Long COVID refers to a long-term multi-system disability syndrome seen in COVID-19 survivors. The US Centers for Disease Control and Prevention (CDC) and National Institutes of Health (NIH) define long COVID as sequelae that extend beyond four weeks after initial infection (Crook et al., 2021). It includes post-acute COVID-19 and post-COVID-19 syndrome. People who have persistent SARS-CoV-2 infection show structural and functional impairment of multiple organ systems, including: respiratory, cardiovascular, haematological, neurological, urinary, gastrointestinal, and musculoskeletal (Sudre et al., 2021). Symptoms include fatigue (47%), dyspnea (32%), myalgia (25%), joint pain (20%), headache (18%), cough (18%), chest pain (15%), olfactory abnormality (14%), taste changes (7%), and/or diarrhea (6%). Heart abnormalities, cognitive impairment, sleep disturbances, post-traumatic stress disorder (PTSD), and concentration problems have also been reported (Aiyegbusi et al., 2021). Table 1 summarizes other reviews on long COVID symptoms (Ahmad et al., 2021; Cabrerai Martimbianco et al., 2021; Ceban et al., 2021; Groff et al., 2021; Long et al., 2021; Lopez-Leon et al., 2021; Michelen et al., 2021; Alkodaymi et al., 2022; Malik et al., 2022; Nguyen et al., 2022). There have also been recent studies on the pathophysiological mechanism of long COVID. Persistent vascular endothelial injury is common in convalescent COVID-19 patients and is not associated with ongoing acute response (Fogarty et al., 2021). Vascular endothelial damage can be caused by long-term viral infection, chronic hypoxia and inflammatory response. This initiates coagulation and microthrombosis, which may lead to various systemic functional impairments and clinical sequelae (Ambrosino et al., 2021; García-Abellán et al., 2021; Peluso et al., 2021). Thrombosis can further aggravate vasculitis, which may further damage various organs. This is consistent with autopsy findings of coagulation disorders/abnormalities in the lungs and critical organ systems following COVID-19. This also indicates that long COVID is essentially a thrombotic sequela. Unfortunately, there is currently no effective treatment for long COVID. Therefore, more effective early treatment is essential to prevent serious COVID-19 disease, lessen the degree of thrombotic damage, and potentially mitigate long-term sequelae, decreasing the burden of long COVID on patients and healthcare systems.

This article first summarizes the manifestations of abnormal coagulation in long COVID and explains the thrombosis mechanism in detail. Extracellular vesicles (EVs) are released by various cell types to transport cargoes (such as mRNA, microRNAs, DNA, lipids, and various proteins) to nearby or distant cells to help maintain their physiological state. Recent studies have shown that SARS-CoV-2 may be transported by EVs to distant tissues and organs (Barberis et al., 2021; Borowiec et al., 2021; Eymieux et al., 2021). In addition, many studies have shown that EVs play an important role in coagulation activation (Guervilly et al., 2021; Rosell et al., 2021). Long COVID often leads to chronic hypoxia with pulmonary vascular changes and decreased lung function (Carfì et al., 2020; Caruso et al., 2021; Cueto-Robledo et al., 2022). Hypoxia also provides conditions under which immune cells produce more inflammatory cytokines (Moasefi et al., 2021; Østergaard, 2021). Ultimately, prolonged viral presence, hypoxia, and inflammatory responses lead to persistent endothelial damage, extensive vascular endotheliitis and thrombosis. Second, we review and analyze the current studies on the dose and timing of antithrombotic therapy. There is substantial evidence that early anticoagulation therapy improves patient outcomes (Terpos et al., 2020; Arslan et al., 2021; Kollias et al., 2021; Rentsch et al., 2021). In acute COVID-19, the importance of controlling viral replication and preventing inflammation is well established. However, early removal of procoagulant substances and protection of the vascular endothelium may be the best means to prevent long-term thrombotic sequelae.

Several studies have tried to quantify the incidence of ongoing thrombosis in patients after discharge ( Table 2 ). Giannis et al. conducted a large-scale (n=4906) statistical analysis of major thromboembolic events in this population. The results showed that 76 patients (1.55%) were diagnosed with venous thromboembolism (VTE), including 44 deep vein thrombosis (0.90%), 42 pulmonary embolism (0.85%), 2 splanchnic vein thrombosis (0.04%), and 3 another vein thromboses (0.06%) (Giannis et al., 2020). Patell et al. showed that the cumulative incidence of thrombosis (including arterial and venous events) at day 30 following discharge was 2.5%; while the cumulative incidence of venous thromboembolism alone at day 30 post discharge was 0.6% (Patell et al., 2020). This shows that discharged COVID-19 patients are still at risk of thrombosis. A study analyzing the serum metabolic profile of 75 previously diagnosed COVID-19 patients 2 months after discharge found that all patients had very high serum concentrations of ferritin and D-Dimer, and 73% had elevated erythrocyte sedimentation rate and CRP (Pasini et al., 2021). Another study showed that plasma samples from Long COVID/PASC (post-acute sequelae of COVID-19) still contain large anomalous (amyloid) deposits (microclots). Various inflammatory molecules were significantly increased in both the supernatant and trapped in the solubilized pellet deposits from Long COVID/PASC samples (Pretorius et al., 2021). A study measuring coagulation indicators 4 months after COVID-19 patients discharge found that the patient samples enhanced thrombin-generating capacity and decreased plasma fibrinolytic potential indicating sustained prothrombotic changes. Increases in plasma factor VIII and PAI-1 levels may be related to the continuous activation of ECs, which may partly explain the hypercoagulable and hypofibrinolytic states (von Meijenfeldt et al., 2021). Korompoki et al. also summarized available evidence on post-acute COVID-19 hematological complications (Korompoki et al., 2022). Overall, these studies have shown that persistent coagulation abnormalities and thrombosis are common in long covid. Other experiments have verified that continuous coagulation activation may lead to abnormal functions in various organs. Post-pulmonary thrombosis syndrome can manifest as persistent thrombosis and long-term functional limitation in long COVID. Pulmonary hypertension, embolism and fibrosis are common sequelae of the lungs (Carfì et al., 2020; Caruso et al., 2021; Cueto-Robledo et al., 2022), which can result in impaired function (diffusing capacity of the lung for carbon monoxide (DLCO), 6-minute walk distances (6MWD), and exercise-induced oxygen saturation) in patients. In summary, the above data indicate that abnormal coagulation is a common manifestation in long COVID, with prolonged coagulation activation, microvascular injury, and thrombosis driving systemic damage in patients.

Anticoagulation is a common treatment for hospitalized patients with COVID-19. Many articles have discussed the optimal dosing and duration of anticoagulant treatment (Jonmarker et al., 2020; Taccone et al., 2020; Atallah et al., 2021; Wahid and Ortel, 2021; Bradbury and McQuilten, 2022). It is certain that thrombosis risk gradually increases with disease progression, necessitating the use of anticoagulants (such as heparin, LMWH and DOAC) in the middle and late stages to inactivated coagulation factors and inhibit re-formation of thrombi. Yet its effects in these patients does not depend on increasing dose. This is likely because the anticoagulants cannot completely remove the large number of thrombi in patients with advanced disease. In contrast, early application of anticoagulants in COVID-19 has shown beneficial results (Terpos et al., 2020; Arslan et al., 2021; Kollias et al., 2021; Rentsch et al., 2021). Arslan et al. found that patients who received LMWH had shorter hospital stays compared with those who did not receive LMWH despite being older, with more comorbidities (such as hypertension, coronary heart disease and cancer) and higher inflammatory markers (C-reactive protein). Early anticoagulation in this study refers to the treatment of patients without any contraindications in early stage COVID-19 infection. We can speculate that early anticoagulation therapy would benefit more patients with advanced age, more underlying comorbidities, and higher inflammatory markers. Another study also found that starting prophylactic anticoagulation within 24 hours of admission reduced 30-day mortality and in-hospital mortality. Evidence of benefit is strongest in patients not admitted to the ICU within 24 hours of admission (Rentsch et al., 2021). Sulodexide is a compound of two glycosaminoglycans (GAGs): a fast-moving heparin fraction (80%) and dermatan sulfate (20%). In addition to being effective in anticoagulation, sulodexide also restores endothelial barrier function. Sulodexide has a lower risk of bleeding than other oral anticoagulants. A study of early outpatient patients with mild COVID-19 has shown that sulodexide is effective in reducing hospitalizations and the need for supplemental oxygen therapy. Patients treated with sulodexide also had lower CRP and D-dimer, inflammation markers and pre-thrombotic status (Gonzalez-Ochoa et al., 2021). In conclusion, the above data provide high-quality evidence for early anticoagulation in COVID-19. Given that early prophylactic anticoagulation in COVID-19 is a new treatment trend, more research is needed to explore which group of patients will benefit the most and determine the duration of treatment.

In Figure 3B of this paper, we propose a hypothesis that in long COVID, EV-delivered virus persistently attacks the systemic system, and coupled with chronic hypoxia and persistent inflammatory response, damages the vascular endothelium. The above factors also lead to PS exposure on the surface of various cells and their derived EVs. These factors influence each other and together promote thrombosis. Early prophylactic anticoagulation in COVID-19 can quickly remove a variety of procoagulant substances, thereby protecting the blood system, surrounding tissues, and organs from damage, inhibiting PS exposure, and avoiding subsequent thrombosis and sequelae ( Figure 3B ).

Some studies suggest that extending venous thromboembolism prophylaxis beyond hospital discharge may be beneficial, but the benefit is limited to high-risk patients with an increased risk of thromboembolism from COVID-19. For example, in one study analyzing 146 patients, 28% were prescribed post-discharge thromboprophylaxis. Its results suggest greater use in patients with higher levels of maximal D-dimer and C-reactive protein after and during ICU admission. Strategies to selectively provide thromboprophylaxis appear to be safe and potentially effective in high-risk patients (Engelen et al., 2021). Another study in high-risk patients (increased risk of venous thromboembolism) hospitalized and discharged for COVID-19 showed that 35 days of rivaroxaban with thromboprophylaxis improved clinical outcomes compared with unextended thromboprophylaxis (Ramacciotti et al., 2022). However, one study reported a low rate of vascular thromboembolic events after discharge in patients with COVID-19 and suggested that thromboprophylaxis should not be routinely used in patients with COVID-19 after discharge (Eswaran et al., 2021). The authors speculate that it is possible that patients at higher risk for vascular thromboembolic events (VaTEs) were on prophylactic anticoagulation at discharge, which may have contributed to the lower rates in the VaTEs group. Many guidelines also recommend the use of anticoagulants after discharge (Barnes et al., 2020; Bikdeli et al., 2020; Moores et al., 2020; Spyropoulos et al., 2020; Vanassche et al., 2020; Cuker et al., 2022). ASH recently issued a conditional recommendation not to use outpatient anticoagulation prophylaxis for discharged COVID-19 patients without suspected or confirmed venous thromboembolism or other indications for anticoagulation (Cuker et al., 2022). The panel judged that both the benefits and harms of thromboprophylaxis after discharge were negligible in absolute terms. Despite a small benefit and reduction in mortality from venous thromboembolism with anticoagulation after discharge, the certainty of the evidence is low. Meanwhile, in COVID-19, there is no high-quality direct evidence that anticoagulation increases the risk of major bleeding complications. However, the panel believes that for patients without COVID-19, there is high-quality indirect evidence that there is an increased risk of major bleeding when anticoagulation is used after hospital discharge. In general, undesirable outcomes outweigh desirable outcomes (Cuker et al., 2022). The CHEST guidelines also recommend that thromboprophylaxis is only recommended for hospitalized patients with COVID-19, not prolonged thromboprophylaxis after hospital discharge (Moores et al., 2020). Other guidelines listed in Table 3 all state that post-discharge prophylaxis should be considered in terms of thrombotic risk and bleeding risk. Of course, these recommendations will be updated in light of changing evidence, but from the current evidence, the use of antithrombotic drugs after discharge requires caution.

Platelets are at the forefront of COVID-19 pathogenesis, as they release a wide variety of molecules (including cytokines, alpha granules, dense granules and EVs) at different stages of the disease (Zaid et al., 2020; Rolla et al., 2021). Furthermore, COVID-19 patients had increased PS exposure in platelet extracellular vesicles (PEVs) (Rolla et al., 2021). Another study showed that SARS-CoV-2+ patients had higher counts of circulating platelet-derived extracellular vesicles (PLT-EVs) compared to healthy controls, with ROC curve analysis showing a sensitivity of 75% and specificity of 74% (Cappellano et al., 2021). SARS-CoV-2 can activate platelets and induce an inflammatory response that produces a wide range of immunomodulatory cytokines, chemokines, and other mediators. Endothelial injury promotes platelet activation, and in turn, chemotaxis of activated platelets recruits leukocytes, increases endothelial inflammation and thrombosis (Rolla et al., 2021). Recently, platelet activation inhibitors have garnered significant interest in COVID-19. A study suggests that antiplatelet therapy (including aspirin, clopidogrel, ticlopidine, prasugrel and ticagrelor) during COVID-19 hospitalization may be associated with a lower risk of death and shorter duration of mechanical ventilation without an increased risk of bleeding (Santoro et al., 2022). Aspirin is a mature drug with multiple effects such as inhibition of viral replication, anticoagulation, antiplatelet aggregation, anti-inflammatory and anti-lung injury (Mohamed-Hussein et al., 2020). Aspirin can inhibit prostaglandin E2 in macrophages and upregulate type I interferon to suppress viral replication. It can also reduce neutrophil aggregation and platelet activation. A study of covid-19 hospitalized patients showed that compared with patients who did not receive antiplatelet therapy, patients receiving aspirin had a significantly lower cumulative incidence of in-hospital death (Meizlish et al., 2021). Another study reported that tirofiban combined with aspirin and clopidogrel can effectively improve the ventilation/perfusion ratio in patients with severe respiratory failure due to COVID-19 (Viecca et al., 2020). There is very little data on combining antiplatelet and anticoagulant drugs in COVID-19. Though the dual mechanisms of antiplatelet and anticoagulation therapy on platelet thrombosis and hypercoagulability induced by COVID-19, may lead to synergistic effects (Matli et al., 2021). However, in hospitalized patients with moderate to severe COVID-19, anticoagulant heparin combined with aspirin may not be enough to inhibit thrombosis and increase bleeding risk (Rizk et al., 2021). There is also a lack of data on aspirin dosing and duration in COVID-19. In conclusion, aspirin can effectively inhibit inflammation, protect the endothelium, and prevent PS exposure after platelet activation. Based on pathophysiological insights, platelets may still represent a promising therapeutic target for COVID-19.

Exploring the pathophysiological mechanism and impact of long COVID thrombosis will help improve understanding of early antithrombotic therapy and better prevent thrombotic sequelae. This article summarizes the effects of persistent viral replication, inflammation, hypoxia, and endothelial injury leading to thrombosis and organ disfunction in the long COVID. The procoagulant effects of EVs and PS exposure caused by injury to circulating blood cells and ECs are highlighted. Although the vaccine is an effective measure to prevent SARS-CoV-2 infection, there are still unmet medical needs. The risk of variants that evade vaccine immunity, vaccine contraindications, immunocompromised persons who respond poorly to vaccines, and the challenge of obtaining vaccines in some areas, result in many Covid-19 patients who need treatment. Research on the use of anticoagulants in early stage COVID-19 is rapid. Many experimental studies have shown that early antithrombosis reduces mortality and improves prognosis. In the future, early preventive antithrombotic therapy may be an important means to better solve COVID-19 sequelae.

CW conceived and wrote the first draft of the article. CY and HJ researched data for the article. XW and VN provided helpful comments and wrote the article. RX provided substantial contribution to discussion of content and wrote the article. JS designed the review, prepared the tables and figures, and wrote the manuscript. All authors read and approved the final manuscript.

This work was supported by grants from the National Natural Science Foundation of China (81670659).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.