Thrombotic Events in COVID-19 Patients and Its Comparison with Other Non-Sars-Cov-2 Respiratory Viruses

Introduction

An increasing number of evidence have reported the association of COVID-19 with increased incidence of thrombotic events. High incidences were initially reported in critically ill COVID-19 patients, but subsequently an increased incidence was also noticed in non-critically ill general ward patients. This has led to a universal recommendation of thromboprophylaxis for all COVID-19 patients by ASH and ISTH. As the data on COVID-19 and thrombosis continue to develop and evolve, we examined the data in two aspects. Firstly, other non-SARS-CoV-2 viral respiratory infections have also been reported to be associated with thrombotic events, be it arterial or venous. Thus, we aimed to compare the thrombotic rates between these two groups of patients directly to hopefully ascertain the actual thrombotic tendency in COVID-19 infections. Secondly, global hemostatic assays such as thromboelastogram and clot waveform analysis (CWA) have been used to demonstrate hypercoagulability in COVID-19 patients, albeit in a small group of patients and only in the critically ill. Incorporating these laboratory results into the management of thromboprophylaxis in COVID-19 is an attractive notion but more data and studies are definitely needed. Here, we evaluate the dynamic changes of hemostatic assays in patients with COVID-19 to better understand the overall coagulation profiles of COVID-19 infection.

Methods

We performed a single center, retrospective cohort study. All consecutive patients admitted to our hospital between 15 January and 10 April 2020 that were tested positive for COVID-19 or other non-SARS-CoV-2 respiratory viruses were included in our study. The main coagulation assays studied were prothrombin time and activated partial thromboplastin time and its associated CWA, min1, min2 and max2.

Findings

We included a total of 181 COVID-19 patients and 165 patients with non-SARS-CoV-2 respiratory viral infections. The respiratory viruses were rhinovirus (n=65), influenza A and B (n=46), adenovirus (n=13), human coronavirus 229E/NL63/OC43 (n=15), human enterovirus (n=3), metapneumovirus (n=6), parainfluenza virus 1 to 4 (n=11), respiratory syncytial virus (n=6) and human bocavirus 1 to 4 (n=0). The median age of COVID-19 patients was 37 (interquartile range [IQR], 30.5-51 years) versus 35 (IQR, 29-51.5) in the non-SAR-CoV-2 respiratory viruses group (P=0.12). Comorbidities, assessed by Charlson score, was also not statistically different between both groups (median score 0 (IQR, 0-1) in both groups, P=0.39). Majority of our patients had relatively mild infection as reflected by the low proportions of them requiring oxygen supplementation (11.0% in COVID-19 vs 4.8% in non-SARS-COV-2, P=0.035). COVID-19 patients had longer hospital stay (7 days (IQR, 5.5-13) vs 3 days (IQR, 2-3), P<0.001) and more required ICU support (5.0% vs 1.2%, P=0.04). Mortality rate was low in both groups. We reported two (1.0 event/1000-hospital-days) and one (1.8 event/1000-hospital-day) thrombotic events amongst COVID-19 group and non-SARS-COV-2 group respectively (P=0.63). All were myocardial infarction and occurred in intensive care unit. No venous thrombotic event was noted. There was no significant difference in all the coagulation parameters throughout the course of mild COVID-19 infection (Table 1). However, CWA parameters were significantly higher in severe COVID-19 infection compared with mild disease (min1: 6.48%/s vs 5.05%/s, P<0.001; min2: 0.92%/s² vs 0.74%/s², P=0.033), suggesting hypercoagulability in severe COVID-19 infection (Table 2 and Figure 1). We also observed that critically ill COVID-19 patients had higher absolute CWA parameters as compared to non-SARS-CoV-2 patients, albeit in small number of patients (Table 3).

Conclusion

The thrombotic rates were low in both groups and did not differ significantly between COVID-19 and Non-SARS-CoV-2 patients. Nonetheless, our analysis of hemostatic parameters demonstrated hypercoagulability in COVID-19 as a dynamic process with the risk highest when the patients are critically ill. These changes in hemostasis could be detected by CWA. With our findings, we suggest that a more individualized thromboprophylaxis approach, considering clinical and laboratory factors, is probably preferred over universal pharmacological thromboprophylaxis for all hospitalized COVID-19 patients and warrants further research.