Maragakis LL, Tucker MG, Miller RG, Carroll KC, Perl TM. Incidence and Prevalence of Multidrug-Resistant Acinetobacter Using Targeted Active Surveillance Cultures. JAMA. 2008;299(21):2513–2514. doi:10.1001/jama.299.21.2513
To the Editor: Recent legislation mandates active surveillance cultures to detect multidrug-resistant (MDR) organisms in hospitalized patients.1 The active surveillance strategy has not been widely applied to MDR Acinetobacter, one of the most difficult gram-negative pathogens to control and treat.2 Universal active surveillance is resource-intensive and may divert resources from other infection control interventions.3 We conducted a prospective cohort study using universal active surveillance cultures to determine the prevalence and incidence of transmission of MDR Acinetobacter and to estimate the effects of targeting active surveillance to patients with recent exposure to a long-term care facility.
Universal active surveillance cultures from the axilla, wounds, sputum, and endotracheal tube aspiration were performed on admission and weekly among a cohort of 1111 adult patients admitted to medical intensive and intermediate care units of the Johns Hopkins Hospital between March and June in 2006. Isolation precautions were implemented for patients with MDR Acinetobacter isolates susceptible to no more than 1 class of antimicrobial agents, excluding colistin. Patients previously known to have the organism were flagged in an administrative database for identification and isolation on readmission. Susceptibility testing was by the Phoenix 100 Automated Microbiology System (Becton-Dickinson, Sparks, Maryland) and agar dilution methods. Patients with missing data (12.5% for absent admission surveillance cultures and 4.8% for location prior to admission) were included in the analysis. Fisher exact test was used to compare categorical variables. All tests were 2-sided and P < .05 was considered statistically significant. Analyses were performed with Stata version 8.2 (Stata Corp, College Station, Texas). Our institutional review board approved the protocol and granted a waiver of informed consent.
Five patients were previously known to be colonized with MDR Acinetobacter, 5 grew the organism within 48 hours of admission, and 3 grew the organism more than 48 hours after admission (none of whom had admission surveillance cultures obtained) (Table 1). Prevalence on admission was 10 of 1223 patients (0.82%; 95% confidence interval [CI], 0.39%-1.50%). Incidence of possible MDR Acinetobacter transmission (growth more than 48 hours after admission and absent surveillance cultures) was 0.43 (95% CI, 0.09-1.3) per 1000 patient days. Colonization pressure (MDR Acinetobacter patient-days divided by total patient-days) was 1.6% (108 of 6934 patient-days). Of 8 newly identified patients with MDR Acinetobacter, 4 did not grow the organism in clinical cultures but were identified by surveillance cultures alone.
A significantly higher proportion of patients with MDR Acinetobacter were admitted from a long-term care facility than patients without MDR Acinetobacter (46.2% vs 3.9%; risk ratio, 18.9; 95% CI, 6.6-54) (Table 2). Patients with MDR Acinetobacter were more likely to have paraplegia than were patients without MDR Acinetobacter (23.1% vs 1.1%; risk ratio, 22; 95% CI, 6.7-72). The majority of patients with MDR Acinetobacter were co-colonized with at least 1 other MDR pathogen: methicillin-resistant Staphylococcus aureus (n = 8, 62%), vancomycin-resistant Enterococci (n = 10, 77%), and extended-spectrum β-lactamase gram-negative bacilli (n = 5, 38%).
The potential consequences of MDR Acinetobacter transmission and infection include a crude mortality of 28% to 58%, prolonged mechanical ventilation, and prolonged stays in the hospital and intensive care unit.4,5 In this cohort, most MDR Acinetobacter was present on hospital admission and the undetected fraction of MDR Acinetobacter was 50%. Limitations include low sensitivity of MDR Acinetobacter surveillance cultures,6 small sample size of patients with MDR Acinetobacter, absent admission surveillance cultures, few wound and endotracheal cultures, and generalizability to other health care settings.
Although active surveillance cultures may be indicated during an outbreak, in general a population at risk should be defined before beginning an active surveillance program to detect MDR organisms. Although universal active surveillance cultures identified otherwise undetected patients with MDR Acinetobacter, screening all admissions required processing a large number of cultures. Targeting active surveillance cultures and isolation precautions to patients with recent exposure to a long-term care facility would have detected most newly identified patients by screening 52 admissions rather than more than 1200. Research is needed to analyze the efficacy and cost-effectiveness of this and other infection control strategies.
Author Contributions: Dr Maragakis had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Maragakis, Miller, Carroll, Perl.
Acquisition of data: Maragakis, Tucker.
Analysis and interpretation of data: Maragakis.
Drafting of the manuscript: Maragakis.
Critical revision of the manuscript for important intellectual content: Maragakis, Tucker, Miller, Carroll, Perl.
Statistical analysis: Maragakis.
Obtained funding: Maragakis, Carroll, Perl.
Administrative, technical, or material support: Tucker, Miller, Perl.
Study supervision: Maragakis, Perl.
Financial Disclosures: Dr Perl reported having served on the advisory board for Pfizer, Replidyne, and 3M; having received recent research funding from 3M and Sage; and having received a speaker honorarium from Ortho McNeil. None of the other authors reported financial disclosures.
Funding/Support: Dr Maragakis was supported by a Research Scientist Development Award (grant No. 5-K01-CI000300) from the Centers for Disease Control and Prevention.
Role of the Sponsor: The Centers for Disease Control and Prevention had no role in the design and conduct of the study; in the collection, management, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.
Additional Contributions: Charles Reuland, ScD; Beryl Rosenstein, MD; Karen G. Davis, RN, MSN; and Karen Haller, RN, PhD, supported the hospital administration aspects of the study; the nurse managers and nurses of the study units obtained surveillance cultures; Tracy Ross, BS, and the microbiology laboratory processed the extra volume of cultures generated by the study; Kathleen Speck, MPH, coordinated the research and provided technical assistance; and Vidhya Gunaseelan, MS, assisted with database construction and data management. All individuals were affiliated with Johns Hopkins medical institutions. Mss Ross and Speck received compensation for their roles in the project. No other individuals received compensation for their assistance.