ORIGINAL ARTICLE
Fluid in the Airway of Nontraumatic Death on Postmortem Computed Tomography Relationship With Pleural Effusion and Postmortem Elapsed Time Masanori Ishida, MD, PhD,*Þ Wataru Gonoi, MD, PhD,* Kazuchika Hagiwara, MD, PhD,* Hidemi Okuma, MD,* Yukako Shintani, MD, PhD,þ Hiroyuki Abe, MD, PhD,þ Yutaka Takazawa, MD, PhD,þ Kuni Ohtomo, MD, PhD,* and Masashi Fukayama, MD, PhDþ MATERIALS AND METHODS
Abstract: To evaluate radiographic features of endotracheal/endobronchial fluid in the airway (FA) observed on postmortem computed tomography (PMCT). We studied 164 subjects who died at our hospital between April 2009 and September 2012. Fluid in the airway was considered positive when fluid was identified in the lumen of 1 of the 2 main bronchi in continuity with a segmental bronchus. Pleural effusion and atelectasis/ consolidation of the lung lower lobes were also evaluated. Fluid in the airway was observed in 60 (71%) of 84 subjects with unilateral or bilateral pleural effusion, and in 44 (55%) of 80 subjects without pleural effusion (P = 0.029). Of the latter, 41 (93%) had atelectasis/consolidation of the lower lung lobes. Among subjects without pleural effusion, average times after death to PMCT of subjects with and without FA were 501 and 314 minutes, respectively (P = 0.01). Time-course analysis showed that cases with FA on PMCT largely correlated with time after death (R2 = 0.7966). Fluid in the airway is frequently observed on PMCT in subjects with pleural effusion or atelectasis/consolidation of the lung. No FA in subjects without pleural effusion correlated to shorter times after death. In addition, FA frequency on PMCT increased over time after death. Key Words: postmortem imaging, computed tomography, pleural effusion, atelectasis, consolidation (Am J Forensic Med Pathol 2014;35: 113Y117)
T
he use of high-resolution imaging techniques such as computed tomography (CT) and magnetic resonance imaging in postmortem investigations is beginning to gain a role as an adjunct to more traditional methods in forensic medicine.1Y3 There are various reports in the literature describing postmortem CT (PMCT) findings in organs such as the brain, lung, heart, and liver.4Y8 Endotracheal/endobronchial fluid is often observed in PMCT, and the features of endotracheal/endobronchial fluid on PMCT have not been described in detail. Our aim was to investigate endotracheal/endobronchial fluid on PMCT in cases of nontraumatic in-hospital death. Manuscript received August 10, 2013; accepted December 1, 2013. *Department of Radiology, Graduate School of Medicine, The University of Tokyo; †Department of Radiology, Mutual Aid Association for Tokyo Metropolitan Teachers and Officials, Sanraku Hospital; and ‡Department of Pathology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. The authors report no conflicts of interest. Reprints: Masanori Ishida, MD, PhD, Department of Radiology, Mutual Aid Association for Tokyo Metropolitan Teachers and Officials, Sanraku Hospital, 2-5 Kandasurugadai, Chiyoda-ku, Tokyo 101-8326, Japan. E-mail: [email protected]. This work was supported in part by a grant from the Japanese Ministry of Health, Labor and Welfare for research into ‘‘Usefulness of Postmortem Images as an Ancillary Method for Autopsy in Evaluation of Death Associated With Medical Practice (2008Y2009).’’ Copyright * 2014 by Lippincott Williams & Wilkins ISSN: 0195-7910/14/3502Y0113 DOI: 10.1097/PAF.0000000000000083
Am J Forensic Med Pathol
&
Volume 35, Number 2, June 2014
Subjects This study was approved by the ethics committee of our institution, and informed consent was obtained from the families of the deceased. We performed unenhanced PMCT and conventional autopsy on 193 subjects who died because of nontraumatic causes in our tertiary-care academic hospital between April 2009 and September 2012. Exclusion criteria were as follows: (a) age younger than 17 years (n = 10); (b) presence of thoracostomy tube(s) (n = 3); (c) performance of cardiopulmonary resuscitation, because cardiac massage and endotracheal intubation are considered to affect the amount of endotracheal/endobronchial fluid (n = 16). All cadavers were placed in the supine position at room temperature from the time of death until PMCT examination. Conventional autopsy was performed soon after PMCT.
Imaging Technique All PMCT studies were performed on a 4-detector-row CT scanner (Robusto; Hitachi Medical Corporation, Tokyo, Japan) in the helical mode in a cephalocaudal direction without using a contrast medium; the cadaver was laid in the supine position with hands placed on either side. The following routine scanning parameters were used for all investigations: 120 kVp, 200 mAs, and 1.25 pitch factor in all cases with 2.5 4-mm collimation and 2.5-mm section thickness. Axial images were reconstructed with a 350-mm field of view and a 512 512 image matrix from the CT scanner at 1.3-mm intervals.
Image Interpretation Images were interpreted in a digital imaging and communications in medicine viewer (zioTerm2009; Ziosoft, Inc, Tokyo, Japan) using 2-dimensional axial data sets. Image analysis was performed by 2 board-certified radiologists who were blinded to clinical information; they interpreted the results by consensus. When fluid was identified in the lumen of at least 1 of the 2 main bronchi in continuity with a segmental bronchus, we defined the finding as positive for fluid in the airway (FA; Fig. 1, AYC). The findings of pleural effusion and atelectasis/ consolidation (Fig. 1B) of the lower lobe of the lung were also evaluated. When the top horizontal level of pleural effusion was higher than the boundary between the anterior and middle column of the vertebral bodies in any of the axial images at the level of the lower lobe, we defined the finding as positive pleural effusion in this study (Fig. 1, C and D). It has been reported that a single area measurement of an axial PMCT image correlates with the volume of pleural effusion.9 On the basis of this, boundary between the anterior and the middle column of the vertebral bodies in any of the axial images may be considered an acceptable approximate indicator of the volume of pleural effusion. The finding of www.amjforensicmedicine.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
113
Am J Forensic Med Pathol
Ishida et al
&
Volume 35, Number 2, June 2014
FIGURE 1. Examples of FA. A, A 78-year-old man scanned 179 minutes after death. Fluid is observed in the main bronchi bilaterally at the inferior tracheobronchial level (a, arrows) and in the absence of the finding of pleural effusion. Lung density is slightly increased. The dotted line shows the level of the boundary between the anterior and the middle column of the vertebral body. B, A 51-year-old man scanned 751 minutes after death. Fluid is observed in the main bronchi along with consolidation of the lungs bilaterally, but no pleural effusion is identified. C, Example of both FA and pleural effusion. An 86-year-old man scanned 287 minutes after death. Fluid is observed in the right main bronchus (arrow) along with bilateral pleural effusion at the lower tracheobronchial level. D, An example of pleural effusion alone. A 75-year-old man scanned 117 minutes after death. Findings of FA are negative, but bilateral pleural effusion is observed. E, Example of neither FA nor pleural effusion. An 81-year-old man scanned 292 minutes after death. Neither FA nor pleural effusion is observed, and virtually no increased density of the lung can be seen.
atelectasis/consolidation was defined as positive when the volume of aerated lung parenchyma of the lower lobe decreased by approximately half.
elapsed time from death to PMCT was also evaluated using the Mann-Whitney U test. P G 0.05 was considered statistically significant.
Statistical Analysis Statistical analyses were performed using GraphPad Prism 6 software (GraphPad Software Inc, San Diego, Calif ). We used the Mann-Whitney U test, and the W2 test to investigate differences between groups with and without pleural effusion and to identify whether FA was related to the absence or presence of pleural effusion on PMCT. The relationship between FA and
114
www.amjforensicmedicine.com
RESULTS Study Population The final study population consisted of 164 adult human patients (109 male, 55 female), with a mean age at death of 66 years (range, 18Y98 years; mean, 63.3 years; median, 71 years). Postmortem computed tomography was performed at 74 to * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Am J Forensic Med Pathol
&
Volume 35, Number 2, June 2014
TABLE 1. Differences Between Groups With and Without Pleural Effusion Pleural Effusion Age, mean (SD), y Sex, male/female, n Elapsed time to PMCT, mean (SD), min FA, %
Negative (n = 80)
Positive (n = 84)
P
62.6 (17.0) 58:22 419 (323)
69.9 (14.0) 51:33 484 (369)
0.003* 0.11 0.25
55
71
0.029*
Statistical analyses were performed using Mann-Whitney U test for age, elapsed time to PMCT, and W2 test for sex and percentage of FA. *Statistically significant.
1921 minutes (mean 452 minutes; median, 314 minutes) after the confirmation of death by physicians. Based on autopsy results, modes of dying were respiratory failure (96 cases), heart failure (20 cases), multiple organ failure (19 cases), hepatic failure (9 cases), septic shock (9 cases), hemorrhagic death (7 cases), and renal failure (4 cases). All data obtained from the subjects in the study regarding the absence or presence of pleural effusion are summarized in Table 1.
Assessment of FA on PMCT The findings of FA on PMCT were found more frequently in patients with unilateral or bilateral pleural effusions than in those without pleural effusion (Table 1). There was a significant difference between the 2 groups (P = 0.029). The pulmonary findings of atelectasis/consolidation of the lower lobe of the lung were identified in more than 90% of subjects with the finding of FA but without pleural effusion. In contrast, these pulmonary findings were identified in less than half of subjects without both FA and pleural effusion (Table 2, Fig. 1E). Among the subjects without pleural effusions, there was a significant difference in the frequency of the pulmonary findings of atelectasis/consolidation between groups with and without the findings of FA (P G 0.0001). Data from subjects negative for pleural effusion are summarized in Table 2.
Relationship to Elapsed Time From Death Among the subjects without pleural effusion on PMCT, there was a significant difference in time elapsed from death to PMCT between the groups with and without findings of FA (P = 0.01). The relationships between percentages of FA on PMCT and elapsed time from death to PMCT are shown in Figure 2. Linear approximations of percentage of FA in the group without pleural effusion and in all subjects show positive trends with increasing elapsed time. With regard to findings such as FA, atelectasis/consolidation, and pleural effusion, Figure 3 shows the relationships between the percentage of findings and elapsed time from death to PMCT in all subjects.
DISCUSSION Hypostasis, the earliest postmortem pathological change, occurs throughout all fluid compartments, tissues, and organs in the body.10 Following the onset of hypostasis, its distribution will shift with repositioning of the body and will blanch with pressure.11,12 When a cadaver is laid in the supine position, hypostasis of the dorsal parts of the lungs is observed under the influence of gravity. After cardiac activity ceases, hypostasis of * 2014 Lippincott Williams & Wilkins
Fluid in the Airway on Postmortem CT
the lungs begins and blood-derived fluid permeates the alveoli via capillaries, possibly because the cessation of cardiac activity results in the shift of large volumes of blood from the systemic circulation into the pulmonary circulation.13 When the mean pulmonary filling pressure increases above the colloid osmotic pressure of the plasma, blood-derived fluid begins to filter out from the capillaries into the interstitial spaces and alveoli. However, an increase in the mean pulmonary filling pressure does not always appear after death. In this case, the postmortem hydrostatic pressure is considered to be the mean circulatory pressure, which does not exceed the colloid osmotic pressure of the plasma.13 Meanwhile, it is possible that transudation into the interstitial spaces and alveoli results in leakage of fluid into the pleural cavity. In reports of autopsies of 260 drowning victims reported in the Japanese literature, little pleural effusion was identified within 24 hours after death, whereas more than 24 hours after death, the volume of pleural effusion increased with time.14 Moreover, it has been reported that, in drowning subjects, the volume of pleural effusion usually observed on PMCT scans performed 2 to 12 days after death is usually quite small.15 These findings suggest that a barrier of visceral pleura may prevent fluid from the alveoli exuding into the pleural cavity during the early postmortem phase. Filling of blood-derived fluid into the interstitial spaces and alveoli results in pulmonary edema and eventually increased density of the lung on PMCT.5,16,17 Increased density of the lung appears more markedly in the posterior part of the lungs, and then, with increasing degrees of hypostasis, the area of increased density in the lungs extends superiorly, forming an approximately horizontal line with a vertical density gradient from the independent to the dependent portions of the lung.18 The blood-derived fluid, observed as an increased density of the lung, flows sequentially upstream from the bronchiole to the trachea over time. Consequently, these characteristic changes are observed as the finding of FA on PMCT. Shiotani et al5 have reported that the findings of an endotracheal/endobronchial air defect on PMCT were observed in 12% of 150 subjects who died because of nontraumatic causes. This result differs from our result of 63% of 164 subjects. One possible reason for this discrepancy could be that in the study of Shiotani et al, PMCT scans were performed as soon as 2 hours after death. A large amount of pleural effusion causes adjacent compressive atelectasis of the lung. We believe that decreased lung volume causes blood-derived fluid in the alveoli to move into central airways such as the bronchus or trachea earlier because of the decrease in the volume of aerated lung parenchyma, which the blood-derived fluid occupies. Supporting this hypothesis, we observed a statistically significant difference in the TABLE 2. Differences in Groups of Negative Pleural Effusion Between With and Without FA FA
Negative (n = 36)
Positive (n = 44)
P
Age, mean (SD), y 60.3 (16.6) 64.5 (15.5) 0.17 Sex, male/female, n 27:9 31:13 0.65 Elapsed time to PMCT, 318 (256) 501 (351) 0.01* mean (SD), min Atelectasis/consolidation, % 44 93 G0.0001* Statistical analyses were performed by using Mann-Whitney U test for age, elapsed time to PMCT, and W2 test for sex and percentage of atelectasis/consolidation. *Statistically significant.
www.amjforensicmedicine.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
115
Am J Forensic Med Pathol
Ishida et al
&
Volume 35, Number 2, June 2014
FIGURE 2. Distribution of the percentage of FA based on pleural effusion. Lines show linear approximations. PE indicates pleural effusion.
frequency of the finding of FA in subjects with pleural effusion on PMCT compared with those without pleural effusion on PMCT. In a similar hypothetical theory, the findings of FA are often observed in subjects with atelectasis/consolidation of the lung, which would result in decreased volume of the aerated lung. In our study, subjects were divided into 2 groups according to the presence of pleural effusion because pleural effusion affects the degree of FA. In the group without pleural effusion, elapsed time from death to PMCT in subjects without FA on PMCT was significantly shorter than that in the subjects with FA on PMCT. Although it is difficult to accurately estimate elapsed time from death to PMCT and a range of times elapsed in this study, elapsed time from death could be estimated to be relatively short (with a mean of approximately 3.5 hours) when neither FA nor pleural effusion was found on PMCT. Regardless of whether pleural effusion is observed on PMCT, the longer the elapsed time from death to PMCT, the higher the percentage of patients with the finding of FA on PMCT. The finding of FA has seldom been observed before death, and we believe that FA develops from the lung parenchyma in an ascending fashion. This would thus explain the consequence that the percentage of findings of FA on PMCT correlates with elapsed time from death to PMCT. In contrast, the findings of pleural effusion or atelectasis/consolidation do not correlate with elapsed time from death to PMCT. There is a possibility that these findings existed before death or that the percentages of these findings on PMCT differ according to the elapsed time from death to PMCT. Unlike the case with FA, these findings might be
observed before death, indicated by the lack of correlation with elapsed time from death to PMCT. The present study has a few limitations. First, we did not take into account the possibility of fluid being derived from the oral cavity. However, when endotracheal/endobronchial fluid in continuity with the oral cavity is observed, there are few likely causes except drowning or choking on eating. Second, potential complicating factors such as underlying disease, cause of death, preservation of cadavers, atmospheric temperature, and humidity were not taken into account. These factors might affect postmortem changes.19,20 Third, the postmortem interval between the cases is similar, and most of the cases were examined some hours after death. It cannot be denied that the fluid was in the airways already before death, and there is another argument that it is somewhat difficult to endorse fully the conclusion of a correlation between the presence of FA and the postmortem delay. Postmortem changes do not follow an even and steady course in all cadavers, and PMCT findings of the lung during the early postmortem period are not always observed. Few articles on the connection between elapsed time from death and PMCT findings have been published.8,21,22 This study will be helpful to elucidate the relationship between PMCT findings and elapsed time from death. Further studies of patients who died in other ways except in-hospital death are needed. We believe that interpretation of PMCT, in additional studies, might provide an important contribution to forensic radiology. In conclusion, the finding of FA on PMCT is frequently observed in the cadavers of patients who experienced nontraumatic
FIGURE 3. Comparison of the percentage of findings among FA, atelectasis/consolidation, and pleural effusion. Lines show linear approximations. PE indicate pleural effusion.
116
www.amjforensicmedicine.com
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Am J Forensic Med Pathol
&
Volume 35, Number 2, June 2014
death with pleural effusion and also in those without pleural effusion who have the finding of atelectasis/consolidation of the lung. A negative finding of FA in subjects without pleural effusion is related to a shorter elapsed time from death to PMCT. The frequency of the finding of FA on PMCT increases over time after death, and it is suggested that endotracheal/ endobronchial fluid, which is recognized as the source of FA, emerges from the peripheral airway. REFERENCES 1. Patriquin L, Kassarjian A, Casserley L, et al. Postmortem whole-body magnetic resonance imaging as an adjunct to autopsy: preliminary clinical experience. J Magn Reson Imaging. 2001;13:277Y287. 2. 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)Va feasibility study. J Forensic Sci. 2003;48:386Y403. 3. Ezawa H, Yoneyama R, Kandatsu S, et al. Introduction of autopsy imaging redefines the concept of autopsy: 37 cases of clinical experience. Pathol Int. 2003;53:865Y873. 4. Levy AD, Harcke HT, Mallak CT. Postmortem imaging: MDCT features of postmortem change and decomposition. Am J Forensic Med Pathol. 2010;31:12Y17. 5. Shiotani S, Kohno M, Ohashi N, et al. Non-traumatic postmortem computed tomographic (PMCT) findings of the lung. Forensic Sci Int. 2004;139:39Y48. 6. Kobayashi T, Shiotani S, Kaga K, et al. Characteristic signal intensity changes on postmortem magnetic resonance imaging of the brain. Jpn J Radiol. 2010;28:8Y14. 7. Jackowski C, Thali MJ, Buck U, et al. Noninvasive estimation of organ weights by postmortem magnetic resonance imaging and multislice computed tomography. Invest Radiol. 2006;41:572Y578.
Fluid in the Airway on Postmortem CT
10. Shepherd R. Simpson’s Forensic Medicine. 12th ed. Boca Raton, FL: CRC Press; 2003. 11. Di Maio VJ, Di Maio DJ. Forensic Pathology. 2nd ed. Boca Raton, FL: CRC Press; 2001. 12. Payne-James J, Busuttil A, Smock W. Forensic Medicine: Clinical and Pathological Aspects. 1st ed. San Francisco, CA: Greenwich Medical Media; 2002. 13. Guyton AC, Hall JE. Textbook of Medical Physiology. 12th ed. Philadelphia, PA: WB Saunders; 2010. 14. Hasegawa I, Nushida H, Vogel H, et al. Post-mortem changes observed in sequential scanning [in Japanese]. Houi-Byouri. 2010;16:97Y102. 15. Levy AD, Harcke HT, Getz JM, et al. Virtual autopsy: two- and three-dimensional multidetector CT findings in drowning with autopsy comparison. Radiology. 2007;243:862Y868. 16. Hedlund LW, Vock P, Effmann EL, et al. Hydrostatic pulmonary edema an analysis of lung density changes by computed tomography. Invest Radiol. 1984;19:254Y262. 17. Scillia P, Delcroix M, Lejeune P, et al. Hydrostatic pulmonary edema: evaluation with thin-section CT in dogs. Radiology. 1999;211:161Y168. 18. Shiotani S, Kobayashi T, Hayakawa H, et al. Postmortem pulmonary edema: a comparison between immediate and delayed postmortem computed tomography. Legal Med (Tokyo). 2011;13:151Y155. 19. Jackowski C, Sonnenschein M, Thali MJ, et al. Intrahepatic gas at postmortem computed tomography: forensic experience as a potential guide for in vivo trauma imaging. J Trauma. 2007;62:979Y988. 20. Singh MK, O’Donnell C, Woodford NW. Progressive gas formation in a deceased person during mortuary storage demonstrated on computed tomography. Forensic Sci Med Pathol. 2009;5:236Y242.
8. Ishida M, Gonoi W, Hagiwara K, et al. Intravascular gas distribution in the upper abdomen of non-traumatic in-hospital death cases on postmortem computed tomography. Legal Med (Tokyo). 2011;13:174Y179.
21. Ishida M, Gonoi W, Hagiwara K, et al. Hypostasis in the heart and great vessels of non-traumatic in-hospital death cases on postmortem computed tomography: relationship to antemortem blood tests. Legal Med (Tokyo). 2011;13:280Y285.
9. Veljkovic B, Franckenberg S, Hatch GM, et al. Quantification of pleural effusion from single area measurements on CT. Emerg Radiol. 2013;20:285Y289.
22. Ishida M, Gonoi W, Hagiwara K, et al. Postmortem changes of the thyroid on computed tomography. Legal Med (Tokyo). 2011;13:318Y322.
* 2014 Lippincott Williams & Wilkins
www.amjforensicmedicine.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
117
Fluid in the Airway of Nontraumatic Death on Postmortem Computed Tomography Relationship With Pleural Effusion and Postmortem Elapsed Time Masanori Ishida, MD, PhD,*Þ Wataru Gonoi, MD, PhD,* Kazuchika Hagiwara, MD, PhD,* Hidemi Okuma, MD,* Yukako Shintani, MD, PhD,þ Hiroyuki Abe, MD, PhD,þ Yutaka Takazawa, MD, PhD,þ Kuni Ohtomo, MD, PhD,* and Masashi Fukayama, MD, PhDþ MATERIALS AND METHODS
Abstract: To evaluate radiographic features of endotracheal/endobronchial fluid in the airway (FA) observed on postmortem computed tomography (PMCT). We studied 164 subjects who died at our hospital between April 2009 and September 2012. Fluid in the airway was considered positive when fluid was identified in the lumen of 1 of the 2 main bronchi in continuity with a segmental bronchus. Pleural effusion and atelectasis/ consolidation of the lung lower lobes were also evaluated. Fluid in the airway was observed in 60 (71%) of 84 subjects with unilateral or bilateral pleural effusion, and in 44 (55%) of 80 subjects without pleural effusion (P = 0.029). Of the latter, 41 (93%) had atelectasis/consolidation of the lower lung lobes. Among subjects without pleural effusion, average times after death to PMCT of subjects with and without FA were 501 and 314 minutes, respectively (P = 0.01). Time-course analysis showed that cases with FA on PMCT largely correlated with time after death (R2 = 0.7966). Fluid in the airway is frequently observed on PMCT in subjects with pleural effusion or atelectasis/consolidation of the lung. No FA in subjects without pleural effusion correlated to shorter times after death. In addition, FA frequency on PMCT increased over time after death. Key Words: postmortem imaging, computed tomography, pleural effusion, atelectasis, consolidation (Am J Forensic Med Pathol 2014;35: 113Y117)
T
he use of high-resolution imaging techniques such as computed tomography (CT) and magnetic resonance imaging in postmortem investigations is beginning to gain a role as an adjunct to more traditional methods in forensic medicine.1Y3 There are various reports in the literature describing postmortem CT (PMCT) findings in organs such as the brain, lung, heart, and liver.4Y8 Endotracheal/endobronchial fluid is often observed in PMCT, and the features of endotracheal/endobronchial fluid on PMCT have not been described in detail. Our aim was to investigate endotracheal/endobronchial fluid on PMCT in cases of nontraumatic in-hospital death. Manuscript received August 10, 2013; accepted December 1, 2013. *Department of Radiology, Graduate School of Medicine, The University of Tokyo; †Department of Radiology, Mutual Aid Association for Tokyo Metropolitan Teachers and Officials, Sanraku Hospital; and ‡Department of Pathology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. The authors report no conflicts of interest. Reprints: Masanori Ishida, MD, PhD, Department of Radiology, Mutual Aid Association for Tokyo Metropolitan Teachers and Officials, Sanraku Hospital, 2-5 Kandasurugadai, Chiyoda-ku, Tokyo 101-8326, Japan. E-mail: [email protected]. This work was supported in part by a grant from the Japanese Ministry of Health, Labor and Welfare for research into ‘‘Usefulness of Postmortem Images as an Ancillary Method for Autopsy in Evaluation of Death Associated With Medical Practice (2008Y2009).’’ Copyright * 2014 by Lippincott Williams & Wilkins ISSN: 0195-7910/14/3502Y0113 DOI: 10.1097/PAF.0000000000000083
Am J Forensic Med Pathol
&
Volume 35, Number 2, June 2014
Subjects This study was approved by the ethics committee of our institution, and informed consent was obtained from the families of the deceased. We performed unenhanced PMCT and conventional autopsy on 193 subjects who died because of nontraumatic causes in our tertiary-care academic hospital between April 2009 and September 2012. Exclusion criteria were as follows: (a) age younger than 17 years (n = 10); (b) presence of thoracostomy tube(s) (n = 3); (c) performance of cardiopulmonary resuscitation, because cardiac massage and endotracheal intubation are considered to affect the amount of endotracheal/endobronchial fluid (n = 16). All cadavers were placed in the supine position at room temperature from the time of death until PMCT examination. Conventional autopsy was performed soon after PMCT.
Imaging Technique All PMCT studies were performed on a 4-detector-row CT scanner (Robusto; Hitachi Medical Corporation, Tokyo, Japan) in the helical mode in a cephalocaudal direction without using a contrast medium; the cadaver was laid in the supine position with hands placed on either side. The following routine scanning parameters were used for all investigations: 120 kVp, 200 mAs, and 1.25 pitch factor in all cases with 2.5 4-mm collimation and 2.5-mm section thickness. Axial images were reconstructed with a 350-mm field of view and a 512 512 image matrix from the CT scanner at 1.3-mm intervals.
Image Interpretation Images were interpreted in a digital imaging and communications in medicine viewer (zioTerm2009; Ziosoft, Inc, Tokyo, Japan) using 2-dimensional axial data sets. Image analysis was performed by 2 board-certified radiologists who were blinded to clinical information; they interpreted the results by consensus. When fluid was identified in the lumen of at least 1 of the 2 main bronchi in continuity with a segmental bronchus, we defined the finding as positive for fluid in the airway (FA; Fig. 1, AYC). The findings of pleural effusion and atelectasis/ consolidation (Fig. 1B) of the lower lobe of the lung were also evaluated. When the top horizontal level of pleural effusion was higher than the boundary between the anterior and middle column of the vertebral bodies in any of the axial images at the level of the lower lobe, we defined the finding as positive pleural effusion in this study (Fig. 1, C and D). It has been reported that a single area measurement of an axial PMCT image correlates with the volume of pleural effusion.9 On the basis of this, boundary between the anterior and the middle column of the vertebral bodies in any of the axial images may be considered an acceptable approximate indicator of the volume of pleural effusion. The finding of www.amjforensicmedicine.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
113
Am J Forensic Med Pathol
Ishida et al
&
Volume 35, Number 2, June 2014
FIGURE 1. Examples of FA. A, A 78-year-old man scanned 179 minutes after death. Fluid is observed in the main bronchi bilaterally at the inferior tracheobronchial level (a, arrows) and in the absence of the finding of pleural effusion. Lung density is slightly increased. The dotted line shows the level of the boundary between the anterior and the middle column of the vertebral body. B, A 51-year-old man scanned 751 minutes after death. Fluid is observed in the main bronchi along with consolidation of the lungs bilaterally, but no pleural effusion is identified. C, Example of both FA and pleural effusion. An 86-year-old man scanned 287 minutes after death. Fluid is observed in the right main bronchus (arrow) along with bilateral pleural effusion at the lower tracheobronchial level. D, An example of pleural effusion alone. A 75-year-old man scanned 117 minutes after death. Findings of FA are negative, but bilateral pleural effusion is observed. E, Example of neither FA nor pleural effusion. An 81-year-old man scanned 292 minutes after death. Neither FA nor pleural effusion is observed, and virtually no increased density of the lung can be seen.
atelectasis/consolidation was defined as positive when the volume of aerated lung parenchyma of the lower lobe decreased by approximately half.
elapsed time from death to PMCT was also evaluated using the Mann-Whitney U test. P G 0.05 was considered statistically significant.
Statistical Analysis Statistical analyses were performed using GraphPad Prism 6 software (GraphPad Software Inc, San Diego, Calif ). We used the Mann-Whitney U test, and the W2 test to investigate differences between groups with and without pleural effusion and to identify whether FA was related to the absence or presence of pleural effusion on PMCT. The relationship between FA and
114
www.amjforensicmedicine.com
RESULTS Study Population The final study population consisted of 164 adult human patients (109 male, 55 female), with a mean age at death of 66 years (range, 18Y98 years; mean, 63.3 years; median, 71 years). Postmortem computed tomography was performed at 74 to * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Am J Forensic Med Pathol
&
Volume 35, Number 2, June 2014
TABLE 1. Differences Between Groups With and Without Pleural Effusion Pleural Effusion Age, mean (SD), y Sex, male/female, n Elapsed time to PMCT, mean (SD), min FA, %
Negative (n = 80)
Positive (n = 84)
P
62.6 (17.0) 58:22 419 (323)
69.9 (14.0) 51:33 484 (369)
0.003* 0.11 0.25
55
71
0.029*
Statistical analyses were performed using Mann-Whitney U test for age, elapsed time to PMCT, and W2 test for sex and percentage of FA. *Statistically significant.
1921 minutes (mean 452 minutes; median, 314 minutes) after the confirmation of death by physicians. Based on autopsy results, modes of dying were respiratory failure (96 cases), heart failure (20 cases), multiple organ failure (19 cases), hepatic failure (9 cases), septic shock (9 cases), hemorrhagic death (7 cases), and renal failure (4 cases). All data obtained from the subjects in the study regarding the absence or presence of pleural effusion are summarized in Table 1.
Assessment of FA on PMCT The findings of FA on PMCT were found more frequently in patients with unilateral or bilateral pleural effusions than in those without pleural effusion (Table 1). There was a significant difference between the 2 groups (P = 0.029). The pulmonary findings of atelectasis/consolidation of the lower lobe of the lung were identified in more than 90% of subjects with the finding of FA but without pleural effusion. In contrast, these pulmonary findings were identified in less than half of subjects without both FA and pleural effusion (Table 2, Fig. 1E). Among the subjects without pleural effusions, there was a significant difference in the frequency of the pulmonary findings of atelectasis/consolidation between groups with and without the findings of FA (P G 0.0001). Data from subjects negative for pleural effusion are summarized in Table 2.
Relationship to Elapsed Time From Death Among the subjects without pleural effusion on PMCT, there was a significant difference in time elapsed from death to PMCT between the groups with and without findings of FA (P = 0.01). The relationships between percentages of FA on PMCT and elapsed time from death to PMCT are shown in Figure 2. Linear approximations of percentage of FA in the group without pleural effusion and in all subjects show positive trends with increasing elapsed time. With regard to findings such as FA, atelectasis/consolidation, and pleural effusion, Figure 3 shows the relationships between the percentage of findings and elapsed time from death to PMCT in all subjects.
DISCUSSION Hypostasis, the earliest postmortem pathological change, occurs throughout all fluid compartments, tissues, and organs in the body.10 Following the onset of hypostasis, its distribution will shift with repositioning of the body and will blanch with pressure.11,12 When a cadaver is laid in the supine position, hypostasis of the dorsal parts of the lungs is observed under the influence of gravity. After cardiac activity ceases, hypostasis of * 2014 Lippincott Williams & Wilkins
Fluid in the Airway on Postmortem CT
the lungs begins and blood-derived fluid permeates the alveoli via capillaries, possibly because the cessation of cardiac activity results in the shift of large volumes of blood from the systemic circulation into the pulmonary circulation.13 When the mean pulmonary filling pressure increases above the colloid osmotic pressure of the plasma, blood-derived fluid begins to filter out from the capillaries into the interstitial spaces and alveoli. However, an increase in the mean pulmonary filling pressure does not always appear after death. In this case, the postmortem hydrostatic pressure is considered to be the mean circulatory pressure, which does not exceed the colloid osmotic pressure of the plasma.13 Meanwhile, it is possible that transudation into the interstitial spaces and alveoli results in leakage of fluid into the pleural cavity. In reports of autopsies of 260 drowning victims reported in the Japanese literature, little pleural effusion was identified within 24 hours after death, whereas more than 24 hours after death, the volume of pleural effusion increased with time.14 Moreover, it has been reported that, in drowning subjects, the volume of pleural effusion usually observed on PMCT scans performed 2 to 12 days after death is usually quite small.15 These findings suggest that a barrier of visceral pleura may prevent fluid from the alveoli exuding into the pleural cavity during the early postmortem phase. Filling of blood-derived fluid into the interstitial spaces and alveoli results in pulmonary edema and eventually increased density of the lung on PMCT.5,16,17 Increased density of the lung appears more markedly in the posterior part of the lungs, and then, with increasing degrees of hypostasis, the area of increased density in the lungs extends superiorly, forming an approximately horizontal line with a vertical density gradient from the independent to the dependent portions of the lung.18 The blood-derived fluid, observed as an increased density of the lung, flows sequentially upstream from the bronchiole to the trachea over time. Consequently, these characteristic changes are observed as the finding of FA on PMCT. Shiotani et al5 have reported that the findings of an endotracheal/endobronchial air defect on PMCT were observed in 12% of 150 subjects who died because of nontraumatic causes. This result differs from our result of 63% of 164 subjects. One possible reason for this discrepancy could be that in the study of Shiotani et al, PMCT scans were performed as soon as 2 hours after death. A large amount of pleural effusion causes adjacent compressive atelectasis of the lung. We believe that decreased lung volume causes blood-derived fluid in the alveoli to move into central airways such as the bronchus or trachea earlier because of the decrease in the volume of aerated lung parenchyma, which the blood-derived fluid occupies. Supporting this hypothesis, we observed a statistically significant difference in the TABLE 2. Differences in Groups of Negative Pleural Effusion Between With and Without FA FA
Negative (n = 36)
Positive (n = 44)
P
Age, mean (SD), y 60.3 (16.6) 64.5 (15.5) 0.17 Sex, male/female, n 27:9 31:13 0.65 Elapsed time to PMCT, 318 (256) 501 (351) 0.01* mean (SD), min Atelectasis/consolidation, % 44 93 G0.0001* Statistical analyses were performed by using Mann-Whitney U test for age, elapsed time to PMCT, and W2 test for sex and percentage of atelectasis/consolidation. *Statistically significant.
www.amjforensicmedicine.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
115
Am J Forensic Med Pathol
Ishida et al
&
Volume 35, Number 2, June 2014
FIGURE 2. Distribution of the percentage of FA based on pleural effusion. Lines show linear approximations. PE indicates pleural effusion.
frequency of the finding of FA in subjects with pleural effusion on PMCT compared with those without pleural effusion on PMCT. In a similar hypothetical theory, the findings of FA are often observed in subjects with atelectasis/consolidation of the lung, which would result in decreased volume of the aerated lung. In our study, subjects were divided into 2 groups according to the presence of pleural effusion because pleural effusion affects the degree of FA. In the group without pleural effusion, elapsed time from death to PMCT in subjects without FA on PMCT was significantly shorter than that in the subjects with FA on PMCT. Although it is difficult to accurately estimate elapsed time from death to PMCT and a range of times elapsed in this study, elapsed time from death could be estimated to be relatively short (with a mean of approximately 3.5 hours) when neither FA nor pleural effusion was found on PMCT. Regardless of whether pleural effusion is observed on PMCT, the longer the elapsed time from death to PMCT, the higher the percentage of patients with the finding of FA on PMCT. The finding of FA has seldom been observed before death, and we believe that FA develops from the lung parenchyma in an ascending fashion. This would thus explain the consequence that the percentage of findings of FA on PMCT correlates with elapsed time from death to PMCT. In contrast, the findings of pleural effusion or atelectasis/consolidation do not correlate with elapsed time from death to PMCT. There is a possibility that these findings existed before death or that the percentages of these findings on PMCT differ according to the elapsed time from death to PMCT. Unlike the case with FA, these findings might be
observed before death, indicated by the lack of correlation with elapsed time from death to PMCT. The present study has a few limitations. First, we did not take into account the possibility of fluid being derived from the oral cavity. However, when endotracheal/endobronchial fluid in continuity with the oral cavity is observed, there are few likely causes except drowning or choking on eating. Second, potential complicating factors such as underlying disease, cause of death, preservation of cadavers, atmospheric temperature, and humidity were not taken into account. These factors might affect postmortem changes.19,20 Third, the postmortem interval between the cases is similar, and most of the cases were examined some hours after death. It cannot be denied that the fluid was in the airways already before death, and there is another argument that it is somewhat difficult to endorse fully the conclusion of a correlation between the presence of FA and the postmortem delay. Postmortem changes do not follow an even and steady course in all cadavers, and PMCT findings of the lung during the early postmortem period are not always observed. Few articles on the connection between elapsed time from death and PMCT findings have been published.8,21,22 This study will be helpful to elucidate the relationship between PMCT findings and elapsed time from death. Further studies of patients who died in other ways except in-hospital death are needed. We believe that interpretation of PMCT, in additional studies, might provide an important contribution to forensic radiology. In conclusion, the finding of FA on PMCT is frequently observed in the cadavers of patients who experienced nontraumatic
FIGURE 3. Comparison of the percentage of findings among FA, atelectasis/consolidation, and pleural effusion. Lines show linear approximations. PE indicate pleural effusion.
116
www.amjforensicmedicine.com
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Am J Forensic Med Pathol
&
Volume 35, Number 2, June 2014
death with pleural effusion and also in those without pleural effusion who have the finding of atelectasis/consolidation of the lung. A negative finding of FA in subjects without pleural effusion is related to a shorter elapsed time from death to PMCT. The frequency of the finding of FA on PMCT increases over time after death, and it is suggested that endotracheal/ endobronchial fluid, which is recognized as the source of FA, emerges from the peripheral airway. REFERENCES 1. Patriquin L, Kassarjian A, Casserley L, et al. Postmortem whole-body magnetic resonance imaging as an adjunct to autopsy: preliminary clinical experience. J Magn Reson Imaging. 2001;13:277Y287. 2. 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)Va feasibility study. J Forensic Sci. 2003;48:386Y403. 3. Ezawa H, Yoneyama R, Kandatsu S, et al. Introduction of autopsy imaging redefines the concept of autopsy: 37 cases of clinical experience. Pathol Int. 2003;53:865Y873. 4. Levy AD, Harcke HT, Mallak CT. Postmortem imaging: MDCT features of postmortem change and decomposition. Am J Forensic Med Pathol. 2010;31:12Y17. 5. Shiotani S, Kohno M, Ohashi N, et al. Non-traumatic postmortem computed tomographic (PMCT) findings of the lung. Forensic Sci Int. 2004;139:39Y48. 6. Kobayashi T, Shiotani S, Kaga K, et al. Characteristic signal intensity changes on postmortem magnetic resonance imaging of the brain. Jpn J Radiol. 2010;28:8Y14. 7. Jackowski C, Thali MJ, Buck U, et al. Noninvasive estimation of organ weights by postmortem magnetic resonance imaging and multislice computed tomography. Invest Radiol. 2006;41:572Y578.
Fluid in the Airway on Postmortem CT
10. Shepherd R. Simpson’s Forensic Medicine. 12th ed. Boca Raton, FL: CRC Press; 2003. 11. Di Maio VJ, Di Maio DJ. Forensic Pathology. 2nd ed. Boca Raton, FL: CRC Press; 2001. 12. Payne-James J, Busuttil A, Smock W. Forensic Medicine: Clinical and Pathological Aspects. 1st ed. San Francisco, CA: Greenwich Medical Media; 2002. 13. Guyton AC, Hall JE. Textbook of Medical Physiology. 12th ed. Philadelphia, PA: WB Saunders; 2010. 14. Hasegawa I, Nushida H, Vogel H, et al. Post-mortem changes observed in sequential scanning [in Japanese]. Houi-Byouri. 2010;16:97Y102. 15. Levy AD, Harcke HT, Getz JM, et al. Virtual autopsy: two- and three-dimensional multidetector CT findings in drowning with autopsy comparison. Radiology. 2007;243:862Y868. 16. Hedlund LW, Vock P, Effmann EL, et al. Hydrostatic pulmonary edema an analysis of lung density changes by computed tomography. Invest Radiol. 1984;19:254Y262. 17. Scillia P, Delcroix M, Lejeune P, et al. Hydrostatic pulmonary edema: evaluation with thin-section CT in dogs. Radiology. 1999;211:161Y168. 18. Shiotani S, Kobayashi T, Hayakawa H, et al. Postmortem pulmonary edema: a comparison between immediate and delayed postmortem computed tomography. Legal Med (Tokyo). 2011;13:151Y155. 19. Jackowski C, Sonnenschein M, Thali MJ, et al. Intrahepatic gas at postmortem computed tomography: forensic experience as a potential guide for in vivo trauma imaging. J Trauma. 2007;62:979Y988. 20. Singh MK, O’Donnell C, Woodford NW. Progressive gas formation in a deceased person during mortuary storage demonstrated on computed tomography. Forensic Sci Med Pathol. 2009;5:236Y242.
8. Ishida M, Gonoi W, Hagiwara K, et al. Intravascular gas distribution in the upper abdomen of non-traumatic in-hospital death cases on postmortem computed tomography. Legal Med (Tokyo). 2011;13:174Y179.
21. Ishida M, Gonoi W, Hagiwara K, et al. Hypostasis in the heart and great vessels of non-traumatic in-hospital death cases on postmortem computed tomography: relationship to antemortem blood tests. Legal Med (Tokyo). 2011;13:280Y285.
9. Veljkovic B, Franckenberg S, Hatch GM, et al. Quantification of pleural effusion from single area measurements on CT. Emerg Radiol. 2013;20:285Y289.
22. Ishida M, Gonoi W, Hagiwara K, et al. Postmortem changes of the thyroid on computed tomography. Legal Med (Tokyo). 2011;13:318Y322.
* 2014 Lippincott Williams & Wilkins
www.amjforensicmedicine.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
117
