Skip to content


Critical Care

Volume 12 Supplement 5

Sepsis 2008

Open Access

Significant decrease of central venous catheter-associated bloodstream infection rates in 38 German intensive care units

  • Sonja Hansen1,
  • Frank Schwab1,
  • Sandra Schneider1,
  • Dorit Sohr1,
  • Christine Geffers1 and
  • Petra Gastmeier1
Critical Care200812(Suppl 5):P7

Published: 18 November 2008


Intensive Care UnitInfection RateCentral Venous CatheterNosocomial InfectionSurveillance Activity


Central venous catheter (CVC)-associated bloodstream infections (BSI) remain a major complication in ICUs. The aim of the study was to evaluate the influence of a structured multimodal intervention programme on the CVC-BSI rate of 38 ICUs in Germany.


ICUs of the 'Krankenhaus Infektions Surveillance System' showing a CVC-BSI rate above the median were asked to implement a 12-month intervention programme starting in April 2007. The intervention included specific evidence-based recommendations for CVC insertion and use, and involved nurses and physicians. Modules were posters, script and advanced training. The modules' content was composed and distributed according to the 'train the trainer' principle by the National Reference Center for Surveillance of Nosocomial Infections. Infection rates were calculated before (January 2005 to June 2006) and during the intervention (May 2007 to March 2008).


Thirty-eight ICUs participated in the study. The ICUs had a median of 11 beds and eight ventilator beds. The majority of ICUs (47%) were affiliated to teaching hospitals; 30% were affiliated to university hospitals. The CVC utilization rate before implementation of the intervention was 69.4 CVC-days per 100 patient-days. The pooled mean CVC-BSI rate was 2.9 CVC-BSI per 1,000 CVC-days. A preliminary analysis of the data obtained during the intervention period showed a decrease of the mean CVC-BSI rate in the participating ICUs (2.2 CVC-BSI per 1,000 CVC-days; relative risk = 0.77, 95% confidence interval = 0.63 to 0.94, P = 0.011), whereas the CVC utilization rate remained almost unchanged (69.0 CVC-days per 100 patient-days).


A structured multimodal intervention programme in addition to ongoing surveillance activities led to a significant decrease of CVC-associated BSI rates.

Authors’ Affiliations

National Reference Center for Surveillance of Nosocomial Infections, Institute of Hygiene, Charite – University Medicine Berlin, Berlin, Germany


© Hansen et al; licensee BioMed Central Ltd. 2008

This article is published under license to BioMed Central Ltd.