Long-term risk of thromboembolic events in patients with post-acute sequelae of COVID-19: A retrospective propensity-matched cohort study

Abstract Background Post-acute sequelae of SARS-CoV-2 infection (PASC), commonly referred to as long COVID, is a clinical syndrome characterized by a range of persistent symptoms that endure beyond the resolution of acute COVID-19. While acute COVID-19 is well known to cause thrombotic complications, the long-term thromboembolic risk associated with PASC remains poorly characterized. Proposed mechanisms include endothelial dysfunction, immune dysregulation, and chronic inflammation. However, existing studies are limited by short follow-up durations and the absence of well-matched control groups. This study addresses those gaps by evaluating long-term thromboembolic outcomes in a large, matched cohort of patients with and without PASC. Methods We conducted a retrospective cohort study using electronic health record data from the TriNetX global federated health research network. Adults aged ≥18 years with confirmed COVID-19 were divided into two cohorts: Cohort A included patients subsequently diagnosed with PASC, and Cohort B included those without any documented post-COVID diagnosis. Propensity score matching (1:1) was applied to balance demographics, comorbidities, and baseline medications, resulting in two matched cohorts of 90,567 patients each. Thrombotic outcomes, including myocardial infarction, cerebral infarction, pulmonary embolism, venous thrombosis, and arterial thrombosis, were assessed over a follow-up period of up to five years. We conducted risk analysis, Kaplan-Meier survival analysis, and count-based analysis to assess outcome frequency, risk, and time to event. Results After matching, baseline characteristics were well balanced between groups (mean age 58 years, 61% female). The risk of thrombotic outcomes was significantly higher in the PASC cohort (17.4%) compared to the non-PASC cohort (10.0%), yielding a risk difference of 7.4% (95% CI: 7.1-7.7), a risk ratio of 1.75 (95% CI: 1.70-1.79), and an odds ratio of 1.90 (95% CI: 1.85-1.96), all p < 0.001. Kaplan-Meier analysis showed lower event-free survival in the PASC cohort (75.8% vs. 81.5%), with a hazard ratio of 1.61 (95% CI: 1.56-1.65, p < 0.001). The average number of thrombotic events per patient was also higher in the PASC cohort (mean 1.33 vs. 0.60, p < 0.001). Median follow-up durations were 988 and 813 days, respectively. Conclusion This research is one of the few studies focusing on the prolonged effects of COVID-19 and PASC on thromboembolic events. In this extensive, multi-institutional study, PASC was associated with a significantly increased long-term risk of thromboembolic events. These findings are especially relevant for patients with existing thrombotic risk factors, such as malignancy, limited mobility, pregnancy, or hormonal therapy. Further research is needed to characterize the underlying mechanisms and evaluate preventive strategies in this emerging high-risk group. Healthcare providers should maintain a high index of suspicion for thromboembolic events in patients diagnosed with PASC.