(P-531) Role of D-dimer on Serious Outcomes Among Patients Hospitalized with COVID-19

Background: Elevated D-dimer levels are a component of COVID-19-associated coagulopathy and may indicate a poor prognosis. The association between D-dimer levels and serious outcomes of COVID-19 remains uncertain.

Objectives: To identify correlates of elevated D-dimer and to evaluate the effect of D-dimer levels on serious outcomes among hospitalized COVID-19 patients.

Methods: This was a retrospective cohort study of patients hospitalized with COVID-19 identified in the Optum EHR COVID Database (January2020-November 2020) using ICD-10 code U07.1 or a positive SARS-CoV-2 viral lab result. Patients were followed from cohort entry (i.e., later of confirmation or admission) for the following serious outcomes: thrombotic events, receipt of critical care, mechanical ventilation, and all-cause mortality. D-dimer levels were determined based on the first result recorded after cohort entry using D-dimer units and categorized as ≤ 250 ng/mL, >250 to 500 ng/mL, >500 to 750 ng/mL, >750 to 1000 ng/mL, and >1000 ng/mL. The associations between D-dimer level and serious outcomes were quantified using relative risks (RRs) and 95% confidence intervals (CIs), adjusting for baseline characteristics (e.g., demographics, comorbidities, and patient-reported medications).

Results: A total of 24,769 patients hospitalized with COVID-19 had a recorded D-dimer level. Factors associated with elevated D-dimer levels included older age (>60 years), African-American race, history of smoking, diabetes, cardiovascular disease (CVD) including hypertension, kidney disease, and prior use of statins or ACEs/ARBs; prevalence of each factor was highest among those with D-dimer >1000 ng/mL. Risks of serious outcomes also increased across D-dimer level categories. Compared to D-dimer level <=250 ng/mL, the adjusted RRs of higher D-dimer level categories ranged from 1.0 (0.9, 1.1) to 2.3 (2.1, 2.5) for thrombotic events, 1.5 (1.3, 1.6) to 3.0 (2.8, 3.2) for receipt of critical care, 1.5 (1.4, 1.7) to 4.3 (3.9, 4.6) for mechanical ventilation, and 1.6 (1.4, 1.8) to 4.7 (4.2, 5.2) for death. Dose-response relationships across D-dimer level categories were also observed for each outcome (all Ps <0.001).

Conclusions: Correlates of elevated D-dimer levels include older age, race, smoking history, diabetes, CVD including hypertension, kidney disease, and prior use of statins or ACEs/ARBs. Elevated D-dimer levels at 4 times the upper normal limit are strongly associated with serious outcomes among patients hospitalized with COVID-19.