Assessment of hyperbaric hyperoxic lung injury in rats

Oxygen is indispensable for human life, but exposure to elevated oxygen partial pressure may also cause damage to the human body, which is also known as oxygen toxicity. The animal model of pulmonary oxygen toxicity (POT) is important for the investigations on the pathogenesis and treatment of POT. Theoretically, exposure to normobaric hyperoxia may also damage the lungs, but the time to lung injury will be very long (an oxygen partial pressure at 75 kPa may cause toxicity to lungs within 24 hours).1 Moreover, the normobaric hyperoxic lung injury is different from the hyperbaric hyperoxic lung injury.2 Thus, in the establishment of POT animal model in the field of diving medicine, the elevation of oxygen partial pressure is employed to shorten the time to lung injury. In our previous studies, an animal model was established by exposing rats to pure oxygen at 250 kPa for 6 hours,3 and results showed this hyperoxic exposure could cause lung edema and significantly increase the pro-inflammatory factors and reactive oxygen species in rat lungs.4 However, these were biochemical findings in this animal model, and the functional parameters (such as lung function and blood gas) have not been determined in this model. In the present study, the same animal model was established in rats, lung function was non-invasively assessed before and after exposure, and the arterial blood was collected for the blood gas analysis. A total of 26 male Sprague-Dawley rats (Experimental Animal Center of Naval Medical University, Shanghai, China; license No. SYXK (Hu) 2017-0019) weighing 212 ± 11 g were used in the present study. Rats were randomly divided into control group (n = 7) and hyperoxia induced lung injury (HILI) group (n = 15). This study was approved by the Ethics Committee of Naval Medical Center (No. NMU20230213). HILI: Animals received hyperbaric oxygen exposure. In brief, animals were placed in a hyperbaric chamber which was first flushed with pure oxygen for about 5 minutes until the concentration of oxygen in the chamber was higher than 95%. Then, animals were pressurized to 250 kPa within 5 minutes, and exposed to hyperbaric oxygen for 6 hours. The oxygen concentration was monitored during the hyperbaric exposure. Blood gas analysis: After exposure, rats were intraperitoneally anesthetized with 1% phenobarbital sodium at 0.3 mL/100 g body mass, and blood was collected from abdominal aorta for the blood gas analysis (i-STAT 1 blood chemistry analyzer, Abbott, FL, USA). The pH value increased markedly immediately after exposure (P < 0.0001, vs. before exposure), but reduced dramatically at 48 hours (P < 0.001, vs. 0 hours) and the pH value at 48 hours was still significantly higher than that before exposure (P < 0.001). The partial pressure of oxygen (PO2) increased significantly immediately after exposure (P < 0.001, vs. before exposure) and then reduced at 48 hours, and the PO2 at 48 hours was still markedly higher than that before exposure (P < 0.001). The partial pressure of carbon dioxide (PCO2) reduced markedly immediately after exposure (P < 0.001, vs. before exposure), and then increased significantly at 48 hours (P < 0.05, vs. 0 hours), but the PCO2 was still markedly lower than that before exposure (P < 0.001). The oxygen saturation (sO2) increased dramatically immediately after exposure (P < 0.0001, vs. before exposure) and the reduced slightly at 48 hours, and the sO2 was still markedly higher than that before exposure (P < 0.0001). The total carbon dioxide reduced significantly immediately after exposure (P < 0.05, vs. before exposure) and further reduced markedly at 48 hours (P < 0.05, vs. 0 hours), and the total carbon dioxide at 48 hours was significantly lower than that before exposure (P < 0.0001). The bicarbonate (HCO3–) reduced slightly immediately after exposure (P > 0.05, vs. before exposure) and further reduced significantly at 48 hours (P < 0.05, vs. 0 hours), and the HCO3– at 48 hours was markedly lower than that before exposure (P < 0.01). The extracellular fluid base excess increased slightly at 0 hours after exposure (P > 0.05, vs. before exposure), but it reduced markedly at 48 hours (P < 0.01, vs. 0 hours), and the extracellular fluid base excess at 48 hours was significantly lower than that before exposure (P < 0.05). The glucose level reduced at 0 and 48 hours after exposure, but there was no marked difference in the glucose level among control and 0 and 48 hours after exposure (P > 0.05). The Na+ level reduced slightly at 0 hours after exposure (P > 0.05, vs. before exposure), but it increased significantly as compared to that before exposure (P < 0.05) and at 0 hours (P < 0.01). The K+ level reduced at 0 and 48 hours after exposure (P > 0.05, vs. before exposure), and there was no marked difference in K+ level at two time points after exposure (P > 0.05). The ionized calcium (iCa2+) level increased at 0 and 48 hours after exposure (P > 0.05, vs. before exposure), and there was no marked difference in iCa2+ level at two time points after exposure (P > 0.05). Both hemoglobin (Hb) level and hematocrit (Hct) increased markedly at 0 hours after exposure (P < 0.01, vs. before exposure), and then reduced dramatically (P < 0.001, vs. 0 hours), but no significant differences were noted in the Hb level and Hct before exposure and at 48 hours (P > 0.05; Figure 1).Figure 1: Blood gas analysis of animals before and after hyperbaric hyperoxia exposure.(A) pH value; (B) PO2; (C) PCO2; (D) sO2; (E) TCO2; (F) HCO3 –; (G) BEecf; (H) Glu; (I) Na+; (J) K+; (K) iCa2+; (L) Hb; (M) Hct. Data are expressed as mean ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 (one-way analysis of variance followed by Tukey post hoc test). BEecf: Extracellular fluid base excess; Glu: glucose; Hb: hemoglobin; HCO3 –: bicarbonate; Hct: hematocrit; iCa: calcium; K: potassium; Na: sodium; pCO2: partial pressure of carbon dioxide; pO2: partial pressure of oxygen; sO2: oxygen saturation; TCO2: total carbon dioxide.Lung function detection: Animals received non-invasive lung function detection (WBP; TOW-INT TECH, Shanghai, China) 1 hour before exposure, and immediately, 24 hours and 48 hours after exposure. The time of inspiration increased significantly at 0 hours after exposure, but it reduced markedly at 24 and 48 hours after exposure (P < 0.01, vs. 0 hours), and there was no significant difference in the time of inspiration at 24 and 48 hours (P > 0.05). The time of expiration reduced slightly at 0 hours after exposure, further decreased at 24 hours, and then remained relatively stable at 48 hours after exposure (P > 0.05, vs. 24 hours). Significant difference was noted in the time of expiration before exposure and at 24 hours (P < 0.05). The peak inspiratory flow remained stable immediately after exposure, but reduced at 24 hours after exposure and then it returned to nearly normal level at 48 hours. There was no marked difference in the peak inspiratory flow at different time points (P > 0.05). The peak expiratory flow markedly increased immediately after exposure (P < 0.01, vs. before exposure), and then reduced markedly at 24 hours (P < 0.05, vs. 0 hours), but it remained relatively stable at 48 hours as compared to that at 24 hours (P > 0.05). The inspiratory time/expiratory time markedly increased immediately after exposure, then reduced dramatically at 24 hours (P < 0.01, vs. 0 hours) and thereafter remained relatively stable at 48 hours (P > 0.05, vs. 24 hours). The respiratory frequency reduced markedly at 0 hours after exposure (P < 0.05, vs. before exposure), but increased dramatically at 24 hours (P < 0.01, vs. 0 hours) and it remained stable at 48 hours as compared to that at 24 hours (P > 0.05). The minute volume increased significantly at 0 hours after exposure (P < 0.05, vs. before exposure), but then reduced at 24 hours (P > 0.05, vs. 0 hours) and it slightly increased at 48 hours as compared to that at 24 hours (P > 0.05). The tidal volume at 0 hours after exposure was significantly higher than that before exposure (P < 0.01), but it reduced dramatically at 24 hours as compared to that at 0 hours (P < 0.01) and then it slightly increased at 48 hours (P > 0.05, vs. 24 hours). The expiratory flow at 50% volume increased significantly at 0 hours (P < 0.01, vs. before exposure), but reduced dramatically at 24 hours (P < 0.05, vs. 0 hours), and it increased markedly at 48 hours (P < 0.01, vs. 24 hours; Figure 2).Figure 2: Lung function examination of animals before and after hyperbaric hyperoxia exposure.(A) TI; (B) TE; (C) PIF; (D) PEF; (E) VOLBAL; (F) f; (G) MV; (H) VT; (I) EF50. Data are expressed as mean ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001 (repeated measurement analysis of variance followed by Tukey post hoc test). EF50: Expiratory flow at 50% volume; f: respiratory frequency; MV: minute volume; PEF: peak expiratory flow; PIF: peak inspiratory flow; TE: time of expiration; TI: time of inspiration; Volbal: inspiratory time/expiratory time; VT: tidal volume.Histological examination: Animals in the HILI group were sacrificed immediately and 1 week after exposure, and the transcardially perfused with normal saline and then with 4% paraformaldehyde. The lung was harvested for the hematoxylin-eosin staining and Masson staining. For hematoxylin-eosin staining, sections were observed under a light microscope. Lung injury was scored using a four-point scale (1 indicating normal and 4 representing the most severe injury)5: (1) alveolar congestion, (2) alveolar hemorrhage, and (3) infiltration or aggregation of neutrophils in the alveolar space or vascular wall, alveolar wall thickening, and/or hyaline membrane formation. Results showed the lung was injured significantly immediately after exposure, which was characterized by thickening and rupture of alveolar septum, infiltration of inflammatory cells, and leakage of red blood cells. At 1 week after exposure, these histological manifestations were still present, but they were improved to a certain extent (Figure 3A). Pathological analysis showed the pathological score increased significantly after hyperoxia exposed (P < 0.001, vs. control group), but it reduced markedly 1 week later (P = 0.001, vs. HILI-0 h group; Figure 3B). One week after exposure, Masson staining showed no significant fibrosis in the lung (Figure 3A).Figure 3: HE staining and Masson staining of lung tissues after hyperbaric hyperoxia exposure.Immediately after exposure, thickening and rupture of alveolar septum, infiltration of inflammatory cells, and leakage of red blood cells were observed. At 1 week after exposure, these histological manifestations were improved to a certain extent (A). No significant fibrosis was observed in the lung 1 week after exposure (A). The pathological score increased significantly after hyperoxia, but it reduced markedly 1 week after exposure (B). The lung microstructure was scored according to the alveolar congestion, alveolar hemorrhage, and infiltration or aggregation of neutrophils in the alveolar space or vascular wall, alveolar wall thickening, and/or hyaline membrane formation. Data are expressed as mean ± SD, and were analyzed by one-way analysis of variance followed by Tukey post hoc test. HE: Hematoxylin-eosin; HILI: hypoxia-induced lung injury.In diving and hyperbaric exposure, people may experience elevated oxygen partial pressure, but prolonged exposure to elevated oxygen partial pressure may cause damage to the lung. Thus, the establishment of an animal model is crucial for the investigation of lung injury under these conditions. In most studies, animals are exposed to normobaric hyperoxia to induce lung injury, mainly in neonatal animals to mimic the bronchopulmonary dysplasia in humans. Crapo6 divided the pathological process after exposure to a lethal dose of hyperoxia (e.g. 100% O2 at 100 kPa) into three phases: initiation phase, inflammatory phase and destructive phase; the pathological process after exposure to a sublethal dose of hyperoxia was composed of initiation phase, inflammatory phase, destructive phase, proliferative phase and fibrotic phase. In the initiation phase, there is no significant evidence of morphologic injury. In the inflammatory phase and destructive phase, the lung structure is disrupted, which is characterized by the infiltration of inflammatory cells, thickening and rupture of alveolar septum, leakage of blood cells into the alveolar space, and so on. In the present study, hematoxylin-eosin staining was employed for the assessment of lung pathology. Our results indicate that exposure to pure oxygen at 250 kPa may cause acute lung injury, which may not induce pulmonary fibrosis. Lung function is an important clinical parameter to assess the lung injury.7 Our results showed the time of inspiration increased but time of expiration reduced significantly, which indicates the increased inspiratory resistance and the reduced expiratory resistance after exposure. These corresponded to the reduced peak inspiratory flow, elevated peak expiratory flow and increased expiratory flow at 50% volume after exposure. In addition, the frequency reduced significantly, but tidal volume and minute volume increased after exposure. The reduced frequency indicates the elevated resistance, and the increased tidal volume and minute volume may suggest the deep breath after lung injury following hyperoxia exposure. These changes in lung function nearly returned to normal at 48 hours, indicating an acute lung injury. The lung injury may directly affect the interchange of gases in the lung, influencing the blood gas, and thus blood gas analysis may be also helpful for the assessment of lung injury. The PaCO2 measurements were different among studies. Some reported significant hypocapnia breathing 100% oxygen at high pressures, but others did not find any significant change. In simulated SCUBA diving and other conditions, PaO2 and PaCO2 were found to increase without hypercapnia. Of note, these findings were from studies in healthy subjects who were exposed to an environment without risk for evident POT. In the present study, blood gas analysis showed that the pH value increased significantly after exposure, corresponding to the elevation of extracellular fluid base excess after exposure, which may be ascribed to the reductions of CO2 partial pressure, total CO2 content and HCO3–. The reduced frequency and deep breath after exposure may suggest the hyperventilation, which may be related to the reduced CO2 partial pressure, total CO2 content and HCO3–. However, the CO2 partial pressure, total CO2 content and HCO3– still remained at markedly lower levels at 48 hours after exposure as compared to those at baseline. The oxygen partial pressure and oxygen saturation increased markedly at 0 hours after exposure, which may be ascribed to the hyperoxia, but they still remained at markedly higher levels at 48 hours after exposure. Divers usually develop dehydration after hyperbaric exposure.8 In our study, the blood Hb level and Hct increased significantly at 0 hours after exposure. The elevated blood Hb level and Hct might be ascribed to the dehydration after exposure. Sengun et al.9 also reported increased Hct after a dive. Hct reflects changes in both red cell mass and plasma volume, and it has been widely used to assess the dehydration level in divers. For divers developing dehydration after a dive, fluid lost from the intravascular space may concentrate the blood and thereafter increase Hct. Of note, in the study of Sengun et al.,9 a significant dehydration was observed during a total dive time although the bottom time was very short, and mild dehydration was recovered despite the long surface decompression time. In addition, splenic contraction may be also related to the increase in Hb and Hct. Splenic contraction is a component of human diving response, and it has been reported that a single contraction of the spleen can expel about 160 mL of red blood cells, which may cause a Hb increase corresponding to a 2.8–9.6% increase in oxygen content.10 Taken together, our results showed short-term (6 hours) exposure to pure oxygen at 250 kPa may cause acute lung injury: the microstructure is disordered, lung function is compromised, and blood gases also change. However, the lung function and most parameters (except for PO2, PCO2, sO2, total carbon dioxide and HCO3–) of blood gases seem to recover within 48 hours. This study was supported by the Talent Development of Deep Blue Project of Naval Medical University (No. 24TPSL0101). Wenwu Liu is an Editorial Board member of Medical Gas Research. He is blinded from reviewing or making decisions on the manuscript. The article was subject to the journal’s standard procedures, with peer review handled independently of this Editorial Board member and his research group.