resistance in enterococci is still a sporadic event in Argentina, as well as it has been shown in other South American countries (Guzman et al., 2000; Rossi et al., 2000). Only three (all of them E. faecium) out of 150 enterococcal isolates were vancomycin resistant and all were linezolid susceptible. The vancomycin resistant isolates were tested at the Reference Center for teicoplanin resistance and showed MICs of 32 ␮g/ml and 64 ␮g/ml suggesting a van A mechanism. Interestingly, only 12 enterococci (8%) were identified as E. faecium, a finding that can explain the low incidence of glycopeptide resistance in our country. Levofloxacin exhibited low activity against enterococci. The viridans group streptococci is frequently disregarded, however it is of great importance, not only due to its etiologic role in bacterial endocarditis, but also in bacteriemic episodes in neutropenic patients. Our finding that only 67.6% of the isolates were susceptible to penicillin and 82.3% to ceftriaxone (drugs commonly used to treat viridans group severe infections) is significant. However our isolates showed penicillin resistance values lower than those presented by Johnson et al. (Johnson et al., 2002). Notably, MIC90 for piperacillin-tazobactam against these isolates was low (4 ␮g/ml). The result of our study showed that piperacillin-tazobactam has a very important activity against the prevalent pathogens in hospital-acquired infections in Argentina. The only important pitfalls are some ESBL producers, amp C derepressed isolates, MRSA and Acinetobacter spp. This behavior justifies the empiric use of piperacillin-tazobactam with the addition of amikacin or vancomycin (according to the presumption of MRSA implication) in the empiric treatment of severe infections particularly in ICU patients who had not received prolonged previous antibacterial therapy. With this strategy the highly active carbapenems can be preserved for multiresistant strains, preventing resistance development to these antibacterial agents (Peterson, 2001; Paterson et al., 2000; Ko, 2002). It must be considered that even piperacillin-tazobactam has not proved to be an important selector for ESBL producers (Paterson et al., 2001) we do not recommend the use of that antibacterial drug for the treatment of infections proven to be due to ESBL producer isolates. Acknowledgments The authors are grateful to Wyeth Laboratories for the support given to this investigation. References Casellas, J. M., & Goldberg, M. (1989). Incidence of strains producing extended spectrum beta lactmases in Argentina. Infection 17, 434 – 436.

Casellas, J. M., Guzma´ n, M., & Pinto, M. E. (1994). The sleeping giant. Antimicrobial resistance. Infect Dis Clin North Am 8, 29 – 45. Casellas, J. M., & Dana, R. (1997). Is sulbactam really the answer for treatment of infections caused by multiresistant strains of Acinetobacter spp.? Enferm Infecc Microbiol Clin 15, 335–36 (in Spanish). Casellas, J. M. (1999). South America, a different continent with different beta-lactamases. International Congress of Chemotherapy, Abst. P3. Birminghan. UK. Casellas, J. M. (2001). Bacterial resı´stance due to extended-spectrum beta lactamases production: A global and Latin American perspective at the hospital scenario. Infec Dis in Clinical Practice, Supl, Nov, 12–16 (in Spanish). Catchpole, C. R., Andrews, J. M., Brenwald, N., & Wise, R. (1997). A reassessment of the in-vitro activity of colistin sulphomethate sodium. J Antimicrob Chemother 39, 255–260. Guzma´ n, M., Casellas, J. M., & Sader, H. (2000). Bacterial resistance in Latin America. The giant is awakening. Infect Dis Clin North Am 14, 67– 81. Heffelfinger, J. D., Dowell, S. F., Jorgensen, J. H., Klugman, K. P., Mabry, L. R., Musher, D. M., Plouffe, J. F., Rakowsky, A., Schuchat, A., & Whitney, C. G. (2000). Management of community-acquired pneumonia in the era of pneumococcal resistance: a report from the DrugResistant Streptococcus pneumoniae Therapeutic Working Group. Arch Intern Med 160, 1399 –1408. Jarlier, V., Nicolas, M., Fournier, G., & Philippon, A. (1988). Extended broad-spectrum beta-lactamases conferring transferable resistance to newer beta-lactam agents in Enterobacteriaceae. Rev Infec Dis 10, 867– 878. Johnson, A. P., Livermore, D. M., & Tillotson, G. S. (2001). Antimicrobial susceptibility of Gram-positive bacteria: what’s current, what’s anticipated? J Hosp Infect 49 (Suppl A), S3–11 Review. Johnson, D. M., Biedenbach, D. J., & Jones, R. N. (2002). Potency and antimicrobial spectrum update for piperacillin/tazobactam (2000): emphasis on its activity against resistant organism populations and generally untested species causing community-acquired respiratory tract infections. Diagn Microbiol Infect Dis 43, 49 – 60. Joshi, M., Bernstein, J., Solomkin, J, Wester, B. A., & Kuye, O. (1999). Piperacillin/tazobactam plus tobramycin versus ceftazidime plus tobramycin for the treatment of patients with nosocomial lower respiratory tract infection. Piperacillin/tazobactam Nosocomial Pneumonia Study Group. J Antimicrob Chemother 43, 389 –397. Ko, W. C., Paterson, D. L., Sagnimeni, A. J., Hansen, D. S., Von Gottberg, A., Mohapatra, S., Casellas, J. M., Goossens, H., Mulazimoglu, L., Trenholme, G., Klugman, K. P., McCormack, J. G., & Yu, V. L. (2002). Community-acquiered Klebsiella pneumoniae bacteremias: Global differences in clinical patterns. Emerg Infec Dis 8 (2), 160 –166. Livermore, D. (1995). Beta-lactamases in laboratory and clinical resistance. Clinical Microbiol Rev 8, 557–584. Me´ ndez, E., Morano, S., Mendosa, A., Copes, A., Mollerach, A., Galarza, P., Pagano, I., Oviedo, C., & Fiorito, S. (1999). Estudio de los fenotipos resistentes de Neisseria gonorrhoeae en u hospital de Santa Fe. Journal of the Biochemical and Biological Sciences of the Litoral University, Santa Fe 3, 31–39 (in Spanish). Montero, A., Ariza, J., Corbella, X., Dome´ nech, A., Cabellos, C., Ayats, J., Tubau, F., Ardanuy, C., & Gudiol, F. (2002). Efficaci of colistin versus beta lactams, aminoglycosides and rifampin as monotherapy in a mouse model of pneumoniae caused by multiresistant Acinetobacter baumannii. Antimicrob Ag Chemother 46, 1946 –1952. Murray, P. R., Baron, E. J., Pfaller, M. A., Tenover, F., Yolken, R. H. (1999). Manual of Clinical microbiology 7th. Edition. American Society for Microbiology. NCCLS (2002). National Committee for Laboratory Standards. Performance standards for antimicrobials susceptibility testing. M100-S11. 21 (1). Wayne (PA), USA. Paterson, D. L., Mulazinoglu, L., Casellas, J. M., Ko, W. C., Goossens, H., Von Gottberg, A., Mohapatra, S., Trenholme, G. M., Klugman, K. P.,

J.M. Casellas / Diagnostic Microbiology and Infectious Disease 47 (2003) 527–537 McCormack, J. G., & Yu, V. L. (2002). Epidemiology of ciprofloxacin resistance and its relationship to extended-spectrum beta-lactamase production in Klebsiella pneumoniae isolates causing bacteremias. Clin Infect Dis 30, 473– 478. Peterson, L. R. (2001). New resı´stance mechanisms are emerging. Consequences on the appropiate use of antimicrobials. Infec Dis in Clinical Practice, Supl. Nov, 4-11 (in Spanish). Rossi, A., Tokumoto, M., Galas, M., Soloaga, R., & Corso, A. (1999). Monitoring antibiotic resistance in Argentina. The WHONET program, 1995-1996. Rev Panam Salud Publica 6 (4), 234 –241. Spanish.

537

Rossi, F., Casellas, J. M., & Guzma´ n, M. (2000). Antimicrobial resistance in gram positive cocci. The role of oxazolidinonas. Infect Dis in Clinical Practice 1, 1–36 (in Spanish). SADEBAC. (2002). Committee for Antibacterial Resistance of SADEBAC Results of the 2001 SIR Surveillance Program. Bulletin of the Argentinean Society for Microbiology. No 1 (in Spanish). Smith Moland, E., Black, J. A., Ourada, J., Reisbig, M. D., Hanson, N. D., & Thomson, K. S. (2002). Occurrence of newer beta lactamases in Klebsiella pneumoniae isolates from 24 U.S. hospitals. Antimicrob Agents Chemther 46, 3837–3842.