Skip to main content

Incidence of thrombotic complications in critically ill ICU patients with COVID-19

Our take —

This study suggests there is a considerable risk of clinically-apparent thromboembolic complications among COVID-19 patients with severe pneumonia, despite systematic administration of thromboprophylaxis. Results may not be generalizable to those with less severe or asymptomatic disease. Analyses of prognostic factors did not appear to be adjusted, thus results should be interpreted with caution. Additional prospective studies including those with less severe disease are warranted.

Study design

Prospective Cohort

Study population and setting

This study included 184 patients with clinically-determined COVID-19 pneumonia who were admitted to the ICUs of 3 hospitals in the Netherlands between March 7 and April 5, 2020.

Summary of Main Findings

The mean age of the study participants was 64 years (76% male). All patients were provided with thromboprophylaxis according to institutional protocols. As of April 5, 2020, 13% had died, 12% had been discharged, and 76% remained hospitalized. The median duration of follow-up was 7 days. Thirty-one patients experienced thrombotic complications over follow-up (81% had a pulmonary embolism, 19% had other venous thromboembolic or arterial thrombotic events), corresponding to a cumulative incidence of 31% (95% CI: 20-41). Older age and the presence of coagulopathy at admission were prognostic of thrombotic complications.

Study Strengths

This study provides one of the first estimations of the incidence of thromboembolic events among patients with confirmed COVID-19 admitted to ICU across 3 different hospital systems in the Netherlands. There was prospective assessment of outcomes, and a detailed characterization of pharmaceutical prophylaxis provided.


The majority of the study participants were still in the ICU by the end of the study period, therefore data on outcomes were incomplete. It is possible the disease trajectories of these patients may differ from those who had an observable thromboembolic event. Any estimates are therefore likely conservative. Moreover, this study may only relate to events that are associated with severe disease. Predictors were identified through automated variable selection, but the full set of  confounding variables considered was not apparent. Because diagnostic tests were only administered with clinical indication of thromboembolism, subclinical events were not included.

Value added

This study provides one of the first characterizations of the incidence of venous and arterial thrombotic events in cases of severe COVID-19 pneumonia across multiple centers in a well-developed healthcare setting.