We read with great interest the elegant study conducted by Patel and colleagues (1) regarding the alterations at pulmonary vessel level in patients with severe coronavirus disease (COVID-19). The authors meticulously presented the combination of physiologic data, the results of high-resolution imaging, and the hematologic observations in a cohort of 39 patients. They showed that the activation of inflammatory and coagulation pathways has a pivotal role in the development of acute respiratory failure induced by COVID-19, demonstrating the great impact of hypercoagulability and reduction of fibrinolysis on the pulmonary vasculature. Such a prothrombotic state finally induces pulmonary (and likely systemic) perfusion abnormalities, heavily contributing to the peculiar phenotype of COVID-19-induced respiratory failure (2). We believe that the results shown by Patel and colleagues (1), highlighting the presence of dilated peripheral lung vessels (roughly two-thirds of patients) and perfusion defects in all patients, are of great importance in cautiously interpreting the results of a recent study on the use of convalescent plasma (CP) in COVID-19. Indeed, this study evaluated the use of CP in more than 5,000 patients with severe or life-threatening COVID-19 (3) and prompted great (and in our opinion excessive) enthusiasm as the authors reported low incidence of serious adverse effects after CP. However, this safety endpoint was evaluated in a particularly short period of observation (4 h), which is far too limited to entirely account for subtle progression of an underlying hypercoagulability state.