Antimicrobial Susceptibilities of Gram-negative Bacilli Isolated from Adult Intensive Care Unit Patients

全文

(1)164. Antimicrobial Susceptibilities of Gramnegative Bacilli Isolated from Adult Intensive Care Unit Patients 1. 2. Wen-Liang Yu, Cheng-Wen Lin , Jen-Heisen Wang, Huei-Shan Cheng 2. Section of Infectious Diseases, Department of Internal Medicine, Committee of Nosocomial Infection Control, 1. China Medical College Hospital, and School of Medical Technology, China Medical College, Taichung, Taiwan, R.O.C.. Antibiotic resistance among pathogens isolated from patients in the intensive care units (ICUs) is a cause of major concern. A prospective in vitro survey of common gram-negative isolates obtained from patients hospitalized in the medical ICU (MICU) and surgical ICU (SICU) was undertaken to document current susceptibility patterns and the differences between both ICUs. One hundred and fifty-two isolates were obtained, including 55 Pseudomonas aeruginosa, 37 Serratia marcescens, 31 Klebsiella pneumoniae, and 29 Acinetobacter baumannii isolates. The distribution frequency of isolates between the MICU and SICU was relatively uniform without statistically significant differences. The activities of gentamicin against P. aeruginosa and S. marcescens from patients in the SICU were significantly higher than those in the MICU. Overall, the most potent compounds against P. aeruginosa included amikacin, ceftazidime, imipenem and ciprofloxacin. Most S. marcescens and K. pneumoniae isolates were only susceptible to moxalactam and imipenem, implicating the existence of extended spectrum lactamase producers. Clinicians should be aware of the multi-drug resistant strains and use the appropriate broad-spectrum cephalosporins. Based on the results of E-tests for bacteremic isolates, ciprofloxacin had good activity against S. marcescens (MIC90 = 0.19 g /mL) and K. pneumoniae (MIC90 = 1 g /mL), and should be restricted in use unless all other cephalosporins are not effective. Imipenem remained the most potent antimicrobial agent against A. baumannii and should be kept as the last resort in the strategy of antimicrobial therapy for patients in the ICUs. ( Mid Taiwan J Med 1999;4:164-70 ). Key words antimicrobial susceptibility, gram-negative bacillus, intensive care unit. INTRODUCTION. severity of illnesses of hospitalized patients. Patients in intensive care units (ICUs) have. and increasing use of invasive devices, the. nosocomial infection rates that are as high as. nosocomial infection rate in many ICUs is. five to 10 times greater than those in general. expected to rise [3]. Surveillance of antibiotic. hospital wards [1,2]. With the increase in. resistance is especially important in ICUs since most nosocomial pathogens are resistant to. Received March 5, 1999.. Revised May 10, 1999.. Accepted May 19, 1999. Address reprint requests to. Wen-Liang Yu, Section of. Infectious Diseases, Department of Internal Medicine China. multiple antibiotics [4,5]. Gram-negative organisms have become the most common pathogens among patients. Medical College Hospital, No 2, Yuh-Der Road, Taichung 404,. hospitalized in the ICUs [2]. The most common. Taiwan, R.O.C.. nosocomial aerobic gram-negative pathogens.

(2) Wen-Liang Yu, et al.. 165. isolated from the ICUs at China Medical. and S. marcescens were identified on the basis. College Hospital (CMCH) in 1997 were. of colonial morphology and standard bio-. Klebsiella pneumoniae, Pseudomonas aerug-. chemical tests [10] and were confirmed using. inosa, Serratia marcescens, and Acinetobacter. the VITEK system (BioMerieux Vitek Inc,. baumannii (unpublished data). We have. Hazelwood, Missouri, USA). P. aeruginosa and. noticed that resistant strains against imipenem. A. baumannii, identifications were confirmed. and ceftazidime are increasing among P.. using the API 20NE system (BioMerieux, La. aeruginosa and the resistant strains against. Balme-les-Grottes, France). Isolates of the same. third-generation cephalosporins (such as. strain from the same patient were not includ-. cefotaxime, ceftriaxone, cefoperazone and. ed.. aztreonam) are also increasing among K.. Antibiotic Sensitivity Testing. pneumoniae, S. marcescens and A. baumannii. Infections in the ICUs caused by resistant pathogens have become an increasing problem with respect to optimal treatment. Conventional standard methods to determine the minimum inhibitory concentrations (MICs) of antimicrobial agents against bacteria include the agar dilution method and broth dilution procedures. Epsilometer (E - test) for determining MICs is a newly established quantitative technique which is a less timeconsuming step using a commercial strip with a continuous concentration gradient of an antimicrobial agent [5]. The E - test results have compared favorably with those of the conventional standard methods in previous reports [6-8]. We conducted a study using Etest to determine the MICs of S. marcescens isolated from blood cultures [9]. In this study, we focused on recent common gram-negative bacilli obtained from various clinical specimens, such as sputum, blood, urine, pus and catheter tip, among adult ICU patients at CMCH. We evaluated in vitro activities of antibiotics commonly used against these organisms.. For the disk diffusion test, a 0.5 McFarland inoculum was placed on to Mueller-Hinton agar (BBL Sensi-Disc, Becton Dickinson), followed by various antibiotic discs [11]. The antibiotics routinely tested were gentamicin, amikacin, piperacillin, aztreonam, ceftazidime, cefotaxime, moxalactam, cefoperazone, and imipenem (BBL Sensi-Disc, Becton Dickinson). For the E-test, a strip with built-in gradient from high to low content of antibiotics in semi-log2 steps (AB Biodisk, Solna, Sweden) were used to determine the MICs of various antibiotics to clinical bacterial isolates [5]. A Mueller-Hinton agar plate was streaked with a 0.5 McFarland standardized suspension to cover the entire surface of the plate. After drying the surface of the inoculated plate, the E-test strip was placed on the surface of the 0 agar, and was incubated at 35 C for 18 to 20. hours. The MIC was read at the point where the zone of inhibition intersected the MIC scale on the strip. Control strains (Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853) were also inoculated. The 10 antibiotic E-test strips included gentamicin, amikacin, ampicillin/clavulanic acid, piperacillin, piperacillin/tazobactam, cefotaxime,. MATERIALS AND METHODS. Bacterial Isolates. aztreonam, ceftazidime, imipenem, and ciprofloxacin.. A total of 152 nosocomial isolates, including P. aeruginosa, K. pneumoniae, S. marcescens,. RESULTS. and A. baumannii, were prospectively. A total of 152 nosocomial isolates of the. obtained from patients hospitalized in the. most common gram-negative bacilli were. ICUs (surgical and medical) at CMCH from. obtained from adult patients hospitalized in. January through June 1998. K. pneumoniae. the ICUs during the study period. The.

(3) Antimicrobial Susceptibilities of Isolates in ICUs. 166. Table 1 . Antibiotic susceptibilities of common clinical isolates from various specimens of patients hospitalized in SICU and MICU P. aeruginosa Antibiotics. Gentamicin Amikacin Piperacillin Cefoperazone Cefotaxime Aztreonam Ceftazidime Moxalactam Imipenem. S. marcescens. K. pneumoniae. A. baumannii. SICU. MICU. Total. SICU. MICU. Total. SICU. MICU. Total. (n = 30). (n = 25). (n = 55). (n = 21). (n = 16). (n = 37). (n = 20). (n = 11). (n = 31) (n = 12). 22 (73)* 26 (87) 18 (60) 18 (60) 0 14 (47) 21 (70) 0 19 (63). 8 (3 2) 17 (68) 17 (68) 14 (56) 0 17 (68) 20 (80) 0 20 (80). 30 (5 5) 43 (78) 35 (64) 32 (58) 0 31 (56) 41 (75) 0 39 (71). 9 (43) 0 14 (67) 8 (50) 3 (14) 0 11 (52) 0 4 (19) 0 8 (38) 9 (56) 12 (57) 11 (69) 19 (90) 13 (81) 21 (100) 16 (100). 9 (24) 7 (35) 3 (27) 12 (59) 7 (35) 5 (45) 3 (8) 5 (25) 3 (27) 11 (30) 7 (35) 3 (27) 4 (11) 7 (35) 3 (27) 17 (46) 7 (35) 3 (27) 23 (62) 7 (35) 3 (27) 32 (86) 16 (80) 9 (82) 37 (100) 20 (100) 11 (100). 10 (32) 12 (39) 8 (26) 10 (32) 10 (32) 10 (32) 10 (32) 25 (81) 31 (100). SICU. 2 2 2 0 0 0 4 0 12. MICU. Total. (n = 17). (n = 29). (17) (17) (17). 3 (18) 3 (18) 4 (24) 0 0 0 (33) 4 (24) 0 (100) 7 (100). 5 5 6 0 0 0 8 0 29. (17) (17) (21). (28) (100). * No. in the parentheses represents percentage. p < 0.05 for comparison between the specimens from SICU and MICU using Chi-squared test or Fisher's exact test. Zero is calculated as 0.5. SICU = surgical intensive care unit; MICU = medical intensive care unit. Table 2.. Antimic robial ac tivit y of 10 c ompounds tested against c ommon bac teremic isolat es f rom adult ICU. patients using E - test P. aeruginosa. S. marcescens. (n = 10). (n = 7) Range. MIC 50 MIC 90. K. pneumoniae. A. baumannii. (n = 15). Antibiotics. MIC 50. MIC 90. Range. MIC 50. MIC 90. Gentamicin Amikacin Amoxicillin/ clavulanate Piperacillin Piperacillin/ tazobactam Cefotaxime Aztreonam Ceftazidime Imipenem Ciprofloxacin. 3 4 256. 6 12 256. 1.5 2 128. 12 32 256. 8 3 32. 256 24 96. 2 1 24. 256 256 256. 256 256 6. 256 256 16. 32 6 3. 4 3. 32 16. 2 1.5. 32 48. 256 1.5. 256 64. 3 1.5. 256 256. 256 3. 256 48. 16 3 2 8 0.25. 128 4 4 16 0.38. 8 2 1.5 1.5 0.094. 256 16 4 32 0.75. 4 32 1.5 0.5 0.125. 192 256 32 1 0.19. 0.38 0.125 0.125 0.19 0.016. 256 256 128 2 0.19. 256 64 48 0.380 0.064. (n = 10). Range. MIC 50. MIC 90. Range. 256 256 64. 24 24 6. 256 256 64. 1.5 0.75 2. 256 256 128. 256 1.5. 256 256. 64 6. 256 128. 8 0.016. 256 256. 256 8 256 1 256 1 0.5 0.19 1 0.032. 256 256 256 1.0 1.5. 12 96 32 0.38 0.38. 256 256 256 0.5 19. 3 3 1.5 0.125 0.094. 256 256 256 32 256. MIC 50 = minimal inhibitory concentrations (mg/L) of 50% isolates; MIC 90 = minimal inhibitory concentrations (mg/L) of 90% isolates;. = the value is greater than the number. majority of bacteria were isolated from the. uniform, except K. pneumoniae was less. surgical ICU (SICU) (55%). The sources of. prevalent in the MICU (Table 1).. isolates were respiratory tract (37.8%), blood. The antimicrobial susceptibilities of these. (27.6%), urinary tract (18.4%), surgical wound. various species from the ICUs are presented in. (12.2%) and intravascular catheter tip (4%).. Table 1. The sensitivities to a subset of the. The 152 microorganisms consisted of 55 P.. antimicrobial agents tested were performed. aeruginosa, 37 S. marcescens, 31 K. pneumo-. using disc diffusion test. The differences of. niae, and 29 A. baumannii isolates. The. sensitivities to individual antibiotics between. distribution frequency of isolates between the. various species from the SICU and MICU were. medical ICU (MICU) and SICU was relatively. similar, except for gentamicin against P..

(4) Wen-Liang Yu, et al.. 167. aeruginosa (p = 0.002) and S. marcescens. DISCUSSION. (p = 0.005) as well as cefoperazone against S.. Nosocomial pathogens found in the ICU. marcescens (p = 0.0006), with significantly. are usually more resistant to antimicrobial. higher activities against SICU isolates.. agents commonly used against these organ-. The rates of resistance for P. aeruginosa to. isms. There is currently little information on. amikacin, ceftazidime, and imipenem were. the prevalence of using the antibiotic. more than 20%. Among the 16 imipenem-. treatment for pathogens occurring in ICUs.. resistant strains of P. aeruginosa, only three. Based on the susceptibility results from. strains were concurrently resistant to cef-. this study, there was a surprising and alarming. tazidime, aztreonam or amikacin. Among the. finding of resistance of P. aeruginosa isolates. 14 ceftazidime-resistant P. aeruginosa isolates,. to imipenem (30%). Imipenem sensitivity was. cross-resistance occurred in aztreonam (11. similar to ceftazidime (75%) and amikacin. isolates), amikacin (4 isolates), and imipenem. (78%) and was only slightly better than. (3 isolates). In contrast, resistance of S.. piperacillin (64%). This might reflect the wide. marcescens and K. pneumoniae to the third. use of imipenem in both ICUs in our hospital.. generation cephalosporins, aminoglycosides,. It is also interesting that there was little cross-. piperacillin, and aztreonam, was markedly. resistance of P. aeruginosa between imipenem. increased in comparing with moxalactam and. and ceftazidime, reflecting different mech-. imipenem. For these organisms, the resistant. anisms of resistance to both drugs. Since no. rate to the third generation cephalosporins. single compound was consistently active. varied between 40% and 90%, whereas. against P. aeruginosa, a combination therapy,. moxalactam resistance was less than 20% and. according to local sensitivity patterns, may be. none was resistant to imipenem. Susceptibility. more effective in a documented P. aeruginosa. for A. baumannii was quite low compared. infection. For example, piperacillin or cef-. with most compounds tested and was 100%. tazidime plus gentamicin could be initiated as. susceptible to imipenem.. empirical therapy for patients with nosocomial. The activity of antimicrobial agents tested. infection in the SICU, while piperacillin,. against all our bacteremic isolates, with MIC 50 ,. aztreonam or ceftazidime plus amikacin is. MIC 90 and the range of MICs values using E-. preferred for patients who acquired infection. test are listed in Table 2. Based on our limited. in the MICU to cover P. aeruginosa infection.. study, imipenem was extremely active against. As shown in our MIC data, the activity of. the majority of isolates, except P. aeruginosa. ciprofloxacin against P. aeruginosa was similar. with a higher MIC 90 value than that of. to those of ceftazidime and aztreonam and. ceftazidime, aztreonam and ciprofloxacin.. was better than those of imipenem and. Ciprofloxacin had a good activity against P.. piperacillin-tazobactam. However, ciprofloxacin. aeruginosa (MIC 90 = 0.38), S. marcescens (. was not available in our hospital before or. MIC 90 = 0.19), and K. pneumoniae (MIC 90 = 1.00).. during the study period. Previous studies. Ciprofloxacin was as active as imipenem and. reported an increasing problem of cipro-. was more active than the third generation. floxacin resistance among gram-negative. cephalosporins (cefotaxime, ceftazidime and. bacteria after its extensive use [12,13].. aztreonam) against both S. marcescens and K.. Ciprofloxacin should be used cautiously and. pneumoniae. Ciprofloxacin was less active. continued surveillance for ciprofloxacin. than imipenem against A. baumannii.. resistance is mandatory after. Piperacillin-tazobactam showed only moderate. this drug .. the release of. activity against P. aeruginosa (MIC 90 = 16) and. In contrast to other third-generation. little activity against S. marcescens, K.. cephalosporins, moxalactam, imipenem, and. pneumoniae, and A. baumannii.. ciprofloxacin. were. active. against. S..

(5) Antimicrobial Susceptibilities of Isolates in ICUs. 168. marcescens and K. pneumoniae, even if the. our study, some strains of A. baumannii may. MIC 90 was higher than that of the wild strains.. be naturally resistant to ciprofloxacin. In. These compounds should be considered when. contrast to the good activity against S.. starting empirical antimicrobial treatment if S.. marcescens, K. pneumoniae, and P. aerug-. marcescens or K. pneumoniae is suggested to. inosa, we do not suggested ciprofloxacin to be. be present. Extended spectrum. - lactamase. used in treatimg severe A. baumannii infec-. (ESBL) producing strains were found in. tion. As a consequence, imipenem remained. Escherichia coli and K. pneumoniae [14].. the most potent agent against this problem. ESBL has the characteristic of inactivating. pathogen and should be kept as the last resort. third-generation cephalosporins but is suscep-. in the strategy of empirical antimicrobial. tible to cephamycins (such as cefoxitin,. therapy in ICUs.. cefmetazole or moxalactam) and imipenem. In conclusion, we have estimated the. [15]. S. marcescens has been reported to have Class I -lactamase, which inactivates all. prevalence of multi-drug resistant strains in. third-generation cephalosporins and cepha-. antimicrobial therapy based on the bacterial. mycins but spares imipenem [16], but it has. epidemiology and sensitivity patterns. If gram-. rarely been. reported to have ESBL. As. negative bacilli are isolated before sensitivity. observed in the present study, however, S.. testing, or are visible in clinical specimens. marcescens is highly susceptible to moxalac-. using Gram’s stain, empirical therapy with. tam (86%) with low susceptibilities to other. piperacillin or ceftazidime plus an amino-. third generation cephalosporins (11% to 62%),. glycoside may be active against most P.. which is similar to ESBL-producing phenotype.. aeruginosa strains. Moxalactam is preferred in. Furthermore, the variable susceptibilities to. empirical use for suggested S. marcescens or. third generation cephalosporins may reflect. K. pneumoniae infection. Ciprofloxacin. variable production of ESBL in vitro. If S.. should be restricted in use unless the. marcescens is really an ESBL-producer, it is. organism proves resistant to cephalosorins.. inappropriate to use any apparently suscep-. Imipenem had better be reserved for A.. tible third-generation cephalosporin to treat. baumannii, ceftazidime-resistant P. aerugi-. severe infection caused by this organism. We. nosa, and other Class I. would not recommend cefoperazone, cefo-. This study establishes a baseline for future. taxime, piperacillin or gentamicin as the initial. epidemiologic and antimicrobial studies.. ICUs and suggested the strategy of empirical. - lactamase producers.. empirical therapy in the present of S. marcescens, due to the low sensitivities in our. ACKNOWLEDGMENTS. MICU. Further investigation should be conducted to identify the presence or absence of ESBL in S. marcescens. A. baumannii has emerged as an impor-. This work was supported by a grant (DMR88-005) from the China Medical College Hospital.. tant nosocomial pathogen in ICUs and is often multi-drug resistant and difficult to treat. We found that amoxicillin-clavulanic acid, piperacillin-tazobactam, ciprofloxacin and all. REFERENCES 1.. mial infections in adult intensive care units. [Review]. third-generation cephalosporins had little activity against the strains of A. baumannii,. Am J Med 1991;91(Suppl 3B):179-84. 2.. of ciprofloxacin against eight isolates of A. baumannii with a MIC 90 of 0.25 mg/L [19]. In. Trilla A. Epidemiology of nosocomial infections in adult intensive care units. [Review] Intensive Care. which is similar with previous studies [1, 18]. In contrast, Liassine et al. reported a good activity. Weinstein RA. Epidemiology and control of nosoco-. Med 1994;20(Suppl 3):1-4. 3.. Jarvis WR, Martone WJ. Predominant pathogens in hospital infections. J Antimicrob Chemother 1992;29(Suppl A):19-24..

(6) Wen-Liang Yu, et al. 4.. 169. Snydman DR. Clinical implications of multi-drug resistance in the intensive care unit. Scand J Infect. of fluoroquinolone-resistant Escherichia coli at a. Dis 1991;78(Suppl):54-63. 5.. Sanders CC, Sanders WE Jr. Beta-lactam resistance in. cancer center. Antimicrob Agents Chemother. gram-negative bacteria: global trends and clinical. 1994;38;681-7.. impact. [Review] Clin Infect Dis 1992;15:824-39. 6.. the DNA gyrase A protein greatly reduces binding. the E test for determining. of fluoroquinolones to the gyrase-DNA complex.. antimicrobial suscepti-. jejuni, and gram-negative bacteria resistant to antimicrobial agents. J Clin Microbiol 1992;30:3243-8. Baker CN, Stocker SA, Culver DH, et al. Comparison. lactam antibiotics. [Review] Scand J Infect Dis Suppl 1991;78:7-16. 15. Jan IS, Hsueh PR, Teng LJ, et al. Antimicrobial. and agar diffusion susceptibility testing techniques. susceptibility testing for Klebsiella pneumoniae. by using a special challenge set of bacteria. J Clin. isolates resistant to extended-spectrum beta-lactam antibiotics. J Formos Med Assoc 1998;97:661-6.. Sanchez ML, Barrett MS, Jones RN. The E-Test. 16. Sanders WE Jr, Sanders CC. Inducible. -lactamases:. applied to susceptibility tests of gonococci,. clinical and epidemiologic implications for use of. multiply-resistant enterococci, and Enterobac-. newer cephalosporins. [Review] Rev Infect Dis. teriaceae producing potent beta-lactamases. Diagn Microbiol Infect Dis 1992;15:459-63. 9.. Antimicrob Agents Chemother 1993;37:126-7. 14. Livermore DM. Mechanisms of resistance to beta-. of the E test to agar dilution, broth microdilution,. Microbiol 1991;29:533-8. 8.. 13. Willmott CJR, Maxwell A. A single point mutation in. Huang M, Baker PN, Banerjee S, et al. Accuracy of bilities of staphylococci, enterococci, Campylobacter. 7.. susceptibility tests, 4th ed. NCCLS, Villanova, Pa. 12. Kern WV, Andriof E, Oethinger M, et al. Emergence. Yu WL, Lin CW, Wang DY. Serratia marcescens bacteremia: clinical features and antimicrobial susceptibility of isolates. J Microbiol Immun Infect 1998;31:171-9.. 10. Gilchrist MJR. Enterobacteriaceae: opportunistic pathogens and other Genera. In: Murray PR, Pfaller MA, Tenover FC, Yolken RH, eds. Manual of Clinical Microbiology. Washington, DC: American Society for Microbiology, 1995:457-64. 11. National Committee for Clinical Laboratory Standards. 1990. Approved. Standard M2-A4.. Performance standards for antimicrobial disc. 1988;10:830-8. 17. Kuah BG, Kumarasinghe G, Doran J, et al. Antimicrobial susceptibilities of clinical isolates of Acinetobacter. baumannii. from Singapore.. Antimicrob Agents Chemother 1994;38:2502-3. 18. Cisneros JM, Reyes MJ, Pachon J, et al. Bacteremia due to Acinetobacter baumannii: epidemiology, clinical findings, and prognostic features. Clin Infect Dis 1996;22:1026-32. 19. Liassine N, Bilie J, Breer C, et al. In vitro activity of cefpirome against microorganisms isolated in haematology, oncology and intensive care units in Switzerland. Scand J Infect Dis 1997;29:615-21..

(7) 170. 1. 2 2. 1997. 1. 1. 6 152. aeruginosa. 37. Serratia marcescens. 31. 55. Pseudomonas. Klebsiella pneumoniae. 29. Acinetobacter baumannii P. aeruginosa. S. marcescens. gentamicin P. a e r u g i n o s a ciprofloxacin imipenem. amikacin, ceftazidime, imipenem S. marcescens. K. pneumoniae. -lactamase E-test. cephalosporins ciprofloxacin (MIC 90 = 1 g/mL). moxalactam. S. marcescens (MIC 90 = 0.19 Imipenem. g/mL). K. pneumoniae. A. baumannii 1999;4:164-. 70. 404 3/5/1999 5/19/1999. 5/10/1999.

(8)