2  Methods

2.1  Study population and design
All patients with laboratory confirmed severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) admitted to the ICU due to acute hypoxemic respiratory failure between 22 March 2020 and 30 April 2020 and with available postmortem needle core biopsy of the lung were eligible for inclusion in this case series. SARS-CoV-2 was diagnosed using real-time reverse transcription polymerase chain reaction (RT-PCR) on sputum and/or bronchial aspirates. Routine ICU management included, among other things, selective digestive tract decontamination (SDD), chloroquine until the Dutch National Institute for Public Health and Environment advised against its use at the end of March 2020, and high dose anticoagulation with low-molecular-weight heparin (LMWH) (nadroparin 87 IE/kg twice daily). Bronchoscopy, with or without lavage, and testing for pulmonary aspergillosis were performed at the discretion of the attending physician. Pulmonary aspergillosis was diagnosed using clinical, radiological and mycological data and included galactomannan (serum and sputum), tracheal or bronchial culture. For galactomannan (GM) testing from bronchoalveolar lavage (BAL) fluid Platelia Aspergillis antigen ELISA (Biorad) was used. Recently, a case definition for influenza associated pulmonary aspergillosis (IAPA) was proposed by an expert panel, which could be used to classify patients with CAPA [14]. Diagnostic criteria include proven infection with clinical symptoms and a GM index of ≥1 on BAL or of ≥0ꞏ5 on serum; or Aspergillus spp. cultured from BAL.

2.2  Postmortem needle core biopsy of the lung
With permission from the patients' families, postmortem core needle biopsies were performed on both lungs within an hour after death. Lung biopsies were ultrasound and CT guided and combined with surface anatomic landmarks as a reference. A minimum of three histologic 14 gauge true cut needle biopsies were performed in both lungs. Biopsy specimens were fixed in formalin, embedded in paraffin, and cut into 3 μm sections. Hematoxylin and eosin (H&E), and periodic acid–Schiff with diastase (PAS-D) stains were applied according to routine procedures.

2.3  Statistical analysis
Descriptive statistics were used to describe patient characteristics. For statistical analysis IBM SPSS version 25.0 was used.
The local Medical Ethics Committee (Medical Research Committees United) approved the study.