Impact of antibiotic use and prescribing policy in a Tunisian intensive care unit

Objective. - The aim of this study was to assess the efficacy of local antibiotic policy in a Tunisian ICU. The predefined primary efficacy objectives were the decrease of antibiotic consumption, reduction of inappropriate antibiotic (ATB) use and antimicrobial resistance. Design. - This prospective intervention study lasted from January 2002 to December 2004. In the first study period or the baseline phase (from January to December 2002) we focused on physician education for ATB prescription practice. The second period concerned intervention (control of all ATB use). Results. - The number of infection episodes significantly decreased from 2002 to 2004; 198 infection episodes in 2002 (1.63 +/- 1.15 episodes/patient) versus 124 in 2003 (1.22 +/- 0.93) (P < 0.0008) versus 121 in 2004 (1.23 +/- 0.8) (P I < 0.0008). The number of ATB/prescription also significantly decreased from 1.85 +/- 1.3 in 2002 to 1.5 +/- 0.9 in 2003 (P = 0.02) and 1.5 +/- 1.4 in 2004 (PI = 0.05). Appropriateness of antibiotherapy improved during the intervention period: 65% in 2002 versus 86% in 2003 (P = 0.0003) and 81% in 2004 (PI = 0.02). The length of antibiotherapy in survivors was considerably reduced: 14.1 +/- 2.9 days in 2002 versus 11.9 +/- 1.2 days in 2003 (P < 10(-5)) and 10.9 +/- 2.5 days in 2004 (P < 10-5) with a significant reduction of antibiotherapy cost and length of stay (20.4 +/- 9 days in 2002 versus 18.3 +/- 6 days in 2003 and 16.9 +/- 8 days in 2004; P = 0.05; P1 = 0.02). There was a significant decrease of carbapenem resistant Enterobacteriaceae esp, Pseudomonas aeruginosa and Acinetobacter baumannii. (c) 2006 Elsevier Masson SAS. Tous droits reserves.