Dexamethasone for Treating Severe Hospital-acquired Pneumonia in Critically Ill Patients With a Proinflammatory Phenotype

Inclusion Criteria: * Hospital-acquired pneumonia (HAP) according to European guidelines (Torres et al. Eur Respir J 2017): Association of two criteria among (body temperature \> 38°C, leukocytosis\>12000 cells per mL, leucopenia \<4000 cells per mL and purulent pulmonary secretions), appearance of a new infiltrate or change in an existing infiltrate on chest radiography, and respiratory sample (Sputum, AET, BAL, mini-BAL or blind BAL) collected for bacteriological diagnosis (results can be pending at inclusion). The diagnosis of HAP can have been made outside of ICU. Diagnosis is done at least 48 hours after hospital admission. * HAP severity defined as a PaO2/FiO2 ratio \< 300 under mechanical ventilation. * Biological systemic inflammatory response defined as CPR≥ 150 mg/L (15 mg/dL)\* * Receiving curative antimicrobial therapy for the current episode of HAP pneumonia for less than 48 hours. * Informed consent from a legal representative, or emergency procedure (when possible, according to national regulation, see below). If it is not possible to obtain the patient consent prior the inclusion (comatose patients), patient consent for the study continuation will be obtained as soon as deemed possible. * Person insured under a health insurance scheme. * Female of childbearing age who agree and who are able to comply with effective contraception for the 28 first days of the study. Exclusion Criteria: * Pregnant women (serum or urine test), breastfeeding women. * Patient under legal protection (incl. under guardianship or trusteeship). * Hypersensitivity to dexamethasone and hypersensitivity to all of its excipients * Ongoing administration of glucocorticoid at the time of randomisation, such as for COVID-19 infection requiring supplemental oxygen therapy * Severe septic shock (norepinephrine \> 0.4 microg/kg/min and serum lactate level greater than 2 mmol/L) at the time of randomisation * Prolonged use of corticosteroids at a mean minimum dose of 0.3 mg/kg/day of prednisone equivalent for \>3 weeks in the past 60 days * Uncontrolled viral (hepatitis,herpes, zona, varicella) or systemic fungal infection * Immunosuppression pre-existing to hospitalisation (severe lymphopenia \< 500 lymphocytes/mm3, hematologic cancer, aplasia, chemotherapy/radiotherapy for cancer within 3 months prior to the inclusion, or anti-graft rejection drug). * Uncontrolled psychotic disorder (acute or chronical) * Patients not expected to survive for more than 48 hours. * Participation in another drug clinical trial : * testing steroids or anti-graft rejection drug or chemotherapy- radiotherapy for cancer * And / Or testing a drug regimen with a known interaction with dexamethasone, * And / Or whose implementation would alter the HAP-DEX 6-month follow-up, notably the collection of the primary outcome. * Situations that may increase the risk associated with study participation or investigational product administration or may interfere with the interpretation of study results and, in the judgment of the investigator, would make the subject inappropriate for entry into this study