ARRS 2022 Abstracts


1381. High NPV of Postmortem Pulmonary Computed Tomography Findings in a Medical Examiner’s Office During the SARS-CoV-2 Pandemic
Authors * Denotes Presenting Author
  1. Jamie Elifritz *; New Mexico Office of the Medical Investigator ; University of New Mexico Hospital
  2. Natalie Adolphi; New Mexico Office of the Medical Investigator
  3. Sarah Lathrop; New Mexico Office of the Medical Investigator
  4. Jennifer Weaver; University of New Mexico Hospital
  5. Lauren Decker; New Mexico Office of the Medical Investigator
  6. Jennifer Febbo; University of New Mexico Hospital
  7. Jonathan Revels; University of New Mexico Hospital
This study aims to determine if Postmortem Computed Tomography (PMCT) pulmonary findings can be used as a triage tool for SARS-CoV-2 infection in a Medical Investigator's Office.

Materials and Methods:
After extensive literature review, a scoring system was developed as a tool to screen decedents for the likelihood of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection based on pulmonary findings at PMCT. Our scoring system was developed after review of the Coronavirus Disease 2019 Reporting and Data System (CO-RADS) scheme with careful attention to the expected findings of postmortem change. The scoring system placed each study interpretation into 1 of 4 groups: 0, cannot assess for COVID-19; 1, low suspicion for COVID-19; 2, intermediate suspicion for COVID-19; and 3, high suspicion for COVID-19. A selection of cases from 3 groups of decedents was randomly presented for interpretation: 1, all RT-PCR COVID positive decedents; 2, decedents with Cause of Death related to viral pulmonary infection prior to 11/2019; and 3, random group of decedents with cause of death not related to viral pulmonary infection prior to 11/2019. Radiologists were blinded to the case groups and read the cases in a random order. Only the chest portion of the cases were reviewed using the axial chest set of images on lung windows. Each case was assigned a COVID-19 score, as above. The case scores were submitted to REDCap. Data were entered into an Excel spreadsheet and analyzed using SAS (Statistical Analysis Software) version 9.2. We calculated percent agreement among the 3 radiologists, as well as positive and negative predictive values based on the known disease status of each case. We also calculated percent agreement for presence of two co-morbidities, and the presence of any significant associations between the presence of co-morbidities and disease status using chi-square tests and a significance level of 0.05.

There were 72 cases included in our study. Our stufy found that when CO-RADS 0 cases were included, the PPV of pulmonary CT findings for SARS-CoV-2 infection was 36.3% and the NPV was 80%. When the CO-RADS 0 cases were excluded, the PPV was 36.4% and the NPV was 100%.

We conclude that PMCT evaluation of the lungs is a powerful tool for triaging decadents for SARS-CoV-2 infection prior to autopsy, especially when CO-RADS 0 cases are excluded. Additionally, we documented how PMCT triage can be used with a true control cohort (cases prior to 11/2019; prior to the pandemic). This information will likely prove to be helpful as quarantine, masking and social distancing measures change over time and previously common pathogens that have been low in prevalence during the pandemic begin to reemerge. Most of the literature examines pulmonary CT findings during the pandemic when SARS-CoV-2 infection rates were high and there was a relative paucity of other pulmonary viral infections for comparison. Our study, to our knowledge, is unique given that we studied control groups that were imaged prior to the pandemic in a postmortem setting.