Antibiotic resistance and molecular typing of Pseudomonas aeruginosa: focus on imipenem

Antibiotic resistance and molecular typing of Pseudomonas aeruginosa: focus on imipenem

Ana Lúcia Peixoto de Freitasand Afonso Luis Barth

Federal University of Rio Grande do Sul, Pharmacy School, Clinical Hospital of Porto Alegre, Cardiology Institute, Porto Alegre, RS

Catholic University of Pelotas, Pharmacy School, Pelotas, RS, Brazil

^{Address to correspondence}

ABSTRACT

Susceptibility tests by disk diffusion and by E-test and molecular typing by macrorestriction analysis were performed to determine the relatedness of Pseudomonas aeruginosa isolates from three distinct hospitals. The resistance profile of 124 isolates to 8 antimicrobial agents was determined in three different hospitals, in Porto Alegre, Brazil. Frequencies of susceptibility ranged from 43.9% for carbenicillin to 87.7% for ceftazidime. Cross-resistance data of imipenem-resistant isolates indicated that most (70%) were also resistant to carbenicillin, although 30% remained susceptible to ceftazidime and cefepime. In general, susceptibility profiles were not able to determine relatedness among isolates of P. aeruginosa. On the other hand, molecular typing by macrorestriction analysis demonstrated high discriminatory power and identified 66 strains among 72 isolates of P. aeruginosa. Imipenem-susceptible isolates were all different. However, identical clones of imipenem-resistant isolates were found in two of the hospitals, despite variable response to other antibiotics. No clustering of infection among the different medical centers was observed. In conclusion, clones of P. aeruginosa did not spread among the different hospitals in our city even though related isolates of imipenem-resistant P. aeruginosa were found.

Key Words: Pseudomonas aeruginosa, antibiotic resistance, imipenem.

Despite improvements in antibiotic therapy, Pseudomonas aeruginosa remains as one of the most prominent Gram-negative bacteria causing hospital-associated infections. P. aeruginosa is intrinsically resistant to a number of antimicrobial agents, frequently including multiple classes of antimicrobial agents [1]. Carbapenens are potent agents for the treatment of infections due to multiresistant pseudomonads. However, nosocomial isolates may easily develop resistance to carbapenens due to reduced uptake of the drug, which leads to outbreaks of multiresistant/carbapenen-resistant strains [2-4]. It is therefore important to develop surveillance programs to determine the epidemiological situation of P. aeruginosa in different hospitals [5].

Outbreaks caused by multiresistant P. aeruginosa have been reported in various nosocomial settings, such as in individual intensive care units (ICU) or other units within a hospital. These outbreaks can be brief or can persist for an extended period of time, with the same strains being identified over a one year period [3,4,6-8].

Typing techniques that establish clonal relationships between individual isolates in hospital settings are warranted in order to recognize nosocomial transmission and hence to guide infection control practice. DNA-based techniques, such as macrorestriction analysis by pulsed field gel electrophoresis, have been successfully applied to the epidemiological study of P. aeruginosa [9].

We examined P. aeruginosa strains isolated from hospitalized patients in Porto Alegre, Brazil, to determine their susceptibility to antibiotics and their epidemiological relatedness.

Materials and Methods

A total of 124 nonreplicate Pseudomonas aeruginosa clinical isolates were recovered from patients from 3 hospitals (HCPA, ISCM, and HSL), in Porto Alegre, Brazil, between September 1998 and June 1999.

Overall, 54% of the isolates were obtained from the respiratory tract, 28% from urine, 6.5% from blood and 11.5% from a variety of other sources (catheters, abdominal secretions, skin).

Identification was based upon production of characteristic pigments (blue and green). Additional biochemical tests used to identify P. aeruginosa included: oxidase, oxidation of glucose on OF-medium, arginine and nitrate, and growth in cetrimide agar [1]. Isolates were checked for purity by plating on MacConkey agar before the susceptibility tests and typing.

The susceptibility tests were performed with the agar disk diffusion method, according to NCCLS guidelines [10]. The following antimicrobial agents were tested: amikacin, aztreonam, carbenicillin, ceftazidime, cefepime, ciprofloxacin, imipenem and ticarcillin (BBL-Becton Dickinson Microbiology, Cockeysville). The resistance rates determined in this study include both intermediate and fully resistant lines. The isolates were classified into antibiotypes according to the susceptibility patterns. Isolates that had an identical response to all antimicrobial agents were considered to belong to the same antibiotype, while a single difference in resistance to any of the antimicrobial agents was considered a distinct profile.

In order to avoid false designation of resistance to imipenem, due to drug degradation during storage [11], isolates that tested resistant by the disk diffusion method were confirmed by the E-test (AB Biodisk, Solna, Sweden).

Macrorestricton analysis, followed by PFGE, was performed as described previously [12]. Briefly, the bacteria was embedded in agarose blocks, digested with restriction endonuclease SpeI (Gibco BRL, USA) and electrophoresed in a CHEF DR II apparatus (Bio-Rad, Richmond, USA). The gels were run at 14^ºC, 5.9 V cm^-1, for 22 hours, with a switch time of 5 to 50 s. A lambda ladder (48.5 Kb, Sigma, USA) was used as a molecular weight marker. The gels were stained with ethidium bromide and the image was acquired on a Chemilmager transiluminator 4000 (Alpha Inntech Corporation). Comparison of macrorestriction profiles was performed by visual analysis.

Isolates were considered to be part of a major clone whenever they had identical macrorestriction profiles (PFGE). They were considered related (subtypes) when there were 1 to 6 differences in fragments (bands) between profiles. Isolates with more than 6 different fragments were considered distinct or unrelated strains [13]. Each major clone was coded with a capital letter and a number was added to each subtype.

Results and Discussion

Among the 124 P. aeruginosa isolates the degree of susceptibility varied from 43.9% for carbenicillin to 87.7% for ceftazidime. Cefepime, ceftazidime and imipenem proved to be effective against most P. aeruginosa clinical isolates (> 80% susceptibility).

The rates of resistance of P. aeruginosa to amikacin, aztreonam, ciprofloxacin, ceftazidime and cefepime were similar to those described in other studies for isolates from Brazil [14-16] and other countries [4, 5, 17-20]. However, in our study ticarcillin and carbenicillin were less effective when compared with most other surveillance studies. Although many reports suggest increasing multiresistance in P. aeruginosa [2-4, 8] only four isolates (3.2%) were resistant to all of the antimicrobial agents, which contrasts with 37% susceptibility to all agents.

We found 20 strains to be resistant to imipenem, most of them with MIC ³ 32 mg/mL. In fact, only one of these isolates had a MIC = 16 mg/mL, which would also be considered resistant according to NCCLS [10]. We therefore found 100% correlation between disk diffusion and E-testing for all imipenem resistant isolates, as also described by Burns et al. [21].

The rate of susceptibility to imipenem (83.7%) was similar to rates found in the SENTRY antimicrobial surveillance program in Brazil and in Latin America [14, 15]. However, a study of 26 Brazilian isolates of P. aeruginosa found less than 70% susceptibility to imipenem [20]. Data from other countries indicates susceptibility to imipenem ranging from 79% [4] to 95% [20].

Cross-resistance data are useful to indicate alternative drugs for treatment. In our study, ceftazidime was the most active antimicrobial agent, and less than 50% of the ceftazidime-resistant isolates were also resistant to imipenem or cefepime (Table 1). Additionally, only 30% of the isolates resistant to imipenem were also resistant to cephalosporins, which indicates that ceftazidime and cefepime may be good options for the treatment of infections due to imipenem-resistant P. aeruginosa.

Bouza and cols. [5] provided cross-resistance data of P. aeruginosa isolated in Spain. They found cross-resistance rates to most of the agents similar to those of our study; imipenem-resistant P. aeruginosa were more resistant to ticarcillin and amikacin in our study. However, the percentage of individual resistance to ticarcillin and amikacin agents was also higher in our population than in the Spanish isolates. Among ceftazidime-resistant isolates Bouza et al. [5] found 80% to be resistant to cefepime, while we found only 47% (Table 1). This may be important since individual resistance to each agent was very similar in the two studies.

The high incidence of cross-resistance to antibiotics of different classes observed in our study is probably due to a combination of multiple, unrelated resistance mechanisms.

We typed 72 isolates by macrorestriction analysis (including those resistant to imipenem, ceftazidime or cefepime) and found 66 distinct strains. There were no common clones for any susceptibility pattern, unless they were also resistant to imipenem.

Several imipenem-resistant isolates were indistinguishable by macrorestriction analysis, despite variations in their susceptibility to other antimicrobial agents. The 20 imipenem-resistant isolates (9 from ISCM, 7 from HSL and 4 from HCPA) had 12 major types of DNA (Table 2).

There were 5 major types of imipenem-resistant P. aeruginosa in ISCM. Two isolates were of type A and were recovered within a period of 4 days, from patients hospitalized in different units. Their susceptibility profiles were very similar, differing only by resistance to aztreonam. Type B included 3 urine isolates, with the same susceptibility profile, but recovered during a longer period of time (18 days). The type C isolates were obtained during a two day period and had a similar susceptibility profile, but were isolated from patients hospitalized in different units. The remaining 2 isolates from ISCM had distinct molecular types. Although they were recovered on the same day, there was no epidemiological correlation between the patients (Table 2). Although laboratory data indicated similar clones of P. aeruginosa in ISCM, thus suggesting patient-to-patient transmission, the epidemiological data do not corroborate this conclusion, since the patients were admitted in distinct units and the isolates were obtained at different periods of time. These isolates may have some characteristic which improves their permanence and dissemination in a hospital environment.

Most of the isolates from HSL (5 out of 7) belonged to the same genotype (figure 1). These isolates were obtained from the respiratory tract of patients in the same unit during a relatively long period (28 days). Unrelated strains were also obtained during this period of time, including one isolated from a patient in the same unit. Almost all isolates of HSL had distinct susceptibility profiles, regardless of their genotype. Since identification of type D P. aeruginosa from HSL was restricted to this period it seems that this clone came from a short duration outbreak.

Isolates from HCPA were all unrelated, according to either DNA or susceptibility profiles even though two isolates were recovered within a two day period from sputum of patients admitted to the same unit.

It is well known that macrorestriction analysis is the most powerful tool for epidemiological studies of P. aeruginosa, while phenotypic markers can lead to misinterpretation of strain relatedness [9, 22]. We found that in some cases both DNA and susceptibility profiles lead to the same conclusion while in others the results were conflicting. For instance, we observed agreement in the case of isolates 21 and 27, which belonged to the same major type (B) and displayed the same antibiotype (Table 2).

However, discrepancy was more frequent than agreement between the different methods. In most cases, susceptibility patterns were not able to indicate strain identity, as seen for HSL isolates belonging to clone D and ISCM isolates belonging to clone A. Furthermore, the susceptibility pattern has low discriminatory power for establishing strain identity. In our study the antibiotype "a" was found in all hospitals, despite the absence of epidemiological relationships among isolates.

We analyzed isolates from distinct hospitals in the same city, during a relatively long period of time and found a lower frequency of resistance to antibiotics used against P. aeruginosa than the rates described in other studies [2-4, 8, 20].

Contrary to other studies that describe major clones of resistant P. aeruginosa spread among several hospitals within a country [3, 4, 7], molecular typing of our isolates revealed mainly unique strains. However, we noted a clustering of infections due to the same genotype among imipenem resistant P. aeruginosa isolates, suggesting possible transmission inside a hospital but not between different medical centers.

Received on 25 July 2001

Revised 15 February 2002

This study was sponsored by "Fundo de Incentivo à Pesquisa e Eventos do Hospital de Clínicas de Porto Alegre" and "Pró-Reitoria de Pós-Graduação da Universidade Federal do Rio Grande do Sul"

• 1. Kiska D.L., Gilligan P.H. Pseudomonas and Bulkholderia In: Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH. Manual of Clinical Microbiology. 7^th ed. Washington: American Society for Microbiology 1999 p.517-25.
• 2. Iaconis P.J., Pitkin D.H., Sheikh W., Nadler H.L. Comparison of antibacterial activities of meropenem and six other antimicrobial against Pseudomonas aeruginosa isolates from North American studies and clinical trials. Clin Infect Dis 1997;24:191-6.
• 3. Tsakris A., Pournaras S., Woodsord N., et al. Outbreak of infections caused by Pseudomonas aeruginosa producing VIM-1 carbapenemase in Greece. J Clin Microbiol 2000;38:1290-2.
• 4. Tassios P.T., Gennimata V., Maniatis A.M., et al. Emergence of multidrug resistance in ubiquitious and dominant Pseudomonas aeruginosa serogroup O:11. J Clin Microbiol 1998;36:879-901.
• 5. Bouza E., Garcia-Garrote F., Cercenado E., et al. Pseudomonas aeruginosa: a survey of resistance in 136 hospitals in Spain. Antimicrob Agents Chemother 1999;43:981-2.
• 6. Bert F., Maubec E., Bruneau B., et al. Multi-resistant Pseudomonas aeruginosa outbreak associated with contaminated tap water in a neurosurgery intensive care unit. J Hosp Infec 1998;39:53-62.
• 7. Panzig B., Schröder G., Gründing M. A large outbreak of multiresistant Pseudomonas aeruginosa strains in north-eastern Germany. JAC 1999;43:415-8.
• 8. Hsueh P., Teng L., Yang P., et al. persistence of a multidrug-resistant Pseudomonas aeruginosa clone in an intensive care burn unit. J Clin Microbiol. 1998;36:1347-51.
• 9. Grundmann H., Schneider C., Hartung D., et al. Discriminatory power of three DNA-based typing techniques for Pseudomonas aeruginosa J Clin Microbiol 1995;33:528-34.
• ¹⁰
  NCCLS - National Committee for Clinical Laboratory Standards. Performance Standards for Antimicrobial disk Susceptibility Tests Pennsylvania, 1999
• 11. Carmeli Y., Eichelberger K., Soja D.,et al. failure of quality control measures to prevent reporting false resistance to imipenem, resulting in a pseudo-outbreak of imipenem-resistant Pseudomonas aeruginosa J Clin Microbiol 1998;36:595-7.
• 12. Kaufmann M.E. Pulsed-field gel electrophoresis. In: Woddford N, Johnson AP. Molecular Bacteriology. Protocols and clinical applications. New Jersey: Humana Press Inc. 1998 p.33-50.
• 13. Tenover F., Arbeit R., Goering R., et al. Interpreting chromossomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995;33:2233-9.
• 14. Sader H.S., Pfaller M.A., Jones R.N., et al. Bacterial pathogens isolated from patients with bloodstream infections in Latin America, 1997: frequency of occurence and antimicrobial susceptibility patterns from the SENTRY Antimicrobial Surveillance Program Braz J Infec Dis 1999;3:97-110.
• 15. Sader H.S., Sampaio J.L.M., Zoccoli C., Jones R.N. Results of the 1997 SENTRY Antimicrobial Surveillance Program in three Brazilian medical centers Braz J Infec Dis 1999;3:63-79.
• 16. Sader H.S., Cerbara E.F., Luz D., Hashimoto A. Evaluation of the cephalosporins, cefepime, cefpirome and ceftazidime, against clinical isolates of imipenem-resistant Pseudomonas aeruginosa Braz J Infec Dis 1999;3:231-7.
• 17. Blondeau J.M., Suter M.E., Borsos S., Misfeld C. Canadian Pseudomonas aeruginosa susceptibility study from 48 medical centers. J Antimicrob Agents 1998;10:297-302.
• 18. Bonfiglio G., Carciotto V., Russo G., et al. Antibiotic resistance in Pseudomonas aeruginosa: an Italian survey. J Antimicrob Agents 1998;41:307-10.
• 19. Bonfiglio G., Marchetti F. In vitro activity of ceftazidime, cefepime and imipenem on 1005 Pseudomonas aeruginosa clinical isolates either susceptible or resistant to beta-lactams Chemother 2000;46:229-34.
• 20. Jones R.N., Pfaller M.A., Marshall A.S., et al. Antimicrobial activity of 12 broad-spectrum agents tested against 270 nosocomial blood stream infection isolates caused by non-enteric gram-negative bacilli: occurrence of resistance, molecular epidemiology and screening for metallo-enzymes. Diagn Microbiol Infect Dis 1997;29:187-92.
• 21. Burns J.L., Saiman L., Whittier S.,et al. Comparison of agar diffusion methodologies for antimicrobial susceptibility testing of Pseudomonas aeruginosa isolates from cystic fibrosis patients. J Clin Microbiol 2000;38:1818-22.
• 22. Grothues D., Koopmann U., Hardt H., Tümmler B. Genome fingerprinting of Pseudomonas aeruginosa indicates colonization of cystic fibrosis siblings with closely related strains. J Clin Microbiol 1988;26:1973-7.

Publication Dates

• Publication in this collection
  18 Mar 2003
• Date of issue
  Feb 2002

History

• Received
  25 July 2001
• Accepted
  15 Feb 2002