Skip to main content


We're creating a new version of this page. See preview

Volume 3 Supplement 1


  • Poster presentation
  • Open Access

Risk factors at icu admission for multi-resistant acinetobacter baumannii colonization and infection and its prediction capability

  • 1,
  • 1,
  • 1,
  • 2,
  • 3,
  • 1,
  • 1,
  • 1,
  • 1,
  • 3,
  • 1,
  • 1,
  • 1,
  • 1 and
Intensive Care Medicine Experimental20153 (Suppl 1) :A127

  • Published:


  • Neutropenic Patient
  • Hospital Ward
  • Acinetobacter Baumannii
  • Multivariable Logistic Regression Analysis
  • Infection Control Measure


Predicting colonization/infection by Acinetobacter baumannii (C-I MRAb) at ICU admission allows optimization of infection control measures and empirical antibiotic treatment.


To describe risk factors for multirresistant Acinetobacter baumannii (defined as resistant to carbapenem) colonization and infection at ICU admission, centered on those that are easy to obtain at admission, without needing access to clinical records, not always avaible at ICU admission. We also study the predictive capability of a model developed with those risk factors.


Retrospective analysis of data collected prospectively from 16950 patients admitted consecutively (stay > 24h) in 151 Spanish ICU participating in ENVIN (National Surveillance Study of Nosocomial Infections in ICU) registry during the period of April-June of 2010. Univariable and multivariable analysis was performed with the next variables: age, gender, illness, origin place, urgent surgery, immunodeficiency, being immunosuppressed, neutropenic patients and skin infections (surgical site -superficial and deep- and skin-soft tissue infections).


Overall, 44 patients with C-I MRAb were detected in the study period. In the multivariable logistic regression analysis gender male (2,99(1,32-6,77)0,008), to have medical illness (3,16(1,42-7,02)0,005), trauma critical patient (4,74(1,44-15,56)0,01), admitted from hospital ward (6,21( 2,49-15,47) < 0,001), or from other ICU (12,92(3,87-43,15) < 0,001), or from a long-term care center (11,95(1,38-102,97)0,02), being immunosuppressed (4,52 (2,2-9,27) < 0,001) and to have skin-soft tissue infection (14,07 (4,26- 46,46) < 0,001) were independent risk factors for C-I MRAb at ICU admission. The resultant predictive model with these variables showed: AUC-ROC 0,85; 95% CI (0,79-0,90) and p-value on the Hosmer-Lemeshow of 0,89.


The risk of being C-I MRAb at ICU admission is 3 times greater in male and medical patients, 4 times in immunosuppressed or in trauma critical patients, 6 times in those admitted from hospital ward, 12 times in patients admitted from other ICU or from a long-term care center and 14 times in those that have skin-soft tissue infection. Our predictive model showed good discrimination but these results are not good enough for ICU setting, being aware of the impact of the false negative results. Risk factors described should be considered for future studies.

Authors’ Affiliations

Hospital Universitario de Burgos, Intensive Care, Burgos, Spain
Hospital Clínico Universitario de Valladolid, Microbiology, Valladolid, Spain
Hospital Universitario de Burgos, Burgos, Spain


  1. Nseir S, Grailles G, et al: Accuracy of American Thoracic Society/Infectious Diseases Society of America criteria in predicting infection or colonization with multidrug-resistant bacteria at intensive-care unit admission. European Society of Clinical Microbiology and Infectious Diseases, CMI. 16: 902-908.Google Scholar
  2. Jianfeng Xie, et al: Value of American Thoracic Society Guidelines in Predicting Infection or Colonization with MultidrugResistant Organisms in Critically Ill Patients. PLOS ONE. 2014, Volume 9 (Issue 3): e89687-1 MarchGoogle Scholar


© Vara Arlanzón et al.; 2015

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.