Am I a coronavirus?

To the editor:

The paper by Su et al. analyzes renal pathologic findings in the kidneys of 26 patients that underwent postmortem exam to understand the anatomic basis of kidney disease in the setting of fatal coronavirus disease 2019 (COVID-19).¹ The authors report the finding of viral particles in the kidney of COVID-19 patients and speculate that direct infection of the kidney by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus causes kidney disease.

Several findings within the manuscript by Su et al. ¹ are presented as definite evidence of specific disease processes without considering alternative explanations. For example, acute tubular injury is reported in all cases, including in patients with normal renal function. Discerning acute tubular injury from postmortem changes is notoriously problematic, as autolysis can mimic and mask acute tubular injury.² Infiltration of inflammatory cells in an arcuate artery is highlighted in a micrograph in which characteristic features of muscular arteries, such as elastic lamina or defined muscular layers, are not apparent (Figure 1d in Su et al. ¹). Distension of small blood vessels by red blood cells is referred to as obstruction, when it may simply represent congestion. Isolated fibrin clots are interpreted as evidence of severe endothelial injury but could also be due to coagulopathy. Most importantly, small vesicular structures identified by electron microscopy are described as viral particles without consideration of other interpretations.

Cells have organelles that can mimic the structure of viral particles, and accurate interpretation of electron micrographs requires integration of morphology and biology. The virus inside renal tubular epithelial cells and podocytes that Su et al. ¹ describe is shown as free particles in the cytoplasm, and not within membrane-bound organelles as would be expected for coronavirus based on in vitro studies and the rare examples of in vivo coronavirus infections reported prior to the current pandemic.3, 4, 5 There is no explanation for why the virus seen by Su et al. ¹ breaks this paradigm, which raises important questions about their interpretation of the micrographs. Cells have many structures comparable in size to the coronavirus, with varying degrees of electron-dense material surrounding and inside these structures. Notable examples include coated vesicles that are responsible for moving cargo into cells and between membrane-bound organelles (e.g., clathrin-coated vesicles and coatamer-coated vesicles).

To support their interpretation of the electron micrographs, Su et al. ¹ present immunofluorescence studies performed on sections of formalin-fixed and paraffin-embedded tissue. The distribution and quality of the positive anti-nucleocapsid protein staining bears striking resemblance to lipofuscin autofluorescence.⁶ Controls were reported to stain as expected, but no images of the controls are provided, nor is there an explanation of what was used as positive and negative controls to validate this antibody for formalin-fixed and paraffin-embedded tissue.

Therefore, in our judgment, Su and colleagues’¹ findings of small vesicular structures that are not conclusively distinguished from cellular vesicles and immunostaining that resembles lipofuscin autofluorescence without adequate controls are not sufficient to establish definitive infection of renal tubular epithelial cells and podocytes by the SARS-CoV-2. More rigorous and definitive studies are required to answer this question. SARS-CoV-2 may in fact infect the kidney and contribute to kidney disease in COVID-19 patients, but this remains an open question in search of an answer.