The overall aim is to compare the composition and spatial heterogeneity of the following in critically ill intensive care unit (ICU) patients: i) immune cell populations and their activation patterns, ii) the surrounding cytokine-chemokine milieu, including trans-compartmental fluxes of these mediators between the lung and bloodstream, and iii) the lung microbiome.

Main hypotheses:

• The immune cell population in bronchoalveolar lavage fluid (BALF) from patients with ARDS is dominated by neutrocytes, while T cells are depleted, and show evidence of hyper-activation and exhaustion
• T cell hyper-activation and exhaustion is specifically compartmentalised to the lungs, and much more pronounced in moderate-to-severe than none-to-mild ARDS
• Cyto- and chemokines derived from pulmonary immune cells are higher in moderate-to-severe than none-to-mild ARDS with a greater release from lungs to the bloodstream, notably of IL-6 and IL-8.
• The differences in T cell profile in BALF, notably the ratio between regulatory T cells and T helper 17 cells, will change with disease severity over time, and can be explained by the presence of tI-IFN antibodies and/or a low microbial diversity of the respiratory tract with low enrichment from the oral cavity.

Background Pulmonary hyperinflammation with neutrocyte and macrophage invasion and T cell depletion are common features of the acute respiratory distress syndrome (ARDS), but it remains to be elucidated whether the T cells in the lungs of patients with ARDS are hyperactivated and/or exhausted, and to which extent this contributes to neutrocyte invasion and thus lung tissue destruction. Furthermore, the pulmonary microbiome has shown a reduction in diversity and interactions in ARDS, which may be important for normal T cell function. At present, immune cell-microbiome interactions and their relation to disease severity and progression have not yet been studied in ARDS.

Overall design In 20 mechanically ventilated patients with none-to-mild and 20 with moderate-to-severe non-COVID-19 ARDS according to the Berlin definition an endotracheal aspirate and BAL fluid (BALF) from separate lung segments will be obtained. Furthermore, an oral and nasal swab and blood samples will be collected. This will be done within 72 hours after intubation, and again after 7-10 days if the patient is still intubated.

Patient's electronic health record The following is obtained after inclusion: diagnosis codes and medication (type and dosage, including vasopressors and sedatives), and smoking history (current/previous/never smoker; pack years); admission time; blood pressure, heart rhythm and heart rate, temperature, ventilator settings, supportive care (dialysis, ECMO), blood tests results at admission and on the study days (blood cell counts, coagulation parameters, renal, liver, and electrolyte panel; arterial and mixed venous blood gases); clinical scores at admission and on the study days (SAPS3, APACHE II, SOFA), death within 30 days

Blood sampling Arterial blood samples are drawn from the patient's invasive arterial catheter (inserted at ICU admission for clinical purposes: for continuous invasive blood pressure monitoring and repeated arterial blood gas collection) immediately before BALF collection.

Bronchoscopy with BALF collection This procedure is performed in a standardized fashion according to current clinical guidelines. Immediately prior to the procedure, an oral swab, nasal swab and an endotracheal aspirate (ETA) are obtained. FIO2 is then increased to 1.0, and the bronchoscopy procedure is performed using a disposable videoscope with an outer diameter of 5.0 mm). Three successive 50-ml aliquots of prewarmed (37°C) isotonic saline are instilled in the medial segment of the right middle lobe, aspirated immediately with low negative suction pressure (< 100 cm H2O), and pooled into a sterile glass container on ice to obtain a BALF specimen.

Afterwards a mini-BAL is performed in the upper and lower lobe of the right lung with a single installation of 20 ml isotonic saline in each lobe with immediately aspiration into a sterile container.

Measurements The composition of the immune cell population, as well as the function and differentiation of various cell lines will be investigated by single-cell RNA sequencing (scRNA-seq) on selected immune cells from BALF, ETA, and blood, and this will be supplemented by bulk RNA sequencing with sample barcoding and multiplexing, giving a detailed expression pattern of all samples.

The composition microbiome in BALF, ETA, and oral swabs will be assessed by targeted amplicon sequencing of the hyper-variable regions 1 through 3 of the 16S subunit of ribosomal RNA gene for bacteria.

Statistical analyses will be performed using R statistical software version 4.1.1 (R Project for Statistical Computing) within RStudio statistical software version 1.4.1717 (RStudio), and p<0.05 considered statistically significant. Inspection of normality and variance homogeneity will be done by creating qq-plots and histograms. The statistical inference tools SPIEC-EASI and HeatMaps will be used, and based on correlational analyses, principal component analyses, including non-hierarchal cluster analysis, will be applied to identify traits in the two groups.

Inclusion Criteria:

• Inclusion criteria - moderate-to-severe ARDS
• Admitted to the ICU at Hvidovre Hospital
• Intubated within the past 72 hours
• Moderate-to-severe ARDS according to the Berlin definition19
• Age ≥ 18 years

Inclusion criteria - none-to-mild ARDS

• Admitted to the ICU at Hvidovre Hospital
• Intubated within the past 72 hours
• None-to-mild ARDS according to the Berlin definition19
• Age ≥ 18 years

Exclusion Criteria:

• ARDS caused by COVID-19
• Absolute contraindications for bronchoscopy
• Untreated malignant arrhythmia
• Documented or suspected intracranial hypertension (intracranial pressure ≥ > 15 mmHg)
• One-lung ventilation
• Severe coagulopathy

Mechanically ventilated adult ICU patients at Hvidovre University Hospital