RETURN TO ABSTRACT LISTING


Objective:
In the setting of international shortages in the forensic pathology workforce, postmortem CT (PMCT) is increasingly used in forensic settings in cause-of-death investigation. With the onset of the COVID-19 pandemic, concern for high transmissibility, morbidity, and mortality led to an abrupt halt in the performance of autopsies at many institutions around the world. Although this led to a loss of valuable pathology and pathophysiology data, the pandemic did reveal an opportunity to apply imaging to cause-of-death investigation in the setting of infectious disease and, by extension, hospital medicine.

Materials and Methods:
In an academic teaching hospital, full-body PMCT was conducted within 48 hours of death for 19 decedents who were consented for unrestricted autopsy and had tested positive for COVID-19. PMCTs were interpreted by fellowship-trained radiologists using a standardized form. PMCT findings were compared to pathology findings of traditional hospital autopsies, antemortem imaging studies, and clinical data from the medical record.

Results:
Ten male and 9 female decedents with either a positive postmortem COVID-19 nasal swab (n = 18) or COVID-19 test during their hospitalization (n = 1) were scanned (age range 34-94 years, average 66 years). Autopsy-determined causes of death were categorized as ‘cardiovascular disease’ (CVD, 4/19, 21%), ‘COVID’ (8/19, 42%), combination ‘CVD/COVID’ (5/19, 26%), and ‘other’ (non-COVID, non-CVD, 2/19, 11%). At least one antemortem CT study from the visit/admission leading up to death was available for 11 decedents, including head CT (6/19), chest CT (9/19), and abdomen/pelvis CT (3/19), all performed 1-14 days prior to scanning (average 7 days). Of these, 5 PMCTs demonstrated worsened findings (primarily worsened lung disease) and 3 PMCTs demonstrated pertinent new findings compared to most recent antemortem CT. For the three cases with new findings, PMCTs demonstrated massive pneumomediastinum (1/3), a known necrotic pancreatic collection with a new adjacent layering hematoma and diffuse air emboli (1/3), and strokes that were not apparent on antemortem head CT on the day of death (1/3).

Conclusion:
While not routinely practiced at most centers, the hospital setting provides a unique atmosphere allowing for rapid postmortem imaging evaluation prior to the onset of significant decomposition, and correlation with autopsy, histopathology, and other data not otherwise accessible in the forensic setting, such as the clinical record and antemortem imaging studies. As in the forensic setting, postmortem imaging in the hospital setting can supplement and complement autopsy, providing valuable information to pathologists and identifying findings that can otherwise be extremely difficult to detect/characterize via traditional autopsy (i.e., pathology related to the presence of air). It can help guide pathologic evaluation and even suggest the acuity of findings in cases where antemortem imaging studies are available; or it can help pathologists make sense of unusual or unexpected findings.