Surgical preparation

The eleven sheep, weighing 35 ± 0.5 kg were surgically prepared for the experimental procedure. After induction of anesthesia with ketamine (Bioiche Pharma, Lake Forest, IL, 15–20 mg/kg) and under isoflurane anesthesia, a 16 G, 24 inch silastic catheter was inserted into the right femoral artery (Intracath, Becton–Dickinson, Sandy, UT, USA). A pulmonary arterial thermodilution catheter, model 131F7 was also positioned in the pulmonary artery through the right common jugular vein (Edwards Lifesciences LLC, Irvine, CA, USA). A silastic (0.062 inch ID) catheter was also inserted into the left atrium through a left thoracotomy in the fifth intercostal space (Dow Corning, Midland, MI, USA). Pre-emptive analgesia was provided with buprenorphine. Postoperative analgesia was maintained with continuous infusion of buprenorphine over 48 h and thereafter as needed.

Experimental procedure

After 5 days of recovery with free access to food and water, tracheostomy (Shiley 10 SCT, Tyco Healthcare, Plesanton, CA) was performed under ketamine/isoflurane anesthesia and analgesia. Pneumonia and sepsis were induced by smoke inhalation and instillation of live P. aeruginosa into the lungs. Briefly, a modified bee smoker was filled with 40 g of smoke from cotton toweling, and attached to the tracheostomy tube via a modified endotracheal tube containing an indwelling thermistor from a pulmonary arterial catheter to monitor the temperature of the smoke. Four sets of 12 breaths of cotton smoke (< 40 °C) were manually delivered. After each set, the arterial carboxyhemoglobin (COHb) level was measured with a 626 CO-Oximeter (Instrumentation Laboratory, Bedford, MA, USA) to ensure that all sheep received a similar level of injury. Immediately after the smoke injury, an amount of 2.5 × 10¹¹ Colony Forming Units (CFU) of live P. aeruginosa (PD-05144, 27317, ATCC) suspended in 30 mL of saline was instilled into the lungs through a bronchoscope. 20 mL were placed into the right lung (10 mL in the main bronchus of the middle lobe, and 10 mL in the main bronchus of the lower lobe) and 10 mL were instilled into the main bronchus of the left lower lobe, thus compensating for the extra lobe of the right lung. A urinary retention catheter (Foley 14 C.R., Bard Inc., Covington, GA) was placed into the urinary bladder via the urethra as well.

After this, a randomization occurred to either treatment with R-100 (n = 5) or placebo (n = 6). The experiment was conducted with two sheep being studied simultaneously side-by-side. Treatment was started 1 h post-injury. The treatment group received R-100 at a dose of 80 mg/kg in 500 mL of 5% dextrose. A bolus (30 min) of 10 mg/kg was followed by the remaining 70 mg/kg as a continuous intravenous infusion over 24 h. The control group was treated with an equivalent amount of 5% dextrose. R-100 was kindly provided by Radikal Therapeutics, Inc. Dosage was based on previous mice studies results (Unpublished data). Mice were given 20, 40 or 80 mg/kg/day of intraperitoneal injection of R-100, which was started 1 h after LPS challenge. Although the severity of injury was attenuated with 20, 80 mg/kg had the most survival benefit (100% 7-day survival, while none of mice survived in non-treated control group). None of the animal groups received antibiotics before (at least 24 h) or during the experiment.

After the injury, the animals were awakened and maintained on mechanical ventilation (Servo Ventilator 300C, Siemens-Elema, Sweden) with a positive end-expiratory pressure (PEEP) of 5 cmH₂O and a tidal volume of 12 mL/kg [17] during the 24 h experimental period.

FiO₂ was adjusted after 1 h and then every 3 h according to arterial blood gas analysis results to, as far as possible, maintain the arterial SaO₂ > 90% and PaO₂ at baseline (~ 100 mmHg). The respiratory rate was also adjusted to keep PaCO₂ between 30 and 40 mmHg. Peak and plateau airway pressures were monitored throughout the experimental period. The animals received fluid resuscitation (Ringer’s lactate, containing sodium 130 mmol/L, chloride 109 mmol/L, lactate 28 mmol/L, potassium 4 mmol/L, calcium 1.5 mmol/L), initially at a rate of 2 mL/kg/h, which was then adjusted to maintain hematocrit at baseline ± 3%. The maximum fluid infusion rate allowed was 15 mL/kg/h.

Plasma and urine samples were taken every 3 h during the 24-h experimental period. All sheep were euthanized at 24 h after the injury.

Measurements

Mean arterial pressure (MAP), left atrial pressure (LAP), central venous pressure (CVP), pulmonary artery pressure (PAP), pulmonary capillary wedge pressure (PCWP) and heart rate (HR) were measured using pressure transducers (model P × 3 × 3; Baxter Edwards Critical Care Division, Irvine, CA) connected to a hemodynamic monitor (IntelliVue MP50; Philips Medizin Systeme Boeblingen, Boeblingen, Germany). Cardiac output (CO) was determined by standard thermodilution method. Stroke volume index (SVI), systemic vascular resistance index (SVRI), pulmonary vascular resistance index (PVRI), left ventricle stroke work index (LVSWI), and right ventricle stroke volume index (RVSWI) were calculated by standard formula.

Arterial and mixed venous blood was drawn at baseline and then every 3 h for blood gas measurements by an analyzer (Premier 300, Instrumental Laboratory, Lexington, MA) and the results were corrected for body temperature.

Urine output and fluid administration were measured at regular time intervals for calculation of fluid balance. Plasma protein concentration was measured by a refractometer.

The airway peak and plateau pressures were recorded from ventilator readouts. The bloodless lung wet-to-dry weight ratio was calculated according to the equation described by Pearce et al. [18].

Pulmonary shunt fraction (Qs/Qt), PaO₂/FiO₂ ratio, and systemic and pulmonary vascular resistance indexes were calculated according to standard formula.

Histological analysis was performed on tissue samples taken from the trachea, bronchioles, and alveoli. Scoring was performed for airway congestion, septal edema, alveolar edema, and alveolar polymorphonuclear neutrophils (PMNs) according to a previously established scoring system [19].

Statistical analysis

The data were compared using analysis of variance (ANOVA) for repeated measures. Bonferroni post hoc tests were performed for comparisons between groups at each time point, and student’s t-tests were performed when appropriate. All values are expressed as mean ± standard error of the mean (SEM). The statistical analysis was performed using GraphPad Prism Software (GraphPad Software, La Jolla, CA, USA). P-values < 0.05 were considered significant.

Baseline variables

There were no differences between the two study groups at baseline within the measured variables. Body weight was similar in both groups (35.4 ± 0.8 kg in the treatment group vs. 35.2 ± 0.9 kg in the control), and there were no differences in body surface area between the groups (0.91 ± 0.01 m² in the treatment group vs. 0.91 ± 0.01 m² in the control). The arterial COHb determined immediately after the smoke-inhalation injury averaged 69.1 ± 2.0% in the control group and 64.4 ± 2.5% in the R-100 group; these levels were not significantly different from each other indicating a similar degree of injury. The core body temperatures similarly increased in both groups and remained comparable. Lactate levels after 3 h were 3.4 ± 1.1 mg/dL in the R-100 group and 2.9 ± 1.0 mg/dL in the control group, indicating tissue hypoxia in both groups. These values rose throughout the experiment and there were no differences between the groups.

Pulmonary gas exchange and global oxygen transport

The PaO₂/FiO₂ (P/F) ratio declined in both animal groups, however at 24 h the R-100 group had a significantly higher P/F ratio than the control group (P < 0.05) (Fig. 1a). The pulmonary shunt fraction (Qs/Qt) increased in both groups implying that the fraction of the blood that was shunted past the lungs was higher as the experiment proceeded. The groups did not differ from each other with respect to mean Qs/Qt means (Fig. 1b).

The peak airway pressure increased in both groups vs. baseline. However, the R-100 group had a significantly lower peak airway pressure than the control group during the last 12 h of the experiment (P < 0.05). The pause airway pressure also increased in both groups, but was lower in the R-100 group (P < 0.05 at 24 h) (Fig. 2).

Cardiopulmonary hemodynamics

Comparisons between the groups with respect to heart rate, cardiac output, central venous pressure, pulmonary capillary wedge pressure left atrial pressure (increased in both groups vs. baseline) and mean arterial blood pressure and systemic vascular resistance index (decreased in both groups vs. baseline) showed no significant differences.

Pulmonary artery pressure increased throughout the experiment in both groups compared to baseline. However, at 24 h after injury, the R-100 sheep had a significantly lower pulmonary arterial pressure than the control sheep (P < 0.05) (Table 1).

Fluid balance and plasma protein content

Arterial hemoglobin and hematocrit levels were stable in both groups throughout the study period. Urine output declined in both groups between hours 12 and 18. The R-100 treated-animals appeared to excrete urine at a higher rate than the control group at 24 h, but the difference was not significant (Fig. 3a). The treated group required less fluid to maintain the hematocrit (Fig. 3b).

The body fluid accumulation tended to be higher in the control group throughout the study period. The vehicle-treated control sheep retained 595 ± 234 mL of fluid between 21 and 24 h vs. 190 ± 45 mL in R-100-treated sheep. A significant difference was noted at 24 h between the two groups (Fig. 4).

The post-mortem wet-to-dry weight ratio was slightly higher in the control group. However no significant differences were found between the groups (6.54 ± 0.53 in the control group vs. 6.03 ± 0.70 in the treatment group).

Histological analysis