Care of the animals, techniques, and procedures were approved by the Animal Care and Use Committee of Chonnam National University Hospital. New Zealand white rabbits weighing 2.76±0.06 kg were anesthetized with an intramuscular injection of a ketamine hydrochloride (30 mg/kg) and xylazine (0.3 mg/kg). An endotracheal tube (3.5 mm internal diameter) was inserted via a tracheostomy. Mechanical ventilation was initiated in the pressure-controlled mode (IV-110B, Sechrist infant ventilator, Sechrist industries, Anaheim, CA, U.S.A.) with a peak inspiratory pressure (PIP) of 15 cm H₂O, a positive end-expiratory pressure (PEEP) of 3 cm H₂O, a frequency of 25 breaths/min, inspiration-to-expiration (I:E) ratio of 1:2 and an inspired oxygen fraction (FIO₂) of 0.40. Anesthesia and paralysis were maintained throughout the experiment by continuous infusions of sodium pentobarbital (2-4 mg/kg/hr) and pancuronium bromide (0.03-0.07 mg/kg/hr). The internal carotid artery was cannulated with a 20-gauge catheter (Custom Product, Abbott Ireland, Sligo, Republic of Ireland) to monitor systemic arterial pressure and heart rate. Arterial blood gas samples were analyzed at 37℃ (Rapidlab 865, Chiron diagnostics corporation, U.K.) and corrected for body temperature. Rectal temperature was monitored and maintained within the range of 37±1℃ using a radiant heating lamp.

After recording baseline hemodynamic and gas exchange, 55 rabbits were randomly assigned to one of three groups (20 high pressure normocapnic group, 20 high pressure hypercapnic group, 15 control group). Control groups were ventilated with same baseline settings during the experiments (PIP of 15 cm H₂O, PEEP of 3 cm H₂O, I:E ratio of 1:2, and a FIO₂ of 0.40). High pressure hypercapnic group (HPHC) was ventilated with PIP of 30 cm H₂O, PEEP of 0 cm H₂O and FIO₂ of 0.40. Carbon dioxide was introduced into the inspiratory limb of the ventilator circuit, as necessary to maintain hypercapnia (PaCO₂, 65 to 75 mmHg). High pressure normocapnic group (HPNC) was ventilated with same setting of HPHC, except normocapnia (PaCO₂, 35 to 45 mmHg). Arterial blood gases were measured at every 1 hr during the experiment. Blood samples for lactate dehydrogenase (LDH) and aspartate aminotransferase (AST) were obtained at every 2 hr and measured by enzyme-linked immunosorbent assay (Olympus, AU 5400, Japan). After completion of the protocol, the heart-lung block was then removed after exsanguinations of the heparinized animals.

The bronchoalveolar lavage fluid (BALF) was harvested from the right lung at the end of the experiment. Through the right mainstem bronchus, 35 mL of saline was infused slowly and withdrawn five times. The BALF was centrifuged and then divided into several aliquots to measure LDH, AST, and interleukin-8 (IL-8) concentrations. IL-8, a potent neutrophil attractant and activator, has been implicated in the neutrophil recruitment in lungs of patients with ARDS (15). Nakamura et al. (16) demonstrated that IL-8 level in BALF was increased and paralleled the BAL neutrophil counts during the lung injury. We examined the level of IL-8 in the BALF by enzyme-linked immunosorbent assay (ELISA).

Heparinization and exsanguinations were used to minimize the effect of residual blood on gravimetric indices. Left lung was physically sectioned into dorsal and ventral regions that were weighed separately to obtain wet weight (WW) and then desiccated in a vacuum oven at 60℃ to dryness for 4, 5 days until a constant dry weight (DW).

The resected lungs were infused with 4% paraformaldehyde solution from a tracheostomy tube. Subsequently, the trachea was clamped and the lungs were placed in 4% buffered paraformaldehyde solution. The lungs were divided into dorsal and ventral regions along a coronal plane connecting the mid-height caudal and cranial portions of the lungs. Subdivision into caudal and cranial sectors was accomplished by sectioning the lungs (in the transverse plane) along upper and lower lobe fissures. To avoid sampling bias, tissue samples of 3 µL were excised (regardless of macroscopic appearance) from the mid-portion of each sector. The samples were sectioned and stained (hematoxylin and eosin). The sections were later characterized using a semiquantitative grading system by a pathologist who was blinded as to the experimental protocol and region of sampling. After viewing approximately ten fields per sector under low and high power, each section was assigned a numerical histologic score ranging from 1 to 4, based on the degree of neutrophilic infiltration, hemorrhage, and edema in the interstitial and alveolar spaces as follows; 1 (normal), normal appearing lung; 2 (mild), minimal congestion, interstitial edema and interstitial neutrophilic infiltrate with occasional red blood cells and neutrophils in the alveolar spaces; 3 (moderate), moderate congestion and interstitial edema with neutrophils partially filling the alveolar spaces but without consolidation; 4 (severe), marked congestion and interstitial edema, with neutrophilic infiltrate nearly filling the alveolar spaces, or with frank lung consolidation. Atelectasis per se was disregarded and not scored as abnormality.

All values are reported as mean±standard deviation (SD), unless specified otherwise. Separate one-way analysis of variance (ANOVA) procedures were used to test for difference in variables. In each group, changes in measured variables over time were analyzed using a repeated-measures analysis of variance. A statistical package, SPSS 12.0 for Windows (SPSS Inc., Chicago, IL, U.S.A.), was used to analyze the data. A value of p<0.05 was considered significant.

There were no significant differences in arterial carbon dioxide tension (PaCO₂), and pH among the groups under baseline conditions. In the HPHC group, PaCO₂ was increased and pH decreased quickly after the induction of carbon dioxide as compared with baseline values and then remained stable throughout the experiment. PaCO₂ remained unchanged in the HPNC and control groups at each time point (Fig. 1A). In the HPNC and control groups, pH tended to be lower slowly throughout the experiment. However, there were no significant differences in pH from baseline values (Fig. 1B).

There were no significant differences in oxygenation and systemic arterial pressure among the groups under baseline conditions. In the HPHC and HPNC groups, systemic arterial pressure trended downward during the experiment as compared with control group. However, there were no significant differences in systemic arterial pressure between HPHC and HPNC group. In the HPHC and HPNC groups, PaO₂ decreased progressively during the experiment. In the HPHC, PaO₂ decreased significantly after 3 hr as compared with baseline values (171.6±7.9 vs. 113.4±12.1 mmHg; p<0.05). In the HPNC group, PaO₂ decreased significantly after 4 hr as compared with baseline values (173.5±11.6 vs. 112.3±11.7 mmHg; p<0.05). At the end of the protocol, however, there was no significant difference in oxygenation between HPHC and HPNC groups (Fig. 2).

Serum levels of AST and LDH elevated progressively during the experiments in injurious ventilator patterns (HPHC and HPNC groups) when compared with the control group. There were no significant differences of serum levels of AST and LDH between HPHC and HPNC groups at each time point (Fig. 3). In the control group, the trend toward elevated levels of AST and LDH over time was not statistically significant when compared with baseline values.

The levels of LDH, AST and IL-8 in the BALF, measured at the end of the experimental period, were significantly higher in the HPHC and HPNC groups when compared with the control group (p<0.05) (Fig. 4). There were no significant difference in BALF LDH, AST, and IL-8 levels between HPHC and HPNC group.

The ratio of wet weight to dry weight (WW/DW) for dependent and nondependent regions was significantly greater in injurious ventilatory pattern (HPHC and HPNC group) than the control group. The WW/DW for dependent regions was highest in the HPHC, intermediate in the HPNC, and lowest in the control group (HPHC: 7.88±1.52; HPNC: 7.42±1.25; control: 5.09±0.47; p<0.05). There were no significant difference in WW/DW for dependent regions between HPHC and HPNC group. The WW/DW for nondependent regions was also the highest in the HPHC, intermediate in the HPNC, and lowest in the control group (HPHC: 6.12±0.97; HPNC: 5.84±1.17; control: 4.21±0.29; p<0.05). There were no significant difference in WW/DW for nondependent regions between HPHC and HPNC group (Fig. 5).

Gross postmortem inspection of the lungs in the HPHC and HPNC group revealed edema and hemorrhage that were most prominent in the dependent regions. Histologic scores for dependent and nondependent regions were significantly higher in injurious ventilator patterns (HPNC and HPHC group) than the control group. However, no significant differences was found in histologic scores for dependent and non-1HPHC and HPNC group (Fig. 6).