Postmortem Examination of Patients With COVID-19

Approximately 15% of individuals affected by coronavirus disease 2019 (COVID-19) develop severe disease, and 5% to 6% are critically ill (respiratory failure and/or multiple organ dysfunction or failure).^1,2 Severely ill and critically ill patients have a high mortality rate, especially with older age and coexisting medical conditions. Because there are still insufficient data on cause of death, we describe postmortem examinations in a case series of patients with COVID-19.

Between April 4 and April 19, 2020, we conducted serial postmortem examinations in patients with proven severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection who died at the University Medical Center Augsburg (Germany). Autopsies were conducted according to published best practice.³ Specimens from lung, heart, liver, spleen, kidney, brain, pleural effusion, and cerebrospinal fluid (CSF) were assessed. Postmortem nasopharyngeal, tracheal, bronchial swabs, pleural effusion, and CSF were tested for SARS-CoV-2 by reverse transcriptase–polymerase chain reaction. This study was approved by the local institutional review board, and written informed consent was obtained from next of kin.

Of 12 consecutive patients with COVID-19 who died, postmortem examinations were conducted in 10. The median age was 79 years (range, 64-90 years); 7 patients were male. All cases tested positive for SARS-CoV-2 by nasopharyngeal swab at time of hospital admission. The median duration from admission to death was 7.5 days (range, 1-26 days). The most frequent initial symptoms included fever, cough, and dyspnea. In 9 patients, infiltrations with ground-glass opacity predominantly in middle and lower lung fields were detected by chest x-ray. Patients had a median of 4 known preexisting comorbidities (range, 0-6), with cardiovascular disease being most frequent (Table). Preexisting structural lung damage (eg, emphysema) was found in 2 patients. None of the patients had thromboembolic events in central vessels at autopsy or prior to death.

In all cases, including 6 patients who did not receive invasive ventilation, disseminated diffuse alveolar damage at different stages (the histopathological correlate of acute respiratory distress syndrome) was the major histologic finding. Diffuse alveolar damage was detectable in all lobes but appeared unevenly distributed with pronounced manifestation in middle and lower lung fields (Figure, A-B). Signs of exudative early-phase acute diffuse alveolar damage with hyaline membrane formation, intra-alveolar edema, and thickened alveolar septa with perivascular lymphocyte-plasmocytic infiltration were consistently found. Organizing-stage diffuse alveolar damage with pronounced fibroblastic proliferation, partial fibrosis, pneumocyte hyperplasia leading to interstitial thickening and collapsed alveoles, and patchy lymphocyte infiltration was the predominant finding. In areas of organizing diffuse alveolar damage, reactive osseous and squamous metaplasia were observed (Figure, C-G). Fully established fibrosis was most prominent in patient 1, ultimately leading to almost complete destruction of pulmonary parenchyma. In 5 patients, minor neutrophil infiltration was indicative of secondary infection and/or aspiration.

Mild lymphocytic myocarditis and signs of epicarditis were detectable in 4 and 2 cases, respectively. Liver histology showed minimal periportal lymphoplasmacellular infiltration and signs of fibrosis. There was no morphologically detectable pathology in other organs. Specifically, no signs of encephalitis or central nervous system vasculitis were found.

At time of autopsy, SARS-CoV-2 was still detectable in the respiratory tracts of all patients. Polymerase chain reaction testing was positive in pleural effusion but negative in all CSF samples.

In this postmortem evaluation of 10 patients with COVID-19, acute and organizing diffuse alveolar damage and SARS-CoV-2 persistence in the respiratory tract were the predominant histopathologic findings and constituted the leading cause of death in patients with and without invasive ventilation. Periportal liver lymphocyte infiltration was considered unspecific inflammation. Whether myoepicardial alterations represented systemic inflammation or early myocarditis is unclear; criteria for true myocarditis were not met. Central nervous system involvement by COVID-19 could not be detected.

This study has limitations, including the small number of cases from a single center and missing proof of direct viral organ infection.

The pulmonary histologic characteristics of COVID-19 resembled those observed in diseases caused by other Betacoronavirus infections such as severe acute respiratory syndrome⁴ and Middle East respiratory syndrome.⁵

Accepted for Publication: May 11, 2020.

Corresponding Author: Rainer Claus, MD, Department of Hematology and Clinical Oncology, University Medical Center Augsburg, Stenglinstrasse 2, 86156 Augsburg, Germany (rainer.claus@uk-augsburg.de).

Published Online: May 21, 2020. doi:10.1001/jama.2020.8907

Author Contributions: Drs Schaller and Claus had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Schaller and Hirschbühl contributed equally as first authors. Drs Märkl and Claus contributed equally as senior authors.

Concept and design: Schaller, Hirschbühl, Braun, Trepel, Märkl, Claus.

Acquisition, analysis, or interpretation of data: Schaller, Hirschbühl, Burkhardt, Märkl, Claus.

Drafting of the manuscript: Schaller, Hirschbühl, Braun, Trepel, Märkl, Claus.

Critical revision of the manuscript for important intellectual content: Schaller, Hirschbühl, Burkhardt, Trepel, Märkl, Claus.

Statistical analysis: Claus.

Obtained funding: Braun.

Administrative, technical, or material support: Schaller, Hirschbühl, Burkhardt, Trepel, Märkl.

Supervision: Claus.

Conflict of Interest Disclosures: None reported.

Additional Contributions: We thank the physicians from the intensive care unit (Michael Wittmann, MD, and Ulrich Jaschinski, MD), and the Department of Radiology (Thomas Kröncke, MD) of the University Medical Center Augsburg. We thank Jürgen Schlegel, MD, from the Department of Neuropathology, School of Medicine, Institute of Pathology, Technical University Munich, for sampling brain tissue. Technical support was provided by Alexandra Martin, AMLT, Christian Beul, AMLT, and Elfriede Schwarz, AMLT, from the Institute of Pathology and Molecular Diagnostics, University Medical Center Augsburg. No compensation was received for their roles in the study.