Rebuttal From Dr David Jimenez et al.

We thank our colleagues for their thoughtful insights in describing the potential role of therapeutic heparin in patients with COVID-19 who are not critically ill. Therapeutic anticoagulation is not a minor intervention and is associated with harm, mainly in the form of major bleeding. For this reason, for instance, there have been decades of research to establish the optimal duration of anticoagulation after an episode of VTE. Therefore, we would like to take an opportunity to defend that therapeutic heparin use might not be generalized to all patients with COVID-19 who are not critically ill.

Our colleagues enthusiastically cite the literature that might support the role of empiric therapeutic anticoagulation in reducing the mortality rate in sepsis outside the COVID-19 setting. In the quoted meta-analysis, the risk ratio for death comparing heparin with placebo or usual care was 0.88 (95% CI, 0.77 to 1.00). After exclusion of the three placebo-controlled trials, the mortality rate was not reduced significantly with the use of heparin (risk ratio, 0.67; 95% CI, 0.34 to 1.30). To our knowledge, none of the existing guidelines or clinical practice statements recommend (or even suggest) the use of therapeutic heparin in this scenario. Specifically for patients with COVID-19 with critical illness, multiplatform randomly controlled trials (mpRCT) that were evaluating the role of therapeutic heparin were stopped because of futility.

Many colleagues, including Tritschler et al, invoke both the time window and the pleiotropic hypotheses to reconcile the apparently contradictory effects of therapeutic anticoagulation for hospitalized (critically ill and not critically ill) patients with COVID-19. According to the first hypothesis, the beneficial effect of therapeutic anticoagulation is diminished in patients with progressively more severe disease. If this were the case, the mpRCT for patients who were not critically ill might have shown a lower effect of therapeutic heparin in the lowest levels of the organ support free days scale. However, as correctly pointed out by our colleagues, this was not the case. The pleiotropic hypothesis suggests that heparin has antiinflammatory and immunomodulatory properties beyond anticoagulation. Although these properties have been postulated many times for specific subgroups of patients who did not have COVID (eg, cancer patients), studies have never been able to demonstrate a clinical benefit so far. According to the prinicipe of parsimony, our interpretation of the results of randomized trials that assess the efficacy and safety of therapeutic anticoagulation for patients both with and without critical illness who are hospitalized with COVID-19 is simple: significant reduction in VTE, significant increase in major bleeding, and no significant reduction in mortality rate. Only in the mpRCT therapeutic heparin showed a significant reduction in the proportion of patients without requirement of organ support. However, (1) it had an open-label design, (2) participants assigned to each trial arm were recalculated based on a single interim analysis in 2020 to favor randomization to the therapeutic dose arm, (3) roughly 40% of the patients who were enrolled the trial were not receiving standard of care treatment (ie, steroids) for COVID-19, (4) there was no significant difference in mortality rates, and (5) since the median value for OSFD was identical in both groups, the study had to report the proportion of patients in each treatment group who survived until hospital discharge without receipt of organ support.

In the recent times, there has been increasing vaccine coverage (with massive reductions in serious disease among populations with high vaccination rates), and the standard of care for patients with COVID-19 has evolved (including, but not limited to, monoclonal antibodies, dexamethasone, or tocilizumab). Also, the newest variants like omicron have milder forms of clinical presentation. Taken all together, we wonder whether the thrombogenic potential of SARS-CoV2 might have decreased. This may tilt the balance of risk vs benefit of therapeutic anticoagulation in coming times.

At this point, International Society of Thrombosis and Haemostasis, CHEST, and the National Institutes of Health suggest the use of therapeutic heparin for patients who are not critically ill, who are hospitalized with COVID-19 and who have low bleeding risk (Table 1 ). If such a treatment is chosen outside of a trial, shared decision-making should be made with patients about potential benefits, as well as existing uncertainties of this choice. We also would like readers to be aware that an American College of Clinical Pharmacy expert panel report on COVID-19 anticoagulation is subject for periodic update because more evidence may emerge in coming times on this very critical issue (ie, FFREEDOM trial [NCT04512079]).

Table 1

High Bleeding Risk

Bleeding within last 30 days that requires acute care setting

History of inherited or acquired bleeding disorder

Hemoglobin < 8 g/dL

Platelet count < 50 × 109/L

Dual antiplatelet agents

Bleeding risk should be individualized and discussed on a case-by-case basis.