Published ahead of print on July 3, 2003, doi:10.1164/rccm.200304-528OC
© 2003 American Thoracic Society
Histologic, Immunohistochemical, and Ultrastructural Findings in Human Blast Lung InjuryDepartment of Forensic Pathology, Institute of Legal Medicine, and Institute of Pathology, University of Hamburg, Hamburg; Institute of Anatomy, University of Kiel, Kiel; and Institute of Legal Medicine, University of Bonn, Bonn, Germany Correspondence and requests for reprints should be addressed to Michael Tsokos, M.D., Department of Forensic Pathology, Institute of Legal Medicine, University of Hamburg, Butenfeld 34, 22529 Hamburg, Germany. E-mail: mtsokos{at}web.de
The objective of this autopsy-based study was to investigate the pathology of human blast lung injury using histology, Fat Red 7B staining, immunohistochemistry, and scanning electron microscopy on lung specimens from eight medicolegal autopsy cases of fatal close-range detonations of chemical explosives. The micromorphologic equivalents of human blast lung injury can be summarized as follows: diffuse alveolar overdistension, circumscribed interstitial hemorrhages showing a cufflike pattern around pulmonary vessels, venous air embolism, bone marrow embolism, and pulmonary fat embolism. Hemorrhages within the lung parenchyma that were present in this study in blast victims without coexisting blunt or penetrating chest trauma must be regarded as potentially life-threatening intrapulmonary bleeding sites in survivors. In addition, the potential clinical importance of the presence of massive pulmonary fat embolism, which has, to the best of our knowledge, not been described previously in human blast lung injury, must be emphasized because pulmonary fat embolism may be a leading cause of the rapid respiratory deterioration with progressive hypoxia and development of acute respiratory distress syndrome in blast victims who survive. Furthermore, this study provides evidence that air embolism presenting in blast victims is not a mere ventilation-induced artifact.
Key Words: acute respiratory distress syndrome • blast lung injury • air embolism • pulmonary fat embolism An explosive is any substance or device capable of a sudden expansion of gas, which upon release of its potential energy creates a pressure wave. Compression of air in front of the pressure wave, which heats and accelerates air molecules, leads to a sudden increase in atmospheric pressure (overpressure) and temperature transmitted into the surrounding environment as a radially propagating shock wave known as the blast wave (1, 2). Injuries directly inflicted by this sudden increase in air pressure after an explosion are referred to as primary blast injuries and involve almost exclusively gas-containing organs such as the lungs, middle ear, and gastrointestinal tract, which are the organs most vulnerable to overpressure (2, 3). Secondary blast injuries result from blast-energized bomb fragments and other displaced objects at the site of explosion such as glass, casing, and masonry causing penetrating trauma (4–7). Tertiary blast injuries occur when the body is accelerated from the blast wave at first and is then abruptly decelerated on rigid objects, thus resulting in blunt force trauma (4, 5, 7). Because injuries inflicted by explosions are mediated by these three different mechanisms, victims usually suffer from a combination of primary blast effects to gas-containing organs, blunt force injuries, penetrating trauma, and burns. However, primary blast injuries are estimated to contribute to 47–57% of the injuries in survivors and to 86% of fatal injuries (2). Of the gas-containing organs, the lung is most susceptible to primary blast effects (5, 8–10), and the extent of lung injury is considered the decisive parameter defining mortality in victims of explosions who survive (3, 11). The diagnosis of "blast lung" is usually made clinically by dyspnea and cough that precede rapid respiratory deterioration with progressive hypoxia and subsequent acute respiratory distress syndrome (ARDS) and is confirmed by chest radiographs typically showing a butterfly appearance with or without pneumothorax on admission and increasing haziness in serial chest radiographs as well as the presence of burn injuries and smoke inhalation of the upper airways seen at bronchoscopy (3, 4, 11–13). In contrast to the large number of studies dealing with the clinical picture and course of near-fatal and fatal events of blast lung injury in the clinical setting, the micromorphologic equivalents of human blast lung injury have not been defined yet. So far, the most comprehensive information about the micromorphologic pattern of blast lung injury derives merely from animal studies (8, 14–16). To obtain further insight into the pathology of human blast lung injury, we undertook the present autopsy-based study using histology, Fat Red 7B staining, immunohistochemistry, and scanning electron microscopy on lung specimens from eight medicolegal autopsy cases of fatal close-range detonations of chemical explosives.
Human Lung Specimens from Explosion-related Fatalities: Blast Lung Group The forensic autopsy files of the Institute of Legal Medicine, University of Hamburg, Germany, and the Institute of Legal Medicine, University of Bonn, Germany, between 1998 and 2001 were reviewed for explosion-related fatalities. Cases in which the thoracic cavity was disrupted and cases with penetrating chest trauma were not included. To avoid putrefaction-associated postmortem artifacts with regard to micromorphology, only cases in which autopsy was performed within 48 hours postmortem were chosen. Eight cases (seven males and one female; individual ages, 34–87 years) were identified and included in the study. In all cases, chemical explosives such as 2,4,6-trinitrotoluene, black powder (potassium nitrate, sulfur, charcoal), liquid gasoline, or natural gas were involved. All detonations were set off in confined spaces at close range to the deceased. In two cases death occurred immediately. In one case the victim died 24 hours after the explosion but without being subjected to any medical assistance. In the remaining cases, survival times varied up to 20 minutes. Resuscitation attempts were not undertaken in any of the cases. The mode of death was homicide in four cases, suicide in three cases, and accident in one case. In each case paraffin-embedded blocks from all lobes of the lung were available. In addition, stored lung tissue fixed in 5% buffered formalin was accessible. The autopsy protocols were reviewed and a standardized protocol was used to register the injury pattern for each case as documented at gross examination.
Control Group
Routine Histology
Fat Red 7B Staining
Immunohistochemistry for Hemoglobin
Scanning Electron Microscopy
Gross Pathology of Human Blast Lung Injury On external examination, lacerations of the skin sporadically interspersed with foreign body material, scattered abrasions and contusions, as well as superficial flash burn injuries together with singeing of head hair and eyebrows and more severe burns restricted to areas of local ignition of clothing were the most frequent findings. Splenic rupture was seen in two cases. Mutilation of the upper extremities accompanied by comminuted fractures of the metacarpals, wrists, and forearms were found in two of the suicide cases. Unilateral pneumothorax was present in two cases.
Routine Histopathology of Human Blast Lung Injury
Considerable congestion of pulmonary arteries, arterioles, veins, venules, and alveolar capillaries accompanied by vascular engorgement was present in four cases. Of these, a mild to moderate amount of eosinophilic deposit filling the alveolar spaces and interalveolar septae was seen in two cases with survival times of a few minutes, and a more intense alveolar as well as interstitial edema was observed in the two other cases (survival time several minutes and 24 hours, respectively). Venous air embolism (Figure 1D) was present in four cases. Pulmonary bone marrow embolism (Figure 1E) was observed in three cases. In two cases, soot aspiration was detected in smaller bronchi (Figure 1F). Apart from the intensity of pulmonary edema, which was positively correlated with individual survival time, no obvious differences were noted in the observed histopathologic features with regard to the individual case characteristics such as age, explosive charge properties, range between detonation and victim, and survival time.
Fat Red 7B Staining
Immunohistochemical Findings in Human Blast Lung Injury Compared with Control Subjects In the four blast lung group cases in which alveolar edema was present, a moderate to strong hemoglobin immunoreactivity of the edema fluid within the alveolar spaces was observed, showing a homogeneous staining pattern (Figure 3A) , whereas no hemoglobin immunoreactivity was detected within the alveolar spaces of the control cases without underlying pathologic alterations or with chronic pulmonary emphysema. In the control cases with hemorrhagic lung edema, a strong hemoglobin immunopositivity restricted to erythrocytes within the alveolar edema fluid was seen (Figure 3B). In contrast, hemoglobin-immunopositive erythrocytic cell structures were not detectable in any of the blast lung group cases expressing homogeneous hemoglobin immunoreactivity within the intraalveolar edema fluid. In both the blast lung group and the control group, hemoglobin-immunopositive erythrocytes were regularly present within larger and smaller pulmonary vessels and alveolar capillaries (Figure 3C).
Ultrastructural Changes in Human Blast Lung Injury Compared with Control Subjects Scanning electron microscopy of blast lungs revealed alveolar ruptures and enlargement of alveolar spaces compared with control subjects (Figures 4A and 4B) . Moreover, thinning of alveolar septae, including small perforations of the alveolar wall measuring between 0.5 and 9 µm in diameter, were visible at higher magnification (Figure 4C). In the four cases from the blast lung group in which alveolar edema was present, interalveolar septae contained amorphous material corresponding to interstitial edema fluid (Figure 4D). Circumscribed intraalveolar and perivascular hemorrhages were visible and confirmed the results obtained by routine histopathology.
So far, histologic and ultrastructural findings in blast lung injury have been described only in experimental animal models, most of them exposing rodents to artificial blast waves (8, 15, 16, 18). Despite a number of clinical reports, mostly dealing with primary blast injuries in civilian victims of terrorist explosions (4, 5, 11–13), the micromorphologic equivalents of human blast lung injury are largely unknown (2). On the basis of the mechanism of energy release, explosives can be classified as chemical, mechanical, or nuclear. Regardless of whether it concerns volatile or solid substances, chemical explosives decompose into gas on detonation. The potential energy release of chemical explosives depends on the rate of decomposition, which is again determined by the chemical compounds used for the explosive (e.g., black powder has a lower rate of decomposition than 2,4,6-trinitrotoluene, which detonates at much higher speeds [19]). The main micromorphologic features of human blast lung injury in the present study can be divided into two distinct patterns: (1) severe alveolar overdistension characterized by alveolar ruptures, thinning of alveolar septae, and enlargement of alveolar spaces, which was found disseminated throughout all sections of the lungs, and (2) subpleural, intraalveolar, and perivascular (cufflike) hemorrhages representing a more circumscribed type of parenchymatous injury that was present in routine histology at least in a few visual fields in all cases. Despite the fact that the detonations derived from different types of chemical explosives in the present study and the micromorphologic observations from different explosion incidents had to be biased, we observed a relatively uniform picture of blast lung injury at the micromorphologic level, a circumstance we assume is attributable to the fact that all explosions had been set off in confined spaces and the detonations occurred at close range to the deceased. It is, however, generally accepted that victims of explosions in confined spaces have a higher incidence of primary blast injury to the lungs and a higher mortality rate in comparison with victims who are injured by explosions in the open air (4, 5, 12). Air embolism is a well recognized complication of blast-induced lung injury and is considered one of the chief factors leading to cardiac dysfunction and immediate death after blast wave exposure (2, 3, 20). It is difficult to estimate the incidence of air embolism subsequent to blast wave exposure in humans because no research data are, to the best of our knowledge, available on this subject and animal studies have provided conflicting results on the incidence, ranging between 0 and 50% (2, 8, 15). However, the proof of air embolism will depend on the diagnostic methodology used to access its occurrence (e.g., autopsy, histology, clinical findings, Doppler technology). In the present autopsy-based investigation, air embolism was observed histologically in four of eight cases. Whether air embolism is caused by mechanical ventilation of blast victims is still a matter of debate (10), but because none of the individuals examined here received any resuscitation attempts or medical assistance at all, our results provide evidence that air embolism in primary blast injury is not a mere ventilation-induced artifact and most probably originates, at least to a certain degree, from blast-induced enlargement of airspace and disruption of alveolar septae and interstitial vessel walls with consecutive absorption into the adjacent pulmonary veins. The manifestation of pulmonary fat embolism as seen here in five of eight cases has, to the best of our knowledge, not been described previously in human blast lung injury. Pulmonary fat embolism develops rapidly in a majority of persons suffering blunt force injury (21) and can even be found histologically in trauma deaths with immediate circulatory arrest, such as occur in airplane crashes (22). We consider this a finding of clinical importance because in survivors of blunt force trauma pulmonary fat embolism is a crucial contributor to the development of ARDS and significantly affects clinical outcome (23–26). In contrast, we consider the histopathologic finding of bone marrow embolism, as seen here in three cases, as unspecific because this finding is a well known phenomenon that can be observed not only in fatalities with a preceding trauma but also in autopsy cases with absence of trauma in the premortem period. The exact mechanisms involved in lung injury from blast wave exposure are not yet completely understood. Studies suggest that high-velocity longitudinal pressure waves propagate through the body, thus resulting in pressure differentials at the interface between tissues with different densities (3, 27–29). Reflection of the stress waves off the mediastinum and the thoracic cavity is considered to cause complex pressure conditions within the lung parenchyma that reinforce pressure differentials at barriers of different densities, thus causing the alveolar septae and capillary walls to rupture (5, 15). We also found circumscribed interstitial hemorrhages showing a cufflike pattern around larger and smaller vessels within the lung parenchyma in routine histology confirmed by scanning electron microscopy, a finding that provides evidence that also in patients suffering blast injuries without coexisting blunt or penetrating chest trauma, pulmonary vessels must be regarded as potentially life-threatening intrapulmonary bleeding sites that might require thoracotomy immediately after admission (30, 31). Whether these perivascular hemorrhages are a result of pressure differentials and therefore a direct primary blast effect, or if these lesions derive from shearing forces affecting the lung parenchyma when the lung is abruptly decelerated on the interior chest wall after acceleration due to the indirect blast wave effect, remains unclear at present. It has been proposed that biochemical changes, such as depletion of pulmonary antioxidant reserves resulting in free radical-mediated oxidative stress and the accumulation of lipid peroxidation products, contribute to blast-induced lung injury (32–34). However, it is controversial whether this oxidative stress is a leading cause of lung injury after blast trauma or occurs as a result of the injury. In addition, evidence suggests that, as alveolar septae and alveolar capillary walls rupture and intraalveolar hemorrhage occurs, red blood cells also rupture and release hemoglobin that in turn amplifies the cascade of events leading to oxidative stress (35). We found a homogeneous hemoglobin immunoreactivity within the intraalveolar edema fluid in the blast lung cases without any hemoglobin-immunopositive erythrocytic cell structures, in contrast to a strong hemoglobin immunopositivity restricted to erythrocytes within the alveolar edema fluid in hemorrhagic (toxic) lung edema from control subjects. It is unknown whether the staining of hemoglobin reflects a pathologic response or is an epiphenomenon. In conclusion, this autopsy-based study displays a relatively uniform picture of the early stages of human blast lung injury from close-range detonations of chemical explosions. The morphologic equivalents consisted of diffuse alveolar overdistension and more circumscribed pulmonary hemorrhages with or without edema, venous air embolism, bone marrow embolism, and pulmonary fat embolism in terms of histopathologic diagnosis. Our data suggest the following: (1) air embolism presenting in blast victims is not a mere ventilation-induced artifact; (2) the presence and extent of pulmonary fat embolism in surviving blast victims may be a major determinant of development of ARDS and influence mortality; and (3) physicians should be aware of occult intrapulmonary bleeding sites in blast victims even when a concomitant blunt or penetrating chest trauma can be ruled out.
Conflict of Interest Statement: M.T. has no declared conflict of interest; F.P. has no declared conflict of interest; S.P. has no declared conflict of interest; B.M. has no declared conflict of interest; K.P. has no declared conflict of interest; E.E.T. has no declared conflict of interest. Received in original form April 16, 2003; accepted in final form June 19, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
