Combined retrospective and prospective cohort study to evaluate the incidence of microbiologically confirmed VAP in mechanically ventilated patients with COVID-19. In the retrospective part, microbiological data are based on bi-weekly surveillance ETA. In the prospective part, microbiological data are based on ETA and BAL performed on VAP suspicion. In the prospective part, immunological and virological analyses will be performed on biological samples (blood, respiratory tract) collected from patients at VAP diagnosis.

Ventilator-associated pneumoniae (VAP) is the most common infection acquired in the intensive care unit (ICU). To date, there is no diagnostic gold standard for VAP, and its diagnosis is based on scores that include radiologic, clinical, laboratory, and microbiologic parameters. In addition, there is no univocal recommendation regarding the type of microbiological diagnostics. Some guidelines suggest the use of noninvasive methods (endotracheal aspiration, ETA) with semiquantitative cultures, while others suggest the collection of distal respiratory samples (bronchoalveolar lavage, BAL) with quantitative cultures. While the former method is characterized by higher sensitivity and lower specificity, the latter in contrast has higher specificity. To date, there is no evidence that one method is superior to the other in terms of clinical outcome.

In patients with severe SARS-CoV-2 infection, COVID-19 disease itself and immunomodulatory therapies have a direct impact on most of the clinical, laboratory and radiologic parameters required to achieve VAP diagnosis. In this setting, a diagnostic approach characterized by higher sensitivity coupled with lower specificity could lead to of a high number of false positives. The greatest risk is that of an overdiagnosis of VAP and a consequent overtreatment, with the related therapeutic toxicity and increased antibiotic resistance.

At the investigators' Hospital, the diagnosis of VAP is based on clinical-radiological suspicion according to the Johanson score (new finding or progression of infiltrates on lung radiography + at least two of the following three clinical criteria: fever > 38°C, leukocytosis or leukopenia, purulent secretions), widely validated in non-COVID patients. Until the end of 2020, microbiological data to confirm the diagnosis of clinically suspected VAP and to guide antibiotic therapy were based on the performance of biweekly surveillance ETA. In view of the limited specificity of this approach in COVID-19 setting, from the end of 2020 patients with SARS-CoV-2 infection and suspected VAP undergo, if clinically possible, to collection of distal respiratory specimens by performing BAL/mini-BAL.

The present prospective-retrospective cohort study aims to evaluate the incidence of microbiologically confirmed VAP with BAL (prospective part) and biweekly surveillance ETA (retrospective part) in mechanically ventilated patients with COVID-19 in the ICU. In the prospective part, immunological and virological analyses will be performed on biological samples (blood, respiratory tract) collected from patients at VAP diagnosis.

The study will last 12 months.

Retrospective part all patients admitted to the ICU for COVID-19-related respiratory failure since the beginning of the pandemics will be included.

Prospective part

Inclusion Criteria:

• Age ≥ 18 years
• Diagnosis of SARS-CoV-2 infection by means of reverse transcription polymerase chain reaction (RT-PCR)
• Respiratory failure requiring mechanical ventilation
• mechanical ventilation ongoing for less than 48h

Exclusion Criteria:

• Age < 18 years
• mechanical ventilation ongoing for more than 48h

Mechanically ventilated patients with severe SARS-CoV-2 infection

Anonymized data will be shared upon written request to the study director