Some reports and observations suggest that infection with SARS-COV-2 creates a profound prothrombotic state with coagulation derangements that results in a higher than expected incidence of venous thromboembolic events (VTE). Several earlier studies cite a VTE incidence of up to 30-40% in hospitalized or intensive care unit (ICU) patients, which is higher compared to historical rates in non-COVID hospitalized patients. These reports spurred a transition to intensified thromboprophylaxis across many institutions, absent rigorous evidence regarding the risks and benefits of such measures. To estimate the actual incidences of VTE (including pulmonary embolism (PE), deep vein thrombosis (DVT)) during hospitalization for COVID-19, we performed a rigorous systematic review of published literature.
We searched the MEDLINE database using the terms [(SARS-CoV2 OR COVID-19 OR Covid19) AND (trial OR series OR cohort OR incidence)] from December 1, 2019 to July 13, 2020. Two authors (JYZ and NGC) screened all study titles and abstracts and excluded those that did not include clinical data on adult patients with COVID-19. The bibliography lists from review articles were screened in the same fashion. The same two authors reviewed the full text of the remaining articles and excluded those that met the following exclusion criteria: 1) not peer-reviewed; 2) not written in English; 3) did not include unique primary data; 4) no defined study time interval; 5) did not describe and quantify hospitalized patients with confirmed diagnosis of COVID-19; 6) did not describe a consecutive case series or consecutive cohort studies or controlled trials with consecutive enrollment of subjects that met specific inclusion criteria; 7) study inclusion criteria were related to the clinical suspicion for VTE; 8) did not describe a method whereby clinically detected VTE was tabulated; 9) did not report the VTE incidence (PE, DVT, or both) among the specified study population. Adjudication of publications with uncertainty in meeting criteria was done by all authors. Data extraction from the selected studies comprised study type and population, VTE incidence and method of diagnosis, as well as thromboprophylaxis regimen and use. We calculated the proportions with 95% confidence intervals of hospitalized COVID-19 patients in whom VTE, PE, and DVT were diagnosed. These proportions were summarized from the entire group and from three pre-specified subsets of studies: those that enrolled fewer than 100 subjects, between 100-1000 subjects, and more than 1000 subjects.
The search of MEDLINE generated 11,983 articles, 11,266 of which were excluded for irrelevance. Of the remaining 717 articles, 664 met our specific exclusion criteria, leaving 52 articles included in the final analysis. The study populations were heterogeneous and included all hospitalized patients, non-ICU hospitalized patients and ICU patients. Screening and diagnostic methods for VTE among the studies varied. Some systematically used Doppler ultrasound screening on all subjects while others used specific diagnostic tests only if DVT or PE were suspected. Thromboprophylaxis regimens varied, whereby most studies employed some form of prophylaxis, a few studies used intensified dosing based on risk stratification. The overall incidences of PE, DVT and VTE among all hospitalized COVID-19 patients were 4.1%, 5.0% and 6.1%, respectively (Table). PE, DVT and VTE proportions reported in the smaller studies (with much wider confidence intervals) were typically three or more times as high as those reported in the larger studies as reflected in a funnel plot (Figure) and Table.
Our comprehensive review revealed incidences of PE, DVT and VTE among hospitalized COVID-19 patients that are comparable to hospitalized patients without COVID-19. Variability among the reported incidences demonstrated a trend toward higher incidences among the smaller studies, suggesting a publication bias towards overestimation in smaller reports. Thus we believe that intensification of thromboprophylaxis on the basis of COVID-19 infection, beyond usual hospital or ICU management protocols, should be done cautiously based on the individualized patient situation. Randomized controlled trials may help provide definitive evidence of VTE incidence and inform the formulation of appropriate thromboprophylaxis algorithms in these patients.
von Drygalski:Biomarin, Bioverativ/Sanofi-Genzyme, Novo Nordisk, Pfizer, Uniqure, Takeda: Consultancy; Hematherix Inc: Membership on an entity's Board of Directors or advisory committees, Other: Cofounder; superFVa; Joint Activity and Damage Examination (JADE) Ultrasound measurement tool: Patents & Royalties.
Asterisk with author names denotes non-ASH members.