Risk Factors for an Outbreak of Multi-Drug-Resistant Acinetobacter Nosocomial Pneumonia Among Intubated Patients *

Rola N. Husni, Lawrence S. Goldstein, Alejandro C. Arroliga, Geraldine S. Hall, Cynthia Fatica, James K. Stoller and Steven M. Gordon Chest 1999;115;1378-1382 DOI 10.1378/chest.115.5.1378 The online version of this article, along with updated information and services can be found online on the World Wide Web at: http://chestjournal.chestpubs.org/content/115/5/1378.full.html

Chest is the official journal of the American College of Chest Physicians. It has been published monthly since 1935. Copyright1999by the American College of Chest Physicians, 3300 Dundee Road, Northbrook, IL 60062. All rights reserved. No part of this article or PDF may be reproduced or distributed without the prior written permission of the copyright holder. (http://chestjournal.chestpubs.org/site/misc/reprints.xhtml) ISSN:0012-3692

Downloaded from chestjournal.chestpubs.org by Chiu Ho on May 26, 2012 1999 American College of Chest Physicians

Risk Factors for an Outbreak of Multi-Drug-Resistant Acinetobacter Nosocomial Pneumonia Among Intubated Patients*
Rola N. Husni, MD; Lawrence S. Goldstein, MD; Alejandro C. Arroliga, MD, FCCP; Geraldine S. Hall, PhD; Cynthia Fatica, RN; James K. Stoller, MD, FCCP; and Steven M. Gordon, MD

Introduction: Acinetobacter baumanii is a Gram-negative coccobacillus that is normally a commensal pathogen but can be a nosocomial pathogen. An epidemiologic study was performed to investigate an outbreak of A baumanii that occurred in our medical intensive care unit (MICU) from March to September 1995. Methods: A case-control study was performed by retrospective chart review, comparing case patients to randomly selected patients who were mechanically ventilated in the MICU for at least 1 week during the outbreak. A case patient was defined as any patient with an Acinetobacter infection in which the epidemic strain was considered to be a pathogen. The epidemic strain was defined by its antibiogram. Case patients and control patients were compared for age, gender, underlying disease, acute physiology and chronic health evaluation III score, length of MICU stay, prior antibiotic use, presence of fever, sepsis, type of pulmonary infiltrate, and outcome. Environmental and handwashing studies also were performed during the period of the outbreak. Molecular typing was performed on available bloodstream isolates. Results: There were 15 cases of A baumanii nosocomial pneumonia. Fifty percent were bacteremic; one chart was unavailable for review. Twenty-nine patients were identified as control patients. The mean age for case patients was 50 (range, 21 to 84). The mean duration of time from admission to the ICU to infection was 12.8 days (range, 4 to 40). Sepsis developed in 35% of the case patients. Forty-three percent of the case patients died during their hospitalization, with two of those deaths attributed to Acinetobacter infection. Univariate analysis showed that prior use of ceftazidime was associated with infection with Acinetobacter (11/14 case patients compared to 11/29 control patients; p < 0.01). Pulsed-field gel electrophoresis revealed two strains to be responsible for the outbreak. Hand washing was performed before patient contact by only 10% of health-care workers, and only 32% washed their hands after patient contact. Conclusion: The use of ceftazidime was associated with an increased risk of nosocomial pneumonia with resistant strains of Acinetobacter. Health-care workers need to improve compliance with hand-washing recommendations. (CHEST 1999; 115:13781382)
Key words: Acinetobacter baumanii; nosocomial infections; pneumonia Abbreviations: APACHE acute physiology and chronic health evaluation; MIC MICU medical intensive care unit mean inhibitory concentration;

cinetobacter small, aerobic, A Gram-negativespp are ubiquitous, moist environbacilli that prefer
1

ments.1 Acinetobacter spp are usually considered to be opportunistic pathogens but have been increasingly reported as the cause of outbreaks of nosocomial pneumonia and bloodstream infections.2 These
*From the Departments of Infectious Diseases (Drs. Husni and Gordon), Pulmonary and Critical Care Medicine (Drs. Goldstein, Arroliga, and Stoller), Infection Control (Ms. Fatica), and Laboratory Medicine (Dr. Hall), Cleveland Clinic Foundation, Cleveland, OH. Manuscript received March 25, 1998; accepted November 13, 1998. Correspondence to: Alejandro C. Arroliga, MD, FCCP, Department of Pulmonary and Critical Care Medicine - G62, Cleveland Clinic Foundation, Cleveland, OH 44195; e-mail: arrolia@ cesmtp.ccf.org
1378

organisms can withstand desiccation and may acquire resistance to many antimicrobial agents, especially -lactam antibiotics.3 We report an outFor editorial comment see page 1226 break of nosocomial Acinetobacter spp infections among patients in our MICU associated with use of a third-generation cephalosporin. Materials and Methods
Background The Cleveland Clinic Foundation is a 950-bed tertiary care hospital with a 17-bed MICU. Infection control personnel noted
Clinical Investigations in Critical Care

Downloaded from chestjournal.chestpubs.org by Chiu Ho on May 26, 2012 1999 American College of Chest Physicians

an increase in the number of nosocomial infections with Acinetobacter baumanii in MICU patients during the spring and summer of 1995. To identify risk factors for infection with this organism, a case-control study was performed. Case Definitions A case patient was defined as any patient in the MICU between March and August 1995 with a nosocomial pneumonia and/or bloodstream infection due to a resistant A baumanii (the epidemic strain). Resistant A baumanii was defined by the following antimicrobial susceptibility pattern: susceptible only to ticarcillin/clavulonate (mean inhibitory concentration ([MIC], 16 g/mL) and imipenem (MIC, 4) intermediate to amikacin (MIC, 32) and resistant to tobramycin (MIC, 8), gentamicin (MIC, 8), ceftazidime (MIC, 16), and ciprofloxacin (MIC, 2). Acinetobacter pneumonia was defined as including the presence of increased respiratory secretions, new lung infiltrate identified on a chest radiograph, and a sputum culture positive for A baumanii. Fever was defined as an oral temperature of 38C, and sepsis syndrome was determined by previously defined criteria.4 Control Patients Control patients were defined as all patients mechanically ventilated in the MICU for 1 week during the epidemic period. Case Ascertainment Infection control personnel reviewed all microbiology cultures to identify all MICU patients with A baumanii isolated from clinical cultures. Medical records of all patients were reviewed to identify case and control patients. Information abstracted included demographic data, clinical presentation, disease severity score,5 underlying diseases, prior antibiotics, treatment, and outcome. Infection Control Practices A review of infection control practices was performed by infection control personnel. Cleaning and disinfecting of flexible bronchoscopes and ventilators, including the use of disposable and reusable items, was observed. In-service education was provided to the physicians, nursing staff, and respiratory therapists in regard to A baumanii infections and body substance isolation.6 Flexible bronchoscopes were routinely cleaned and sterilized by a trained equipment technician (Steris process; Steris Corporation; Mentor, OH). Ventilators were cleaned and disinfected within the Department of Respiratory Therapy, and reusable items were cleaned and sterilized in the Central Service Department. All practices were in compliance with our current policy and procedure. Hand-Washing Survey A hand-washing survey was conducted in the MICU by infection control personnel during the outbreak period. This survey consisted of six 2-h observation periods (three on the day shift and three on the evening shift), during which all health-care providers (physicians, nurses, and respiratory therapists) were unaware that they were being observed regarding compliance with hand washing and glove use as appropriate for body substance isolation.

Culture and DNA Preparation for Pulsed-Field Gel Electrophoresis of A baumanii Isolates After bacterial growth of 18 to 22 h, the isolates were inoculated into sterile saline to a density approximately equal to McFarland number 4, sonicated, and centrifuged. The pellet was resuspended in 1 mL of cold cell-suspension buffer (Bio-Rad; Hercules, CA). Agarose gel plugs were prepared with the sonicated isolates and digested with SmaI, a restriction enzyme. The digested DNA plugs were placed in wells of agarose gel molds, and electrophoresis was carried out in a contour-clamped homogenous electric field apparatus (Gene-Path; Bio-Rad), using a No. 12 program that allowed for the separation of molecular weights of 5 to 300 kb. Gels were stained with ethidium bromide and were photographed under ultraviolet light. The restriction patterns were analyzed following criteria by Tenover et al.7 If there were identical banding patterns, the isolates were said to be identical; if there were one to three band differences between isolates, they were considered to be closely related. If they differed by more than three bands, they were said to be unrelated. Data Analysis The data was subjected to statistical analysis using (Epi-info, version 6.0; Centers for Disease Control and Prevention; Atlanta, GA). Differences between case patients and control patients were analyzed with the 2 test and Fishers exact test (two-tailed). A p value of 0.05 was chosen to assign statistical significance.

Results Fifteen cases of A baumanii pneumonia were identified in MICU patients between March and August 1995 that met the case definition (Fig 1). One patient was excluded from the study because the chart was not available for review. Characteristics of the Case Patients The mean age of the case patients was 50 (range, 21 to 84 years), and 79% were male (Table 1). Indications for MICU admission included respiratory failure (eight patients), sepsis (three), cardiac tamponade (one), encephalitis (one), and acidosis (one) (Table 2). Underlying chronic diseases included cirrhosis (four), collagen vascular disease (three), chronic renal failure (one), AIDS (one), and leukopenia (one). The mean acute physiology and chronic health evaluation (APACHE) III score on admission to the MICU was 95.3. Clinical manifestations included fever (79%), sepsis syndrome (35%), and leukocytosis (mean WBC count, 14,634/mL; range, 700 to 26,360). All patients were intubated and mechanically ventilated. Radiographic evidence of pneumonia was present in all patients at the time of the Acinetobacter infection. Lobar infiltrates occurred in seven patients (50%), and seven had diffuse infiltrates. Acinetobacter was isolated from respiratory secretions in 12 patients (79%). Seven
CHEST / 115 / 5 / MAY, 1999

1379

Downloaded from chestjournal.chestpubs.org by Chiu Ho on May 26, 2012 1999 American College of Chest Physicians

Figure 1. Epidemic curve of patients with A baumanii pneumonia in an MICU, Cleveland Clinic Foundation, 1995.

patients had concomitant bacteremias, accounting for 15% of all bacteremias in the MICU during the study period. The mean interval from admission to the MICU to the first respiratory culture for Acinetobacter was 12.8 days (range, 4 to 40 days). All case patients were on antimicrobial agents at the time the Acinetobacter infections were identified. The most common antimicrobials that patients received at the time of Acinetobacter infections were vancomycin (100%), ceftazidime (78%), and ciprofloxacin (50%). Antimicrobial treatment for Acinetobacter infections included administration of ticarcillin/clavulonate (64%), imipenem (21%), and a combination of both (15%). Amikacin was used in three patients. The mean duration of antimicrobial therapy for Acinetobacter infections was 26 days (range, 2 to 71 days). Nine patients (64%) were successfully treated for Acinetobacter infections, while three patients failed therapy and two were lost to follow-up. Six patients (43%) died during their hospital admissions. In two patients the death was directly attributed to Acinetobacter sepsis, and in two other patients Acinetobacter infection contributed to death. Case-Control Study Twenty-nine patients who underwent mechanical ventilation for at least 7 days without developing A
Table 1Characteristics of Case and Control Patients
Case Patients (n 14) Mean age, yr Men APACHE III score Admitted for acute respiratory failure Vancomycin received Ceftazidime received Died 50 11 (79%) 95.3 8 (57%) 14 (100%) 11 (78%) 6 (43%) Control Patients (n 29) 60 12 (41%) 92.4 20 (68%) 22 (76%) 11 (37%) 11 (32%) p Value 0.4 0.05 0.79 0.5 0.08 0.01 0.9

baumanii infection or colonization were included in the case-control study. The mean age of control patients was 60 years, and 41% were male. Indications for MICU admission were acute respiratory failure (20 patients), sepsis (4), cardiac disease (2), gastrointestinal disorders (2), and acidosis (1). The mean APACHE III score on admission to the MICU was 92.4. Twenty-two patients (76%) received vancomycin and 11 (37%) received ceftazidime during their MICU admissions. The only significant risk factor associated with Acinetobacter infection was prior use of ceftazidime (11/14 case patients vs 11/29 control patients; p 0.01; odds ratio, 6; 95% confidence intervals, 1.1 to 35; Table 1). There were no significant differences between case patients and control patients in age, gender, indications for MICU admission, underlying diseases, use of other antimicrobials, or the presence of sepsis syndrome. Hand-Washing Survey Hand washing was practiced before patient contact in only 10% of the 111 physicians, nurses, and respiratory therapists observed. After patient contact, the rate of hand washing was 32%. Donning of gloves was practiced by all respiratory therapists, but only 13 (72%) removed their gloves immediately after completing the patient encounter. Pulsed-Field Gel Electrophoresis Assay Of the four A baumanii isolates (all in the bloodstream) available for pulsed-field gel electrophoresis, two distinct banding patterns were identified.

Discussion Our study found an association between the prior administration of ceftazidime and the subsequent development of A baumanii infections. This finding is in agreement with an earlier study by Mulin et al8 that associated the use of third-generation cephalosporins with colonization and infection by A baumanii. Other studies have noted an association between other antibiotic classes and Acinetobacter infection. One study, interestingly, found an association between the use of cefazolin and Acinetobacter infection but not with higher generation cephalosporins.9 Recently, in a report of three case-control studies on a retrospective cohort study, it was found that epidemic infections may coexist with endemic infections of A baumanii.10 Previous use of a fluoroquinolone antibiotic (perfloxacin, ofloxacin, norfloxacin, or ciprofloxacin) was a risk factor found for the development of endemic A baumanii infection. Other risk factors for the development of nosocomial AcinetoClinical Investigations in Critical Care

1380

Downloaded from chestjournal.chestpubs.org by Chiu Ho on May 26, 2012 1999 American College of Chest Physicians

Table 2Characteristics of Case Patients

Patient No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 *A I Age, yr 22 74 43 33 45 62 62 53 39 65 67 84 37 20 Days Intubated Before Infection 24 9 40 12 9 4 8 14 15 10 11 7 9 8 Underlying Disease Pneumonia, respiratory failure Renal failure, pancytopenia ARDS, respiratory failure Respiratory failure, pulmonary hemosiderosis Respiratory failure, colon carcinoma Sepsis, cirrhosis Sepsis, cirrhosis Sepsis, cirrhosis, renal failure Respiratory failure, pneumonia, encephalitis Cirrhosis, chronic renal failure, acidosis Alveolar hemorrhage, collagen vascular disease, respiratory failure Collagen vascular disease, Pneumocystis carinii pneumonia, respiratory failure Cardiac tamponade, collagen vascular disease AIDS, Pneumocystis carinii pneumonia, respiratory failure, sepsis syndrome Prior Use of Antibiotics* A,C,Ci,G,I,M,V,T,Ak,T/S C,Ci,V A,C,Ci,E,G,T/S,V V,T,To,C E,Cef,A,G,M,T,V C,V,Ci V,C,To C,Ci,V A,C,Cef,Ci,V,Cl V,G V,T,E,C C,E,V,T/S C,V T/S,V,Ci,To Outcome Survived Survived Survived Survived Survived Survived Died Died Survived Survived Died Died Died Died

ampicillin; Ak amikacin; c ceftazidime; cef ceftizoxime or cefuroxime; Ci ciprofloxacin; E erythromycin; G gentamicin; imipenem; M metronidazole; T ticarcillin/clavulanic acid; To tobramycin; T/S trimethoprim/sulfamethoxazole; V vancomycin.

bacter pneumonia in intubated patients included recent neurosurgery, head trauma, large aspiration, and ARDS. A case-control study by Baraibar et al11 examined patients who developed A baumanii infections and compared the infections to other etiologies of ventilator-associated pneumonia; the study could not detect risk factors common to both groups. That patient population differed from ours in that the ICUs were combined medical-surgical units, with 10 of 12 patients being admitted post-trauma or postoperatively. Seventy-seven percent of patients in the Baraibar study had been on antibiotics in the 10 days preceding the development of ventilator-associated pneumonia. Lortholary et al12 noted that 32 of 40 patients (75%) in an ICU who were colonized or infected with Acinetobacter had previously taken antimicrobials, although in their experience only previous infection and high severity of illness were statistically significant risk factors for Acinetobacter infection. Because of the difficulty in identifying a specific cause of infection in this patient population, the choice of antimicrobials is often empiric. At the time of the outbreak, ceftazidime and ticarcillin/clavulanic acid were the antimicrobials most commonly used for empiric Gram-negative rod coverage in our MICU. This was the most common indication for case patients to receive ceftazidime. In our study, all of the case patients and most of the control patients (76%) had received vancomycin for empiric Grampositive bacteria coverage. The p value for this comparison was not significant, and it is doubtful that the prior use of vancomycin plays any role in the

development of Acinetobacter infection. Our study did not confirm the findings of Lortholary et al12 that a high disease-severity score was a risk factor for Acinetobacter infection. Because Acinetobacter resistance is acquired rather than inherent, prior exposure to antibiotics plays a role in the subsequent acquisition of resistant Acinetobacter.2 All strains of Acinetobacter identified in our epidemic were susceptible to imipenem, which is consistent with prior reports that show rare but increasing resistance to carbapenams.13 In contrast, Marques et al14 found that 70% of isolates of Acinetobacter were resistant to at least two extended-spectrum penicillins and two aminoglycosides but that imipenem and an aminoglycoside resulted in at least partial synergy in all of the highly resistant isolates. Unfortunately, detailed descriptions of imipenem-resistant outbreaks are unavailable. However, the account of one outbreak in New York noted that the epidemic of imipenem-resistant Acinetobacter occurred after imipenem had been heavily prescribed for an outbreak of ceftazidime-resistant Acinetobacter.15 Single-strained nosocomial outbreaks may originate from a common environmental source. We cannot rule out transmission from patient to patient via the hands of health-care workers, especially in light of the handwashing survey that documented poor adherence to the hand-washing policy by health-care workers before and after patient contact. However, we did not identify a reservoir for Acinetobacter in the MICU, and molecular typing suggested that at least two epidemic strains were present during the outbreak. The availability of
CHEST / 115 / 5 / MAY, 1999

1381

Downloaded from chestjournal.chestpubs.org by Chiu Ho on May 26, 2012 1999 American College of Chest Physicians

only four bacteria for molecular typing is a limitation of our study and leaves open the possibility that more strains were responsible for the outbreak, but it does not invalidate the finding that some of the outbreaks of Acinetobacter infections may be caused by several strains.10 Our study identified prior use of ceftazidime as a risk factor for acquiring an infection with a resistant Acinetobacter sp. The use of ceftazidime as a firstline antibiotic in treating nosocomial infections has increased over the past years because of the increased incidence of resistant organisms like Pseudomonas spp. Now, it seems that ceftazidime has been overused, selecting out even more resistant Gramnegative bacteria. Therefore, the empiric use of these agents should be reserved for instances where there is a high probability of Pseudomonas aeruginosa infections. Another goal we recommend is better attention to good hand-washing practices among health-care workers. References
1 Gerner-Smidt P. Taxonomy and epidemiology of Acinetobacter infections. Rev Med Microbiol 1995; 6:186 195 2 Towner KJ. Clinical importance and antibiotic resistance of Acinetobacter spp. J Med Microbiol 1997; 46:721746 3 Bergogne-Berezin E, Decre D, Joly-Guillou ML. Opportunistic nosocomial multiply resistant bacterial infections: their treatment and prevention. J Antimicrob Chemother 1993; 32(suppl A):39 47 4 Bone RC, Fisher CJ Jr, Clemmer TP, et al. Sepsis syndrome: a valid clinical entity. Methylprednisolone severe sepsis study group. Crit Care Med 1989; 17:389 393

5 Knaus WA, Wagner DP, Draper EA, et al. The APACHE III prognostic system: risk prediction of hospital mortality in critically ill hospitalized patients. Chest 1991; 100:1619 1636 6 Lynch P, Jackson MM, Cummings J, et al. Rethinking the role of isolation practices in prevention of nosocomial infections. Ann Intern Med 1987; 107:243246 7 Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsedfield gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995; 33:22332239 8 Mulin B, Talon D, Viel JF, et al. Risk factors for nosocomial colonization with multiresistant Acinetobacter baumanii. Eur J Clin Microbiol Infect Dis 1995; 14:569 576 9 Beck-Sague CM, Sinkowitz RL, Chinn RY, et al. Risk factors for ventilator-associated pneumonia in surgical intensive care unit patients. Infect Control Hosp Epidemiol 1996; 17:374 376 10 Villers D, Espaze E, Coste-Burel M, et al. Nosocomial Acinetobacter baumanii infections: microbiological and clinical epidemiology. Ann Intern Med 1998; 129:182189 11 Baraibar J, Correa H, Mariscal D, et al. Risk factors for infection by Acinetobacter baumanii in intubated patients with nosocomial pneumonia. Chest 1997; 112:1050 1054 12 Lortholary O, Fagon JY, Hoi AB, et al. Nosocomial acquisition of multiresistant Acinetobacter baumanii: risk factors and prognosis. Clin Infect Dis 1995; 20:790 796 13 Swartz MN. Hospital-acquired infections: diseases with increasingly limited therapies. Proc Natl Acad Sci USA 1994; 91:2420 2427 14 Marques MB, Brookings ES, Moser SA, et al. Comparative in vitro antimicrobial susceptibilities of nosocomial isolates of Acinetobacter baumanii and synergistic activities of nine antimicrobial combinations. Antimicrob Agents Chemother 1997; 41:881 885 15 Urban C, Go E, Mariano N, et al. Effect of sulbactam on infections caused by imipenem-resistant Acinetobacter calcoaceticus biotype anitratus. J Infect Dis 1993; 167: 448 451

1382

Clinical Investigations in Critical Care

Downloaded from chestjournal.chestpubs.org by Chiu Ho on May 26, 2012 1999 American College of Chest Physicians

Risk Factors for an Outbreak of Multi-Drug-Resistant Acinetobacter Nosocomial Pneumonia Among Intubated Patients * Rola N. Husni, Lawrence S. Goldstein, Alejandro C. Arroliga, Geraldine S. Hall, Cynthia Fatica, James K. Stoller and Steven M. Gordon Chest 1999;115; 1378-1382 DOI 10.1378/chest.115.5.1378 This information is current as of May 26, 2012
Updated Information & Services Updated Information and services can be found at: http://chestjournal.chestpubs.org/content/115/5/1378.full.html References This article cites 15 articles, 10 of which can be accessed free at: http://chestjournal.chestpubs.org/content/115/5/1378.full.html#ref-list-1 Cited Bys This article has been cited by 17 HighWire-hosted articles: http://chestjournal.chestpubs.org/content/115/5/1378.full.html#related-urls Permissions & Licensing Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://www.chestpubs.org/site/misc/reprints.xhtml Reprints Information about ordering reprints can be found online: http://www.chestpubs.org/site/misc/reprints.xhtml Citation Alerts Receive free e-mail alerts when new articles cite this article. To sign up, select the "Services" link to the right of the online article. Images in PowerPoint format Figures that appear in CHEST articles can be downloaded for teaching purposes in PowerPoint slide format. See any online figure for directions.

Downloaded from chestjournal.chestpubs.org by Chiu Ho on May 26, 2012 1999 American College of Chest Physicians