Review articleDiagnostic accuracy of postmortem imaging vs autopsy—A systematic review
Introduction
Half a century ago the rate of clinical autopsies in Europe and the USA was around 60%, but since then, clinical autopsy rates have drastically declined, and is today less than 10% (e.g. [1], [2], [3]).
The autopsy has long been regarded as the “gold standard” for retrospective quality assessment of clinical diagnoses [4]. Studies comparing clinical diagnoses and autopsy findings have revealed major discrepancies in ¼ or more of deceased patients undergoing autopsy [5], [6], [7]. Low autopsy rates may conceal medical malpractice, thereby preventing an important quality assurance indicator in health care [8]. Further, low autopsy rates will decrease the reliability of the cause-of-death statistics, in turn decreasing the usefulness of this statistics for health care planning and research. In the individual death, the family may be provided with wrongful or insufficient information regarding the cause of death, the underlying disease, etc. Hence, the decrease in clinical autopsy rate has negative consequences for the family, future patients, health care, and the society as a whole.
The reasons for this decline are manifold, and include the non-reimbursement of autopsies, clinician’s fear of medico-legal problems, reluctance of the family and/or of health care personnel, more deaths occurring at long-term care facilities, shortage of clinical pathologists, judicial principles, inadequate and delayed communication of autopsy results to clinicians, the requesting of autopsies being delegated to junior medical staff, organ retention issues, and advances in laboratory diagnostic technology – wrongfully – result in the belief among clinicians that the autopsy has become redundant.
If alternatives such as postmortem imaging techniques should turn out to be reliable in a situation where the autopsy rate is (too) low, these techniques could contribute to a more reliable cause-of-death diagnostic process. Particular virtues of postmortem imaging are the following; non-invasive, more rapid diagnostic process, and higher availability of radiologists than of clinical pathologists – further increased by telemedicine techniques. Imaging techniques also enables permanent documentation and meticulous reexamination, and would also enable increased quality assurance in health care by analyses of cases which currently are not autopsied at all.
The diagnostic accuracy of the postmortem imaging techniques is, however, not yet clarified, and as yet there is no evidence they are able to adequately replace the conventional autopsy. Hence, it is of importance to evaluate the scientific basis of postmortem imaging as a substitute for clinical autopsy, and as an adjunct for additional information to an autopsy.
The main objectives of this systematic review were to determine (i) the diagnostic accuracy of post-mortem imaging for different findings in populations which otherwise would be subjected to a (clinical) autopsy or no autopsy at all, and (ii) what additional information imaging can provide to a clinical and a medico-legal autopsy in different populations.
Section snippets
Protocol and registration
This systematic review was conducted at The Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU) and published in April 2015 as a report in Swedish [www.sbu.se/201501]. SBU uses a peer-reviewed protocol including pre-specified objectives in accordance with standards in HTA assessments (http://www.sbu.se/upload/ebm/metodbok/SBUsHandbok.pdf). The protocol is available from SBU upon request.
Eligibility criteria
The criteria for eligibility were:
- •
Population: Fetuses, children and
Study selection
The literature search generated 2600 records, and of these 340 original papers were of potential relevance and read in full text. 269 of these were excluded (178 other research question, 87 not original data, 4 duplicates), and the remaining 71 were assessed for risk of bias. No report was deemed to have low risk of bias, 22 to have moderate risk, and 49 to have a high risk (Fig. 1).
Study characteristics, risk of bias within studies
The 22 included studies with moderate risk of bias [10], [11], [12], [13], [14], [15], [16], [17], [18], [19],
Summary and evidence
With the advantage of being rapid and non-invasive [32] and documenting the exact in situ state of body organs, postmortem imaging techniques such as computed tomography (PMCT) and magnetic resonance imaging (PMMR) have become increasingly used in forensic medicine during the last decade [33], [34].
This systematic literature review reveals, however, that the current scientific literature does not allow us to determine the diagnostic accuracy of postmortem imaging, whether used in conjunction
Conflicts of interest
No author declares any conflict of interest.
Funding
The project was conducted at by the Swedish Agency for Health Technology Assessment and Assessment of Social Services.
Acknowledgements
We gratefully acknowledge the administrative and secretarial support by Ms Anna Attergren Granath, and the literature searches by documentation specialist Ms Hanna Olofsson. Valuable comments were provided by professor Niels Lynnerup and professor Jytte Banner, Copenhagen, professor emeritus Christer Sundström, Uppsala, and associate professor Carl Winskog, Adelaide.
References (37)
Diagnostic errors in medicine: a case of neglect
Jt. Comm. J. Qual. Patient. Saf.
(2005)- et al.
Diagnostic accuracy of post mortem MRI for abdominal abnormalities in foetuses and children
Eur. J. Radiol.
(2015) - et al.
Comparison of whole-body post mortem 3D CT and autopsy evaluation in accidental blunt force traumatic death using the abbreviated injury scale classification
Forensic Sci. Int.
(2013) - et al.
Interobserver agreement of the injury diagnoses obtained by postmortem computed tomography of traffic fatality victims and a comparison with autopsy results
Forensic Sci. Int.
(2013) - et al.
Brain volume estimation from post-mortem newborn and fetal MRI
NeuroImage Clin.
(2014) - et al.
Detection of coronary thrombosis after multi-phase postmortem CT-angiography
Leg. Med.
(2013) - et al.
New evidence for old lore − urinary bladder distension on post-mortem computed tomography is related to intoxication
Forensic Sci. Int.
(2013) - et al.
Estimation of heart weight by post-mortem cardiac magnetic resonance imaging
J. Forensic Radiol. Imaging.
(2013) - et al.
Rib fractures at postmortem computed tomography (PMCT) validated against the autopsy
Forensic Sci. Int.
(2013) Special issue on postmortem imaging 2013
Forensic Sci. Int.
(2013)
Gunshot injuries detected by post-mortem multislice computed tomography (MSCT): a feasibility study
Legal Med.
The vanishing nonforensic autopsy
N. Engl. J. Med.
Declining autopsy rate in a French hospital: physician’s attitudes to the autopsy and use of autopsy material in research publications
Arch. Pathol. Lab. Med.
Discrepancies between clinical and autopsy diagnosis and the value of post mortem histology; a meta-analysis and review
Histopathology
Changes in rates of autopsy-detected diagnostic errors over time: a systematic review
JAMA
Who is at risk for diagnostic discrepancies? Comparison of pre- and postmortal diagnoses in 1800 patients of 3 medical decades in East and West Berlin
PLoS One
The value of autopsies in the era of high-tech medicine: discrepant findings persist
J. Clin. Pathol.
Investigation of medical intervention with fatal outcome: the impact of post-mortem CT and CT angiography
Radiol. Med.
Cited by (35)
Autopsy Method
2022, Encyclopedia of Forensic Sciences: Volume 1-4, Third EditionA review on the evolution and characteristics of post-mortem imaging techniques
2020, Forensic ImagingCitation Excerpt :Finally, experts should also be cautious when considering the information gathered within the current literature, since the scientific publications available on virtual autopsy are mainly based on a small volume of cases studied, the minimum number of cases is from 1 to 3 [88], in which a comparative analysis of MSCT, MRI and traditional autopsy was performed; and the maximum number in some cases being 182 [20]. Furthermore, previous systematic reviews have shown that there is a significant variability in the design of past studies, with differences not only in the sample size but also on the experimental design, the imaging process, the definition of the standard method(s), the terminology and the blinding [54,83-85]. It is agreed among experts that current studies are insufficient and further development and testing, as well as proper standardisation of the experimental protocols, is still required, as pointed by Eriksson et al and other authors [54,83-86,89].
The Impact of COVID-19 on Criminal Investigations and Proceedings in Sweden – A Snapshot of Practitioners’ Realities
2020, Forensic Science International: SynergyCitation Excerpt :A similarity between what the Swedish pathologists report and what is reported from Italy, is that full autopsies are more rarely conducted (for known COVID-19 cases in Sweden) and this, potentially, entails a risk in relation to the quality of the pathologists’ work, for example because important information may be overlooked with more limited examinations. In the long run, it is possible that new methods will have to be developed to enable full autopsies and the potential of training as well as technology such as Virtual Autopsy Tables [43,44] and potentially robotic autopsies, inspired by robotic surgeries [45,46] may need to be evaluated by experts in the field. Another partial similarity between Sweden and other countries is found in the Courts, since Swedish as well as American and Canadian judges report an increased usage of video technology.
Spleen measurements with reference to cause of death and spleen weight estimation: A study on postmortem computed tomography
2019, Journal of Forensic Radiology and ImagingCitation Excerpt :Its size, weight and macroscopic features, like congestion, pallor and the appearance of the splenic capsule, are usually of forensic relevance, as acute organ responses or chronic diseases can be reflected in the spleen's macroscopic appearance and may contribute to determination of cause of death [1]. Postmortem computed tomography (PMCT) is an invaluable supplementary tool to traditional autopsy [2,3]. Organ radiodensity can be measured in Hounsfield Units (HU) [4] on PMCT and routine measurements can be conducted giving organ-specific information [5].
Role of PMCT as a triage tool between external inspection and full autopsy – Case series and review
2018, Journal of Forensic Radiology and ImagingCitation Excerpt :A summary and literature review pertaining to strategic decisions with regard to autopsy is presented on Table 1. Postmortem imaging for medicolegal triage based on PMCT is a promising decision making tool [1,12,58–66]. As that, it can be analyzed within that context with the goal to identify both adequate use and relevant research subjects to improve evidence-based decision making.