Objective
To test whether a reduction of third generation cephalosporin (3GC) use has a sustainable positive impact on the high endemic prevalence of 3GC resistant K. pneumoniae and E. coli.

Design
Segmented regression analysis of interrupted time series was used to analyse antibiotic consumption and resistance data 30 months before and 30 after the intervention.

Setting
Surgical intensive care unit (ICU) with 16-bed unit in a teaching hospital.

Intervention
In July 2004, 3GCs were switched to piperacillin in combination with a β-lactamase-inhibitor as standard therapy for peritonitis and other intraabdominal infections.

Results
Segmented regression analysis showed that the intervention achieved a significant and sustainable decrease in the use of 3GCs of −110.2 daily defined doses (DDD)/1,000 pd. 3GC use decreased from a level of 178.9 DDD/1,000 pd before to 68.7 DDD/1,000 pd after the intervention. The intervention resulted in a mean estimated reduction in total antibiotic use of 27%. Piperacillin/tazobactam showed a significant increase in level of 64.4 DDD/1,000 pd, and continued to increase by 2.3 DDD/1,000 pd per month after the intervention. The intervention was not associated with a significant change in the resistance densities of 3GC resistant K. pneumoniae and E. coli.

Conclusion
Reducing 3GCs does not necessarily impact positively on the resistance situation in the ICU setting. Likewise, replacing piperacillin with β-lactamase inhibitor might provide a selection pressure on 3GC resistant E. coli and K. pneumoniae. To improve resistance, it might not be sufficient to restrict interventions to a risk area.

• Quantitative data - Quantitative data were used during the intervention

• Human - All activities regarding organisations and individuals involved in health products and services

• Healthcare delivery