Don’t Do Anything! Just Stand There!

FOR RELATED ARTICLE, SEE PAGE 1182

The available literature has reported high rates of VTE, predominantly pulmonary embolism (PE), in hospitalized patients with coronavirus disease 2019 (COVID-19) pneumonia, particularly in critically ill patients admitted to the ICU. Many patients develop thrombotic complications despite the use of prophylactic, or at times, therapeutic doses of anticoagulation. This has prompted many institutions to develop intensified thromboprophylaxis protocols that recommend higher doses of anticoagulants than are typically used in hospitalized medical patients, including therapeutic dosing for patients in the ICU.¹ We do, however, have existing, well-developed, evidence-based guidelines regarding thromboprophylaxis in medical patients.² To change practice in patients with COVID-19, we need to be certain the that rate of VTE is indeed higher in COVID-19 patients compared with similarly ill non-COVID-19 patients. We also need evidence that higher doses of anticoagulants will be effective. We need to feel confident that increased thromboprophylaxis will be safe.

Like many critically ill patients, patients with COVID-19 are clearly at increased risk of macrothrombosis, because they exhibit all three components of Virchow’s triad. Pathologic reports have shown that they also exhibit significant microthrombosis or immunothrombosis, related to hypoxemia, endothelial injury, and inflammation.³ Emerging data regarding the discrepancy between the rates of DVT and PE, and the fact that many of the documented cases of PE occur in the absence of DVT and are located in the more peripheral pulmonary arteries,⁴ have led to the hypothesis that there may be a unique PE phenotype in COVID-19 patients. These are likely thrombi and not emboli, that is, immunothrombosis is likely much more prominent than originally recognized. One report of 66 ICU patients, all of whom received standard-dose thromboprophylaxis, noted only a 5% rate of VTE that was not thought to be either line related or immunothrombosis.⁴ The role of immunothrombosis in other forms of ARDS is well established, but the ideal therapeutic approach, including the use of systemic or inhaled heparins, has not been.⁵ This may be why some small reports note high rates of VTE even on full-dose anticoagulation.⁶ It therefore may follow that targeting upstream therapies, such as antiviral and immunomodulating agents, to reduce the development of immunothrombosis will prove more efficacious than downstream attempts to suppress the coagulation system. Platelet inhibition also may be a potential more effective target.^{3 , 7}

The risk of bleeding in patients with COVID-19 is emerging. In a series of 92 ICU patients with COVID-19, all of whom received either prophylactic or full-dose anticoagulation, the authors reported a 34% incidence of VTE but also a 21% incidence of major hemorrhagic events, most of which (84%) occurred in patients on full-dose anticoagulation. Only 50% of those patients had a confirmed thrombosis.⁸ In a retrospective review of hospitalized patients with COVID-19, the authors noted that patients on therapeutic doses of anticoagulation had higher rates of major bleeding, whereas patients on prophylactic doses did not.⁹ A multicenter retrospective study reported the rate and severity of hemostatic and thrombotic complications of 400 hospital-admitted COVID-19 patients (144 critically ill) primarily receiving standard-dose prophylactic anticoagulation. They noted an overall bleeding rate of 4.8 % (7.6% in critically ill) and a major bleeding rate of 2.3% (5.6% in critically ill).¹⁰

In the absence of pending randomized data, how do we balance the benefit and risk in determining the optimal dosing of thromboprophylaxis? In this issue of CHEST, Jimenez and colleagues¹¹ share a systematic review and meta-analysis regarding the rates of VTE and bleeding in patients with COVID 19. This is a well-done study that follows appropriate methodology for high-quality systematic reviews and statistical approach to the meta-analysis. Despite the limitations imposed by the nature of the primary literature, several important lessons could be gleaned. The rate of VTE is consistent across cohorts whether no, standard, or higher doses of anticoagulation are used, but the rate of bleeding significantly increases with intermediate- or full-dose anticoagulation. When isolated distal DVT, catheter-associated DVT, and isolated subsegmental PE events are excluded, the overall rate of VTE is much lower (and not different from that of similarly ill non-COVID hospitalized patients). This highlights the role of immunothrombosis and the need for therapies that target the virus and immune response, rather than the coagulation system and macrothrombosis. In addition, using an inverse variance fixed effects model, the rate of VTE was 4.8%, whereas the rate of bleeding was 9.4%, bringing into question the practice of increased intensity of thromboprophylaxis in these patients.

We cannot underestimate the degree of uncertainty that clinicians are facing during this pandemic. We may be faced with the temptation to choose intervention over caution when confronted with ill patients and limited data. I would implore us to continue to give the great standard of care we have been trained to give.¹² We need more evidence to change our practice, particularly when it could potentially cause harm.