Efficacy of Cotrimoxazole as a De-escalation Treatment of Ventilator-Associated Pneumonia in Intensive Care Unit

Multicentre randomized non-inferiority trial comparing cotrimoxazole to standard antibiotic therapy for enterobacterial VAP. Selection of patients will be done by physicians in ICU. All clinically suspected VAP will be confirmed with a lung sample (preferably bronchoalveolar lavage or protected distal specimen, otherwise endotracheal aspiration). Patients with a microbiologically confirmed VAP due to an Enterobacteriaceae susceptible to cotrimoxazole and at least one antibiotic of the empiric antibiotic therapy (based on international recommendations) will be included. After written informed consent, they will be randomized (1:1), using a computer-generated randomization scheme of various-sized blocks, stratified by presence of septic shock at VAP diagnosis and by presence of COVID-19 pneumonia on ICU admission, through a centralized 24 hours internet service (CleanWEB™) to cotrimoxazole, or best standard of care (either a beta-lactam or a fluoroquinolone), after randomization for a total duration of 7 days (including empiric initial appropriate treatment). Posology and modalities of antibiotic administration will be optimized based on most recent recommendations for ICU patients. Because antibiotic therapy will be variable in the control group, single or double blind is not appropriate. Daily follow-up until death or ICU discharge or day 28 will be performed (vital status, antibiotic therapy, new infection, Clostridium-difficile infection). Clinical (arterial blood gas, temperature, haematology, tracheal secretions) and radiological cure (chest X-ray) will be assessed at Day 7. Systematic MDR bacteria screening will be performed weekly and at ICU discharge. Vital status will be assessed at day 90. Alive patients leaving ICU before 90 days will be contacted by phone (if discharge at home) or by interview at hospital (if transferred in a different ward). Assessment of the clinical and radiological cure by an independent committee (1 specialist in infectious disease and 1 intensivist), blinded of the randomization arm (PROBE methodology).

Inclusion criteria

• Adult patients hospitalized in an ICU
• Under mechanical ventilation for at least five days
• Microbiologically confirmed VAP preferably on a distal lung sample (bronchoalveolar lavage or protected distal specimen) otherwise endotracheal aspiration
• Enterobacteriaceae susceptible to cotrimoxazole, and for polymicrobial VAP, all bacteria susceptible to cotrimoxazole
• 5) Treated for at least 24 hours by an appropriate empiric antibiotic therapy (at least one effective antibiotic from the initiation of treatment for this VAP episode), and for polymicrobial VAP, all bacteria susceptible to empiric antibiotic therapy
• Stability of haemodynamic (stability or decrease in catecholamine dose) and respiratory (stability or improvement of FIO2) parameters

Exclusion criteria

• Haemodynamic instability (increasing dose of a catecholamine in the last 24 hours)
• Contra-indication to cotrimoxazole:

• allergy,
• advanced liver insufficiency,
• renal dysfunction with clearance <15 mL/min/1.73 m² without hemodialysis
• G6PD deficiency
• history of hypersensitivity to one of the components (in particular, hypersensitivity to sulphonamides
• known macrocytic anemia defined by VGM >
• treatment with methotrexate
• Infection requiring prolonged antibiotic-therapy (pleural empyema, lung abscess, necrotizing pneumonia, etc…)
• Cystic fibrosis
• Immunosuppression (neutropenia, HIV with CD4 lymphocytes below 200/mm3, immunosuppressive therapy or corticosteroid therapy >0.5 mg/kg/j before ICU admission)
• Cardiac arrest without awakening
• Moribund state (patient likely to die within 24h)
• Limitation of life support (comfort care applied only) at the time of screening
• Enrolment to another interventional study on VAP care/management
• Pregnancy or breastfeeding
• Subject deprived of freedom, subject under a legal protective measure
• No affiliation to any health insurance system
• Refusal to participate to the study (patient or legal representative or family member or close relative if present)
• Patients previously included in the study