Colistin is used as an elective treatment of infections of multi drug resistant gram negative bacteria. Until now colistin is used in the therapeutic regimen of these infections intravenous or nebulized. There are a plenty of studies about the efficacy of nebulized colistin in the therapy of pseudomonas aeruginosa pneumonia in patients with cystic fibrosis and in the therapy of ventilator associated pneumonia in ICU. On the other hand there are only a few studies about the use of nebulized colistin in the prevention of ventilator associated pneumonia whereas the role of nebulized colistin in the prevention of severe forms of pneumonia such as VAP due to multi drug resistant gram negative bacteria are limited.

This double blinded randomized trial aim to investigate the effect of nebulized colistin on the incidence of patients with due to gram negative bacteria in the ICU compared to nebulized normal saline.

INTRODUCTION Pneumonia in the intensive care unit (ICU) constitutes the most frequent infection and is most often associated with the application of the mechanical ventilation. VAP (Ventilator - Associated Pneumonia) is defined as the nosocomial pneumonia in a patient on mechanical ventilatory support (by endotracheal tube or tracheostomy) for more than 48 hours. The risk for such complication is proportionally increased with the duration of hospitalization. The symptoms include the appearance of pulmonary infiltrate, fever, leukocytosis and purulent tracheobronchial secretions although this symptomatology is not always present. VAP is usually caused by nosocomial species and especially by gram negative bacteria such as Acinetobacter baumannii, Pseudomonas aeruginosa, Klebsiella pneumoniae.

VAP is frequently complicated with the entry of bacteria in the circulation of blood resulting in bacteremia and sepsis. In this respect, this disorder is associated with substantial morbidity and devastating costs of hospitalization. Notably, the cost of one episode of VAP is estimated in 57000 $ per occurrence.

Inhaled antibiotics achieve high (both peak and trough) concentrations in respiratory secretions. Endobronchial antibiotics have been given initially in cystic fibrosis patients for the prevention of Pseudomonas aeruginosa pneumonia. More recently, some studies reported on the use of endobronchial antibiotics in non-cystic fibrosis patients for VAP treatment. There are only a few, relatively small, one-centre studies about the prevention of VAP; in those studies Colistin was given by instillation or nebulization and decreased the incidence of Gram-negative pneumonia.

AIM To study prospectively the effect of nebulized colistin on the incidence of Ventilator Associated Pneumonia due to gram negative bacteria in the ICU.

STUDY DESIGN Parallel two group blinded randomized controlled clinical trial. The trial medication will be commenced within 12 hours of randomization. The treatment group will receive nebulized colistin (500000 international units of colistimethate sodium, three times a day) for 10 days.

The control group will receive normal saline 0.9% for 10 days. PRIMARY OUTCOME The incidence of VAP due to gram-negative bacteria at the 28th day following randomization INTERVENTIONS Experimental group: nebulized colistin (colistimethate sodium, 500000 international units, three times a day) for 10 days Control group: nebulized Normal saline 0.9%, three times a day for 10 days TIMELINE All patients undergoing invasive mechanical ventilation will be assessed for eligibility after 48h of invasive mechanical ventilation. Patients will be included and randomized and a tracheal aspirate, blood culture, rectal swab will be sampled.

Patients will receive their first treatment delivery within 12h following randomization.

Patients will be assessed daily until extubation or day 28 - whichever occurs first - for suspicion of ventilator associated pneumonia (VAP) according to accepted criteria (see 'Definitions'), clinical pulmonary infection score computation, systemic antibiotic delivery, acute kidney injury, ventilator associated events. Weaning from mechanical ventilation will occur based on treating physician decision (an informative-training session on 'weaning' will take place in all centres at the beginning of the trial).

RANDOMIZATION AND BLINDING Randomization (based on tables of random numbers) will be centralized and stratified on centre and systemic antibiotic delivery the day of randomization.

Study drug and placebo will be prepared by blinded staff not involved in included patients' care. Blinding will be maintained throughout the study except in case of a deliberated decision on unblinding.

SAMPLE SIZE The hypothesis is based on data of a previous trial on this topic (Karvouniaris et al 2016); the incidence of VAP due to gram negative bacteria was 25% in the control group (normal saline group). For a reduction to 12,5%, a sample of 152 patients is necessary in each arm of the study, with a probability of a two-tailed Type I error of 0.05 and a power of 80%. This hypothesis represents an important relative risk reduction which is clinically meaningful in case of a positive trial. The population of the study will be at least of 304 patients (152 in each group) as it has been calculated by the following analysis: All statistical analyses will be performed or supervised by E Zintzaras, Professor of Biometry Head, Department of Biomathematics, University of Thessaly School of Medicine, Larissa, Greece.

FEASIBILITY The project has been endorsed by two research centres (University Hospital of Larissa and Evaggelismos University Hospital of Athens) with significant expertise in randomized clinical trials in the area of VAP prevention, mechanical ventilation and sepsis.

The cooperation of the two centres resulted in the following published trials:

Vasopressin, steroids, and epinephrine and neurologically favorable survival after in-hospital cardiac arrest: a randomized clinical trial. Mentzelopoulos SD, Malachias S, Chamos C, Konstantopoulos D, Ntaidou T, Papastylianou A, Kolliantzaki I, Theodoridi M, Ischaki H, Makris D, Zakynthinos E, Zintzaras E, Sourlas S, Aloizos S, Zakynthinos SG. JAMA. 2013 Intermittent recruitment with high-frequency oscillation/tracheal gas insufflation in acute respiratory distress syndrome.Mentzelopoulos SD, Malachias S, Zintzaras E, Kokkoris S, Zakynthinos E, Makris D, Magira E, Markaki V, Roussos C, Zakynthinos SG.Eur Respir J. 2012 Mar;39(3):635-47.

One of the centres has experience in studies regarding VAP prevention:

Nebulised colistin for ventilator-associated pneumonia prevention.Karvouniaris M, Makris D, Zygoulis P, Triantaris A, Xitsas S, Mantzarlis K, Petinaki E, Zakynthinos E.Eur Respir J. 2015 Dec;46(6):1732-9.

Effect of pravastatin on the frequency of ventilator-associated pneumonia and on intensive care unit mortality: open-label, randomized study.Makris D, Manoulakas E, Komnos A, Papakrivou E, Tzovaras N, Hovas A, Zintzaras E, Zakynthinos E.

Crit Care Med. 2011 Nov;39(11):2440-6. EXPECTED DURATION OF THE STUDY

• Enrollment: 24 months
• Duration of the study for a participant: 28 days
• Total duration of the study: 30 months

All patients undergoing invasive mechanical ventilation through an endotracheal tube or tracheostomy will be considered for inclusion and recorded on a screening log. Patients will be assessed for eligibility within the first 48h of invasive ventilation.

Inclusion criteria

1. - Age ≥ 18 years the day of inclusion
2. - Mechanical ventilation through an endotracheal tube > 48h
3. - Written informed consent of patient or proxy Exclusion criteria

1. Community acquired pneumonia or other nosocomial pneumonia or tracheobronchitis 2. Known colistin resistant bacteria in tracheo-bronchial aspirate's culture 3. Known allergy to colistin 4. Patient scheduled for extubation within the next 24h 5. Patient mechanically ventilated > 96h 6. Known or suspicion of pregnancy at the time of inclusion 7. Pregnant or breastfeeding women 8. Participation in other studies with inhaled or systemic antibiotics or in other VAP related studies.