High risk of venous thromboembolism in hospitalized COVID-19 patients

Introduction and methods

Studies from China have demonstrated increased coagulation activation in COVID-19 patients. Findings included increased D-dimers (0.5 mg/L or higher) in 46-63% of patients, mild thrombocytopenia and prolonged prothrombin time [1, 2]. Coagulation activation in COVID-19 patients appeared to correlate with severe disease course, including admission to the ICU and death. For example, as compared to patients who survived COVID-19, patients who died had higher D-dimers on admission and increased levels of D-dimers during hospital stay [3]. None of the previous studies, however, described the number of COVID-19 patients with thrombotic complications. Since the pandemic spread of the SARS-CoV2 virus, anecdotal reports have described a high incidence of thrombotic complications in COVID-19 patients.

The present study evaluated the incidence and risk factors of venous thromboembolism (VTE) in COVID-19 patients (n=198) admitted to the Amsterdam University Medical Centers, and were categorized as IC-patients (n=74) or ward patients (n=124), between March 2 and April 12, 2020. Patients diagnosed with COVID-19 during hospital stay for other medical conditions were not included. All COVID-19 patients were on standard thrombosis prophylactic care, and from April 3 onwards IC-patients received a dose of the low-molecular-weight heparin nadroparin twice as high as compared to the dose of ward patients. Primary outcome was defined as diagnosis of VTE, including distal or proximal DVT, PE or venous thrombosis at other sites including catheter-related thrombosis. Secondary outcome was defined as symptomatic VTE, excluding events detected by bilateral leg ultrasound screening.

Main results

• ICU patients had higher D-dimer levels on admission than ward patients (median 2.1 mg/L vs. 1.1 mg/L, P=0.006).

• Regarding all patients, cumulative incidences of VTE were 15% at 7 days (95% CI, 9.3-22) and 34% at 14 days (95% CI, 23-46). Cumulative incidences of symptomatic VTE were 11% at 7 days (95% CI, 5.8-17) and 23% at 14 days (95% CI, 14-33).

• VTE analyzed as time-varying variable was significantly associated with death (HR, 3.3; 95% CI, 1.2-8.9), also after adjustment for age, sex, and ICU stay (adjusted HR, 2.9; 95% CI, 1.02-8.0).

• Comparing ICU patients with ward patients, the proportion of patients with VTE was significantly higher for ICU patients than for ward patients (39% vs. 3.2%; subdistribution HR, 7.3; 95% CI, 2.5-21). Similarly, more ICU patients had symptomatic VTE (24% vs. 3.2%; SHR, 3.8; 95 CI, 1.3-12).

• Cumulative incidences of VTE in ICU patients was 25% at 7 days (95% CI, 15-36) and 48% at 14 days (95% CI, 33-61). Cumulative incidences of symptomatic VTE in ICU patients was 15% at 7 days (95% CI, 7.8-25) and 31% at 14 days (95% CI, 19-44).

• Besides ICU stay, other risk factors associated with VTE in univariable regression analyses were lower lymphocyte count (SHR, 0.59 per 1*109/L increase; 95% CI, 0.37-0.93), higher neutrophil-to-lymphocyte ratio (SHR, 2.4 per unit increase; 95% CI, 1.5-3.7), and higher D-dimer level (SHR, 1.8 per 1 mg/L increase; 95% CI, 1.3-2.4).

Conclusion

References

Find this article online at J Thromb Haemost