DOI: 10.1148/radiol.2433061009
(Radiology 2007;243:862-868.)
Virtual Autopsy: Two- and Three-dimensional Multidetector CT Findings in Drowning with Autopsy Comparison1
Angela D. Levy, COL, MC, USA,
H. Theodore Harcke, COL, MC, USA,
John M. Getz, BS,
Craig T. Mallak, CAPT, MC, USN,
James L. Caruso, CAPT, MC, USN,
Lisa Pearse, LCDR, MC, USN,
Aletta A. Frazier, MD, and
Jeffrey R. Galvin, MD
1 From the Department of Radiologic Pathology, Armed Forces Institute of Pathology, Washington, DC (A.D.L., A.A.F., H.T.H., J.R.G.); Office of the Armed Forces Medical Examiner, Armed Forces Institute of Pathology, Rockville, Md (L.P., C.T.M., J.M.G., J.L.C.); Department of Radiology, University of Maryland School of Medicine, Baltimore, Md (A.A.F., J.R.G.); and Department of Radiology and Nuclear Medicine, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Rd, Bethesda, MD 20814-4799 (A.D.L., H.T.H.). Received June 11, 2006; revision requested August 14; revision received August 30; accepted September 28; final version accepted October 20. Supported by a grant from the Defense Advanced Research Projects Agency.
Address correspondence to A.D.L. (e-mail: alevy{at}usuhs.mil).
 |
ABSTRACT
|
|---|
Purpose: To retrospectively determine the multidetector computed tomographic (CT) virtual autopsy findings of death by drowning in comparison with autopsy findings.
Materials and Methods: The institutional review board of the Armed Forces Institute of Pathology approved this HIPAA-compliant study and did not require informed consent by the next of kin. Total-body multidetector CT was performed, immediately prior to routine autopsy, in 28 consecutive male subjects (mean age, 24.2 years) who died of drowning and a control group of 12 consecutive male subjects (mean age, 50.8 years) who died of sudden death from atherosclerotic coronary artery disease. Images were evaluated for the presence of fluid and sediment in the paranasal sinuses and airways, mastoid air cell fluid, frothy fluid in the airways, pulmonary opacity (ground-glass opacity or airspace consolidation), interlobular septal thickening, and gastric distention and contents (fluid or sediment). Image findings were compared with findings from autopsy reports and photographs.
Results: All drowning subjects had fluid in the paranasal sinuses and mastoid air cells and had ground-glass opacity within the lungs. Twenty-six subjects (93%) had fluid in the subglottic trachea and main bronchi. Fourteen subjects (50%) had high-attenuation sediment in the subglottic airways. Frothy fluid in the airways was present in six subjects (21%). Twenty-five (89%) of the drowning subjects had pulmonary ground-glass opacity with septal lines, which was mild with apical and perihilar distribution in 12 subjects, severe and diffuse in nine, posterior and basilar in three, and limited to the apices in one (not assessed in three of 28 subjects because of decomposition). Control subjects showed mastoid cell fluid (25%), sinus fluid (83%), subglottic airway fluid (92%), and pulmonary ground-glass opacity (100%) but did not have evidence of frothy airway fluid or high-attenuation sediment in the airways.
Conclusion: The multidetector CT finding of frothy airway fluid or high-attenuation airway sediment is highly suggestive of drowning; multidetector CT findings of pan sinus fluid, mastoid cell fluid, subglottic tracheal and bronchial fluid, and ground-glass opacity within the lung at multidetector CT are supportive of drowning in the appropriate scenario.
|
INTRODUCTION
|
|---|
The determination of drowning as the cause of death for a body that is found in water is imperative in forensic investigation because becoming submerged in water may be a secondary rather than a primary event. Anatomic findings at autopsy to support the diagnosis of drowning include the presence of frothy fluid in the airways or lungs, hyperinflated and congested lungs, fluid in the paranasal sinuses, watery fluid in the stomach, and dilated and engorged right-sided cardiac chambers and great vessels. The frothy fluid in the airways and lungs is a proteinaceous exudate of intraalveolar edema and surfactant mixed with water of the drowning medium (1). Hyperinflated lungs at autopsy are defined as lungs that are expanded with air such that they fill the pleural cavity and encroach on the mediastinum or cross the midline such that they touch one another when the thoracic cavity is opened and the sternum is removed. Congested or edematous lungs are also characteristic and typically result in combined lung weights of greater than 1000 g (2). Because inhaled and aspirated fluid enters the paranasal sinuses as it passes through the nasal cavity, the finding of sphenoidal and ethmoidal sinus fluid is supportive of the diagnosis of drowning. Although many consider the findings of pleural effusion, watery fluid in the stomach, and dilated engorged right-sided cardiac chambers and great vessels as nonspecific, they are helpful supportive findings when other anatomic findings of drowning are present. In addition, the finding of sand or sediment in the paranasal sinuses and airway is an important marker of inhalation and aspiration of sediment-laden water and is clinically important in near drowning because it may cause airway injury and inflammation resulting in the need for prolonged mechanical ventilation (3).
Until recently, our knowledge of the imaging manifestations of drowning has been limited to reports describing the chest radiograph findings in near drowning (4–7). Kim et al (8) reported the findings of bilateral patchy or diffuse ground-glass opacity in the lungs of near drowning subjects at thin-section computed tomography (CT). Thali et al (9) published virtual autopsy (multiplanar, three-dimensional anatomic display) findings of death by drowning in two subjects and described the spectrum of paranasal sinus fluid, patchy airspace disease, right atrial enlargement, and upper gastrointestinal fluid. The purpose of our study was to retrospectively describe the multidetector CT virtual autopsy findings of death by drowning in comparison with autopsy findings.
|
MATERIALS AND METHODS
|
|---|
Study Group
The medical examiners' database at the Armed Forces Institute of Pathology was searched for cases from January 2005 through April 2006 and yielded 37 drowning subjects who underwent multidetector CT scanning prior to conventional autopsy. Exclusion criteria included extensive resuscitation with intubation prior to death (n = 5) or blunt force injury to the chest (n = 4), because each of these processes may affect the pathologic changes in the airways and pulmonary parenchyma. The final study group consisted of 28 male subjects (age range, 19–35 years; mean age, 24.2 years) who had drowned. Imaging and autopsy were performed 2–12 days (mean, 3.9 days) following death. Twenty-three of the 28 subjects drowned while submerged in motor vehicles that accidentally overturned into inland canals. Of the other five subjects, one died while attempting to rescue persons in one of the submerged vehicles, one was found in a swimming pool, one drowned while snorkeling, and two accidentally fell into an inland canal. This Health Insurance Portability and Accountability Act–compliant study was performed with the approval of the institutional review board of the Armed Forces Institute of Pathology and did not require informed consent from the next of kin.
Control Group
The control group included consecutive subjects with sudden cardiac death who had a final autopsy diagnosis of atherosclerotic coronary artery disease and in whom autopsy was performed between October 2005 and April 2006 and multidetector CT scanning was performed prior to autopsy. Of the 14 subjects in this group, two had been intubated and resuscitated and therefore were excluded from the study. The final control group included 12 men (age range, 40–66 years; mean age, 50.8 years). Imaging and autopsy were performed 2–7 days (mean, 4.0 days) after death.
CT and Autopsy Technique
Prior to autopsy, all subjects were imaged with whole-body multidetector CT (LightSpeed 16; GE Medical Systems, Milwaukee, Wis). Subjects were scanned while fully clothed, with their arms at their sides. Two series of images were obtained: Series 1 was obtained from the skull vertex to the most distal point allowable by table travel (distal femur or proximal tibia and fibula), and series 2 was obtained from the distal femur to the toes. Subjects were scanned with (a) 16 x 5-mm collimation (16 detectors with 5-mm section thickness), pitch of 1.375:1, rotation speed of 0.6 second, and table speed of 27.5 mm per rotation or (b) 16 x 5-mm collimation, pitch of 0.562:1, rotation speed of 0.6 second, and table speed of 11.2 mm per rotation. No contrast material was administered. Images were retrospectively reconstructed at the CT console to a section thickness of 1.25 mm prior to being sent to a workstation (Advantage, software version 4.2; GE Medical Systems).
Autopsies were performed by six board-certified forensic pathologists (including C.T.M., J.L.C.) with 8–19 years of experience who were from the medical examiners' office of our institution. Complete dissection and gross examination of the intracranial contents, oral cavity, neck, chest, heart, mediastinum, abdomen, and pelvis were performed in each subject. The sphenoidal sinus was evaluated for the presence of fluid in each case. Organ weights and body cavity fluid volumes were recorded in all cases. Toxicologic examination was performed in each case. The final determination of drowning as the cause of death was made on the basis of data from the scene investigation and a combination of supportive autopsy findings, which included cerebral edema, laryngeal edema, hyperinflated lungs (defined at autopsy as lungs crossing midline upon opening of the chest), congested or edematous lungs, fluid and/or sediment in the subglottic airways, frothy fluid in the airways, fluid and/or sediment within the sphenoidal sinus, and evidence of swallowed water. Each autopsy report included digital photographs of the external examination of the body to document cutaneous manifestations of immersion, personal identifying features, abrasions, wounds, and pertinent photographs of internal organ abnormalities. The data from the autopsies were obtained from the final autopsy report and autopsy photographs by a radiologist (A.D.L.), epidemiologist (L.P.), and project manager (J.M.G.).
Image Interpretation
Retrospectively, four radiologists (A.D.L., H.T.H., A.A.F., J.R.G.) with 8–30 years of experience interpreted the multidetector CT virtual autopsy scans. The date, mechanism of submersion, and final diagnosis at autopsy were known at the time of image interpretation. Final image interpretation was reached in consensus. Images were interpreted at a workstation (Advantage; GE Medical Systems) by using two-dimensional transverse, coronal, oblique, and sagittal data sets, as well as three-dimensional volume-rendered images.
Multidetector CT images were evaluated to determine the presence or absence of fluid and/or sediment in the paranasal sinuses and fluid in the mastoid air cells. The oral cavity and upper airways were evaluated for the presence or absence of fluid, frothy fluid, or high-attenuation sediment. The subglottic trachea and bronchi were evaluated for the presence of fluid and sediment. The lungs were evaluated for ground-glass opacity, interlobular septal thickening, and parenchymal consolidation. If ground-glass opacity was present, it was visually assessed with multiplanar two- and three-dimensional images for overall distribution (diffuse, apical, perihilar, peripheral, or basilar) and severity (mild, moderate, or severe). The pleural spaces were assessed for evidence of effusion or pneumothorax, and if present, the effusion or pneumothorax was visually estimated to be small, moderate, or large. The abdomen was evaluated for fluid or high-attenuation sediment in the stomach and for evidence of gastric distention. The remainder of the scan was evaluated for incidental abnormalities in the head and neck, mediastinum, solid and hollow abdominal and pelvic viscera, soft tissues, and skeletal structures. After image review, two- and three-dimensional data sets were reviewed and compared with findings from the final autopsy reports and photographs by two radiologists (A.D.L., H.T.H.) and the project manager (J.M.G.). Final autopsy findings and photographs were compared with multidetector CT findings.
|
RESULTS
|
|---|
Paranasal Sinuses and Mastoid Air Cells
All drowning subjects had fluid in the mastoid air cells bilaterally and in all paranasal sinuses (Fig 1). Seven subjects (25%) had high-attenuation material dependently layering in both maxillary sinuses and the sphenoidal sinus on multidetector CT images that was consistent with inhaled sediment (Fig 2). In the control group, three subjects (25%) had evidence of mastoid cell fluid, 10 subjects (83%) had sinus fluid (five with fluid in all paranasal sinuses and five with fluid in only some of the paranasal sinuses), and two subjects (17%) had no evidence of sinus fluid. There were no subjects in the control group with high-attenuation sediment in the airways or sinuses.
View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a: Upper airway and sinus features of drowning. (a) Transverse multidetector CT image at the level of the mastoid air cells shows bilateral patchy fluid within the mastoid air cells (black arrows) and frothy fluid in the nasopharynx (white arrow). (b) Sagittal multidetector CT image shows frothy fluid filling the posterior nasal cavity, nasopharynx, and oropharynx. Fluid is in the frontal and sphenoidal sinuses.
|
|
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b: Upper airway and sinus features of drowning. (a) Transverse multidetector CT image at the level of the mastoid air cells shows bilateral patchy fluid within the mastoid air cells (black arrows) and frothy fluid in the nasopharynx (white arrow). (b) Sagittal multidetector CT image shows frothy fluid filling the posterior nasal cavity, nasopharynx, and oropharynx. Fluid is in the frontal and sphenoidal sinuses.
|
|
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a: Sediment aspiration in drowning. (a) Transverse multidetector CT image of the head shows high-attenuation dependent-layering sediment (arrows) in the maxillary and sphenoidal sinuses. (b) Transverse and (c) coronal multidetector CT images of the chest show fluid-filled main bronchi with high-attenuation dependent-layering sediment (arrows). (d) Autopsy photograph of resected distal trachea, carina, and proximal main bronchi shows tan sediment in the lumen of the airways.
|
|
View larger version (78K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b: Sediment aspiration in drowning. (a) Transverse multidetector CT image of the head shows high-attenuation dependent-layering sediment (arrows) in the maxillary and sphenoidal sinuses. (b) Transverse and (c) coronal multidetector CT images of the chest show fluid-filled main bronchi with high-attenuation dependent-layering sediment (arrows). (d) Autopsy photograph of resected distal trachea, carina, and proximal main bronchi shows tan sediment in the lumen of the airways.
|
|
View larger version (87K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c: Sediment aspiration in drowning. (a) Transverse multidetector CT image of the head shows high-attenuation dependent-layering sediment (arrows) in the maxillary and sphenoidal sinuses. (b) Transverse and (c) coronal multidetector CT images of the chest show fluid-filled main bronchi with high-attenuation dependent-layering sediment (arrows). (d) Autopsy photograph of resected distal trachea, carina, and proximal main bronchi shows tan sediment in the lumen of the airways.
|
|
View larger version (129K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2d: Sediment aspiration in drowning. (a) Transverse multidetector CT image of the head shows high-attenuation dependent-layering sediment (arrows) in the maxillary and sphenoidal sinuses. (b) Transverse and (c) coronal multidetector CT images of the chest show fluid-filled main bronchi with high-attenuation dependent-layering sediment (arrows). (d) Autopsy photograph of resected distal trachea, carina, and proximal main bronchi shows tan sediment in the lumen of the airways.
|
|
Trachea and Main Bronchi
Twenty-six (93%) of the drowning subjects had fluid in the subglottic trachea and main bronchi. In the two subjects who did not have fluid in the subglottic trachea and main bronchi, there was fluid in the second-order bronchi bilaterally in both subjects, and one had high-attenuation sediment in the second-order bronchi, in addition to fluid. Fourteen (50%) of the drowning subjects had high-attenuation sediment in the distal trachea and main bronchi (Fig 2). Frothy fluid characterized by heterogeneous low-attenuation fluid admixed with rounded foci or air was identified in the airways in six (21%) drowning subjects (Fig 1). In the control group, subglottic airway fluid was present in 11 (92%) subjects. There were no subjects in the control group with frothy airway fluid or high-attenuation sediment in the airways.
Lungs and Pleural Space
Twenty-five (89%) of the drowning subjects showed pulmonary ground-glass opacity with septal lines and occasionally spared secondary pulmonary lobules. The ground-glass opacity was mild with an apical and perihilar distribution in 12 subjects (Fig 3), severe with dense consolidation diffusely in nine, mild with posterior basilar distribution in three, and mild with an apical distribution in one. In the remaining three drowning subjects, the lungs could not be assessed because of an advanced state of decomposition (Fig 4). However, in one of the subjects with decomposed lungs, there was high-attenuation sediment in the mainstem and second-order bronchi, which indicated that aspiration of sediment-laden water occurred prior to death (Fig 4). Small bilateral pleural effusions were present in 68% (19 of 28) of all drowning subjects. Small bilateral pneumothoraces were present in three of the drowning subjects without advanced decomposition. The three subjects with advanced decomposition had moderate-sized pneumothoraces bilaterally.
View larger version (186K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a: Pulmonary findings in drowning. (a) Coronal multidetector CT image shows apical predominance of moderate diffuse ground-glass opacity throughout the entire lung. There are areas sparing throughout. (b) Three-dimensional volume rendering of the lungs shows apical predominance. (c) Lungs in situ at conventional autopsy (different subject than in a and b) show congestion, edema, and hyperinflation with extension of the lungs across midline. Black anthracotic pigment is also present.
|
|
View larger version (157K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b: Pulmonary findings in drowning. (a) Coronal multidetector CT image shows apical predominance of moderate diffuse ground-glass opacity throughout the entire lung. There are areas sparing throughout. (b) Three-dimensional volume rendering of the lungs shows apical predominance. (c) Lungs in situ at conventional autopsy (different subject than in a and b) show congestion, edema, and hyperinflation with extension of the lungs across midline. Black anthracotic pigment is also present.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c: Pulmonary findings in drowning. (a) Coronal multidetector CT image shows apical predominance of moderate diffuse ground-glass opacity throughout the entire lung. There are areas sparing throughout. (b) Three-dimensional volume rendering of the lungs shows apical predominance. (c) Lungs in situ at conventional autopsy (different subject than in a and b) show congestion, edema, and hyperinflation with extension of the lungs across midline. Black anthracotic pigment is also present.
|
|
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a: Findings in a drowning subject with severe decomposition. (a) Transverse multidetector CT image of the head shows brain liquefaction and high-attenuation sediment in the maxillary, ethmoidal, and sphenoidal sinuses (arrows). There is postmortem decompositional gas throughout all tissues and within the vasculature. (b) Coronal multidetector CT image of the chest shows high-attenuation sediment at the carina and in the main bronchi (arrow) and branches of the left upper lobe bronchus (arrowhead). Cystic spaces within the lungs and gas in the vasculature and soft tissues represent advanced decomposition.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b: Findings in a drowning subject with severe decomposition. (a) Transverse multidetector CT image of the head shows brain liquefaction and high-attenuation sediment in the maxillary, ethmoidal, and sphenoidal sinuses (arrows). There is postmortem decompositional gas throughout all tissues and within the vasculature. (b) Coronal multidetector CT image of the chest shows high-attenuation sediment at the carina and in the main bronchi (arrow) and branches of the left upper lobe bronchus (arrowhead). Cystic spaces within the lungs and gas in the vasculature and soft tissues represent advanced decomposition.
|
|
In the control group, subglottic tracheal and bronchial fluid was present in 11 (92%) subjects. Bilateral pulmonary ground-glass opacity was present in all control subjects. The pattern of pulmonary ground-glass opacity was severe, bilateral, and diffuse in six control subjects and was moderate to severe, bilateral, and patchy in six control subjects.
Stomach
Gastric distention was present in 25 (89%) of the drowning subjects. Six subjects with gastric distention had evidence of high-attenuation sediment in the stomach, which indicated that sediment-laden fluid had been swallowed prior to death. In the control group, all subjects had a collapsed stomach containing fluid admixed with gastric contents. There were no subjects in the control group with high-attenuation material in the stomach.
Incidental Findings
Incidental findings in the drowning population on multidetector CT images included calcified hilar and pulmonary granuloma in two subjects and mild esophageal dilatation and wall thickening of unknown cause in one subject. In the control population, one subject had incidental urolithiasis and a 2-cm gastric mass that was histologically confirmed to be a gastrointestinal stromal tumor.
Multidetector CT and Autopsy Comparison
All drowning subjects and control subjects had evidence of pulmonary congestion and edema at autopsy, with combined lung weights ranging from 840 to 2210 g in the drowning population and from 1040 to 1860 g in the control population. In the drowning group, the presence of high-attention material in the sinuses in seven (25%) subjects, in the subglottic trachea and bronchi in 14 (50%), and in the stomach in six (21%) at multidetector CT was confirmed at autopsy. Frothy fluid was present in six (21%) of the drowning subjects at multidetector CT compared with five (19%) subjects at autopsy. Minor blunt force injuries such as facial and extremity abrasions and minor contusions were identified in 15 (54%) subjects at autopsy and could not be identified at multidetector CT. A scalp laceration in one subject was confirmed at both multidetector CT and autopsy. There were two subjects with occult fractures (one at the skull base and one nondisplaced rib fracture) identified at autopsy that could not be identified at multidetector CT.
|
DISCUSSION
|
|---|
Many authors have sought more specific findings to diagnose drowning, such as the detection of diatoms, serum biochemical analysis, and comparison of organ weights (10–13). However, to date, the most reliable method to determine that a death was due to drowning is to compare the anatomic findings with the scene investigation and exclude other causes of death. The accurate diagnosis of drowning is of utmost importance in the forensic investigation of a body found in water when there is suspicion that the body may have been placed in the water after death from another cause.
Although our study results show remarkable consistency in the constellation of findings at the multidetector CT virtual autopsy evaluation of death by drowning—pan sinus fluid, mastoid cell fluid, subglottic tracheal and bronchial fluid, and pulmonary ground-glass opacity—there is overlap with our control group of subjects with sudden cardiac death in the findings of sinus fluid, subglottic tracheal and bronchial fluid, and pulmonary ground-glass opacity. Cardiogenic pulmonary edema is the most likely cause of ground-glass opacity in our control group. Cardiogenic pulmonary edema in sudden cardiac death results from transudation of fluid, from the pulmonary capillary bed into the alveoli, bronchi, and distal trachea, that is indistinguishable on multidetector CT images from fluid inhaled into the airways and alveoli during drowning. Two multidetector CT findings distinguished drowning subjects from the control group subjects: frothy airway fluid and high-attenuation sediment. The presence of high-attenuation sediment in drowning subjects depends on the body of water in which the drowning occurred. High-attenuation sediment represents sediment from silt-laden freshwater or sand from saltwater. We did not have the opportunity to perform compositional analysis of the sediment to determine the cause of the high attenuation. Because our subjects drowned in a variety of water types in multiple geographic locations, the true frequency of this finding is unknown. Tracheobronchial calcification should be considered in the differential diagnosis of high-attenuation airway sediment at multidetector CT; however, the multiplanar capability of multidetector CT easily enables the distinction between intraluminal high-attenuation sediment and tracheobronchial wall calcification.
The patterns of pulmonary ground-glass opacity identified in the drowning deaths in our study are very similar to those reported by Kim et al (8) in cases of near drowning. Moderate ground-glass opacity with septal lines and occasionally spared secondary pulmonary lobules with a predominantly apical and perihilar distribution was the most common pattern we observed. Kim et al reported central distribution as the most common pattern in near drowning (8). The ground-glass opacity with septal lines and diffuse consolidation we observed most likely represents a spectrum of mild to severe pulmonary edema, which is nonspecific and cannot be distinguished from other causes of pulmonary edema.
While hyperinflation of the lungs at opening of the chest is an autopsy finding indicative of drowning, we could not develop a criterion for this finding on multidetector CT images. Within the closed thorax, an equivalent finding to hyperinflated lungs could not be observed with confidence. This may be a sign of drowning at conventional autopsy that is not applicable to virtual autopsy.
Older medical literature supports the hypothesis that death by drowning may occur with or without fluid aspiration on the basis of whether there was evidence for substantial pulmonary edema (14–16). More recently, the existence of drowning without fluid aspiration (dry drowning) has been disputed (17). Twenty-six (93%) of our drowning subjects had fluid in the subglottic trachea and main bronchi, but two of our subjects lacked fluid in the central airways, which indicated that large amounts of fluid were not inhaled. These latter two subjects inhaled smaller amounts of fluid that were observed in the second-order bronchi and may conceivably represent "dry drowning" described in the older medical literature.
There are several limitations to our study. Twenty-three of our 28 subjects drowned in a closed vehicle. Although the constellation of multidetector CT findings in these subjects was indistinguishable from those in the subjects of the other drowning scenarios, the latter group was a small percentage of our patient population, and larger numbers may show differences. Imaging and autopsy were performed 2–12 days after death, and some of our findings may be related to postmortem decomposition. Specifically, lung congestion and pleural fluid may be seen as normal postmortem findings, especially in nontraumatic deaths. However, in these circumstances, combined lung weights rarely exceed 500–600 g, and pleural effusions are usually quite small in volume (10–20 mL). In all of our cases, combined lung weights exceeded the normal postmortem weights, supporting the autopsy findings of lung congestion and multidetector CT findings of ground-glass opacity due to drowning. The delay in scanning and autopsy also may have affected the distribution of pulmonary ground-glass opacity observed and thereby contributed to the apical and posterior distribution we observed in some cases. In addition, we did not have the opportunity to obtain histologic verification of the pulmonary findings observed at multidetector CT, which would have enhanced our understanding of the findings. However, experienced forensic pathologists rendered the cause of death as drowning in all cases.
Limitations in our study design may have affected the interpretation of the multidetector CT images and conclusions. Namely, consensus reading and unblinded reviewers may have influenced interpretation of subtle findings. A control group of subjects who had undergone postmortem submersion would have provided a more equivalent comparison of findings than did the control group of subjects with sudden cardiac death. In postmortem submersion, fluid and sediment theoretically may enter air passages passively, but there are not data established to support this hypothesis. As such, further study is warranted to compare the multidetector CT features of premortem and postmortem submersion.
In the routine practice of forensic pathologic evaluation and death investigation, there are many instances in which a body is recovered from water days or weeks after death, and the cause of death is determined to be drowning on the basis of the circumstances surrounding the death and an external examination of the body, with no formal autopsy performed. In many of these cases, multidetector CT has the potential to establish or exclude drowning as a cause of death when frothy fluid or high-attenuation sediment is present in the airways. In the absence of frothy fluid or sediment, multidetector CT is nonspecific but may provide anatomic findings that support the diagnosis of drowning in the appropriate scenario when all other causes of death have been excluded. Furthermore, multidetector CT virtual autopsy may be useful as a pre-autopsy triage tool in mass casualty scenarios or may add additional anatomic information to a cause of death rendered by external examination or limited autopsy.
In conclusion, the multidetector CT finding of frothy airway fluid or high-attenuation airway sediment is highly suggestive of drowning. The multidetector CT findings of pan sinus fluid, mastoid cell fluid, subglottic tracheal and bronchial fluid, and ground-glass opacity within the lungs on multidetector CT are supportive of drowning in the appropriate scenario.
|
IMPLICATIONS FOR PATIENT CARE
|
|---|
- Multidetector CT virtual autopsy may help to establish drowning as a cause of death when frothy fluid or high-attenuation sediment is present in the airways in a body that is found in water.
- Multidetector CT findings may provide support for the diagnosis of drowning when other causes of death have been excluded by means of limited autopsy or external examination of the body only.
- Multidetector CT virtual autopsy may be useful as a pre-autopsy triage tool in mass casualty scenarios or may add additional anatomic information to a cause of death rendered by means of external examination or limited autopsy.
|
FOOTNOTES
|
|---|
The opinions and assertions contained herein are the private views of the authors, and are not to be construed as official, or as reflecting the view of the Departments of the Army, Navy, Air Force, or Defense.
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, A.D.L.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, A.D.L., J.L.C.; clinical studies, A.D.L., H.T.H., C.T.M., J.L.C., A.A.F.; experimental studies, A.D.L.; statistical analysis, H.T.H., J.M.G., L.P.; and manuscript editing, H.T.H., J.M.G., C.T.M., J.L.C., L.P., A.A.F., J.R.G.
|
References
|
|---|
- Gordon I. The anatomical signs in drowning: a critical evaluation. Forensic Sci 1972;1(4):389–395.[CrossRef][Medline]
- Kringsholm B, Filskov A, Kock K. Autopsied cases of drowning in Denmark 1987–1989. Forensic Sci Int 1991;52(1):85–92.[CrossRef][Medline]
- Dunagan DP, Cox JE, Chang MC, Haponik EF. Sand aspiration with near-drowning: radiographic and bronchoscopic findings. Am J Respir Crit Care Med 1997;156(1):292–295.[Abstract/Free Full Text]
- Bonilla-Santiago J, Fill WL. Sand aspiration in drowning and near drowning. Radiology 1978;128(2):301–302.[Abstract]
- Hunter TB, Whitehouse WM. Fresh-water near-drowning: radiological aspects. Radiology1974;112(1):51–56.[Medline]
- Rosenbaum HT, Thompson WL, Fuller RH. Radiographic pulmonary changes in near-drowning. Radiology 1964;83:306–313.[Medline]
- Wunderlich P, Rupprecht E, Trefftz F, Thomsen H, Burkhardt J. Chest radiographs of near-drowned children. Pediatr Radiol 1985;15(5):297–299.[CrossRef][Medline]
- Kim KI, Lee KN, Tomiyama N, et al. Near drowning: thin-section CT findings in six patients. J Comput Assist Tomogr 2000;24(4):562–566.[CrossRef][Medline]
- Thali MJ, Yen K, Schweitzer W, et al. Virtopsy, a new imaging horizon in forensic pathology: virtual autopsy by postmortem multislice computed tomography (MSCT) and magnetic resonance imaging (MRI)—a feasibility study. J Forensic Sci 2003;48(2):386–403.[Medline]
- Azparren JE, Fernandez-Rodriguez A, Vallejo G. Diagnosing death by drowning in fresh water using blood strontium as an indicator. Forensic Sci Int 2003;137(1):55–59.[CrossRef][Medline]
- Azparren JE, Vallejo G, Reyes E, Herranz A, Sancho M. Study of the diagnostic value of strontium, chloride, haemoglobin and diatoms in immersion cases. Forensic Sci Int 1998;91(2):123–132.[CrossRef][Medline]
- Hadley JA, Fowler DR. Organ weight effects of drowning and asphyxiation on the lungs, liver, brain, heart, kidneys, and spleen. Forensic Sci Int 2003;133(3):190–196.[CrossRef][Medline]
- Hurlimann J, Feer P, Elber F, Niederberger K, Dirnhofer R, Wyler D. Diatom detection in the diagnosis of death by drowning. Int J Legal Med 2000;114(1-2):6–14.[CrossRef][Medline]
- Modell JH. Drowning. N Engl J Med 1993;328(4):253–256.[Free Full Text]
- Modell JH, Gaub M, Moya F, Vestal B, Swarz H. Physiologic effects of near drowning with chlorinated fresh water, distilled water and isotonic saline. Anesthesiology 1966;27(1):33–41.[CrossRef][Medline]
- Modell JH, Moya F, Newby EJ, Ruiz BC, Showers AV. The effects of fluid volume in seawater drowning. Ann Intern Med 1967;67(1):68–80.[Medline]
- Modell JH, Bellefleur M, Davis JH. Drowning without aspiration: is this an appropriate diagnosis? J Forensic Sci 1999;44(6):1119–1123.[Medline]
This article has been cited by other articles:
|
|
|
|
 
S. Grabherr, E. Gygax, B. Sollberger, S. Ross, L. Oesterhelweg, S. Bolliger, A. Christe, V. Djonov, M. J. Thali, and R. Dirnhofer
Two-Step Postmortem Angiography with a Modified Heart Lung Machine: Preliminary Results
Am. J. Roentgenol.,
February 1, 2008;
190(2):
345 - 351.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
