Antimicrobial resistance is a major threat worldwide and now concerns last-ressource antibiotics such as carbapenems. As the resistance to carbapenems is directly due to their use, their spare has become a public health emergency. Their efficacy in ventilator-associated pneumonia has never been compared to other classes of antibiotics such as piperacillin-tazobactam which can be an alternative to carbapenems.

The rising antimicrobial resistance has led to more than 33,000 deaths in Europe in 2015. Among them, extended-spectrum beta-lactamase-producing Enterobacteriaceae (ESBL-E) are the most frequent in Europe and carbapenems are recommended as a first line treatment by the guidelines despite the fact they are last-resource agents. Nevertheless, the overuse of carbapenems triggered the emergence of carbapenems-resistant Enterobacteriaceae (CRE). Alternatives to carbapenems are needed to treat ESBL-E infections efficiently without selecting CRE. One strategy described during the last few years is to guide the empirical antimicrobial therapy upon the fecal carriage of ESBL-E. In fact, ESBL-E fecal carriers are more often colonised in the lungs with ESBL-E and have more subsequent ESBL-E infections than non-carriers. However, the positive predictive value of ESBL-E fecal carriage for subsequent ESBL-E is only of 40% despite a negative predictive value of almost 100%. Besides, a before-after cohort study with and without ESBL-E fecal carriage screening exhibited a decrease of carbapenems consumption without any clinical harm. Several studies compared carbapenems vs alternatives after ESBL-E documentation and did not find any clinical harm either but carbapenems were almost always used before documentation. At the best of our knowledge, no study prospectively investigated an alternative to carbapenems for the empirical antimicrobial therapy (before documentation) of ventilator-associated pneumonia despite those encouraging data. This study aims to assess the relevance of piperacillin-tazobactam as the empirical antimicrobial therapy in case of a ventilator-associated pneumonia among ESBL-E fecal carriers. The treatment will be then adapted according to the susceptibility profile. The follow-up of the patients will last until 28 days after their inclusion and until their discharge from intensive care unit if it occurs later. Bacterial susceptibility to the antimicrobial treatment and clinical outcomes will be recorded.

Inclusion Criteria:

• Patient above 18 year-old admitted to intensive care unit
• ESBL-E fecal carriage according to current screening recommendations
• Suspicion of ventilator-associated pneumonia according to ICU society guidelines

Exclusion Criteria:

• Septic shock according to Sepsis-3 classification
• Neutropenia (neutrophils count < 500/mm3)
• Known fecal carriage of Carbapenemase-producing Enterobacteriaceae or multi-drug resistant A. baumanii during the past 6 months.
• Infection with a bacteria resistant to piperacillin-tazobactam during the past 6 months
• Treatment with piperacillin-tazobactam in the 10 previous days
• Proven hypersensitivity to penicillin or tazobactam
• Pregnancy or breastfeeding
• Curatorship or guardianship
• Prisoners
• No health insurance