Background In April 2015, a national project was created to reduce the rate of patients with nosocomial infections by multiresistant bacterias (MRBs). This project had several recommendations, including ones for improving the use of antibiotics, especially against MRBs.

Purpose To analyse the impact of the project on the consumption of antibiotics in our intensive care unit (ICU) during the first months after initiation of the project; and also to assess its economic impact and the number of patients colonised/infected by multiresistant Acinetobacter baumannii (MAB), the most important MRB in our ICU.

Material and methods Retrospective, observational study to compare two periods of time (April–September 2015 vs April–September 2014). The number of defined daily dose per 100 admissions (DDD/100A) was used to evaluate consumption of the following antibiotics: glycopeptides, linezolid, daptomycin, tigecycline, colistin and carbapenems.

The average cost of these drugs was used to do the economic assessment; we did not considerate either the indirect costs or the possible variation in the number of admissions between the two periods. We supposed that infected/colonised patients by MAB were those that had a positive microbiological test for Acinetobarter baummanni that was resistant to three or more families of antibiotics, including carbapenems.

Results Overall antimicrobial consumption was reduced by 45.4% (56.3 vs 30.7 DDD/100A) and costs decreased by 32,9% (42783€ vs 28685€). All studied antibiotics reduced their consumption: 55% for carbapenems (20 vs 9 DDD/100A), 7.4% for linezolid (2.7 vs 2.5 DDD/100A), 56.5% for daptomycin (2.3 vs 1 DDD/100A), 50% for tigecycline (1.8 vs 0.9 DDD/100A) and 48.1 for colistin (16.2 vs 8.4 DDD/100A).

There were the same numbers of patients (n = 13) with infection/colonisation with MAB in both studied periods.

Conclusion Avoiding the use of unnecessary broad spectrum antibiotics and/or a shorter treatment period could reduce the selective pressure and number of MRBs. In addition, this also could lead to an important saving.

Implementation of the project has reduced the use of all studied antibiotics for the treatment of MRBs, but no significant differences were found in the number of patients infected/colonised by MAB. This could be because more time is needed to detect this difference.

No conflict of interest.