ASSOCIATION OF PRIVATE ISOLATION ROOMS WITH VENTILATOR-ASSOCIATED ACINETOBACTER-BAUMANII PNEUMONIA IN A SURGICAL INTENSIVE-CARE UNIT

OBJECTIVE: To determine the rates and routes of Acinetobacter baumanii colonization and pneumonia among ventilated patients in a surgical in tensive care unit (SICU) before and after architectural modifications. DESIGN: A nonsequential study comparing two groups of patients. AU is olates from systematic and clinical samples were genotyped by pulsed-f ield gel electrophoresis (PFGE). Records of patients hospitalized duri ng the first and second periods were reviewed and findings were compar ed. Between the two periods, the SICU was remodeled from enclosed isol ation rooms and open rooms to only enclosed isolation rooms with handw ashing facilities in each room. SETTING AND PATIENTS: All patients hos pitalized and mechanically ventilated for more than 48 hours in the 15 -bed SICU of the University Hospital of Besancon (France). RESULTS: Fo r the first and second periods, the rates of colonization were, respec tively, 28.1% and 5.0% of patients (P < 10(-7); relative risk [RR], 2. 23; 95% confidence interval [CI95], 1.8-2.75) and the specific rates o f bronchopulmonary (BP) colonization were, respectively, 9.1 and 0.5 p er 1,000 days of mechanical ventilation (P < 10(-5)). Seven major PFGE isolate types were identified, 4 of which were isolated from 44 of th e 47 colonized or infected patients. Logistic regression analysis show ed that colonization was not associated with patient characteristics. CONCLUSION: Conversion from open rooms to isolation rooms may help con trol nosocomial BP tract acquisition of A baumanii in mechanically ven tilated patients hospitalized in an SICU (Infect Control Hosp Epidemio l 1997;18:499-503).