We aim to show that systematic ultrasonography performed in ventilated patients suspected of ventilation-acquired pneumonia could improve the accuracy of diagnostic of pneumonia, and helps defining the diagnostic of tracheobronchitis when lower respiratory tract infection is considered.

Chest CT scan is often performed before or just after admission in ICU, and usually show abnormalities that are revealed later on standard radiographs.

This last exam is traditionally considered as the gold standard to prove new pulmonary infiltrates, but the correlation with parenchymal consolidation is pretty low, and lead to over-diagnosing pneumonia, thus leading to a massive and maybe sometimes unconsidered prescription of antibiotic therapy.

Lung ultrasonography conducted systematically within the 3 first days after suspcion of pneumonia could help making the difference between real infection-linked lesions, and banal abnormalities following the hydric inflation of intra-thoracic organs, for instance pulmonary edema or pleural effusion.

An independent evaluation using lung ultrasound, and analysis of CT scan acquisition when performed, compared with the physician in charge of the patient appreciation by suggesting him to provide his own probability of pneumonia upon routine clinical and biological datas.

Diagnostic of VAP and tracheobronchitis is often difficult in UC under mechanical ventilation, and usually occuring quite early after initial phase of hemodynamic instability, and characterized by needs of massive fluids supports and drugs infusion. The goldstandard is still standard chest X-ray, providing a picture of the whole abnormalities due to cardiac failure, and cardiogenic pulmonary edema. The high rate of water in chest of the patients under ventilation assistance and hemodynamic support is responsible of a misinterpretation of the abnormalities visualized on the radiographs. It is often considered that Lung ultrasonography is useful to appreciate the involvement of pleural effusion, explaning etiologies of hypoxemia, where cardiac failure is excluded by the simultaneous ultrasonographic evaluation. However, it is not well recognized that pulmonary parenchyma can be explored through echography, and that it provides a lot of information about the amount of liquid inside it, and the default of aeration. Thus, it can in real time help the physician to guide the therapeutics and manage the ventilation better. Moreover, the suspicion of infection, clinically and biologically leads to the early prescription of antibiotics, given that the radiography is abnormal. Knowing that there is now parenchymal consolidation but rather effusion or a certain degree of pulmonary edema could help avoiding treating falsely a respiratory infection.

The same thoughts can be held concerning Chest CT Scan, an exam often realized at the early stage of management in a critical situation. We can take for granted that when occurring in the 48 first hours of resuscitation, this exam shows in a certain number of cases preexisting abnormalities, that are revealed severla hours later by Chest radiograph, when the suspicion of respiratory infection acquired under ventilation emerge. If we take into account that these abnormalities seen on radiographs are just correlated to those that could be seen on CT, but with a certain delay, and are not de novo, it could lead to an overestimating of ventilation acquired pneumonia, because the criteria of a new radiographic infiltrate won't be valid anymore. The diagnostic of tracheobronchitis could then be more appropriated in a certain number of situations.

Our aim is to verify retrospectively by an adjudication committee, that this early CT Scan, within the 2 first days after admission if patient is ventilated and/or suspicion of lower respiratory tract infection, and a systematic lung ultrasonographic evaluation, provided by an independent operator, could change our appreciation of the frequency of ventilation-acquired pneumonia, comparing to the appreciation of th physician in charge of the patients.

We are thus conducting a repeated evaluation, at day 0, day 3 and day 7 with ultrasonography, in order to give a probability of pneumonia or tracheobronchitis by the echographist operator, and suggesting the physician to give his own probability based on clinical and biological routine datas. We then measure the rate of agreement between the two parts, to see how far the systematic evaluation using pleural ultrasonography could help defining the probability of infection, and validating the diagnostic of pneumonia. The impact should be to improve the accuracy within the first days 2 or 3 days after suspicion, when repeated, if possible by the same operator. One of the main benefit could be the reduction or better reevaluation of antibiotic therapy, if diagnostic of pneumonia is rejected secondary, eventually shorter course of treatment could be chosen, even if this point deserve to be evaluated in further studies.

Inclusion Criteria:

• Immunocompetent subject,
• fever and/or purulent tracheal secretions and/or hyperleukocytosis or leukopenia, associated with a positive microbiological sample (within broncho-alveolar leakage, endotracheal aspiration, or distal sample),
• invasive mechanical ventilation initiated since at least 48hours, and suppose to be maintained for at least 48hours

Exclusion Criteria:

• Immunocompromised-patients defined by ; HIV treated or not, patient under corticotherapy, immunotherapy, inflammatory systemic diseases, solid organ transplant, solid tumor treated or not, bone marrow transplant or stem cells graft, hematological malignancy known or under treatment,
• moribond,
• ventilation expected to last less than 48hours,
• minor-aged patients,
• no social insurance or isolation,
• mental disability making the understanding of the purpose of too difficult.

Every patient place under mechanical ventilation at least for 48 hours, suspected of lower respiratory tract infection, based on CDC criterias of ventilatior-acquired pneumonia.