Introduction : The treatment of critically-ill childrenin pediatric intensive care units ( PICUs) is associated with better outcomes, such as lower mortality rates; thus, we hypothesized that the incidence of healthcare-associated infections would also be lower in PICUs. Acinetobacter baumannii is an important nosocomial pathogen associated with mortality among critically-ill patients. We compared the incidence of A. baumannii infection among pediatric patients treated only in PICUs with those treated in adult ICUs or multiple ICUs. Methods : We retrospectively reviewed the medical and laboratory records of 300 critically-ill patients of the Department of Pediatrics aged under 18 years who were admitted to the ICUs of a tertiary referral hospital between July 2013 and June 2016. We investigated all kinds of specimens obtained after 48 hours of ICU admission to identify culture results positive for A. baumannii during ICU admission. The cases transferred from one ICU to another were not regarded as separate cases, and patients who were discharged from the ICU within 48 hours or transferred from the neonatal ICU were excluded. Results : A total of 184 ICU admissions were included in this study, of whom 111 were treated only in the PICU ( PICU group), and 73 were treated in the adult ICU or multiple ICU (AICU group). There were no significant differences between the PICU and AICU groups regarding patient characteristics at ICU admission except the mean age (4.9 years vs. 7.4 years, respectively) and comorbidities. The PICU group had a higher prevalence of respiratory disease (54.1% vs. 35.6%), while the AICU group had a higher prevalence of hemato-oncological diseases (33.3% vs. 50.7%) and cardiovascular diseases (11.7% vs 23.3%). During ICU admission, there were no differences between both groups regarding length of ICU stay and mechanical ventilator care, but the AICU group required more medical devices such as central venous catheter, urinary catheter, and drainage. A. baumannii was isolated 6 times in the PICU group (5.4%) and 11 times in the AICU group (15.1%); thus, the incidence densities of positive culture results for A. baumannii were 2.31 and 8.57 per 1000 ICU-days, respectively. Conclusions : Critically-ill children treated only at PICUs had a lower incidence of A. baumannii infection.