Three-dimensional cinematic rendering of fetal skeletal dysplasia using postmortem computed tomography.

Abstract

As parental consent for invasive perinatal autopsy is increasingly declined, postmortem cross-sectional imaging techniques have gained popularity as tools for confirming antenatal findings and highlighting new abnormalities. In addition to being a non-invasive method, one of the major benefits of imaging is that it allows acquisition of three-dimensional (3D) volumetric datasets which can be postprocessed to produce a variety of imaging reconstructions, such as the traditional volume-rendering (VR) technique. Recently, a novel software termed ‘cinematic rendering’ (CR) (syngo.via, Siemens Healthineers, Forchheim, Germany) was developed and approved by the US Food and Drug Administration for clinical use, which provides a more ‘photorealistic’ and detailed imaging reconstruction. Whilst CR technology bears several similarities to VR, it utilizes a more complex light modeling algorithm incorporating information from multiple light paths and predicted photon-scattering patterns1, 2. Several recent publications have demonstrated the superior realism and expressiveness of CR over VR techniques in musculoskeletal3-5 and vascular diseases, including in the case of a pregnant patient with Loeys–Dietz syndrome6. Given that this technique has been shown to perform best when there are greatest contrast differences between tissues, we applied the software to two postmortem computed tomography datasets which were acquired as part of the perinatal death investigation in two fetuses with lethal skeletal dysplasia. The first case was of a 22-week female fetus that underwent termination of pregnancy due to a prenatal diagnosis of suspected skeletal dysplasia (Figure 1). Radiographically, the diagnosis was in keeping with thanatophoric dysplasia. Both the VR and CR reconstructions demonstrated key characteristics of shortened, curved long bones, curved femora with trident acetabula and narrow thoracic cavity. However, the bone reconstruction using the CR method appeared less pixelated and the paler soft-tissue CR overlay (Figure 1c) appeared more realistic and comparable to fetal skin than the red VR overlay (Figure 1a). The second case was of a stillborn 39-week male fetus with osteogenesis imperfecta (OI) Type II (Figure 2). The mother had not attended any antenatal appointment and the diagnosis was made at postmortem imaging. Both the CR and VR techniques demonstrated the classical features of OI with numerous fractured bones with callus formation leading to bowed, shortened limbs and ‘beaded’ appearance of the ribs. The fetus was also noted to have a left inguinal hernia containing small bowel loops. Again, the CR images provided more photorealistic detail of the internal organs and skeletal findings, with less pixelation compared to the VR reconstructions. In conclusion, CR appears to represent a promising method for providing more realistic 3D reconstructions of fetal postmortem imaging for skeletal dysplasia over traditional VR techniques, without any loss of detail relating to the underlying pathology. It can produce ‘sanitized’ images of the fetus, which could be helpful in communicating the findings to the parents, or as an educational tool for training healthcare professionals7. Further work comparing formally the acceptability and benefit of CR techniques vs those of standard two-dimensional images and 3D printed models is required. S.C.S. is supported by a UK Research and Innovation Fellowship and Medical Research Council (MRC) Clinical Research Training Fellowship (Grant Ref: MR/R002118/1). This award is jointly funded by the Royal College of Radiologists (RCR). O.J.A. is funded by a National Institute for Health Research (NIHR) Career Development Fellowship (NIHR-CDF-2017-10-037). The authors receive funding from the Great Ormond Street Children's Charity and the Great Ormond Street Hospital NIHR Biomedical Research Centre. This article presents independent research funded by the MRC, RCR, NIHR and the views expressed are those of the author(s) and not necessarily those of the NHS, MRC, RCR, the NIHR or the Department of Health. The data that support the findings of this study are available from the corresponding author upon reasonable request.