Risk assessment model for venous thromboembolism in hospitalized patients with cancer and COVID-19

CoVID-TE risk assessment model (RAM) may help in real-time data-driven decisions on thromboprophylaxis initiation in hospitalized patients with cancer and COVID-19.

Key Takeaway

  • In this retrospective cohort study of hospitalized patients with cancer and COVID-19, incidences of venous thromboembolism (VTE) and arterial thromboembolism (ATE) were 7.6% and 3.9%, respectively.
  • The risk of VTE was higher in patients admitted to the intensive care unit (ICU) and in those receiving recent anti-cancer systemic therapy.
  • The newly derived VTE RAM (CoVID-TE score) stratified patients into 2 different risk categories, with modest discrimination:
    • In low-risk group (score 0–2), 4.1% had VTE and 2.3% had pulmonary embolism (PE).
    • In high-risk group (score 3+), 11.3% had VTE and 5.5% had PE.
  • CoVID-TE RAM may help in real-time clinical decisions on thromboprophylaxis initiation in hospitalized patients with cancer and COVID-19.

Why This Matters

  • The exact thrombotic risk is not known in hospitalized patients with cancer and COVID-19.
  • This retrospective study estimated the incidence of VTE and ATE in patients with COVID-19 and cancer requiring hospitalization and derived a simple RAM in these patients.

Study Design

  • The COVID-19 and Cancer Consortium registry (NCT04354701) is an ongoing multi-center effort to evaluate the clinical-pathologic factors and disease course in patients with COVID-19 and cancer.
  • Inclusion criteria: Adult patients with an active or previous diagnosis of cancer; a laboratory-confirmed SARS-CoV-2 test between March 17, 2020, and November 30, 2020
  • Exclusion criteria: Patients who did not reside within the United States or Canada; no assessable thrombotic complication status within 90 days (13 weeks); not hospitalized at baseline; poor data quality (quality score ≥5), or follow-up <30 days
  • Data were recorded at baseline around the time of COVID-19 diagnosis and then at 30 days, 90 days, and 180 days after diagnosis.
  • Primary outcomes: VTE defined as PE, deep vein thrombosis (DVT), or thrombosis not otherwise specified (NOS); and ATE defined as myocardial infarction or ischemic stroke (CVA)
  • Secondary outcomes: Frequency of PE and/or DVT, PE only, or CVA only

Key Results

  • Overall, 2,804 hospitalized patients (median age, 70 years; male, 54%) were included and median follow-up was 42 days (interquartile range [IQR], 21–90).
  • VTE occurred in 213 (7.6%) patients (PE, 113 [4.0%] patients), and ATE occurred in 109 (3.9%) patients (CVA, 45 [1.6%] patients); most of these events occurred within 30 days of hospitalization.
  • Incidence of all thrombotic complications was ~2-fold higher in severely ill patients admitted to ICU vs moderately ill patients admitted to wards (VTE, 14.1% vs 6.3%; ATE, 7.3% vs 3.2%).
  • Incidence of VTE but not ATE was higher in patients receiving recent anti-cancer systemic therapy (VTE, 10.0%; ATE, 3.1%) vs those not receiving recent therapy (VTE, 5.8%; ATE 4.0%).
  • Variables significantly associated with VTE: Recent anti-cancer systemic therapy (odds ratio [OR] = 1.58; 95% confidence interval [CI]: 1.16–2.14), VTE history (OR = 1.89; 95% CI: 1.21–2.95), and direct ICU admission (OR = 2.62; 95% CI: 1.89–3.64)
  • A simplified RAM was derived for VTE named CoVID-TE (Cancer subtype high to very-high risk by original Khorana score [+1], VTE history [+2], ICU admission [+2], D-dimer elevated on admission [+1], Therapy [recent systemic therapy last 3 months] [+1], and Ethnicity non-Hispanic [+1]).
  • The RAM stratified patients into low risk (0–2 points, n = 1423) vs high risk (3+ points, n = 1034); VTE occurred in 4.1% vs 11.3%, respectively (c statistic 0.67, 95% CI: 0.63–0.71; Hosmer-Lemeshow test P value 0.90).
  • In a sensitivity analysis, CoVID-TE RAM showed similar discrimination for VTE prediction after excluding patients with anticoagulant use before COVID-19 diagnosis (c-statistic, 0.69 [95% CI: 0.64–0.74]).
  • Patients classified as high risk vs low risk for VTE had higher risk for overall bleeding (10.0% vs 4.3%) and mortality (29.3% vs 19.5%).


  • Limitations inherent to retrospective study (potential for unmeasured confounding, selection bias, and underreporting of outcomes).
  • Heterogeneity in the protocol for the prevention and diagnosis of thrombosis could have impacted the actual rates from individual sites.
  • Anticoagulation heterogeneous use before admission and the low proportion of very high-risk cancer types might limit the generalizability of RAM.
  • Time-to-event or competing risks analysis was not performed.
  • Novel model needs to be tested and validated in an external cohort.
    1. Li A, Kuderer NM, Hsu CY, Shyr Y, Warner JL, Shah DP, et al. The CoVID-TE risk assessment model for venous thromboembolism in hospitalized patients with cancer and COVID-19. J Thromb Haemost. 2021: doi: 10.1111/jth.15463. Epub ahead of print. PMID: 34260813.