Epidemiologic, Clinical, and Economic Evaluation of an Outbreak of Clonal Multidrug-ResistantAcinetobacter baumanniiInfection in a Surgical Intensive Care Unit

Abstract

Objectives.To determine risk factors for acquisition of multidrug-resistant (MDR)Acinetobacter baumanniiinfection during an outbreak, to describe the clinical manifestations of infection, and to ascertain the cost of infection.Design.Case-control study.Setting.Surgical intensive care unit in a 400-bed urban teaching hospital and level 1 trauma center.Patients.Case patients received a diagnosis of infection due toA. baumanniiisolates with a unique pattern of drug resistance (ie, susceptible to imipenem, variably susceptible to aminoglycosides, and resistant to all other antibiotics) between December 1, 2004, and August 31, 2005. Case patients were matched 1 : 1 with concurrently hospitalized control patients. Isolates' genetic relatedness was established by pulsed-field gel electrophoresis.Results.Sixty-seven patients met the inclusion criteria. Case and control patients were similar with respect to age, duration of hospitalization, and Charlson comorbidity score. MDRA. baumanniiinfections included ventilator-associated pneumonia (in 56.7% of patients), bacteremia (in 25.4%), postoperative wound infections (in 25.4%), central venous catheter-associated infections (in 20.9%), and urinary tract infections (in 10.4%). Conditional multiple logistic regression was used to determine statistically significant risk factors on the basis of results from the bivariate analyses. The duration of hospitalization and healthcare charges were modeled by multiple linear regression. Significant risk factors included higher Acute Physiology and Chronic Health Evaluation II score (odds ratio [OR], 1.1 per point increase;P= .06), duration of intubation (OR, 1.4 per day intubated;P<.01), exposure to bronchoscopy (OR, 22.7;P= .03), presence of chronic pulmonary disease (OR, 77.7;P= .02), receipt of fluconazole (OR, 73.3;P<.01), and receipt of levofloxacin (OR, 11.5;P= .02). Case patients had a mean of $60,913 in attributable excess patient charges and a mean of 13 excess hospital days.Interventions.Infection control measures included the following: limitations on the performance of pulsatile lavage wound debridement, the removal of items with upholstered surfaces, and the implementation of contact isolation for patients with suspected MDRA. baumanniiinfection.Conclusions.This large outbreak of infection due to clonal MDRA. baumanniicaused significant morbidity and expense. Aerosolization of MDRA. baumanniiduring pulsatile lavage debridement of infected wounds and during the management of respiratory secretions from colonized and infected patients may promote widespread environmental contamination. Multifaceted infection control interventions were associated with a decrease in the number of MDRA. baumanniiisolates recovered from patients.