Lack of resources.

The study is designed to evaluate sensitivity, specificity and accuracy of non-bronchoscopic bronchoalveolar lavage and endotracheal aspirate in comparison with bronchoscopic bronchoalveolar lavage (gold standard), as methods for etiologic diagnosis of ventilator associated pneumonia and their impact in morbimortality.

Ventilator associated pneumonia is defined as one that occurs 48-72h after tracheal intubation, is classified into early and late and is responsible for a mortality of 24-50% to almost 70% in some studies.

The clinical suspicion of ventilator associated pneumonia is based on the finding of new or progressive pulmonary opacity associated with two or more variables such as: fever, leukocytosis or leukopenia, and purulent pulmonary secretions.

Some studies show that the etiologic diagnosis reduces spectrum of antimicrobials and bacterial resistance, reducing hospital stay and time of mechanical ventilation, as well as treatment costs. The collection of material for quantitative culture ensures greater specificity however there is no standard method.

The investigator's hypothesis is that collection of material to confirm the etiologic diagnosis of ventilator associated pneumonia, held by bronchoscopic bronchoalveolar lavage, define the etiology in a larger number of patients when compared to the endotracheal aspirate and non-bronchoscopic bronchoalveolar lavage.

The primary goal is evaluate and compare the sensitivity and specificity of noninvasive methods (endotracheal aspirate and non-bronchoscopic bronchoalveolar lavage) in comparison with invasive (bronchoscopic bronchoalveolar lavage) regarding the collection of material for etiologic diagnosis.

The secondary objectives are determine the prevalence of ventilator associated pneumonia and antimicrobial susceptibility patterns of bacteria isolated, helping to update empirical antibiotic therapy protocols.

This is a prospective comparative observational study on etiologic methods of diagnosis. All the patients, with clinical suspicion, will be submitted to collection of tracheal aspirate, bronchoscopic and non-bronchoscopic bronchoalveolar lavage. These procedures are already incorporated in the daily practice of our intensive cara unit and are routinely performed for investigation of patients with compatible framework of ventilator associated pneumonia, as availability of bronchoscopist.

Whereas an analysis of variance will be performed to compare the three interventions, with a minimum difference between the treatment means, a power of 0.95 and a significance level of 0.05, with an estimation error of 0.1, will be required a total sample of 72 patients.There will be only one group of 72 patients, in which will be held in all the three samples of the study.

Inclusion Criteria:

• Older than 18 and clinical suspicion of VAP;

Exclusion Criteria:

• Age less than 18 years;
• Introduction or modification of antibiotic therapy due to associated diagnosis in the first 3 days of the start of the protocol;
• No agreement with informed consent or for any other reason, expose the desire to leave the protocol at any time.

All patients older than 18 years admitted to the ICU of Hospital de Clínicas de Uberlândia with suspected VAP will be included in the project. from february 2015 to complete the total of 72 patients required, since provided that agree with the terms of informed consent.

We are still collecting data.