Abstract
BACKGROUND: Coronavirus disease 2019 (COVID-19) increases the risk of venous thromboembolism, requiring monitoring of low molecular weight heparin (LMWH) via a time-consuming, costly and often unavailable test - anti-factor Xa (anti-Xa). An affordable, rapid point-of-care alternative, the thromboelastogram, is available, but performance comparisons to anti-Xa are lacking. OBJECTIVE: This study evaluated the relationship between anti-Xa and thromboelastogram in patients with COVID-19 receiving LMWH. METHODS: This was a retrospective study of patients with COVID-19 receiving LMWH at Chris Hani Baragwanath Academic Hospital, Johannesburg, South Africa, between November 2020 and January 2021. Blood samples tested with thromboelastogram and anti-Xa were drawn at three timepoints (one prior to and two after administration of LMWH). Thromboelastogram parameters comprised reaction time (R-time; onset of testing to the start of clot formation), kinetics time (K-time; start of clot formation until the clot reached 20 mm), and thromboelastogram coagulation index (overall coagulation status of whole blood). RESULTS: Forty-two patients with COVID-19 (15 male and 27 female) met the study criteria. There was a statistically significant, low to moderate correlation (Spearman's correlation coefficient [r (s) 0.43, p = 0.014]) between anti-Xa and thromboelastogram coagulation index. A statistically significant moderate correlation (r (s) 0.52, p = 0.002) between anti-Xa and R-time, and a statistically significant low correlation (r (s) 0.35, p = 0.049) between anti-Xa and K-time, were found. All correlations were 48 h post admission. CONCLUSION: Thromboelastogram coagulation index, R-times and K-times had a statistically significant association with anti-Xa levels in patients with COVID-19. Further research is required regarding their clinical utility. WHAT THIS STUDY ADDS: Thromboelastograms may represent a more cost-effective and accessible option to the conventional anti-Xa test in patients receiving LMWH. However, future research with larger sample sizes, varying disease profiles, and severity of illness is required.