Abstract
Purpose
To investigate the interest of post-mortem ultrasonography in the diagnosis of pathological background, and manner and cause of death.
Methods
Post-mortem ultrasonography exams were carried out on 38 fresh human adult cadavers referred to the Department of Forensic Medicine and Pathology (Garches, France). Data obtained from ultrasonography were independently compared with further forensic autopsy findings.
Results
Two important limitations relative to ultrasound utilization appeared: hyper-echoic abdominal and thoracic walls, with gas distension of the whole digestive tube and subcutaneous tissues (due to precocious putrefactive gas releasing); and difficulty in accessing lateral and posterior structures (i.e. liver, spleen, kidneys, lung bases, aorta) due to rigor mortis and evident non-compliance of the subject. Post-mortem diagnoses (moderate ascites, gallbladder stones, bladder globe, chronic kidney disease, cirrhosis, thyroid gland cysts and hypertrophy, intrauterine device), were strongly limited. False negative diagnoses comprised fatty liver, pleural effusion, thoracic aortic dissection, and focal organ and/or soft tissues lesions (for example, wounds or infarcts).
Conclusion
According to the results, post-mortem ultrasonography seems to have a very limited role for forensic purposes. Other post-mortem utilizations are cited, proposed, and discussed.
Introduction
Post-mortem imaging examinations are increasingly used in forensic medicine as a preliminary element of autopsy, especially multidetector computed tomography (MDCT) or conventional radiology, as a complement or substitute to complete body opening. 1 Since the 1950s, ultrasonography has been the classic radiological exam commonly used for living individuals, 2 but nothing has yet been described about the usefulness of post-mortem ultrasonography for forensic purposes (i.e. diagnosis of circumstances and/or causes of death) as a complement (preliminary exam) or a full substitution to a classical and complete autopsy. 3
Material and methods
A total of 38 fresh cadavers (period between death and examination being less than 48 hours) were examined under ultrasonography with an ACUSON 128XP/10 SIEMENS® by a well-trained radiologist. All individuals (28 males and 10 females) were adults (from 15 to 81 years old). Males were younger than females, with median and interquartile range of ages 46.5 [40–58.5] and 60.5 [54–70], respectively. The origins of the bodies were forensic cases (n = 37), and hospital population (n = 1). All investigations took place within 6 months. Data obtained from ultrasonography were compared with further autopsy findings. We compared the frequency of the number of diagnosis obtained from ultrasonography, with those from a classical and complete autopsy, using a non-parametric Wilcoxon sign test on the differences. To estimate the performance in identifying lethal lesions of the ultrasonography, positive and negative predictive value (PPV, NPV) were calculated using classical autopsy as the reference. All procedures were independently carried out.
According to the post-mortem character of this research (data analysis and comparison, no primary human experimentation or research), and as all data were anonymously indexed and compared, no previous consent (written or verbal) was necessary according to the highest standards of local French ethics laws and principles.
Results
Frequencies of the number of diagnosis related to diagnosis methods are presented in Figure 1. The median of diagnosis findings was higher with classical autopsy (median = 4) than with ultrasonography (median = 0). The difference between diagnosis findings with classical autopsy and ultrasonography was statistically significant (Wilcoxon sign test, p < 0.0001). These results were not modified when considering males and females separately.
Frequencies of the number of diagnoses related to the examination technique utilised (autopsy or post-mortem ultrasonography).
Among the 31 lesions identified by ultrasonography, all were non-lethal leading to a PPV equal to zero. In addition, classical autopsy results showed that these 31 non-lethal lesions correspond to true non-lethal lesions (NPV = 1). These findings suggest that ultrasonography has a very low sensitivity in identifying lethal lesions (but it could eventually be useful in the rare case where classical autopsy identified 2 or more lethal lesions).
List of all 38 studied cases with indications of sex, age, manner of death, autopsy and ultrasonography findings.
Post-mortem diagnoses were possible, but strongly limited due to the excessive gas artifacts related to subcutaneous and/or visceral putrefaction. Indeed, precocious gas release within vascular structures may occur as soon as the 5th hour following death,4,5 not only in deep structures close to the intestinal system, but also in subcutaneous vessels (for example superficial thyroid veins: Figure 2).
Superficial thyroid vein with early putrefaction gas (autopsy performed 7 hours after death).
The following were observed by ultrasonography: moderate ascitis, gallbladder stones, bladder globe and strong urinary retention (Figure 3(a)), chronic kidney disease, cirrhosis and steatosis, focal liver lesion (Figure 3(b)), thyroid gland nodule (Figure 3(c)) and hypertrophy, and intrauterine device.
(a) Sonogram of a bladder full of urine; (b) sonogram of the liver showing multiple gas bubble due to mixed vascular and visceral early putrefaction; (c) sonogram of the thyroid with an hypoechoic nodule of the right lobe; (d) sonogram of the liver with the gallbladder and artifacts caused by early vascular and visceral putrefaction gas bubble.
False negative diagnoses (i.e. lethal and non-lethal lesions visible at the moment of the autopsy, but missed at the ultrasonography) comprised liver steatosis, pleural effusion, thoracic aortic dissection, and focal organ and/or soft tissues lesions (for example wounds or infarcts).
Indeed, two important limitations relative to ultrasound utilization appeared:
Hyper-echoic abdominal and thoracic walls, with gas distension of the whole digestive tube and subcutaneous tissues (due to precocious putrefactive gas releasing: Figure 3(d)); Difficulty in accessing lateral and posterior structures (i.e. liver, spleen, kidneys, lung bases, aorta) due to rigor mortis and evident non-compliance of the subject.
Due to strong technical differences, such artifacts do not exist after multi-slice computed tomography (MSCT) and/or magnetic resonance imaging (MRI) examination of cadavers, and these last still conserve a diagnostic superiority relative to ultrasonography as previously shown by anatomo-radiological confrontations.6–9
Conclusion
In conclusion, except for occasional evident diagnosis such as a massive pericardial effusion 10 (although the inability of ultrasonography to detect slight pericardial effusions in 5 of our cases), or experimental evaluation of the lung density at the time of autopsy, 11 post-mortem ultrasonography seems to have a very limited role for forensic purposes. However, ultrasonography has been proposed as an ideal tool for the evaluation of skin thickness in the context of face reconstruction processes.12–15
One classical forensic utility of such a technique could be image-guided biopsy, as part of minimal invasive autopsy, 16 as recently demonstrated in the context of pediatric applications. 17 Further analyses are necessary to evaluate this potential role.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Declaration of conflicting interests
The authors declare that they do not have any conflict of interest.
Ethical approval
According to the post-mortem character of this research (data analysis and comparison, no primary human experimentation or research), and as all data were anonymously indexed and compared, no previous consent (written or verbal) was necessary according to the highest standards of local French ethics laws and principles.
Guarantor of the work
Philippe Charlier, MD, PhD, is the guarantor for the work.
Contributorship statement
P.C. wrote the entire manuscript, and carried out all autopsies; P.F.C. performed the ultrasonography; I.H.C. did the correlation through ultrasonography and autopsy data; all other authors did significant modifications within the manuscript during the editorial phases.