CP-160 Carbapenem de-escalation therapy for intra-abdominal infection

Background The Antimicrobial Stewardship Programme promotes strategies to improve antibiotic prescriptions, optimise clinical outcomes, minimise costs and avoid adverse effects. It is also recommended for the prevention and decrease in the appearance of emerging resistant bacteria. One of the goals of this programme is therapy de-escalation with broad spectrum antibiotics. Purpose To analyse carbapenem prescriptions and de-escalation therapy in intra-abdominal infections, and determine the impact of de-escalation on hospital stay and inhospital mortality. Material and methods This prospective observational study of carbapenem prescriptions and de-escalation performance was conducted in a third level hospital between 1 August 2013 and 31 July 2014. Data were gathered on the number of carbapenem prescriptions and patients characteristics, carbapenem treatment duration, culture requests, de-escalation performance, length of hospital stay and mortality rate. The oncology–haematology, traumatology, neurosurgery and neurology departments were excluded. Results 489 prescriptions for 437 patients were recorded during this period. Mean age of patients was 65.3 years; 57.7% were men. The median Charlson Index score was 4 (2–6). 76.5% of inpatients were from the surgical department and 7.6% had sepsis. 78.9% of prescriptions were for monotherapy. 64.4% of carbapenems were prescribed as firstline therapy and 35.6% as rescue therapy. 68.9% of microbiological cultures were requested and 50.4% were positive. The most prescribed carbapenems were ertapenem (44.4%) and imipenem (30.7%). De-escalation was performed in 31.9% of cases, and 53.6% in the presence of positives cultures versus 46.4% of negative cultures(p=0.418). Median duration of carbapenem therapy was 6 (4–9) days, 5 days in the de-escalation group versus 6 days in the no de-escalation group(p=0.006). Length of hospital stay was 10 (6–20) days, 10 days in de-escalation group versus 12 days in non-de-escalation group (p=0.052). Total inhospital mortality was 10.8%; in de-escalation group 4.7% versus 13.9% in the non-de-escalation group(p=0.003). Conclusion The de-escalation of carbapenem therapy reduces patient mortality, exposure to carbapenems treatment and length of hospital stay. No conflict of interest