From the Department of Medicine, University of Texas Health Science Center at San Antonio, and the Medical Service, South Texas Veterans Health Care System, San Antonio, Texas. Received July 4, 2006; accepted July 4, 2006; electronically published August 24, 2006. Infect Control Hosp Epidemiol 2006; 27:889-892 2006 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2006/2709-0001$15.00. Antimicrobial resistance continues to emerge in our hospitals. Despite the predominance of gram-positive healthcare-associated infections in the past 2 decades, gram-negative pathogens continue to be a concern because of the high mortality from these infections among seriously ill patients. Although several alternative antimicrobials active against multidrugresistant gram-positive pathogens are now available, multidrug resistant gram-negative pathogens continue to emerge and new alternative therapies are not available. Preventive measures, then, become all the more important. Pseudomonas aeruginosa, Klebsiella pneumoniae, and Enterobacter species are among the most problematic gram-negative pathogens with emerging resistance. Four articles in this month’s journal address important issues in the prevention of the emergence of resistance in these pathogens. Lautenbach et al. and Fortaleza et al. study risk factors associated with imipenem resistance and ceftazidime resistance 3 in P. aeruginosa. The setting of the study by Fortaleza et al. was a 400-bed general teaching hospital in Campinas, Brazil. Cases in which imipenem-resistant P. aeruginosa strains were recovered (108 patients) and in which ceftazidime-resistant strains were recovered (56 patients) were selected during the years 1999-2002. Case-control studies showed that independent risk factors associated with imipenem resistance in their hospital were transfer from another hospital (perhaps a marker for longer time at risk), receipt of hemodialysis, and use of imipenem, amikacin, or vancomycin. Only transfer from another hospital and amikacin use were independent risk factors for recovery of ceftazidimeresistant P. aeruginosa; ceftazidime use was not. The setting of the study by Lautenbach et al. was a 625bed tertiary care hospital in Philadelphia, Pennsylvania. Cases in which imipenem-resistant P. aeruginosa strains were recovered (142 patients) were identified during the years 19992000. Results of a case-control study showed that fluoroquinolone use, and not imipenem use, was an independent risk factor for recovery of an imipenem-resistant strain. Why was imipenem use a risk factor in one study and not in the other? Both were case-control studies that evaluated imipenem resistance in P. aeruginosa at large teaching hospitals. There was a similar distribution of anatomical sites from which the clinical specimens were isolated and a similar amount of intensive care unit exposure in both studies. Recent studies have discussed the importance of the method used to select control subjects in case-control studies of antibiotic resistance. In the study by Lautenbach et al., the question being asked was “Among clinical isolates of P. aeruginosa, what are the risk factors associated with imipenem resistance in a given isolate?” Thus, patients colonized or infected with imipenem-resistant P. aeruginosa were classified as case patients and patients colonized or infected with imipenem-susceptible P. aeruginosa were classified as control subjects. The study by Fortaleza et al. was designed to evaluate risk factors associated with the recovery of imipenemresistant and ceftazidime-resistant P. aeruginosa isolates from patients admitted to the authors’ hospital. Case patients were those from whom imipenem-resistant or ceftazidime-resistant strains of P. aeruginosa were recovered, and control subjects matched to case patients were selected from the patients admitted to the same ward. The questions asked were different, and thus the procedures for selection of control subjects were different. Harris et al. have reviewed the methodologies for casecontrol studies of antibiotic resistance that may lead to varying results with respect to risk factors. Both Lautenbach et al. and Fortaleza et al. acknowledged the impact of methodology and chose control subjects on the basis of their intended question. The novel case-case-control study design described by Kaye et al. addresses some limitations in study design and has been used to study risk factors for colonization or infection with imipenem-resistant P. aeruginosa among hospitalized patients. Findings of the study by Kaye et al. showed that receipt of imipenem, piperacillin-tazobactam, vancomycin, and/or aminoglycosides were risk factors for isolation of imipenem-resistant P. aeruginosa. The odds ratio for the risk of isolation of imipenem-resistant P. aeruginosa, how-
Read full abstract