Respiratory tract pathogens isolated from patients hospitalized with suspected pneumonia in Latin America: frequency of occurrence and antimicrobial susceptibility profile: results from the SENTRY Antimicrobial Surveillance Program (1997-2000)

Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311

www.elsevier.com/locate/diagmicrobio

Respiratory tract pathogens isolated from patients hospitalized with suspected pneumonia in Latin America: frequency of occurrence and antimicrobial susceptibility profile: results from the SENTRY Antimicrobial Surveillance Program (1997-2000) Ana C. Galesa,*, He´lio S. Sadera, Ronald N. Jonesb,c a

Laborato´rio Especial de Microbiologia Clı´nica, Division of Infectious Diseases, Universidade Federal de Sa˜o Paulo, Brazil b The JONES Group/JMI Laboratories, North Liberty, IA, USA c Tufts University School of Medicine, Boston, MA, USA Received 10 July 2002; accepted 19 September 2002

Abstract In spite of the recent medical advances, lower respiratory tract infections are still the most frequent infectious causes of mortality worldwide. The objective of this study was to determine the frequency of occurrence and antimicrobial susceptibility of bacterial isolates collected from hospitalized patients with pneumonia in Latin American medical centers during the first four years of the SENTRY Program. The five most frequently isolated species were (n/%): Pseudomonas aeruginosa (659/26.3%), Staphylococcus aureus (582/23.3%), Klebsiella pneumoniae (255/10.2%), Acinetobacter spp. (239/9.6%), and Enterobacter spp. (134/5.4%). P. aeruginosa demonstrated high rates of resistance to most of the antimicrobials tested. Against P. aeruginosa, the most active agents were meropenem (MIC50, 1 ␮g/ml; 71.6% susceptible), amikacin (MIC50, 4 ␮g/ml; 71.0% susceptible), and piperacillin/tazobactam (MIC50, 16 ␮g/ml; 70.4% susceptible). Imipenem (MIC50, 1 ␮g/ml; 84.1% susceptible) and meropenem (MIC50, 2 ␮g/ml; 84.9% susceptible) were the most active agents against Acinetobacter spp. followed by tetracycline (MIC50, ⱕ4 ␮g/ml; 52.3% susceptible). Although the broad-spectrum cephalosporins had demonstrated excellent in vitro activity against Klebsiella pneumoniae isolates (MIC50s range, ⱕ 0.12 to 0.25 ␮g/ml), elevated rates of resistance (46.3%-58.5%) were observed. Approximately 44.0% and 29.0% of K. pneumoniae and E. coli isolates were considered ESBL producers based on NCCLS criteria, respectively. Overall, the prevalence of methicillin-resistant S. aureus was 46.2%. The most active drugs against this pathogen were vancomycin, teicoplanin, linezolid and quinupristin/dalfopristin. In summary, the SENTRY Antimicrobial Surveillance Program has detected a high prevalence of methicillin-resistant S. aureus and multidrug resistant non-fermentative Gramnegative bacilli isolated from respiratory tract specimens of hospitalized patients with pneumonia in Latin America. Our results emphasize the importance of local surveillance programs in correctly guiding empiric therapy and local intervention programs in attempt to reduce antimicrobial resistance. © 2002 Elsevier Science Inc. All rights reserved.

1. Introduction Community-acquired pneumonia (CAP) remains a common cause of serious disease. In spite of the recent advances in the field, including the identification of new pathogens, new methods of microbial detection, new antimicrobial agents, and effective vaccines, lower respiratory infections are still the most frequent causes among infectious mortality worldwide. In 1998, the World Health Organization re-

* Corresponding author. Tel.: ⫹(5511) 50812819; fax: ⫹(5511) 5571 5180. E-mail address: [email protected] (A.C. Gales).

ported ⬎ 3.7 million deaths due to lower respiratory infections [WHO, 1998]. In the United States, it has been estimated that more than 5 million cases of CAP occur annually, and nearly 1.1 million of these require hospitalization [Garibaldi, 1985; Niederman et al., 1998]. In the outpatient setting, the mortality rate of pneumonia remains low (⬍ 5%); however, when patients with CAP require hospitalization, the mortality increases to an average of 12% [Marston et al., 1997]. According to the National Nosocomial Infections Surveillance (NNIS) report, pneumonia accounted for approximately 15% of all hospital-associated infections and was the second most common nosocomial infection after those

0732-8893/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 7 3 2 - 8 8 9 3 ( 0 2 ) 0 0 4 9 9 - 6

302

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311

of the urinary tract [CDC, 1997a; NNIS, 2001]. In the intensive care units, it usually ranks number one [Richards et al., 1999; NNIS, 2001]. Nosocomial pneumonia has been associated with high fatality rates. Crude mortality rates of 20-50% and attributable mortality rates of 30-33% have been reported [Leu et al., 1989; Lynch & Niederman, 1997; Mandell & Campbell, 1998]. Analyses of pneumonia-associated morbidity have shown that pneumonia could prolong hospitalization by 4-9 days increasing the direct cost of hospital stay [Leu et al., 1989]. Because of its reported frequency, associated high fatality rate, and attendant costs, nosocomial pneumonia is a major infection control problem. Early clinical suspicion and prompt empiric antimicrobial therapy are mandatory in patients with severe pneumonia since increase in the morbidity and mortality rates have also been associated with delay in the initiation of optimal antimicrobial therapy [CDC, 1997a; Lynch & Niederman, 1997; Mandell & Campbell, 1998; Torres et al., 1990]. The identification and antimicrobial susceptibility of bacterial pathogens responsible for lower respiratory infections are usually not available to the physicians in the beginning of treatment. Thus, the results of local microbiologic surveillance studies could be useful in guiding the selection of the most adequate empiric treatment regimen. The SENTRY Antimicrobial Surveillance Program was initiated in early 1997 with the purpose of monitoring patterns of antimicrobial resistance among various types of infection. Currently, the program includes more than 80 medical centers in North America, South America, Europe, Africa, Asia, and Western Pacific regions. The data generated by this program assume relevant importance principally in developing regions such as Latin America, where extensive surveillance studies have not been previously conducted. The main purpose of this study was to report the frequency of occurrence and antimicrobial susceptibility of bacterial isolates collected from hospitalized patients with pneumonia in the Latin American medical centers during the first four years of the SENTRY Program.

2. Materials and Methods 2.1. Clinical isolate collection The participating medical centers were instructed to submit 100 consecutive bacterial pathogens considered to be the cause of pneumonia in hospitalized patients, i.e., nosocomial acquired pneumonias or serious community-acquired pneumonia, which required hospitalization for treatment. Isolates were determined to be clinically significant based on local participant criteria. Only one clinical isolate per patient was evaluated. All isolates were shipped to the monitoring laboratory on charcoal transport swabs.

2.2. Bacterial strains A total of 2,502 bacterial isolates were collected from the Latin American medical centers between 1997 and 2000 including 1,186 high-quality sputum, 1,134 invasive pulmonary and 68 pleural fluid samples. Referral of isolates began in July of each year, the beginning of the respiratory season in the South Hemisphere, and continued for six months or until 100 samples were collected. All isolates were identified at the participating institution by the routine methodology in use at each laboratory. Upon receipt at the monitoring laboratory, isolates were subcultured onto blood agar to ensure viability and purity. Confirmation of species identification was performed with the Vitek (bioMe´ rieux Vitek, St Louis, MO) or conventional methods as required. 2.3. Medical centers The participant medical centers included eleven cities in seven countries: Sa˜ o Paulo (1997-2000), Rio de Janeiro (1997-98), Floriano´ polis (1997-2000), and Porto Alegre (1999-2000) in Brazil; Buenos Aires and San Isidro in Argentina (1997-2000); Santiago in Chile (two sites; 19972000); Montevideo in Uruguay (1997); Medelin in Colombia (1997-2000); Mexico City in Mexico (two different sites; 1997-2000); and Caracas in Venezuela (1998-2000). 2.4. Susceptibility testing Antimicrobial susceptibility testing was performed and interpreted following the guidelines for reference broth microdilution method as described by the NCCLS [NCCLS, 2000 and 2002]. The minimal inhibitory concentrations (MICs) were defined as the lowest antimicrobial concentration able to totally inhibit bacterial growth. The diverse antimicrobial agents were obtained from the respective manufacturers. These agents included piperacillin, piperacillin/tazobactam, ticarcillin, ticarcillin/clavulanic acid, ceftriaxone, ceftazidime, cefepime, imipenem, meropenem, ciprofloxacin, levofloxacin, gatifloxacin, amikacin, gentamicin, tobramycin, tetracycline, and trimethoprim-sulfamethoxazole. Dry-form microdilutiuon panels and broth for inoculation were purchased from Trek Inc. (Westlake, OH, USA). Testing of quality control strains Escherichia coli ATCC 25922 and 35218, Staphylococcus aureus ATCC 29213, Pseudomonas aeruginosa ATCC 27853, Enterococcus faecalis ATCC 29212, Haemophilus influenzae ATCC 49247 and 49766, and Streptococcus pneumoniae ATCC 49619 was performed for quality assurance purposes. Isolates of Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis with increased MICs (ⱖ 2 ␮g/ml) for ceftazidime and/or ceftriaxone and/or aztreonam were considered as possible extended-spectrum ␤-lactamase (ESBL)-producing phenotypes according to NCCLS criteria [NCCLS, 2002]. Isolates of Acinetobacter spp. and Pseudo-

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311 Table 1 Frequency of the 2,502 bacterial pathogens causing pneumonia in patients hospitalized in the Latin American medical centers (SENTRY Antimicrobial Surveillance Program, 1997–2000). Rank

Organism

No. of isolates (%)

1 2 3 4 5 6 7 8 9 10 11

Pseudomonas aeruginosa Staphylococcus aureus Klebsiella pneumoniae Acinetobacter spp.a Enterobacter spp.b Streptococcus pneumoniae Escherichia coli Haemophilus influenzae Serratia spp.c Stenotrophomonas maltophilia Other 9 speciesd

659 (26.3) 582 (23.3) 255 (10.2) 239 (9.6) 134 (5.4) 122 (4.9) 114 (4.6) 98 (3.9) 71 (2.8) 41 (1.6) 187 (7.5)

a Acinetobacter spp.: A. anitratus (16), A. baumannii (181), A. calcoaceticus (13), A. lwoffii (two strains), and Acinetobacter spp. (27). b Enterobacter spp.,: E. cloacae (69), E. aerogenes (40), E. asburieae (two strains), E. gergoviae (one strain), E. sakazakii (one strain), and Enterobacter spp. (21) c Serratia spp.: Serratia marscencens (64), Serratia fonticola (one strain), and Serratia spp. (6) d Includes Enterococcus spp. (49 strains); coagulase-negative staphylococci (30 strains); Proteus spp. (35 strains); Moraxella catarrhalis (27 strains); Citrobacter spp. (20 strains); Burkholderia cepacia (15 strains); Morganella morganii (eight strains); Salmonella spp. (two strains); and Providencia spp. (one strain).

monas aeruginosa exhibiting MICs ⱖ8 ␮g/ml were considered non-susceptible to carbapenems [NCCLS, 2002]. 2.5. Statistical analysis Analysis was performed using SPSS Version 9.0 and values of p ⬍ 0.05 were considered statistically significant.

3. Results and Discussion Table 1 lists the 10 most frequent bacterial pathogens isolated from patients with pneumonia hospitalized in the Latin American medical centers between 1997 and 2000. A total of 2,502 strains were tested and the most frequent pathogen was Pseudomonas aeruginosa (26.3%) followed by Staphylococcus aureus (23.3%), Klebsiella pneumoniae

303

(10.2%), Acinetobacter spp. (9.6%), and Enterobacter spp. (5.4%). Gram-negative genera constituted nearly 70.0% of the recovered pathogens. Overall, there was no significant change in the rank order of the top five pathogens over the years with the exception of a significant increase in the prevalence of Haemophilus influenzae isolates in 1999 (p ⬍ 0.01-Table 2). The high numbers of H. influenzae isolates collected in the Argentinean and Chilean medical centers could explain this fact. Table 3 shows the occurrence of the five most frequent pathogens causing pneumonia in the distinct Latin American medical centers. In general P. aeruginosa or S. aureus was the most frequent species isolated. H. influenzae ranked among the top five pathogens in the Argentinean and Chilean medical centers. In fact, it ranked as number two in one of the Chilean medical centers. Acinetobacter spp. was encountered among the top five pathogens in medical centers located in Argentina, Brazil, Chile, Colombia, Uruguay, and Venezuela. It is worthwhile to notice that the reported distribution of etiologic agents causing pneumonia varied between medical centers located in the same country; for instance, S. pneumoniae ranked as one of the top five pathogens in just one of the Brazilian medical centers. These variations could be attributed to differences in patient population and hospital epidemiology. The distribution of pathogens in Latin America differ significantly from the SENTRY Program results from North America, where H. influenzae and S. pneumoniae consistently ranked among the five most frequent pathogens and Acinetobacter spp. was isolated in less than 3.0% of the cases [Jones et al., 2000; Mathai et al., 2001]. Nosocomial bacterial pneumonia is frequently polymicrobial and Gram-negative bacilli are usually involved [NNIS, 1996; CDC, 1997a]. However, Staphylococcus aureus and other Gram-positive cocci, including Streptococcus pneumoniae, have emerged as significant causes of pneumonia in hospitalized patients [Schleupner et al., 1992; American Thoracic Society, 1995; NNIS, 1996]. In addition, Haemophilus influenzae has been isolated from mechanically ventilated patients with pneumonia that occurs within 48-96 h after intubation [Rello et al., 1992]. The high prevalence of Gram-negative nonfermentative bacilli, Enterobacteriaceae and methicillin-resistant S. aureus in this study may indicate that the majority of our isolates origi-

Table 2 Occurrence of the six most frequent pathogens causing pneumonia in patients hospitalized in the Latin American medical centers according to the year of isolation (SENTRY Antimicrobial Surveillance Program, 1997–2000). Year (Total No. of isolates)

Organism No. of isolates (%) P. aeruginosa

S. aureus

K. pneumoniae

Acinetobacter spp.

Enterobacter spp.

H. influenzae

1997 (547) 1998 (678) 1999 (691) 2000 (586) Total (2502)

150 (27.4) 191 (28.2) 155 (22.4) 163 (27.8) 659 (26.3)

128 (23.4) 171 (25.2) 153 (22.1) 130 (22.2) 582 (23.6)

53 (9.7) 75 (11.1) 64 (9.3) 63 (10.8) 255 (10.2)

65 (11.9) 75 (11.1) 48 (6.9) 51 (8.7) 239 (9.6)

38 (6.9) 34 (5.0) 31 (4.5) 31 (5.3) 134 (5.4)

2 (0.4) 6 (0.9) 66 (9.5) 24 (4.1) 98 (3.9)

304

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311

Table 3 Occurrence of the five most frequently isolated pathogens causing pneumonia in hospitalized patients in the distinct Latin American medical centers (SENTRY Antimicrobial Surveillance Program, 1997–2000). Species

Medical centers (Nation) 39 (ARG)a

40 (ARG)a

41 (BRA)a

42 (CHI)a

43 (CHI)a

44 (COL)a

45 (MEX)a

46 (BRA)a

47 (URU)a

48 (BRA)a

49 (VEN)a

56 57 (MEX)a (BRA)a

Number of organisms (%) Gram-negative bacilli Acinetobacter spp. Citrobacter spp. E. coli Enterobacter spp. H. influenzae K. pneumoniae P. aeruginosa Serratia spp. Gram-positive cocci S. aureus CoNSc S. pneumoniae Enterococcus spp. Total number of isolates tested

55 (14.9) —b —b —b 15 (4.1) 28 (7.6) 83 (22.4) —b

—b —b 18 (6.0) —b 20 (6.6) 47 (15.6) 85 (28.1) —b

—b —b —b 15 (8.2) —b 24 (13.2) 60 (33.0) 16 (8.8)

15 (7.0) —b 14 (6.5) —b 9 (4.2) —b 31 (14.5) —b

—b —b —b 8 (3.7) 44 (20.4) 15 (6.9) 22 (10.2) —b

10 (7.0) —b —b 11 (7.7) —b 29 (20.4) 33 (23.2) —b

—b 6 (6.6) 9 (9.9) 6 (6.6) —b —b 37 (40.7) —b

25 (7.7) —b —b 21 (6.5) —b 26 (8.0) 99 (30.6) —b

13 (25.0) —b —b —b —b 4 (7.7) 8 (15.4) 3 (5.8)

70 (18.7) —b —b —b —b 34 (9.1) 131 (35.0) —b

18 (11.6) —b —b 22 (14.2) —b 25 (16.1) 27 (17.4) —b

—b —b —b —b —b —b —b —b

—b —b 3 (1.8) —b —b 12 (7.1) 42 (24.7) —b

99 (26.8) —b —b —b 370

43 (14.2) —b —b —b 302

17 (9.3) —b —b —b 182

79 (36.9) —b —b —b 214

70 (32.4) —b —b —b 216

15 (10.6) —b —b —b 142

8 (8.8) —b —b —b 91

100 (30.9) —b —b —b 324

15 (28.8) —b —b —b 52

52 (13.9) —b —b 27 (7.2) 374

10 (6.5) —b —b —b 155

2 (9.5) 7 (33.3) —b 5 (23.8) 21

72 (42.4) —b 19 (11.2) —b 170

a

ARG, Argentina; BRA, Brazil; CHI, Chile; COL, Colombia; MEX, Mexico; URU, Uruguay; VEN, Venezuela. Bacterial species not detected among the top five pathogens. c CoNS, coagulase-negative staphylococci. b

nated from patients with nosocomial pneumonia rather than patients with severe community-acquired pneumonia who needed hospitalization. The most common site of Acinetobacter nosocomial infection is the lower respiratory tract, especially in mechanically ventilated patients, where Acinetobacter spp. has been recognized as a typical pathogen in late-onset ventilator-associated pneumonia [Kollef et al., 1995; Forster & Daschner, 1998]. The isolation of Acinetobacter spp. in respiratory specimens may reflect colonization rather than infection [Struelens et al., 1993]. The high frequency of occurrence of Acinetobacter spp. isolates in this study could reflect true infections since Acinetobacter spp. has also been confirmed as an important pathogen causing bloodstream infections in Latin America [Sader et al., 1999]. The exact reason for the high prevalence of Acinetobacter infections in determinate regions has not been elucidated, but it has been associated with conditions of high humidity and temperatures, which would facilitate the maintenance and spread of Acinetobacter in the hospital environment [McDonald et al., 1999]. Antimicrobial susceptibility patterns of the four most frequent Gram-negative bacilli causing pneumonia in the Latin American medical centers are summarized in Table 4. P. aeruginosa was the most prevalent pathogen causing pneumonia in the Latin American region monitored by the SENTRY Program. In this region, based on the antimicrobial susceptibility results, the best therapeutic agents would be meropenem (MIC50, 1 ␮g/ml; 71.6% susceptible), amikacin (MIC50, 4 ␮g/ml; 71.0% susceptible), and piperacil-

lin/tazobactam (MIC50, 16 ␮g/ml; 70.4% susceptible). Although ciprofloxacin had been the most potent quinolone tested against the P. aeruginosa isolates, the susceptibility rate (59.5%) to this compound was similar to those of levofloxacin and gatifloxacin. Among the cephalosporins, cefepime (MIC50, 8 ␮g/ml; 61.2% susceptible) and ceftazidime (MIC50, 4 ␮g/ml; 61.2% susceptible) showed similar potency and spectrum against P. aeruginosa. Long periods of hospitalization, previous use of third-generation cephalosporins, and concurrent chronic obstructive pulmonary disease have been described as the most significant predictors of colonization/infection by P. aeruginosa in patients with nosocomial pneumonia [Talon et al., 1998]. Meropenem (MIC50, ⱕ0.06 ␮g/ml) and imipenem (MIC50, ⱕ0.5 ␮g/ml) were the most active agents tested against the K. pneumoniae isolates. After the carbapenems, the fluoroquinolones were the most potent compounds against K. pneumoniae. The susceptibility rates to gatifloxacin, the most active fluoroquinolone, reached 89.4%. Two isolates of K. pneumoniae resistant to carbapenems were detected in Argentinean and Brazilian medical centers, respectively. Their mechanisms of carbapenem resistance are currently being investigated. Acinetobacter spp. was the fourth most frequent Gramnegative bacilli genus isolated. Decreased activity of all compounds tested was observed among the Acinetobacter spp. isolates evaluated. Imipenem (MIC50, 1 ␮g/ml; 84.1% susceptible) and meropenem (MIC50, 2 ␮g/ml; 84.9% susceptible) were the most active agents against Acinetobacter

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311

305

Table 4 Antimicrobial activity and spectrum of drugs tested against the four most prevalent gram-negative pathogens causing pneumonia in hospitalized patients in the Latin American medical centers (SENTRY Antimicrobial Surveillance Program 1997–2000). Antimicrobial class/agent

Pathogens (No. of isolates) P. aeruginosa (659) MIC50/90 (␮g/ml)

␤-lactams Amoxicilin clavulanate Aztreonam Cefazolin Cefepime Cefoxitin Ceftazidime Ceftriaxone Cefuroxime Imipenem Meropenem Piperacillin Piperacillin/Tazobactam Ticarcillin Ticarcillin/Clavulanate Aminoglycosides Amikacin Gentamicin Tobramycin Fluoroquinolones Ciprofloxacin Gatifloxacin Levofloxacin Others Tetracycline Trimethoprim/ Sulfamethoxazolec

%Sb

K. pneumoniae (255) MIC50/90 (␮g/ml)

%Sb

Acinetobacter spp. (239) MIC50/90 (␮g/ml)

%Sb

Enterobacter spp. (134) MIC50/90 (␮g/ml)

%Sb

—a 16/⬎16 —a 8/⬎16 —a 4/⬎16 —a —a 2/⬎8 1/⬎8 32/⬎128 16/⬎64 64/⬎128 64/⬎128

—a 42.9 —a 61.2 —a 61.2 —a —a 69.2 71.6 63.9 70.4 53.4 54.3

4/⬎16 ⬍0.12/⬎16 8/⬎16 ⬍0.12/⬎16 4/32 0.5/⬎16 ⱕ0.5/⬎32 4/⬎16 0.25/0.5 ⱕ0.06/0.12 16/⬎128 4/⬎64 ⬎128/⬎128 8/⬎128

62.7 60.4 53.7 76.1 81.2 63.1 61.2 54.1 99.6 99.2 52.2 68.6 4.7 56.5

—a ⬎16/⬎16 —a ⬎16/⬎16 —a ⬎16/⬎16 ⬎32/⬎32 —a 1/⬎8 2/⬎8 ⬎128/⬎128 ⬎64/⬎64 ⬎128/⬎128 ⬎128/⬎128

—a 8.1 —a 26.4 —a 16.7 8.4 —a 84.9 84.1 10.9 16.7 11.7 13.8

⬎16/⬎16 0.25/⬎16 ⬎16/⬎16 0.12/⬎16 ⬎32/⬎32 0.5/⬎16 ⱕ0.25/⬎32 ⬎16/⬎16 0.5/2 0.06/0.12 8/⬎128 4/⬎64 8/⬎128 16/⬎128

3.0 64.9 6.0 93.3 5.2 63.4 63.4 40.3 100.0 100.0 54.5 65.7 53.7 53.0

4/⬎32 4/⬎16 1/⬎8

71.0 60.2 60.2

2/32 ⱕ1/⬎16 1/⬎8

84.7 71.8 59.8

⬎32/⬎32 ⬎8/⬎8 ⬎8/⬎8

20.5 23.8 31.1

2/32 ⱕ1/⬎16 1/⬎8

86.6 82.1 69.4

⬍0.5/⬎2 2/⬎4 1/⬎4

59.5 54.3 58.0

0.25/⬎2 0.06/4 ⱕ0.5/4

86.7 89.4 87.5

⬎2/⬎2 ⬎4/⬎4 ⬎4/⬎4

18.8 25.1 21.8

0.25/⬎2 0.06/4 ⱕ0.5/⬎4

83.6 86.6 85.8

⬎8/⬎8 ⬎1/⬎1

2.0 3.2

ⱕ4/⬎8 ⱕ0.5/⬎1

69.8 72.4

ⱕ4/⬎8 ⬎2/⬎2

52.3 23.4

ⱕ4/⬎8 ⱕ0.5/⬎1

70.9 73.5

a

Antimicrobial agent with no spectrum of activity against this pathogen. %S, percentage of susceptibility defined by the NCCLS criteria [2002]. c Isolates exhibiting MICs ⱕ 0.5/9.5 ␮g/ml were classified as susceptible to this association. b

spp. followed by tetracycline (MIC50, ⱕ4 ␮g/ml; 52.3% susceptible). Although a similar potency was observed among antimicrobial compounds belonging to the same class, tobramycin (31.1% susceptible), cefepime (26.4% susceptible), and gatifloxacin (25.1% susceptible) were the most active representatives of these classes, respectively. Acinetobacter spp. has assumed an important role as an etiologic agent of nosocomial infections, especially in Latin America. This pathogen is often resistant to multiple antimicrobial agents, making it even more difficult to treat serious infections. In addition, high mortality rates have been reported among patients who developed Acinetobacter spp. nosocomial-acquired pneumonia [Baraibar et al., 1997; Forster & Daschner, 1998]. ␤-lactam resistance rates were high among the 134 Enterobacter spp. isolates evaluated. Only 65.5% of these isolates were susceptible to piperacillin/tazobactam (MIC50, 4 ␮g/ml). The third-generation cephalosporins, ceftriaxone and ceftazidime, showed high potencies (MIC50, ⱕ 0.25 ␮g/ml and 0.5 ␮g/ml), but demonstrated elevated resistance rates (29.1% and 32.1%, respectively). Among the ␤-lactams, only cefepime and the carbapenems were active

against ⬎90% of isolates tested. This fact probably indicates that most of cephalosporin-resistant Enterobacter spp. isolates overproduce AmpC ␤-lactamases. The antimicrobial activity of the agents tested against the 582 S. aureus isolates is shown in Table 5. Overall, the prevalence of methicillin-resistant S. aureus was 53.8%. However, the resistance rates to methicillin varied from 0.0 to 78.5% among the participating medical centers. The highest rate of methicillin-resistant S. aureus was observed in Chile (77.2%) followed by Brazil (50.6%)⬎ Argentina (47.9%) ⬎ Uruguay (40.0%) ⬎ Mexico (37.5%) ⬎ Colombia ⫽ Venezuela (0.0%). The methicillin-resistance rates varied slightly among medical centers located in the same country except for the Brazilian medical centers. Medical centers located in the Brazilian South region showed resistance above 63.0%, while medical centers located in the Southeast region showed rates around 35.0%. It was already reported that isolates from patients with pneumonia demonstrated the highest overall rates of methicillin resistance independently of the geographic region [Diekema et al., 2001]. Vancomycin (MIC50, 1 ␮g/ml), teicoplanin (MIC50, 1

306

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311

Table 5 Antimicrobial activity and spectrum of drugs tested against Staphylococcus aureus, the most prevalent gram-positive pathogen, causing pneumonia in hospitalized patients in the Latin American medical centers (SENTRY Antimicrobial Surveillance Program, 1997–2000). Antimicrobial class/agent

Oxacillin Amoxicillin/clavulanate Cefazolin Ceftriaxone Cefepime Gentamicin Ciprofloxacin Gatifloxacin Erythromycin Clindamycin Quinupristin/dalfopristin Chloramphenicol Rifampin Tetracycline Trimethoprim/sulfametoxazole Teicoplanin Vancomycin Linezolid a

S. aureus (No. tested) Oxacillin-susceptible (268)

Oxacillin-resistant (314)

Total (582)

MIC50/90 (␮g/ml)

% Susc.

MIC50/90 (␮g/ml)

% Susc.

MIC50/90 (␮g/ml)

% Susc.

0.5/1 1/2 ⱕ2/ⱕ2 4/4 2/4 0.5/0.5 0.25/0.5 0.06/0.12 0.5/⬎8 0.25/0.25 0.25/0.5 8/16 0.25/0.25 ⱕ4/8 ⱕ0.5/ⱕ0.5 0.5/1 1/1 2/2

100.0 100.0 99.6 98.9 98.9 95.9 95.5 100.0 77.6 96.3 100.0 84.0 97.8 89.2 99.6 100.0 100.0 100.0

⬎8/⬎8 ⬎16/⬎16 ⬎16/⬎16 ⬎32/⬎32 ⬎16/⬎16 ⬎16/⬎16 ⬎2/⬎2 2/4 ⬎8/⬎8 ⬎8/⬎8 0.5/1 ⬎16/⬎16 2/⬎2 ⬎8/⬎8 ⬎2/⬎2 1/2 1/2 2/2

0.0 3.2a 3.2a 1.0a 2.2a 3.5 2.9 72.6 1.6 3.8 100.0 22.9 40.8 37.3 35.5 99.7 100.0 100.0

⬎8/⬎8 ⬎16/⬎16 ⬎16/⬎16 ⬎32/⬎32 ⬎16/⬎16 ⬎16/⬎16 ⬎2/⬎2 2/4 ⬎8/⬎8 ⬎8/⬎8 0.5/1 8/⬎16 ⱕ0.25/⬎2 ⱕ4/⬎8 ⱕ0.5/⬎2 1/2 1/1 2/2

46.2 47.3a 47.6a 46.0a 46.7a 46.0 45.5 85.2 36.6 46.4 100.0 51.0 67.0 61.2 64.9 99.8 100.0 100.0

Oxacillin-resistant strains should be considered resistant to all ␤-lactams in spite of in vitro susceptibility [NCCLS 2002].

␮g/ml), linezolid (MIC50, 2 ␮g/ml), and quinupristin/dalfopristin (MIC50, 0.25 ␮g/ml) were the most active compounds against S. aureus isolates, and, except for teicoplanin, inhibited 100.0% of the S. aureus isolates. Interestingly, one isolate of S. aureus, which was resistant to methicillin, showed intermediate resistance to teicoplanin (MIC, 16 ␮g/ml). It was isolated from a 84 year-old male patient who was hospitalized in the ICU from an Argentinean medical center in 1998. To our knowledge, isolation of S. aureus with reduced susceptibility to glycopeptides from Argentina has not been reported. The clinical usefulness of glycopeptides has been jeopardized by several recent reports of clinical isolates of methicillin-resistant S. aureus exhibiting reduced susceptibility to glycopeptides [CDC, 1997b; Hiramatsu et al., 1998]. Recently, four methicillinresistant S. aureus strains showing reduced susceptibility to glycopeptides were isolated from patients hospitalized in a burn unit of a Brazilian hospital. Pulsed-field gel electrophoresis characterized these isolates as belonging to the Brazilian endemic clone [Oliveira et al., 2001]. Gatifloxacin exhibited reasonable in vitro activity against the 314 methicillin-resistant S. aureus isolates tested (72.6% susceptibility). On the other hand, rifampin (40.8% susceptibility), trimethoprim/sulfamethoxazole (37.5% susceptibility) and tetracycline (37.3% susceptibility) showed lower activity against this pathogen. The 268 methicillin-susceptible S. aureus strains remained fully susceptible to other ␤-lactams and gatifloxacin. Among the non-␤-lactam drugs, tetracycline (89.2%) demonstrated the lowest susceptibility rate against the methicillin-susceptible S. aureus followed

by gentamicin (95.9%) ⬍ rifampin (97.8%) ⬍ levofloxacin (98.3%), and ⬍ trimethoprim/sulfamethoxazole (99.6%). Overall, 25.4%, 43.9%, 35.5% of the E.coli, K. pneumoniae, and P. mirabilis isolates recovered from patients with pneumonia expressed the ESBL-phenotype according to the NCCLS criteria, respectively. Variations in the percentage of isolates showing ESBL-phenotype were noticed across the years and the medical centers (Tables 6 and 7). Among the K. pneumoniae and P. mirabilis isolates showing the ESBL phenotype, the variations of the ESBL percentage during the years did not reach statistical significance. In contrast, a trend toward decreasing numbers of ESBL-producing E. coli phenotypes was noticed between 1997 and 1998 (p ⬍ 0.01). The increase in the percentage of E. coli exhibiting the ESBL phenotype in the year 1999 compared to 1998 was not statistically significant (p ⬎ 0.05). The comparison of the ESBL-phenotype occurrence rates among the Latin American medical centers is shown in Table 7. Generally, the ESBL phenotype rates varied among medical centers located in the same country. The ESBL phenotype was more frequently detected among K. pneumoniae isolates, except for the medical centers 40 (Argentina) and 45 (Mexico), where the ESBL phenotype was more frequently isolated among E. coli strains. Different ESBLs hydrolyze ␤-lactams to varying degrees [Bush, Jacoby, Medeiros, 1995]. The distribution of the preferred ␤-lactam substrates for the detection of ESBL isolates in the Latin American region is shown in Table 8. Overall, independent of country and species, the ␤-lactams tested, az-

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311 Table 6 Occurrence and trends in the Latin American region for respiratory isolates having the ESBL phenotypea (SENTRY Antimicrobial Surveillance Program, 1997– 00). Species

Year

Total No. of isolates

No. of isolates showing ESBL phenotypea (%)

Escherichia coli

1997 1998 1999 2000 Total 1997 1998 1999 2000 Total 1997 1998 1999 2000 Total

33 24 34 23 114 53 75 64 63 255 3 11 10 7 31

10 (30.3) 3 (12.5) 10 (29.4) 6 (26.0) 29 (25.4) 20 (37.7) 25 (33.3) 31 (48.4) 36 (57.1) 112 (43.9) 0 (0) 5 (45.5) 5 (50.0) 1 (14.3) 11 (35.5)

Klebsiella pneumoniae

Proteus mirabilis

a The ESBL phenotype was defined according to the NCCLS criteria [2002].

treonam, ceftazidime, and ceftriaxone, showed similar ability for detecting ESBL phenotype isolates. Ceftazidime was able to detect 147 of 152 (96.7%) possible ESBL producers, while both aztreonam and ceftriaxone were able to detect 146 of 152 (96.0%) possible ESBL producers. Among E. coli isolates, aztreonam demonstrated the best performance for detecting the ESBL phenotype in Argentina and Brazil, respectively. On the other hand, ceftriaxone was the best

307

substrate for detection of the ESBL phenotype among P. mirabilis in the Argentinean medical centers. It is interesting to note that Latin American centers exhibiting a high prevalence of ESBL producers among K. pneumoniae isolates did not show high rates for P. mirabilis. The only centers that had a high prevalence of ESBL among P. mirabilis were those which had experienced outbreaks. Two separate clusters were detected previously. They occurred in Chile and Argentina during 1997 and 1998, respectively. The evaluation of the cluster strains showed that they shared a related pulsed field-gel electrophoresis (PFGE) and a unique ␤-lactamase isoeletric (pI) pattern (5.4, 7.6, and 7.9). The pI 7.9 was attributed to CTX-ase-M-2 production, an ESBL with a high specificity for ceftriaxone and aztreonam. Although the ESBL-producing P. mirabilis strains isolated from Chile had shown a distinct PFGE pattern from those of the Argentinean isolates, they also had a band focusing at pI 7.9, which probably indicates the spread of CTX-ase-M-2 [Gales et al., 1999]. ESBL derivatives from the CTX-M genes generally hydrolyze aztreonam and ceftriaxone at higher rates than ceftazidime [Bauernfeind et al., 1992]. This fact probably explains why ceftriaxone was the best substrate for detection of the ESBL phenotype among P. mirabilis strains isolated in Argentina. The number of ESBL phenotype isolates detected in E. coli and Klebsiella pneumoniae by the SENTRY Antimicrobial Surveillance Program varied widely among the Latin American medical centers. Most of them had a high frequency of occurrence of the ESBL

Table 7 Comparison of ESBL-phenotype occurrence rates among the Latin American medical centers (SENTRY Antimicrobial Surveillance Program, 1997–2000). Medical Center

Argentina 39 40 Brazil 41 46 48 57 Chile 42 43 Colombia 44 Mexico 45 56 Uruguay 47 Venezuela 49

E. coli

K. pneumoniae

P. mirabilis

Total No. of isolates

No. of ESBL (%)

Total No. of isolates

No. of ESBL (%)

Total No. of isolates

No. of ESBL (%)

28 10 18 37 0 20 11 0 17 14 3

10 (35.7) 1 (10.0) 9 (50.0) 11 (29.7) 0 (0) 6 (30.0) 5 (45.5) 0 (0) 4 (23.5) 4 (28.6) 0 (0)

75 28 47 96 24 26 34 12 23 8 15

34 (45.3) 17 (60.7) 17 (36.2) 52 (54.2) 6 (25.0) 17 (65.4) 22 (64.7) 7 (58.3) 9 (39.1) 7 (87.5) 2 (13.3)

16 5 11 1 0 0 0 0 10 6 4

5 (31.3) 0 (0) 5 (45.5) 1 (100.0) 0 (0) 0 (0) 1 (100.0) 0 (0) 4 (40.0) 3 (50.0) 2 (50.0)

7 9 9 0

1 (14.3) 2 (22.2) 2 (22.2) 0 (0)

29 9 9 0

10 (34.5) 1 (11.1) 1 (11.1) 0 (0)

0 0 0 0

0 (0) 0 (0) 0 (0) 0 (0)

2

0 (0)

04

1 (25.0)

0

0 (0)

1 (7.1)

25

5 (20.0)

4

0 (0)

14

308

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311

Table 8 Distribution of the preferred ␤-lactam substrate for detection of ESBL isolates in the Latin American region (SENTRY Antimicrobial Surveillance Program, 1997–2000). Nation

E. coli ESBL No. /Total No.

Argentina Aztreonan Ceftriaxone Ceftazidime Brazil Aztreonan Ceftriaxone Ceftazidime Chile Aztreonan Ceftriaxone Ceftazidime Colombia Aztreonan Ceftriaxone Ceftazidime Mexico Aztreonan Ceftriaxone Ceftazidime Uruguay Aztreonan Ceftriaxone Ceftazidime Venezuela Aztreonan Ceftriaxone Ceftazidime a

K. pneumoniae No. detected (%)

10/28

ESBL No. / Total No.

P. mirabilis No. detected (%)

34/75 10 (100.0) 9 (90.0) 10 (100.0)

11/37

52/96

1/1

9/23

1/7

5/10

10/29

0/0

1/9

—a —a —a 0/0 —a —a —a

1 (100.0) 1 (100.0) 1 (100.0) 1/4

—a —a —a 1/14

4 (80.0) 4 (80.0) 2 (40.0)

10 (100.0) 9 (90.0) 10 (100.0)

2 (100.0) 2 (100.0) 2 (100.0) 0/2

1 (100.0) 1 (100.0) 1 (100.0)

9 (100.0) 9 (100.0) 9 (100.0)

1 (100.0) 1 (100.0) 1 (100.0) 2/9

4 (80.0) 5 (100.0) 3 (60.0)

51 (98.0) 52 (100.0) 52 (100.0)

4 (100.0) 4 (100.0) 4 (100.0)

0/0 —a —a —a

1 (100.0) 1 (100.0) 1 (100.0) 5/25

1 (100.0) 1 (100.0) 1 (100.0)

No. detected (%)

5/16 33 (97.0) 32 (94.1) 34 (100.0)

11 (100.0) 11 (100.0) 10 (90.9) 4/17

ESBL No. / Total No.

0/4 3 (60.0) 4 (80.0) 5 (100.0)

—a —a —a

ESBL-producing Proteus mirabilis isolates were not detected.

phenotype comparable only to those reported by Asian Pacific medical centers [Winokur et al., 2001]. Non-fermentative Gram-negative bacilli, especially Acinetobacter spp. and P. aeruginosa, represent a real problem in certain geographic regions, such as Latin America. Previous surveillance studies have demonstrated geographic differences in the antimicrobial susceptibility patterns of these pathogens [Gales et al., 2001a; Gales et al., 2001b]. The occurrence of carbapenem-resistant Acinetobacter spp. and P. aeruginosa is displayed in Tables 9 and 10. Overall, 14.6% of the Acinetobacter spp. collected from the respiratory tract were resistant to carbapenems. Carbapenemresistant Acinetobacter isolates were detected in only three Latin American centers, which were located in Argentina, Brazil, and Venezuela. Although there were variations in the percentage of carbapenem resistance over the years, they were not statistically significant. Due to the number of isolates forwarded to the monitoring laboratory, clusters might have occurred in an Argentinean medical center (number 39) in 1998 and 2000, and in a Brazilian medical center (number 48) in 1997 and 1998. Epidemic clusters of multidrug resistant Acinetobacter spp. were previously reported in the Argentinean and Brazilian medical centers

[Lewis et al., 2000]. The reduction in the number of carbapenem-resistant Acinetobacter spp. isolates referred by the Brazilian medical center led us to suspect that infection control measures had been implemented in the respective medical center. In contrast to what was observed for Acinetobacter spp. isolates, P. aeruginosa strains resistant to carbapenems were encountered in all Latin American medical centers except one, with resistance rates varying from 3.0% (Colombia) to 59.5% (Brazil). The number of P. aeruginosa resistant to carbapenems increased significantly over the years, especially for meropenem (p ⬍ 0.01). The hyperproduction of AmpC ␤-lactamases coupled with an alteration in the bacterial outer membrane are probably the most common mechanisms of carbapenem resistance [Livermore, 2000]. Thus, it is expected that isolates of P. aeruginosa exhibiting resistance to imipenem show crossresistance to meropenem. However, due to differences in the potency, isolates of P. aeruginosa can be categorized as resistant to imipenem (low level of resistance) but susceptible to meropenem since the latter has been recognized as the most potent carbapenem commercially available against P. aeruginosa [Sader & Gales, 2001]. In addition, the loss of a specific porin, D2, might confer resistance only to

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311

309

Table 9 Occurrence and trends in the Latin American Region for Acinetobacter spp. and Pseudomonas aeruginosa respiratory isolates showing resistance to carbapenemsa (SENTRY Antimicrobial Surveillance Program, 1997– 00). Species

Acinetobacter spp.

P. aeruginosa

a

Year

No. of isolates resistant to carbapenemsa (%)

Total No. of isolates

Imipenem

Meropenem

1997 1998 1999 2000 Total 1997 1998

65 75 48 51 239 150 191

6 (9.2) 14 (18.7) 5 (10.4) 10 (19.6) 35 (14.6) 36 (24.0) 33 (17.3)

6 (9.2) 14 (18.7) 4 (8.3) 10 (19.6) 34 (14.2) 17 (11.3) 34 (17.8)

1999

155

42 (27.1)

36 (23.2)

2000 Total

163 659

51 (31.3) 162 (24.6)

52 (31.9) 139 (21.1)

Medical Center (No. of isolates resistant to carbapenems) 48 (6) 39 (6), 48 (7), 49 (1) 48 (1), 49 (4) 39 (9), 48 (1) 39 (15), 48 (15), 49 (5) 39 (4), 40 (4), 41 (10), 45 (7), 47 (1), 48 (10) 39 (1), 40 (4), 41 (4), 43 (1), 44 (1), 45 (3), 46 (2), 48 (16), 49 (2) 39 (7), 40 (6), 42 (1), 43 (3), 44 (1), 46 (5), 48 (16), 49 (1), 57 (2) 39 (4), 40 (5), 42 (1), 46 (2), 48 (36), 49 (3), 57 (4) 39 (16), 40 (19), 41 (14), 42 (2), 43 (4), 44 (1), 45 (10), 46 (9), 47 (1), 48 (78), 49 (6), 57 (6)

Carbapenem resistance defined by the NCCLS criteria [2002].

imipenem. In Latin America, most of P. aeruginosa isolates were resistant to both carbapenems. Isolates of P. aeruginosa resistant to meropenem but susceptible to imipenem were not commonly observed. Two mechanisms of resistance have been responsible for this phenotype: the overexpression of the efflux MexAB-OmpM system and the production of metallo-enzymes such as IMP-6, which hydrolyzes meropenem at higher rates than imipenem [Livermore, 2002]. Isolates resistant to meropenem and suscep-

tible to imipenem were found in two Latin American medical centers located in Brazil (medical center 57) and Colombia (medical center 44). The clonal dissemination of epidemic carbapenem-resistant P. aeruginosa and Acinetobacter spp. strains in the Latin American medical centers, as previously reported, might be responsible, at least in part, to the overall high rates of resistance [Gales et al., 2001a; Gales et al., 2001b; Lewis et al., 2000]. In summary, the SENTRY Antimicrobial Surveillance

Table 10 Comparison of occurrence rates of carbapenem-resistant Acinetobacter spp. (CRAS) and Pseudomonas aeruginosa (CRPA) isolates collected from patients with pneumonia in the Latin American medical centers (SENTRY Antimicrobial Surveillance Program, 1997–2000). Nation Medical Center

Argentina 39 40 Brazil 41 46 48 57 Chile 42 43 Colombia 44 Mexico 45 56 Uruguay 47 Venezuela 49 a

Acinetobacter spp. Total No. of isolates 68 55 13 110 14 25 70 1 16 15 1

P. aeruginosa a

No. of CRAS (%) Imipenem

Meropenem

15 (22.1) 15 (27.3) 0 (0) 15 (13.6) 0 0 15 (21.4) 0 0 0 0

15 (22.1) 15 (27.3) 0 (0) 14 (12.7) 0 0 14 (20.0) 0 0 0 0

Total No. of isolates

No. of CRPA (%) Imipenem

Meropenem

168 83 85 331 60 98 131 42 53 31 22

35 (20.8) 16 (19.3) 19 (22.4) 103 (31.1) 14 (23.3) 9 (9.2) 78 (59.5) 2 (4.8) 6 (11.3) 2 (6.5) 4 (18.2)

29 (17.3) 13 (15.7) 16 (18.8) 91 (27.5) 7 (11.7) 9 (9.2) 69 (52.7) 6 (14.3) 5 (9.4) 2 (6.5) 3 (13.6)

10

0

0

33

1 (3.0)

2 (6.1)

4 —

0 —

0 —

37 —

10 (27.0) —

5 (13.5) —

13

0

0

8

1 (12.5)

1 (12.5)

18

5 (27.8)

5 (27.8)

27

6 (22.2)

6 (22.2)

CRAS: carbapenem-resistant Acinetobacter spp.; CRPA: carbapenem-resistant P. aeruginosa.

310

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311

Program has detected a high prevalence of methicillinresistant S. aureus and multidrug resistant non-fermentative Gram-negative bacilli isolated from respiratory tract specimens of hospitalized patients with pneumonia in Latin American medical centers. Our results emphasize the importance of local surveillance programs in guiding empiric therapy, and local intervention programs attempting to reduce antimicrobial resistance in this geographic region.

Acknowledgments The co-authors express their gratitude to the following investigators in Latin America active during this study: J. Smayevsky (Buenos Aires, Argentina), J.M. Casellas (Buenos Aires, Argentina), J.L.M. Sampaio (Rio de Janeiro, Brazil), V. Prado (Santiago, Chile), P. Garcia (Santiago, Chile), J.A. Robledo (Mendelin, Colombia), J.S. Osornio (Vasco de Quiroga, Mexico), C.M. Zoccoli (Florianopolis, Brazil), H. Bangnulo (Montivideo, Uruguay), H. Sader (Sao Paulo, Brazil), M. Guzman (Caracas, Venezuela), J.M. Presno-Bernal (Mexico D.F., Mexico), A.L. Barth (Porto Alegre, Brazil), M.S. Rangel-Frausto (Mexico D.F., Mexico), and J. Ribeiro (Brasilia-DF, Brazil). The SENTRY Program was funded by an educational/research grant from Bristol-Myers Squibb.

References American Thoracic Society (1995). Hospital-acquired pneumonia in adults: diagnosis, assessment of severity, initial antimicrobial therapy, and preventative strategies. Am J Respir Crit Care Med, 153, 1711–1725. Baraibar, J., Correa, H., Mariscal, D., Gallego, M., Valles, J., & Rello, J. (1997). Risk factors for infection by Acinetobacter baumannii in intubated patients with nosocomial pneumonia. Chest, 112, 1050 –1054. Bauernfeind, A., Casellas, J. M., Goldberg, M., Holley, M., Jungwirth, R., Mangold, P., et al (1992). A new plasmidic cefotaximase from patients infected with Salmonella typhimurium. Infection, 20, 158 –163. Bush, K., Jacoby, G. A., & Medeiros, A. A. (1995). A functional classification scheme for ␤-lactamases and its correlation with molecular structure. Antimicrob Agents Chemother, 39, 1211–1233. Center for Diseases Control Prevention (1997a). Guidelines for prevention of nosocomial pneumonia. MMWR, 46, 6 –7. Center for Diseases Control Prevention (1997b). Staphylococcus aureus with reduced susceptibility to vancomycin–United States, 1997. MMWR, 46, 765–766. Diekema, D. J., Pfaller, M. A., Schmitz, F. J., Smayevsky, J., Bell, J., Jones, R. N., & Beach, M. (2001). Survey of infections due to Staphylococcus species: frequency of occurrence and antimicrobial susceptibility of isolates collected in the United States, Canada, Latin America, Europe, and the Western Pacific region for the SENTRY Antimicrobial Surveillance Program, 1997–1999. Clin Infect Dis, 32 (Suppl 2), S114 –132. Forster, D. H., & Daschner, F. D. (1998). Acinetobacter species as nosocomial pathogens. Eur J Clin Microbiol Infect Dis, 17, 73–77. Gales, A. C., Jones, R. N., Forward, K. R., Lin˜ ares, J., Sader, H. S., & Verhoef, J. (2001a). Emerging importance of multidrug-resistant Acinetobacter spp. and Stenotrophomonas maltophilia as pathogens in seriously ill patients: geographic patterns, epidemiologic features and

trends from the SENTRY Antimicrobial Surveillance Program (1997– 99). Clin Infect Dis, 32 (Suppl 2), 104 –113. Gales, A. C., Jones, R. N., Turnridge, J., Rennie, R., & Ramphal, R. (2001b). Characterization of Pseudomonas aeruginosa: ocurrence rates, antimicrobial susceptibility patterns, and molecular typing in the global SENTRY Antimicrobial Surveillance Program, 1997–1999. Clin Infect Dis, 32 (Suppl 2), S146 –155. Gales, A. C., Lewis, M. T., Casellas, J. M., Prado, V., Samayevsky, J., & Jones, R. N. (1999). Rapidly Emerging Extended-Spectrum ␤-lactamase (ESBL)-producing Proteus mirabilis. Results of SENTRY Antimicrobial Surveillance Program 1997–1998. In: Program of 39th Interscience Conference on Antimicrobial Agents and Chemotherapy. P. 170. Abstract 1487. San Francisco, September 26 –29. Garibaldi, R. A. (1985). Epidemiology of community-acquired respiratory tract infections in adults. Incidence, etiology, and impact. Am J Med, 78, 32–37. Hiramatsu, K. (1998). The emergence of Staphylococcus aureus with reduced susceptibility to vancomycin in Japan. Am J Med, 104 (Suppl 5A), 7–10. Jones, R. N., Croco, M. A., Kugler, K. C., Pfaller, M. A., & Beach, M. L. (2000). Respiratory tract pathogens isolated from patients hospitalized with suspected pneumonia: frequency of occurrence and antimicrobial susceptibility patterns from the SENTRY Antimicrobial Surveillance Program (United States and Canada, 1997). Diagn Microbiol Infect Dis, 3, 115–125. Kollef, M. H., Silver, P., Murphy, D. M., & Trovillion, E. (1995). The effect of late-onset ventilator-associated pneumonia in determining patient mortality. Chest, 108, 1655–1662. Leu, H. S., Kaiser, D. L., Mori, M., Woolson, R.F., & Wenzel, R. P. (1989). Hospital-acquired pneumonia. Attributable mortality and morbidity. Am J Epidemiol, 129, 1258 –1267. Lewis, M. T., Gales, A. C., Sader, H. S., Pfaller, M. A., & Jones, R. N. (2000). Frequency of occurrence and antimicrobial susceptibility patterns for pathogens isolated from Latin American patients with a diagnosis of pneumonia: Results form the SENTRY antimicrobial surveillance program (1998). Diagn Microbiol Infect Dis, 37, 63–74. Livermore, D. M. (2002). Multiple mechanisms of antimicrobial resistance in Pseudomonas aeruginosa: our worst nightmare? Clin Infect Dis, 34, 634 – 640. Lynch, J. P., & Niederman, M. S. (1997). Controversies in respiratory infections. Semin Respir Infect, 12, 69 –270. Mandell, L. A., & Campbell, G. D. Jr. (1998). Nosocomial pneumonia guidelines. An international perspective. Chest, 113, 188S–193S. Marston, B. J., Plouffe, J. F., File, T. M. Jr., Hackman, B. A., Salstrom, S. J., Lipman, H. B., Kolczak, M. S., & Breiman, R. F. (1997). Incidence of community-acquired pneumonia requiring hospitalization. Results of a population-based active surveillance Study in Ohio. The Community-Based Pneumonia Incidence Study Group. Arch Intern Med, 157, 1709 –1718. Mathai, D., Lewis, M. T., Kugler, K. C., Pfaller, M. A., & Jones, R. N. (2001). Antibacterial activity of 41 antimicrobials tested against over 2773 bacterial isolates from hospitalized patients with pneumonia: I–results from the SENTRY Antimicrobial Surveillance Program (North America, 1998). Diagn Microbiol Infect Dis, 39, 105–116. McDonald, L. C., Banerjee, S. N., & Jarvis, W. R. (1999). Seasonal variation of Acinetobacter infections: 1987–1996. Nosocomial Infections Surveillance System. Clin Infect Dis, 29, 1133–1137. National Committee for Clinical Laboratory Standards. (2000). Methods for dilution antimicrobial susceptibility test for bacteria that grow aerobically, fourth edition. Approved standard M7-A5. National Committee for Clinical Laboratory Standards. Wayne, PA: NCCLS. National Committee for Clinical Laboratory Standards. (2002). Performance standard for antimicrobial susceptibility testing. Document M100-S12. Wayne, PA: NCCLS. National Nosocomial Infections Surveillance (NNIS) Report, Dta Summary from October 1986 –April 1996, Issued May 1996. (1996). Am J Infect Control, 24, 380 –388.

A.C. Gales et al. / Diagnostic Microbiology and Infectious Disease 44 (2002) 301–311 National Nosocomial Infections Surveillance (NNIS) System Report, Data Summary from January 1992–June 2001, issued August 2001. (2001). Am J Infect Control, 29, 404 – 421. Niederman, M. S., McCombs, J. S., Unger, A. N., Kumar, A., & Popovian, R. (1998). The cost of treating community-acquired pneumonia. Clin Ther, 20, 820 – 837. Oliveira, G. A., Dell’Aquila, A. M., Masiero, R. L., Levy, C. E., Gomes, M. S., Cui, L., Hiramatsu, K., & Mamizuka, E. M. (2001). Isolation in Brazil of nosocomial Staphylococcus aureus with reduced susceptibility to vancomycin. Infect Control Hosp Epidemiol, 22, 443– 448. Rello, J., Ricart, M., Ausina, V., Net, A., & Prats, G. (1992). Pneumonia due to Haemophilus influenzae among mechanically ventilated patients: incidence, outcome, and risks factors. Chest, 102, 1562–1565. Richards, M. J., Edwards, J. R., Culver, D. H., Gaynes, R. P., & the National Nosocomial Infections Surveillance System. (1999). Nosocomial Infections in medical intensive care units in the United States. Crit Care Med, 27, 887– 892. Sader, H. S., & Gales, A. C. (2001). Emerging Strategies in infectious Diseases: Carbapenems and trinems in development. Drugs, 61, 553– 564. Sader, H. S., Pfaller, M. A., Jones, R. N., Doern, G. V., Gales, A. C., Winokur, P. L., & Kugler, K. C. (1999). Bacterial Pathogens Isolated from Patients with Bloodstream Infections in Latin America, 1997: Frequency of occurrence and antimicrobial susceptibility patterns from

311

the SENTRY Antimicrobial Surveillance Program. Braz J Infect Dis, 3, 97–110. Schleupner, C. J., & Cobb, D. K. (1992). A study of the etiologies and treatment of nosocomial pneumonia in a community-based teaching hospital. Infect Control Hosp Epidemiol, 13, 515–525. Struelens, M. J., Carlier, E., Maes, N., Serruys, E., Quint, W. G., & van Belkum, A. (1993). Nosocomial colonization and infection with multiresistant Acinetobacter baumannii: outbreak delineation using DNA macrorestriction analysis and polymerase chain reaction-fingerprinting. J Hosp Infect, 25, 15–32. Talon, D., Mulin, B., Rouget, C., Bailly, P., Thouverez, M., & Viel, J. F. (1998). Risks and routes for ventilator-associated pneumonia with Pseudomonas aeruginosa. Am J Respir Crit Care Med, 157, 978 –984. Torres, A., Aznar, R., Gatell, J. M., Jimenez, P., & Gonzales, J. (1990). Incidence, risk and prognosis factors of nosocomial pneumonia in mechanically ventilated patients. Am Rev Resp Dis, 142, 523–528. Winokur, P. L., Canton, R., Casellas, J. M., & Legakis, N. (2001). Variations in the prevalence of strains expressing an extended-spectrum beta-lactamase phenotype and characterization of isolates from Europe, the Americas, and the Western Pacific region. Clin Infect Dis, 32 (Suppl 2), 94 –103. World Heath Organization. (1998). Causes of annual death worldwide– 1998. Geneva: World Heath Organization.