Clinical differentiation of anticoagulant and non-anticoagulant properties of heparin.

I thank Lindahl and Li for their thoughtful comments on the non‐anticoagulant properties of heparin. Heightened awareness of hypercoagulability has made heparin part and parcel of the COVID‐19 management algorithms. In addition, reports of prophylactic anticoagulation failure have triggered several trials in which escalated doses of heparin are compared with standard doses with the aim of preventing thrombotic complications. At this juncture, we do need to consider where do the non‐anticoagulant properties of heparin fit in the COVID‐19 clinical context?

As the letter suggests and based on several experimental studies, heparin has several non‐anticoagulant properties that are likely to benefit symptomatic COVID‐19 patients. These include anti‐inflammatory, anticomplement, antiplatelet, and endothelial quiescence in addition to even being antiviral. , But, we do have to bear in mind that many of these data are from research laboratories and have not found their way into mainstream clinical use. This may be due to a lack of awareness of these various “outside the box” properties of heparin. Also, these functions are always considered as ancillary to the anticoagulant role rather than considered on their own accord. In other words, the non‐anticoagulant properties are only considered in patients who are anticoagulated for proven thrombosis or having high risk for thrombosis. In addition, concerns of bleeding dissuade heparin use in patients with an increased risk of bleeding—for example, critically ill septic patients—which are exactly the patients for whom the non‐anticoagulant properties may have an important role. Heparins without anticoagulant function (heparin analogues) may be the answer here but once again we need proof of efficacy in clinical studies. Until then, we have to continue using conventional heparin, while trying to identify ways to demonstrate the non‐anticoagulant effects of heparin clinically.

One of the well‐known complications of COVID‐19 is acute lung injury, caused by the virus itself and virus‐induced inflammation. Activated leukocytes in the setting of inflammation degrade glycocalyx, the physiological endothelial protective barrier. This allows transmigration of the cells relevant to the inflammatory process causing vascular leakage, the severity of which can correlate with the degree of hypoxemia. In experimental animals with sepsis, both unfractionated and low molecular weight heparins abrogated this complication. The clinically relevant question here is when should we intervene with heparin because inflammation in the early stages is clearly a host defence mechanism, but, if continued unabated, could lead to irreversible damage. The early, physiological response is clearly part of host defence and unlikely to be harmful but the latter stages, during which the markedly increased vascular leakage has stopped adequate gas exchange, would be pathological. Hence, considering heparin administration at early signs of hypoxia may be the most appropriate time for exploiting its anti‐inflammatory, protective function. It would of course be ideal to have biomarkers which can “detect” the increased vascular leakage to aid appropriate timing of heparin commencement.

It's not just the wrong timing behind the failure of heparin trials for acute respiratory distress syndrome (ARDS) effectiveness, but probably also the dosing regimen. If early intervention is possible, prophylactic doses of heparin may be adequate. Prophylactic doses of heparin rarely cause increased bleeding even when mild to moderate renal impairment or thrombocytopenia (>25 × 109/L) complicate the clinical scenarios in which the non‐anticoagulant properties may be desirable. On the other hand, if the inflammatory process has progressed, much higher—even therapeutic—doses of heparin may be required, which clearly has an increased bleeding risk in these complex patients. Needless to say, one of the reasons for the increased bleeding in critically ill patients is the marked endothelial dysfunction and leakage, which would obviously be compounded by therapeutic heparin administration.

How may the non‐anticoagulant properties of heparins be relevant to the COVID‐19 patient? If we examine the patient journey, symptoms from the SARS CoV‐2 usually start in the pre‐hospital setting. Persistent fever for 48 to 72 hours and/or extreme fatigue could be considered as signs of continued inflammation in COVID‐19. If these symptoms are associated with a decrease in oxygen saturation; it may be prudent to commence prophylactic doses of heparin to benefit from its anti‐inflammatory properties. In the current pandemic, many of these patients may not attend hospitals until they are much more symptomatic, when the continued inflammation would progress and have a significant impact on their prognosis. Once the patient reaches the hospital, current trials look at intensifying heparin doses even in the absence of thrombosis but we need to worry about higher bleeding risks with this approach with possibly little anti‐inflammatory benefit. It may be better to continue prophylactic doses of heparin and consider additional anti‐inflammatory agents to maximize therapeutic benefits with minimal drug‐related complications.

There have been suggestions for considering the more practical, direct oral anticoagulants (DOACs) instead of parenteral heparins in the pre‐hospital COVID‐19 setting. Although a more convenient and reasonable approach, we need evidence for their non‐anticoagulant properties, which is currently lacking but an interesting research prospect. Similarly, questions arise about post‐discharge thromboprophylaxis of COVID‐19 patients who did not develop a thrombotic complication—are they protected from future complications (including thrombosis) with the use of low‐dose DOACs or prophylactic dose low molecular weight heparins? Hopefully, the several ongoing studies will provide us effective answers soon.

CONFLICT OF INTEREST

None.