Venous Thromboembolism in Hospitalized Critical and Noncritical COVID-19 Patients: A Systematic Review and Meta-analysis

TH Open. 2021 Jul 6;5(3):e286-e294. doi: 10.1055/s-0041-1730967. eCollection 2021 Jul.


Introduction Venous thromboembolism (VTE) has been observed as a frequent complication in patients with severe novel coronavirus disease 2019 (COVID-19) infection requiring hospital admission. Aim This study was aimed to evaluate the epidemiology of VTE in hospitalized intensive care unit (ICU) and non-ICU patients. Materials and Methods PubMed was searched up to November 13, 2020, and updated in December 12, 2020. We included studies that evaluated the epidemiology of VTE, including pulmonary embolism (PE) and/or deep vein thrombosis (DVT), in patients with COVID-19. Results A total of 91 studies reporting on 35,017 patients with COVID-19 was included. The overall frequency of VTE in all patients, ICU and non-ICU, was 12.8% (95% confidence interval [CI]: 11.103-14.605), 24.1% (95% CI: 20.070-28.280), and 7.7% (95% CI: 5.956-9.700), respectively. PE occurred in 8.5% (95% CI: 6.911-10.208), and proximal DVT occurred in 8.2% (95% CI: 6.675-9.874) of all hospitalized patients. The relative risk for VTE associated with ICU admission was 2.99 (95% CI: 2.301-3.887, p <0.001). DVT and PE estimated in studies that adopted some form of systematic screening were higher compared with studies with symptom-triggered screening. Analysis restricted to studies in the 5th quintile of sample size reported significantly lower VTE estimates. Conclusion This study confirmed a high risk of VTE in hospitalized COVID-19 patients, especially those admitted to the ICU. Nevertheless, sensitivity analysis suggests that previously reported frequencies of VTE in COVID-19 might have been overestimated.

Keywords: COVID-19; anticoagulation; venous thromboembolism.

Grant support

Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.