Despite major advances in imaging technologies, in the settings of both prenatal diagnosis and neonatal medicine, the final gold standard for the identification of underlying pathophysiological mechanisms and detection of associated abnormalities continues to remain the autopsy in many scenarios. Numerous studies have now reported the clinical usefulness of perinatal postmortem examination with highly consistent results—namely that in around 30% of cases, clinically significant additional information is obtained at autopsy that was not available during life1–3, with the likelihood of gaining such information being related to the specialist status of the pathologist and other factors such as the length of time between death and autopsy4.
However, in practice, offering an optimal perinatal postmortem service may not be straightforward, since this may involve both transfer of the body to a specialist center and delay due to the need for specialist referral. Furthermore, the superimposed effects of postmortem autolysis and decomposition, in association with the inherent fragility of the fetal brain requiring adequate formalin fixation prior to examination, may result in a significant number of cases submitted for perinatal autopsy being associated with suboptimal examination, particularly of the central nervous system—around 20% in one series5. It is now more than a decade since several studies suggested the possible usefulness of postmortem magnetic resonance imaging (MRI) in this setting, but although several case reports and small series have described specific anomalies using this technique, it has failed to gain widespread acceptance6–10.
However, in recent years in the UK, several events have led to a renewed interest in pursuing the development of the ‘minimally invasive autopsy’. While it is recognized that a full postmortem examination is likely to provide the greatest likelihood of determining both the cause or mechanism of death and all specific abnormalities present, some parents and doctors have been increasingly reluctant to consider the possibility of autopsy examination, following the widespread publicity regarding fetal and infant organs being historically retained in some hospitals11. Public pressure in the UK has therefore led to the introduction of new legislation regarding all aspects of the examination and use of human tissue, the Human Tissue Act12, and, in addition, a revision of Her Majesty's Coronial system13, which investigates all deaths of unknown cause and a range of other deaths including some perinatal cases. These changes have heightened awareness of the issues surrounding removal and retention of tissue and increased recognition that some religious groups may wish for a rapid funeral, thus driving the need to critically examine both the role of postmortem examination and how it is carried out.
Since a significant component of a traditional autopsy involves macroscopic examination of internal organs, use of a radiological technique that can provide good soft-tissue imaging, such as a MRI, is a good potential candidate to augment the traditional autopsy or modify its practice. In terms of achieving excellent tissue visualization, there are several advantages to postmortem imaging compared to such imaging in life. First, and perhaps most important, there is no movement in the region of interest (ROI) and therefore no motion artifacts and no limits on acquisition time. Secondly, since the ROI in perinatal cases is small, the receiver coil can be placed close to the area to be imaged with corresponding increased resolution. Finally, the ROI can be placed in the optimum position and there is no concern regarding exposure dosage. Using appropriately modified acquisition protocols to optimize visualization in the postmortem setting, excellent tissue detail can be obtained of many structures (Figure 1), and several studies have now suggested the superiority of postmortem MRI examination over traditional autopsy for confirmation and visualization of some central nervous system lesions5, 14. An increasingly important additional advantage of the use of such cross-sectional imaging modalities is the acquisition of a large raw dataset, which can subsequently be manipulated and reanalyzed in the light of future clinical information, or by another specialist as appropriate.
Although MRI can provide excellent soft tissue visualization (Figure 1), it remains suboptimal for the assessment of bony structures, whereas computerized tomography (CT) scanning can provide excellent visualization of bone, is quick to perform, and also similarly allows the aquisition of data for further three-dimensional (3D) reconstruction as required. Early experience suggests that excellent anatomical visualization can be obtained by a combination of postmortem MRI and CT scanning, and with increasing experience in these techniques it is further likely that additional sequences or techniques, such as diffusion weighted MRI or MR spectroscopy, may allow additional relevant information to be obtained from postmortem imaging studies.
With these potential advantages in mind, it is highly likely that such postmortem imaging techniques will rapidly become standard practice in centers of excellence. However, it should be noted that no data thus far exist from large-scale studies regarding the accuracy of anatomical or pathological diagnosis based on postmortem imaging techniques versus traditional autopsy examination, although such studies are currently in progress. The major disadvantage of this ‘minimally invasive’ approach, however, is the lack of availability of tissue for subsequent microscopic histopathological examination or other ancillary techniques, which often provide the most important clinical information in these cases. It is possible that image-guided needle tissue biopsy could be performed during postmortem imaging to allow collection of tissue samples, since needle-core biopsy is being increasingly used for the primary investigation of a large range of pediatric conditions in life with sufficient diagnostic material obtained15, but the adequacy of such sampling in the postmortem setting has not yet been established, in particular regarding the potential issues of sample representativeness.
Like all new techniques, as a consequence of the issues of both optimizing sequences for data acquisition, and interpretation of findings in the postmortem setting compared to during life, there will be an initially steep learning curve for those involved in the development of this service. Appropriate interpretation will in the first instance require the use of these imaging techniques to be validated by the currently accepted gold standard of the traditional autopsy, but at present it remains uncertain as to which medical professionals should take on this role. The fetal medicine specialist is experienced in examining and interpreting two-dimensional (2D), and more recently 3D, images and has good knowledge of the major malformations which may be present, but little experience of postmortem changes and other organ-specific pathology. The pediatric pathologist is a specialist in the examination of fetuses and infants and in particular the correct interpretation of additional investigations and histopathological sampling in the postmortem setting, but the majority of pathologists have no training or expertise in the interpretation of diagnostic imaging modalities. Finally, radiologists with an interest in MRI are the best equipped to optimize the technical aspects of image acquisition, image interpretation and manipulation, but few are trained in fetal medicine or pathology that may affect the fetus and infant. Only time will tell in what form this subspecialty will evolve, but it is highly likely that during the next decade, adjunctive postmortem MRI will become the standard of care in specialist perinatal pathology departments.
In summary, although there are significant potential advantages to the use of postmortem MRI/CT imaging, because of their associated inherent limitations it is highly unlikely that such postmortem imaging will entirely replace traditional autopsy examination in the foreseeable future. However, it is almost certain that the use of such techniques will rapidly become an integral part of postmortem examination of the fetus and infant, and may significantly dictate the type and extent of further examination required for a given case. In addition, such ‘minimally invasive’ autopsy examination may provide at least some further clinically relevant additional information for those parents who decline traditional autopsy. However, it is imperative that appropriate interpretation of imaging findings in the postmortem setting are validated before there is widespread adoption of this modality.