The diagnosis of drowning, as a respiratory. Influence of postmortem findings on lung weight of drowning victims

Transcription

1 Rom J Leg Med [22] [2014] DOI: /rjlm Romanian Society of Legal Medicine Influence of postmortem findings on lung weight of drowning victims Leticia Rubio 1,*, Fernando Martín-Cazorla 2, María-Jesús Gaitán 1, Valentín Ramos 2, Ignacio Santos 1 Abstract: The identification of fatal mechanisms involving drowning represents a difficult task for the forensic pathologists. Here, we analysed macromorphological, histopathological and microbiological findings that influence the lung weight of drowning. Then, we also tried to characterize drowning victims with low-weight lungs. Following exclusion criteria, of the 78 drowning cases finally selected, 18% exhibited a lung weight lower than 1000 g. Higher lung weights, mainly found in males, younger subjects and bodies found in seawater than freshwater, were correlated with external and tracheobronchial foam and alveolar macrophages. In males, pulmonary oedema was associated with higher lung weight but alveolar oedema was related to lower lung weight. Following exclusion criteria and microscopic findings (alveolar oedema and diatoms) we could identify a 3.8% of drowning cases with lungs weighting less than 1000 g. These results could help us to characterise cases of drowning as a consequence of a profound asphyxiation without liquid aspiration. Key Words: forensic science and pathology, drowning, lung weight, diatom. The diagnosis of drowning, as a respiratory impairment caused by the penetration of liquid media into the airways from submersion/ immersion, is one of the most difficult tasks for forensic pathologists [1]. These difficulties are related to the variety of fatal mechanisms involving drowning, such as hypoxemia, hypercapnia, hypothermia, laryngospasm and fluid and electrolyte disturbances [2]. Additionally, specific pathomorphological and microbiological findings (pleural effusion and surfactant) can usually help in the positive diagnosis of immersion death but often may be poor [2-3]. A controversy has surrounded the variable volumes of drowning media found in lungs during drowning because the absence of liquid cannot exclude the possibility of drowning. Therefore, autopsy findings of no significant aspiration of liquid in bodies found in water have led to the introduction of the concept of dry drowning [4]. Although a group of international experts at the World Congress on Drowning in Amsterdam [5] gathered to review the definition of drowning and elaborate a guide for the drowning process, recommendations for the significance of dry drowning were not addressed. Thus, there is currently little knowledge about the factors and findings that characterise lungs with low weight and the mechanisms involved, such as profound asphyxiation and acute cardiac death. Consequently, the incidence of drowning has been questioned, suggesting that bodies found in water with apparently normal lungs may conceal more cases of natural death than previously recognised [4, 6]. In the present study, we statistically analysed individual and circumstantial factors as well as macromorphological, histopathological and microbiological findings with regard to the lung weight 1) Universidad de Málaga, Departamento de Medicina Legal y Forense, Málaga, Spain * Corresponding author: MD, PhD, Universidad de Málaga, Departamento de Medicina Legal y Forense, Avenida Louis Pasteur 32, 29071, Tel: , Fax: , Málaga, Spain, lorubio@uma.es 2) Instituto de Medicina Legal, Servicio de Patología Forense, Málaga, Spain 89

2 Rubio L. et al Postmortem factors on drowning in drowning victims. A differential diagnosis of these multivariable analyses could help to characterize cases of drowning victims with low-weight lungs. MATERIALS AND METHODS Ethics statement The research procedures described in the present study were approved by the Research and Ethics Committee of the Institute of Legal Medicine of Malaga and were conducted according to the principles expressed in the Declaration of Helsinki (1983). Selection criteria A total of 200 deaths by drowning, autopsied and diagnosed in the Service of Forensic Pathology of the Institute of Legal Medicine of Malaga during , were retrospectively obtained from the database of the Department of Forensic and Legal Medicine of the University of Malaga (Spain). Forensic pathologists carried out multiple diagnostic criteria of exclusion used as guidelines in the current forensic practice, including autopsy and morphological and histopathologic findings (external and tracheobronchial foam, Paltauf ecchymosis, pulmonary oedema, aqueous emphysema, alveolar haemorrhage, rupture of alveolar septa, alveolar oedema, alveolar macrophages, etc.) and the diatom test in watertight organs such as bone marrow and liver as a good compromise in arriving at a conclusion of death by drowning (Principles and Rules relating to Medico-Legal Autopsy Procedures of the Council of Europe Committee of Ministers, Recommendation No. R(99)3; coe.int/t/dg3/healthbioethic/texts_and_documents/ RecR(99)3.pdf). Then, the following exclusion criteria were applied in order to obtain a normalized cohort of cases of drowning: victim under 18 years old, significant putrefactive changes, postmortem interval > 6 days and resuscitation attempts. Finally, we obtained 78 drowning victims to be analysed for the present study. Description and data analysis The 78 drowning cases were analysed using the Table 1. Lung weight in male and female victims of verified and nonverified drowning Nonverified Verified All Lung Weight N % N % N % < 1000 g g Total Male < 1000 g g Total Female < 1000 g g Total following relevant variables: 1) individual factors, including gender, age, salinity of water, postmortem interval, heart weight, coronary alteration, and concentration of alcohol (>0.3 g/l) or other drugs (benzodiazepines, opioids and cannabinoids) in blood and/or urine; 2) circumstantial factors, including drowning verification (witnessed or with a farewell message) and season and manner of death (aetiology); 3) postmortem macromorphological findings, including lung weight (combined weight of both lungs), external foam, frothy fluid in the conduction airway (tracheobronchial foam), pleural effusion and water in the stomach; 4) histopathological findings, including Paltauf ecchymosis, pulmonary oedema, aqueous emphysema, alveolar haemorrhage, rupture of alveolar septa, alveolar oedema, and alveolar macrophages; and 5) microbiological findings, including diatoms in closed organs, such as bone marrow and lungs. Statistical analysis Data were analysed using GraphPad Prism v.5 (GraphPad Software Inc., La Jolla, CA, USA). The results are described as percentage or the mean ± standard deviation (S.D.). All of the variables described above were statistically correlated to the lung weight. Thus, data obtained from the 78 drowning cases were analysed by both Pearson s correlation and multivariable logistic regression tests for linear dependence using lung weight as the dependent variable. P < 0.05 was considered to be significant. RESULTS From 2004 to 2011, 200 deaths by drowning occurred in the province of Malaga (Spain) in fresh water and seawater. After exclusion criteria were applied, 78 cases were included; 56 cases were males and 22 cases were females, with a mean age of 57.79±17.85 and 68.45±12.6 years, respectively. Most cases were unverified (70.5%) versus 29.5% of verified deaths by drowning (Table 1). When the manner of death was analysed, we found that accidental drowning death was the most common. When gender was analysed, suicide was Table 2. Lung weight in male and female victims of drowning, by type of water Seawater Freshwater All Lung weight N % N % N % < 1000 g g Total Male < 1000 g g Total Female < 1000 g g Total

3 Romanian Journal of Legal Medicine Vol. XXII, No 2(2014) identified for a majority of female victims (63.7%) in contrast to the 8 cases of suicide (14.3%) and 47 cases of accidents (83.9%) described in males. The percentage of victims who drowned in fresh water and seawater was 62.23% and 30.77%, respectively (Table 2). We observed that liquid penetration into the stomach occurred in 57.7% of victims, but only 10 of them had 500 ml. Additionally, 30.3% of males and 18.2% of females analysed presented a blood alcohol concentration higher than 0.3 g/l. Regarding postmorten interval, most victims were found before 24 hours postmortem (93.59%) (Table 3). Morphological findings indicated the presence of external foam in 49 victims (62.8%). Histopathological findings exhibited a high incidence of rupture of alveolar Table 3. Distribution of the postmortem interval in male and female victims of drowning Postmortem interval Ranges N % 12 hours MALE > 12 to 24 hours > 24 to 48 hours > 48 hours 0 0 Total hours FEMALE > 12 to 24 hours > 24 to 48 hours 0 0 > 48 hours 0 0 Total hours ALL > 12 to 24 hours > 24 to 48 hours > 48 hours 0 0 Total septa (95.6%) and alveolar oedema (88.02%). In contrast, we found a low number of victims with lungs containing alveolar macrophages (38.5%). All cases of drowning showed the presence of the same spread of diatom species in lungs and bone marrow compared to those found in the water. The majority of presumed victims of drowning (82%) showed a lung weight higher than 1000 g; lungs weighing less than 1000 g were found in 18% of the victims (Table 4). Lung weight differed significantly between genders (p < ) (Table 5, Fig. 1A-C). In total, 7.1% of the total male victims presented lung weights less than 1000 g in contrast to 45.5% of total female victims. The prominent difference in lung weight between genders Table 4. Distribution of lung weight in male and female victims of drowning Lung weight Ranges N % 750 to <1000 g Male 1000 to <1500 g to <2000 g g Total to <1000 g to <1500 g Female 1500 to <2000 g g 0 0 Total to <1000 g to <1500 g ALL 1500 to <2000 g g 6 8 Total Table 5. Correlation between the lung weight and multiple factors and postmortem findings in male and female victims of drowning Correlation with lung weight Total Male Female Pearson r P value* Pearson r P value* Pearson r P value* Gender (female vs. male) < Age (years old) Freshwater vs seawater Verified vs unverified Postmortem interval (<12 vs. 12) Heart weight (grams) Coronary alteration Water in stomach Foam -0, Tracheobronchial foam Paltauf ecchymosis Pleural effusion Pulmonary edema Aqueous emphysema Alveolar hemorrhage Breaking alveolar septa Alveolar edema Alveolar macrophages Alcohol in blood (>0.3 g/l) Other drugs Etiology of drowning * Based on Pearson-correlation r test. A P<0.05 was considered significant 91

4 Rubio L. et al Postmortem factors on drowning Figure 1. Correlation between lung weight and gender, age, water salinity, external foam or tracheobronchial foam in male and female victims of drowning. Statistical significance is described in Table 5. 92

5 Romanian Journal of Legal Medicine Vol. XXII, No 2(2014) led us to analyse men and women separately. We found a marked lung weight decreased in older subjects (p = ) (Table 5, Fig. 1D-F), with a significant increase in lung weight in younger male victims (p = 0.02), but this correlation was not detected in female victims (p = 0.075). Interestingly, drowning in seawater led to a higher lung weight than drowning in fresh water (p = , Table 5, Fig. 1G). Male victims who drowned in seawater showed significantly higher lung weights than drowning in freshwater (p = 0.046, Fig. 1H), but we did not find a correlation between the lung weight of female victims and salinity of the water (Fig. 1I). Drowning was verified in only one of the cases with lung weight < 1000 g but in 22 cases with lung weight 1000 g (p = 0.043). In contrast, we found that female victims with verified drowning showed higher lung weights than unverified drowning (p = 0.049), but there was no correlation in male victims (Table 5). The presence of external foam and tracheobronchial foam was significantly correlated with increased lung weight (p = and p = 0.027, respectively) (Table 5; Fig. 1J-L and M-O, respectively). Increased lung weight was significantly related with foam and tracheobronchial foam in male victims (p = and p = 0.001, respectively), but we did not observe statistical significance in female victims (p = 0.053). Lungs with alveolar macrophages weighed more than lungs without alveolar macrophages (p = 0.037, Table 5, Fig. 2A-C). Alveolar macrophages and pulmonary oedema were found in male victims with higher lung weights (p = and p = 0.009, respectively, Fig, 2B, E); in contrast, alveolar oedema was found to be associated with lower lung weight in male drowning victims (p = 0.043, Fig. 2K). Interestingly, alveolar haemorrhage in female victims was related with higher lung weight (Fig. 2I). However, differences in lung weight were not correlated with water in the stomach (p = 0.473), pleural effusion (p = 0.112) or broken alveolar septa (p = 0.716) (Table 5). The presence of alcohol in the blood of male victims tended to be related to higher lung weight but it was no significant (p = 0.066, Fig. 2N). Moreover, when lung weight was compared with the presence of others drugs in the blood and urine, such as benzodiazepines, opioids and cannabinoids, we detected higher lung weight related with toxicity in male victims (p = 0.028, Table 5, Fig. 2Q). In the present study, we also characterized the cases of drowning victims with low-weight lungs following a differential diagnosis. First, from 78 cases of drowning, we selected 14 lungs (18%) that weighed less than 1000 g. All these low-weight lungs had values of lung/body weight (L/B) ratio lower than 19.5 g/kg and values of lung/heart weight (L/H) ratio lower that 3.8 g/kg (Azparren et al., 2007). Second, we performed several exclusion criteria of the cases of low-weight lungs of persons found in water following 1) apparently normal lung characteristics, 2) no macroscopic evidence of penetration of liquid into their airways (external and tracheobronchial foam) and 3) no signs of violence. Following these criteria we could identify 3 cases (21.4%) of lungs weighing <1000 g (3.8% of total drowning cases). In these 3 cases, we could not find evidences of histopathological alterations in lungs, heart and other organs. For instance, they did not show pleural effusion, aqueous emphysema or alveolar haemorrhage and macrophages. In addition, no relevant cardiac microscopic pathology was really found that could explain the cause of the death by a heart disease. Moreover, the medical history of these victims didn t show relevant diseases. Interestingly, it should be noted that the microscopic analysis of these 3 cases with apparently normal lung characteristics revealed broken alveolar septa, alveolar oedema and the same spread of diatom species in lungs and peripheral organs, including bone marrow, compared to those diatoms found in the water. Thus, these drowning-related findings let us to suggest as drowning these 3 cases of apparently normal lung weighting less than 1000 g (3.8% of total drowning cases). DISCUSSION In the present study, we described individual and circumstantial factors associated with 78 deaths by drowning in the province of Malaga. Then, we analysed the influence of these factors and other macromorphological, histopathological and microbiological findings on lung weight. The key results indicated that higher lung weight was correlated with the verification of drowning deaths, bodies found in seawater and the presence of foam, tracheobronchial foam and alveolar macrophages in airways. A differential analysis of the macromorphological, histopathological and microbiological findings in the 78 deaths by drowning lets us to identify 3 cases of drowning with low-weight lungs. These victims with apparently normal lungs weighing less than 1000 g didn t show macroscopic signs of liquid penetration, macroscopic and microscopic evidences of natural death or signs of violence. But they specifically showed broken alveolar septa, alveolar oedema and the same spread of diatom species in lungs and bone marrow compared to those found in the water. Lung weight depended significantly on the gender and age of the drowning victims, with higher lung weight in males than females and in younger subjects. The contribution of these factors to the lung weight must be partially considered; it may be smaller than statistically obtained because the lung volume is directly related to the physical constitution of the victims, their vital pulmonary capacity and cardiac output in combination with the agonal period of time [7]. Another contributory factor to 93

6 Rubio L. et al Postmortem factors on drowning Figure 2. Correlation between lung weight and alveolar macrophages, pulmonary oedema, alveolar haemorrhage, alveolar oedema, and alcohol (>0.3 g/l) or other drugs (benzodiazepines, opioids and cannabinoids) in blood in male and female victims of drowning. Statistical significance is described in Table 5. 94

7 Romanian Journal of Legal Medicine Vol. XXII, No 2(2014) be considered is the amount of water aspirated relative to the survival time or vital activity. The difference in lung weight between fresh water and seawater observed in the present study suggests that fresh water may contribute to shortening the survival time during drowning [2, 7]. In an experimental model of rats, Locali and collaborators [8] found a higher amount of macrophages in rat lungs after drowning induction by salty water in comparison with fresh water. Although these authors did not observe higher lung weights in salty water-drowned rats, we can hypothesise a close relationship between increased lung weight, drowning in seawater and the presence of alveolar macrophages. Another study [9] showed a positive correlation between the lung weight and left cardiac serum levels of biochemical markers (calcium, sodium, chloride and magnesium) in saltwater drowning but an inverse correlation for pericardial fluid and serum chloride in fresh water drowning. In contrast to a previous study by Zhu et al. [7] that described a gradual postmortem interval-dependent decrease in lung weight, we did not find a correlation between these factors. This discrepancy could be related to the postmortem interval used because these authors used a postmortem interval between 4.5 hours and 8 days, while in the present study, we considered a postmortem interval up to 6 days. The presence of diatoms in the bone marrow and lung has been considered by many authors [10] as an additional sign of liquid penetration into the lungs and consequential drowning. In this study, diatoms were found in the lungs and bone marrow in all cases of drowning (including lungs weighing less than 1000 g), so it can be considered that there was at least some liquid penetration. Regarding other circumstantial factors, we confirmed that suicide and accidental deaths were predominant, and a large number of drowning cases (27%; > 0.3 g/l) were alcohol related [10-12]. Few studies have analysed drowning victims with no significant penetration of liquid in airways. Mechanisms that could explain this particular type of drowning are controversial; some authors have suggested lethal hypoxia caused by a reflex laryngospasm from liquid aspiration [13]. Additionally, the association of long QT syndrome with accidental drowning has been reported [14, 15]; therefore, we should not exclude death from heart disease, although we found evidence to confirm death by drowning. In this study, we did not consider an association of long QT syndrome with drowning deaths because it has a very low frequency [15]. However, the existence of no objective evidence suggest a completely absent or, at least, a very low incidence of drowning without aspiration [6]. A precise description of macromorphologic, histopathologic and microscopic findings in drowning lungs has been performed in the present study to clarify this issue. Several studies have established a correlation between lungs weighing less than 1000 g and dry lung [3, 16]. However, we suggest that other variables associated with low lung weight need to be considered to exclude signs of liquid penetration or natural death. We identified that 3.8% of drowning victims with apparently normal lungs weighing less than 1000 g did not show macroscopic signs of liquid penetration, such as external and tracheobronchial foam, signs of violence and macroscopic and microscopic signs that could conceal natural death in water. We also analysed the lung-heart weight ratio (L/H) (Zhu et al., 2003b) and lung-body weight ratio (L/B) [18], which were previously described as indexes that could indicate death by drowning. We observed that the indexes obtained in our 3 cases of low-weight lung were under the limits established for the identification of drowning cases with liquid penetration (< 3.8 g/kg for L/H and < 19.5 g/kg for L/B). These findings suggested that these 3 lungs did not have the typical characteristics of drowning lungs, evidence of violence nor signs of natural death. Therefore, it was necessary to consider additional histopathological and microbiological findings in the diagnosis of drowning such as broken alveolar septa, alveolar oedema and the same spread of diatom species in lungs and bone marrow compared to those obtained in the water where the body was found. There are disagreements between experts regarding the incidence of normal lungs in drowning deaths without liquid aspiration. Some authors have speculated that the percentage of drowning deaths without liquid aspiration could be 10-15%, while other authors have suggested that it must be much lower than believed [4, 6]. For example, Lunetta and collaborators [10] showed that this type of lung represents 1.4% of total cases analysed in their study. In conclusion, two results obtained in the present study are highlighted: 1) Higher lung weight was correlated with the verification of drowning deaths, bodies found in seawater and the presence of foam, tracheobronchial foam and alveolar macrophages into airways. These parameters, which were correlated with the lung weight in drowning victims, can help forensic pathologists to better understand the mechanisms involved in respiratory impairment with liquid penetration into the airways. 2) The exclusion criteria of no signs of liquid penetration, violence and natural death, and the finding of microscopic evidences, such as broken alveolar septa, alveolar oedema and the presence of diatoms in the lungs and bone marrow, are evidence that could help us to characterize cases of drowning without relevant liquid penetration in victims whose apparently normal lungs weighted less than 1000 g. 95

8 Rubio L. et al Postmortem factors on drowning Conflicts of interest. The authors state no conflicts of interest. Acknowledgment. The authors are indebted to Dr. Jose Luis Palomo (Head of the Service of Forensic Pathology) for kindly advising the present study. References 1. Piette MH, De Letter EA. Drowning: still a difficult autopsy diagnosis. Forensic Sci Int. 2006;163: Zhu BL, Ishida K, Quan L, et al. Pulmonary immunohistochemistry and serum levels of surfactant-associated protein A in fatal drowning. Leg Med. 2002;4: Morild I. Pleural effusion in drowning. Am J Forensic Med Pathol. 1995;16: Modell JH, Bellefleur M, Davis JH. Drowning without aspiration: is this an appropriate diagnosis? J Forensic Sci. 1999;44: Idris AH, Bert R, Bierens J, et al. Timerman Recommended guidelines for uniform reporting of data from drowning: the Ulstein Style. Circulation 2003;108: Orlowski JP, Spizlman D. Drowning: rescue, resuscitation and reanimation. Pediatr Clin North Am 2001;48: Zhu BL, Quan L, Li DR, et al. Postmortem lung weight in drownings: a comparison with acute asphyxiation and cardiac death. Leg Med. 2003;5: Locali RF, Almeida M, Oliveira-Júnior IS. Use of the histopathology in the differential diagnosis of drowning in fresh and salty water: an experimental model establishment in rats. Acta Cir Bras. 2006;21: Maeda H, Zhu BL, Ishikawa T, et al. Analysis of postmortem biochemical findings with regard to the lung weight in drowning. Leg Med. 2009;11:269S 272S. 10. Lunetta P, Modell JH, Sajantila A. What is the incidence and significance of "Dry-Lungs" in bodies found in water? Am J Forensic Med Pathol. 2004;25: Lunetta P, Smith GS, Pentillä A, et al. Unintentional drowning in Finland : a population-based study. Int J Epidemiol. 2004;33: Papadodima SA, Athanaselis SA, Skliros E, et al. Forensic investigation of submersion deaths. Int J Clin Pract. 2010;64: Kinght B. Immersion deaths. In: Knight B, editors. Forensic Pathology. London: Edward Arnolds; Ackerman MJ, Tester DJ, Porter CJ, et al. Molecular diagnosis of the inherited long-qt syndrome in a women who died after neardrowning. N Engl J Med. 1999;341: Lunetta P, Levo A, Männikkö A, et al. Death in bathtub revised whith molecular genetics: a victim with suicidal traits and a LQTS gene mutation. Forensic Sci Int. 2002;130: Kringsholm B, Filskov A, Kock K. Autopsied cases drowning in Denmark Forensic Sci Int. 1991;52: Zhu BL, Quan L, Ishida K, et al. Lung-heart weight ratio as a possible index of cardiopulmonary pathophysiology in drowning. Leg Med. 2003;5: Azparren JE, Cubero C, Perucha E, et al. Comparison between lung weight and blood strontium in bodies found in seawater. Forensic Sci Int. 2007;168: