where m is the mass of the aluminum weight (in kilograms), g is gravitational acceleration (9.8 m/s2), and h is the height of weight above the Lexon platform (in meters). Calculations assumed that all the potential energy of the weight was transferred to the animal, neglecting frictional dissipation. The heights for the hollow cylindrical weight above the chest were calculated to generate external chest impact energies of 1.8, 2.0, 2.2, 2.45, and 2.7 J.
Arterial oxygenation (arterial partial pressure of oxygen, Pao2) was measured at 48 h in all groups, and additionally at 8 min, 4, and 24 h for animals injured with 2.45 J of impact energy. At a given time, subgroups of animals were allowed to breathe 98% O2 for 5 min (fraction of inspired oxygen [Fio2] = 0.98). Halothane anesthesia was then induced, and a midline incision made through the peritoneum. Blood (0.5 mL) was collected from the descending aorta in a heparinized syringe, followed by analysis with an ABL5 blood gas analyzer (Radiometer America, Westlake, OH).
Pulmonary P-V mechanics were measured immediately after blood samples were collected at 48 h post-contusion. A 14-gauge steel cannula was inserted into the trachea of a halothane-anesthetized rat through a 2-cm ventral midline neck incision, and was secured with a silk suture. The animal was then exsanguinated by transection of the abdominal inferior vena cava. Air was injected into the lungs at a rate of 25 mL/min by a syringe pump connected to the tracheal cannula. Inflation pressure was monitored continuously by an in-line pressure transducer connected to an Apple PowerBook G4 (Apple Computer, Cupertino, CA) equipped with a National Instruments data acquisition board (Austin, TX) and software written by the laboratory in Lab VIEW 6.0 (National Instruments). When pressure reached 40 cm H2O, the syringe pump was reversed and deflation pressures were monitored. Volumes were calculated based on the rate of injection or withdrawal, and were normalized to kilogram body weight.
After measurements of mechanics, a ventral midline incision was made through the sternum and the lung vasculature was flushed by injecting 20 mL of Hanks balanced salt solution into the beating right ventricle. BAL was then performed by alternating injection and collection of 5 × 10 mL of 37°C normal saline through the tracheal cannula. Recovered BAL fluid was centrifuged at 1500g at 4°C for 3 min to pellet cells, and the supernatant analyzed for albumin content (described below). The cell pellet was resuspended in 4 mL of phosphate-buffered normal saline + 0.1% sodium azide, and total numbers of erythrocytes (RBCs) and leukocytes (WBCs) were determined using a Multisizer 3 Coulter Counter (Beckman Coulter, Fullerton, CA). Differential cell counts were performed on cytocentrifuge preparations (Cytospin 3; Shandon Southern Instruments, Sewickley, PA) stained with Diff-Quik (Baxter, Detroit, MI).
Albumin concentrations in BAL fluid were determined by ELISA with a polyclonal rabbit anti-mouse albumin antibody (provided by Dr. Daniel Remick, University of Michigan, Ann Arbor, MI), a horseradish peroxidase-labeled goat anti-rabbit immunoglobulin G (BD Biosciences Pharmingen, San Diego, CA), and rat albumin (Sigma, St. Louis, MO) as a standard.
Histopathological evaluations were blinded and were performed by an experienced laboratory pathologist. Pulmonary and cardiac tissue were obtained from rats over a 48-h period after initial chest contusion with maximal sublethal impact energy of 2.45 J. For histological assessments of pulmonary tissue, the lungs were gradually inflated with 1% formalin, and tissue sections were stained with hematoxylin and eosin (H&E). Histological assessments of cardiac tissue also used H&E staining, and included evaluations of all four chambers of the heart in multiple sections.
Data are expressed as mean ± sem (standard error of the mean) for n animals per group. One-way ANOVA with Dunn’s post hoc test was used for intergroup pairwise comparisons. Values for correlations such as r value and P value (Fisher’s r-z test) were ascertained with the Statview software program.
Three of the 39 injured rats failed to survive (Table 1). Two of 6 rats in the 2.7-J chest impact energy group died as a result of major intrathoracic bleeding from lung lacerations, as determined at necropsy. In addition, 1 of 9 rats in the 2.2-J group died of intrathoracic bleeding from a lacerated liver without injury to the heart. All three rats that failed to survive died without waking from anesthesia immediately after the initial blunt trauma insult. Rats that received external blunt chest impact energies of 2.45, 2, and 1.8 J all survived (Table 1). Mortality in the 2.7-J impact energy group (2 of 6 animals) exceeded a 15% limit that was chosen prospectively in consultation with the IACUC to define sublethal lung contusion injury. Thus, additional experiments used 2.45 J as the maximal sublethal chest impact energy condition.
All surviving animals demonstrated no evidence of intrathoracic or intraabdominal bleeding or visceral injury, and radiography and physical evaluation did not reveal any associated rib fractures. In addition, injured animals exhibited normal movement, activity, and behavior in room air throughout the 48-h period after lung contusion. More detailed gross observations were made in rats injured with a chest impact energy of 2.45 J. Gross examination at death indicated that, at least superficially, both lungs were affected by contusion in a relatively uniform pattern in injured animals. Areas of contusion were acutely hemorrhagic (e.g., at 4 h). By 24 h postinjury, the contused lungs had an altered gross appearance, with prominent purplish surface discolorations. By 48 h, areas of contused lung appeared to be brownish in color. In contrast to the lungs, there was no evidence of significant cardiac trauma on gross examination.
Histological sections were examined to assess the severity of pulmonary and cardiac tissue injury in rats receiving the maximal sublethal chest impact energy of 2.45 J. At 4 h and 12 h post-contusion, pulmonary tissue showed diffuse areas of intraalveolar hemorrhage with disruption of alveoli (Fig. 2A). Perihilar tissue was also involved, with affected areas of parenchyma in many cases extending to the visceral surface of the pleura. Similar patterns of acute injury were apparent bilaterally, although quantitative morphometrics were not done. At 24 h post-contusion, a predominantly neutrophilic pattern of leukocyte infiltration was apparent in the alveolar spaces, and significant atelectasis was also observed (Fig. 2B). At 48 h, thickening of the alveolar lining with a continuing leukocytic infiltration was apparent along with intraalveolar edema (Fig. 2C). Tissue examination of all four chambers of the heart did not reveal any substantial disruption of cardiac muscle at any time (Fig. 2D–F at 12 and 24 h post-contusion).
The ratio of Pao2/Fio2 was determined at 48 h post-contusion in rats with sublethal lung contusion injury (1.8–2.45 J). Before blood sampling from the abdominal aorta, animals breathed 98% oxygen to fix Fio2 at 0.98. This brief period of high Fio2 breathing highlighted differences in hypoxemia between experimental groups, and also facilitated subsequent absorption atelectasis to standardize P-V compliance measurements. Pao2/Fio2 ratios at 48 h in rats given chest impact energies of 1.8–2.45 J were equivalent to uninjured controls (e.g., Table 2 for 2.45 J). However, Pao2/Fio2 ratios at 8 min, 4 h, and 24 h post-contusion in rats receiving the 2.45-J level of impact indicated severe acute arterial hypoxemia, with Pao2/Fio2 ratios of 109 ± 14 mm Hg at 1 h and 169 ± 25 mm Hg at 4 h after contusion (Table 2). Significant arterial hypoxemia was also evident at 24 h post-contusion in rats injured with 2.45 J of chest impact energy (Table 2).
Albumin levels in BAL were measured to assess permeability injury to the alveolocapillary membrane, as in previous studies of acute inflammatory lung injury (15–19). At 48 h post-contusion, injured rats showed significant energy-dependent increases in BAL albumin (r value = 0.62 and P value < 0.001 by Fisher’s r-z test) (Fig. 3). Albumin levels in BAL at 48 h post-contusion were 211 ± 53 μg/mL for rats receiving 2.45 J of chest impact energy, and 146 ± 34 μg/mL for those receiving 2.2 J, compared with 22.7 ± 3 μg/mL in controls (Fig. 3). Additional time-course studies showed that albumin levels in BAL from rats injured with 2.45 J of chest impact energy were increased to an even greater extent in the first 24 h post-contusion (Table 2). Early increases in BAL albumin levels were consistent with the severe arterial hypoxemia observed in these animals at 8 min and 4 h post-contusion (Table 2).
The cellular response in BAL varied with both chest impact energy and time. At 48 h post-contusion, the total numbers of leukocytes (WBCs) in BAL increased as chest impact energy increased (Fig. 4). Animals injured with 2.45 or 2.2 J had the most total WBCs per milliliter in BAL fluid (5.18 ± 3.89 × 107 and 3.44 ± 2.43 × 107, respectively, compared with 9.19 ± 1.13 × 106 in noninjured controls; Fig. 4). Additional studies in animals injured with 2.45 J of chest impact energy showed that the numbers of neutrophils in BAL were significantly increased at 24 h post-contusion (Table 2). Animals injured with 2.45 J of impact energy also had increased numbers of RBCs in BAL compared with uninjured controls at 8 min, 4 h, and 24 h post-contusion, but with no significant differences between values at these time points (Table 2).
Quasistatic P-V inflation/deflation measurements were done at 48 h post-contusion in rats given different levels of chest impact energy (1.8–2.45 J) (Fig. 5). Both inflation and deflation lung volumes at fixed pressures more than approximately 8 cm H2O were decreased in injured animals relative to controls. Decreases in lung volumes were most pronounced in rats injured with the two highest sublethal chest impact energies investigated (2.2 and 2.45 J) (Fig. 5).
This study has defined a reproducible, sublethal model of isolated bilateral lung contusion in rats induced by focused external blunt chest trauma. Rats receiving 1.8–2.7 J of external chest impact energy were examined. To develop a model that would be most applicable for future mechanistic studies on the evolution and pathophysiology of blunt trauma-induced lung contusion, an arbitrary upper limit of 15% mortality was prospectively chosen for maximal sublethal injury. This mortality limit was low to guarantee substantial numbers of survivors for lung injury assessments, whereas still allowing for some animal deaths in response to a significant blunt chest trauma. Rats injured with 2.7 J of chest impact energy had mortality exceeding this limit (Table 1), whereas rats receiving 2.45 and 2.2 J of impact energy had the most severe sublethal injury. Injured rats had no rib fractures on chest radiographs, and bilateral lung contusion was present on gross or histological examination without substantial disruption of cardiac muscle tissue (Fig. 2).
Isolated lung contusion injury was found to be dependent on time in addition to impact energy. Rats injured with 2.45 J of external chest impact energy had significantly increased levels of albumin and inflammatory leukocytes in BAL at 48 hours post-contusion (Figs. 3 and 4), as well as decreased inflation/deflation lung volumes at fixed pressure (Fig. 5). Hypoxemia had significantly improved in these injured animals at 48 hours, but time course studies for rats receiving 2.45 J of impact energy showed substantial acute reductions in Pao2/Fio2 ratios that met criteria for clinical ALI and/or ARDS at 8 minutes, 4 hours, and 24 hours (Table 2). Albumin levels in BAL from rats injured with 2.45 J of chest impact energy were significantly increased over controls at 8 minutes, 4 hours, 24 hours, and 48 hours post-contusion, with peak increases occurring at the earlier times (Table 2). Rats receiving 2.45 J of chest impact energy also had an acute inflammatory response, with increased numbers of neutrophils at 24 hours post-contusion (Table 2). This cellular response became more lymphocytic by 48 hours, consistent with a transition from an acute injury to a subacute or reparative stage. Histological analysis also confirmed the presence of neutrophils at 24 hours (Fig. 2).
Blunt chest trauma is a common problem in the care of critically ill trauma patients (20), and thoracic trauma accounts for 20%–25% of adult deaths caused by trauma (21). Lung contusion is the most frequently diagnosed intrathoracic injury resulting from blunt trauma (1,2), and is an important risk factor for the development of other conditions such as pneumonia and ALI/ARDS. Various large-animal models for lung contusion have been developed, including studies in canines, swine, and monkeys (4–13) (Table 3). In addition, a rat model of combined lung and heart contusion is available (22), and a blast-induced lung contusion model in mice has also been reported (23). The present rat model has the advantages of being specific in terms of lung contusion, and also uses an impact induction method that is highly relevant for clinical blunt thoracic trauma (e.g., as occurs in motor vehicle accidents). As described in Table 3, many of the features and responses found in our model are consistent with lung contusion injury in humans and in large-animal models.
One important feature of the current rat model is that it involves isolated (or at least relatively isolated) lung contusion injury without substantial cardiac or abdominal trauma. Wang et al. (22) reported an animal model of thoracic trauma without protection of the heart in which mortality was primarily associated with blunt myocardial injury (Table 3). In the present study, cardiac trauma was minimized by the use of a special platform and precordial shield that directed the energy of external blunt chest impact toward the lateral aspects of the chest and away from the heart. This approach is based on the work of Stockman and Roscher (14) in 1977, which described blunt trauma injury in rats caused by a falling weight in a related system. The current model differs from this previous model in the materials and specific design of the impact delivery apparatus and how it was positioned on the chest (e.g., positioning was relative to the marked xiphoid process to enhance reproducibility) (Table 3). Cardiac output and other variables of cardiac dysfunction were not measured in the present study, because these are not specific to blunt cardiac trauma. The hypoxemia and pulmonary hemodynamic changes (specifically acute reactive pulmonary hypertension) associated with lung injury have been shown in many animal studies to seriously affect cardiac function (13,26–28). In addition, in the present study, we examined a much more extensive range of physiological, histopathological, and BAL assessments of lung contusion injury, and also investigated its progression over time and its severity as a function of external chest impact energy (Tables 1 and 2 and Figs. 2–5).
Many current animal models of chest contusion are affected by how the injury is induced (Table 3). For example, one common method of inducing lung injury in such models involves firing an empty handgun into the ipsilateral lung after opening the thoracic cavity (7,11). This mechanism of injury induction in an opened chest is very dissimilar to that found in real-life situations, and may significantly alter the associated details of injury and inflammation. Other available models of chest contusion do involve closed thoracic trauma, but include continuing mechanical ventilation at the time of injury that can influence the specifics of pulmonary pathology (12,13). In our experiments, rats were not ventilated before, during, or after lung contusion.
In addition to incorporating a clinically relevant method of injury induction, future studies of lung injury in the rat model can also be expected to benefit from the greater availability of specific molecular probes for investigating the inflammatory response in rodents (mice, rats) compared with many large-animal species.
Our primary emphasis in this study was on defining the technical variables and basic characteristics of the rat model of isolated lung contusion rather than on detailed mechanistic investigations of pathophysiology. However, the results display several features consistent with other forms of acute inflammatory pulmonary injury in animals. Energy-dependent increases in the levels of albumin in BAL are consistent with increasing levels of acute injury to the integrity of the alveolocapillary membrane (Table 2, Fig. 4). Injury to the alveolar epithelium as well as the capillary endothelium is indicated, because albumin had to be present in the alveolar lumen to be accessible to lavage. The fact that albumin levels in BAL were greatest early in the course of injury at 1 and 4 hours post-contusion, and had begun to recover by 48 hours (Table 2), shows that substantial tearing or gross disruption of the alveolocapillary membrane did not occur in injured animals. The transient nature of the arterial hypoxemia observed in injured rats (Table 2, Fig. 3) also indicates that isolated lung contusion from blunt chest trauma is at least partly reversible.
The results in rats reported here support the paradigm that a severe but isolated lung contusion could lead to transient injury with a relatively rapid recovery. Our laboratory has previously reported that transient acute pulmonary injury in small animals also occurs in response to other insults such as the aspiration of acid (15–18). In contrast, we and others have demonstrated that a second insult to the lung (“two-hit hypothesis”) exacerbates inflammatory pulmonary injury and allows it to become progressive (29). Many patients with blunt chest trauma experience a transient clinical course with rapid recovery, whereas others exhibit disease that is severe and progressive (2,6,24,25). The latter clinical picture of profound, persistent hypoxemia may reflect lung contusion injury that is exacerbated by secondary events such as gastric aspiration that are also associated with the development of ALI/ARDS. In addition to having utility in future mechanistic studies on isolated lung contusion, the rat model reported herein could also be adapted to help investigate specific interactions between contusion injury and additional insults such as gastric aspiration.
1. Miller PR, Croce MA, Bee TK, et al. ARDS after pulmonary contusion: accurate measurement of contusion volume identifies high-risk patients. J Trauma 2001;51:223–8.
2. Cohn SM. Pulmonary contusion: review of the clinical entity. J Trauma 1997;42:973–9.
3. Schild HH, Strunk H, Weber W, et al. Pulmonary contusion: CT vs. plain radiograms. J Comput Assist Tomogr 1989;13:417–20.
4. Border JR, Hopkinson BR, Schenk WG Jr. Mechanisms of pulmonary trauma: an experimental study. J Trauma 1968;8:47–62.
5. Geller E, Khaw BA, Strauss HW, et al. Technetium-fibrinogen lung scanning in canine lung contusion. J Trauma 1984;24:611–8.
6. Fulton RL, Peter ET. The progressive nature of pulmonary contusion. Surgery 1970;67:499–506.
7. Nichols RT, Pearce HJ, Greenfield LJ. Effects of experimental pulmonary contusion on respiratory exchange and lung mechanics. Arch Surg 1968;96:723–30.
8. Obertacke U, Neudeck F, Majetschak M, et al. Local and systemic reactions after lung contusion: an experimental study in the pig. Shock 1998;10:7–12.
9. Malkusch W, Hellinger A, Konerding M, et al. Morphometry of experimental lung contusion: an improved quantitative method. Anal Cell Pathol 1995;8:279–86.
10. Erickson DR, Shinozaki T, Beekman E, Davis JH. Relationship of arterial blood gases and pulmonary radiographs to the degree of pulmonary damage in experimental pulmonary contusion. J Trauma 1971;11:689–94.
11. Hellinger A, Konerding MA, Malkusch W, et al. Does lung contusion affect both the traumatized and the noninjured lung parenchyma? A morphological and morphometric study in the pig. J Trauma 1995;39:712–9.
12. Davis KA, Fabian TC, Ragsdale DN, et al. Endogenous adenosine and secondary injury after chest trauma. J Trauma 2000;49:892–8.
13. Moomey CB Jr, Fabian TC, Croce MA, et al. Cardiopulmonary function after pulmonary contusion and partial liquid ventilation. J Trauma 1998;45:283–90.
14. Stockmann U, Roscher R. [Pulmonary contusion after blunt chest trauma: an experimental model (author’s translation)]. Thoraxchir Vask Chir 1977;25:211–3.
15. Davidson BA, Knight PR, Helinski JD, et al. The role of tumor necrosis factor-alpha in the pathogenesis of aspiration pneumonitis in rats. Anesthesiology 1999;91:486–99.
16. Knight PR, Druskovich G, Tait AR, Johnson KJ. The role of neutrophils, oxidants, and proteases in the pathogenesis of acid pulmonary injury. Anesthesiology 1992;77:772–8.
17. Knight PR, Rutter T, Tait AR, et al. Pathogenesis of gastric particulate lung injury: a comparison and interaction with acidic pneumonitis. Anesth Analg 1993;77:754–60.
18. Kennedy TP, Johnson KJ, Kunkel RG, et al. Acute acid aspiration lung injury in the rat: biphasic pathogenesis. Anesth Analg 1989;69:87–92.
19. Nader-Djalal N, Knight PR, Davidson BA, Johnson K. Hyperoxia exacerbates microvascular lung injury following acid aspiration. Chest 1997;112:1607–14.
20. Shorr RM, Crittenden M, Indeck M, et al. Blunt thoracic trauma: analysis of 515 patients. Ann Surg 1987;206:200–5.
21. LoCicero J 3rd, Mattox KL. Epidemiology of chest trauma. Surg Clin North Am 1989;69:15–9.
22. Wang ND, Stevens MH, Doty DB, Hammond EH. Blunt chest trauma: an experimental model for heart and lung contusion. J Trauma 2003;54:744–8.
23. Knoferl MW, Liener UC, Seitz DH, et al. Cardiopulmonary, histological, and inflammatory alterations after lung contusion in a novel mouse model of blunt chest trauma. Shock 2003;19:519–25.
24. Kollmorgen DR, Murray KA, Sullivan JJ, et al. Predictors of mortality in pulmonary contusion. Am J Surg 1994;168:659–63.
25. Schreiter D, Reske A, Scheibner L, et al. [The open lung concept: clinical application in severe thoracic trauma]. Chirurg 2002;73:353–9.
26. Schuster DP, Perez JE, Trulock EP, et al. Cardiac dysfunction during acute lung injury induced by oleic acid in dogs. Am Rev Respir Dis 1986;133:519–25.
27. Steenbergen C, Deleeuw G, Rich T, Williamson JR. Effects of acidosis and ischemia on contractility and intracellular pH of rat heart. Circ Res 1977;41:849–58.
28. Teplinsky K, O’Toole M, Olman M, et al. Effect of lactic acidosis on canine hemodynamics and left ventricular function. Am J Physiol 1990;258:H1193–9.
© 2005 International Anesthesia Research Society
29. Knight PR, Davidson BA, Nader ND, et al. Progressive, severe lung injury secondary to the interaction of insults in gastric aspiration. Exp Lung Res 2004;30:535–57.