A clinically relevant model of acute respiratory distress syndrome in human-size swine

ABSTRACT Despite over 30 years of intensive research for targeted therapies, treatment of acute respiratory distress syndrome (ARDS) remains supportive in nature. With mortality upwards of 30%, a high-fidelity pre-clinical model of ARDS, on which to test novel therapeutics, is urgently needed. We used the Yorkshire breed of swine to induce a reproducible model of ARDS in human-sized swine to allow the study of new therapeutics, from both mechanistic and clinical standpoints. For this, animals were anesthetized, intubated and mechanically ventilated, and pH-standardized gastric contents were delivered bronchoscopically, followed by intravenous infusion of Escherichia coli-derived lipopolysaccharide. Once the ratio of arterial oxygen partial pressure (PaO2) to fractional inspired oxygen (FIO2) had decreased to <150, the animals received standard ARDS treatment for up to 48 h. All swine developed moderate to severe ARDS. Chest radiographs taken at regular intervals showed significantly worse lung edema after induction of ARDS. Quantitative scoring of lung injury demonstrated time-dependent increases in interstitial and alveolar edema, neutrophil infiltration, and mild to moderate alveolar membrane thickening. This pre-clinical model of ARDS in human-sized swine recapitulates the clinical, radiographic and histopathologic manifestations of ARDS, providing a tool to study therapies for this highly morbid lung disease.


Preprocedure preparation and hemodynamic monitoring
Swine were vaccinated from the vendor for porcine reproductive and respiratory syndrome, Mycoplasma hyopneumoniae, Haemophilus parasuis, Swine Influenza (H1N1, H3N2), parvovirus, leptospirosis, Erysipelas spp., E. coli, Pasteurella pneumotropica, and Bordetella bronchiseptica. Five days prior to shipment, animals were treated with oxytetracycline (8lb/ton) in food. Animals were housed at Columbia in an AAALAC-accredited facility. All conducted studies were approved by the IACUC at Columbia University.
Swine underwent general anesthesia via intramuscular induction with tiletamine/zolazepam (5 mg kg −1 , Zoetis). A 10Fr urinary catheter was placed into the urethra or percutaneously into the bladder, if necessary, to monitor urine output. Peripheral and central venous catheters and arterial catheters were placed percutaneously under ultrasound guidance. A 5Fr introducer (Cook Medical) was placed in the right external jugular vein to establish central venous access for subsequent ECMO cannulation if needed. An 8.5Fr introducer sheath (Arrow) was placed in the left external jugular vein and a 7Fr pulmonary arterial catheter (Edwards) was floated into the pulmonary artery. Placement was confirmed by chest radiograph. A 7Fr dual-or triple-lumen catheter (Arrow) was also placed in the left external jugular vein for medication infusion. A 20G arterial micropuncture kit (Cook Medical) was used to access the femoral artery and subsequently upsized to a 6-7Fr 20cm introducer sheath for continuous arterial blood pressure monitoring, and to establish arterial access for Veno-Arterial (VA)-ECMO, if required. Continuous infusion with vasopressors (norepinephrine 0.125-1.0 mcg kg −1 hr −1 , phenylephrine 1 mcg kg -1 min -1 , and dopamine 4-6 mcg kg -1 min -1 ) via central venous catheter was initiated and continued as needed to maintain mean arterial blood pressure ≥ 55 mmHg.

ARDS induction
Gastric aspiration injury was induced by previously established injury methods. [1][2][3][4] Briefly, the tip of a bronchoscope (Ambu® aScope TM 4 Broncho Slim 3.8/1.2) was positioned sequentially 1cm distal to the carina into the right and left mainstem bronchi and standardized gastric contents (30-50 mL; pH 2) were delivered to the bilateral lungs. The location of the tip of the bronchoscope was confirmed visually prior to delivery of gastric contents. After delivery, gastric contents remained in the lungs and bronchial alveolar lavage was not performed to allow development of acute lung injury. Lipopolysaccharide (LPS) from Escherichia coli O55:B5 (10 μg kg −1 , Sigma) in 100 mL normal saline was infused via central intravenous catheter over 30 to 60 minutes. ARDS 0hr was defined as the timepoint at which the PaO2/FIO2 ratio was first less than 150 mmHg following LPS infusion.

Use of paralytics and extracorporeal membrane oxygenation (ECMO)
Pancuronium bromide (initial bolus of 0.1 mg kg -1 followed by continuous infusion of 0.1-0.5 mg kg -1 , Hospira) was used for paralysis in animals with severe hypoxia or hypercarbia. Depth of anesthesia was verified by continuous hemodynamic monitoring and vital signs response to painful stimuli every 15 minutes from initiation of paralysis until 2 hours after cessation of the paralytic infusion.
ECMO was used as rescue therapy to attempt to prevent mortality in severely ill swine prior to planned study endpoint. Cannulation for Veno-Venous (VV)-ECMO was achieved via either 20Fr dual-lumen Avalon (Maquet) in the right external jugular vein, or with a 23Fr outflow cannula in the femoral vein and 17Fr inflow cannula in the

Experimental endpoint
Pre-defined experimental endpoints were ARDS 6hr (n=3) and up to ARDS 48hr. Animals were euthanized prior to experimental endpoint if their clinical status deteriorated despite maximal medical treatment, including vasopressor support, ECMO, and/or cardiopulmonary resuscitation per Advanced Cardiac Life Support, and in consultation with Columbia University Institute of Comparative Medicine veterinarians. Animals were euthanized with an overdose of pentobarbital sodium (100 mg kg -1 , Euthasol, Virbac). Death was confirmed via cessation of heartbeat, spontaneous respiration, and lack of corneal reflex by a veterinarian.

Histopathologic analysis of lung injury
Tissue samples were collected from lung segments were immediately fixed in cold phosphate-buffered 4% paraformaldehyde (ThermoScientific) for 24-48 h. Samples were then embedded in paraffin, sectioned at 3 μm or 5 μm thickness, and stained with hematoxylin and eosin (H&E) by the Department of Molecular Pathology at Columbia University Medical Center. H&E slides were randomly numbered prior to pathologic review by an experienced pulmonary pathologist under light microscopy without reference to experimental endpoints. Slides from normal swine lung tissue were included within the blinded set of slides for review. A previously described lung injury severity score which includes airway polymorphonuclear cells per high-power field (hpf), alveolar polymorphonuclear cells per hpf, alveolar edema, interstitial infiltrate (lymphocytes and neutrophils in the interstitium around vessels and airways and in alveolar septa and pleura), and interstitial edema (perivascular and peribronchial spaced expanded with edematous fluid) was applied (Table S7). 1 Immunohistochemical staining Lung sections were de-paraffinized, placed in boiling citrate buffer (pH 6.0) for antigen retrieval, and blocked with 10% normal goat serum in phosphate-buffered saline for 2 h at room temperature. Next, primary antibodies were diluted 1:100, applied, and incubated for 12 h at 4 °C or 4 h at room temperature. Secondary antibodies were diluted 1:200 and incubated for 1 h at room temperature. Sections were mounted in Vectashield Mounting Medium with DAPI (Sigma), and coverslips were applied. Images were obtained using an Olympus FSX100 microscope. Immunofluorescence stains for CD31 (Abcam), EpCAM (Abcam), pro-surfactant protein C (Abcam), P-selectin (Abcam), zonula occludens-1 (Abcam), and zonula occludens-3 (Abcam). A complete list of antibodies and dilutions used is provided in Table S8.