Select COVID-19 Patients May Benefit From Postdischarge Anticoagulation

The risk of venous thromboembolic (VTE) events in patients discharged after a COVID-19 hospitalization is relatively low, but the presence of certain features may justify consideration of extended thromboprophylaxis, researchers suggest.

Specifically, patients with a history of VTE, a D-dimer level above 3 μg/mL, or a predischarge C-reactive protein (CRP) level greater than 10 mg/dL had greater odds of VTE events within the first 90 days after leaving the hospital, lead author Pin Li, PhD (Henry Ford Health System, Detroit, MI), and colleagues report in a study published online this week in JAMA Network Open.

On the other hand, patients who received anticoagulation, particularly at therapeutic doses (as opposed to lower prophylactic doses), had markedly reduced odds of VTE events.

“We agree that routine anticoagulation is not recommended for every COVID patient after their hospital discharge,” study co-author Wei Zhao, MD, PhD (Ascension St. John Hospital, Detroit), told TCTMD. “We propose that anticoagulation may be considered in the high-risk populations, and we have to make sure that their bleeding risk is low.”

That stance is consistent with guidance from the US National Institutes of Health (NIH), which discourages routine use of VTE prophylaxis at discharge but states that extended anticoagulation “can be considered” in patients with a low bleeding risk and high thrombosis risk according to established risk scores.

But Tracy Wang, MD (Duke Clinical Research Institute, Durham, NC), indicated that it remains unclear whether this is a valid approach.

“We are seeing lower rates of thromboembolism compared with earlier during the pandemic, but this is still a sick population,” she told TCTMD via email. “Looking at this study, even if you survive to discharge, the incidence of 90-day mortality is still 4%. Some of this may be explained by known VTE/[arterial thromboembolism], some of these may be undiagnosed thromboembolic complications. I think there’s also a lot more work needed to understand what kind of anticoagulants to use, how are they dosed, and how long we would need to use them for.”

Higher Risks Seen in Certain Patients

Once it became clear that COVID-19 comes with a greater risk of arterial and venous thrombosis, studies were launched to determine how antithrombotic therapies should be deployed to counteract it. In some centers, patients have been discharged on short courses of anticoagulation after discharge. It’s not clear, though, that this strategy is of benefit considering the low rate of VTE events observed after patients leave the hospital. “Universal prescription of postdischarge anticoagulation in patients with COVID-19 offers marginal clinical benefits and may cause harm in patients at high risk of bleeding,” Li et al note.

To explore the potential utility of a more-targeted approach, the investigators examined data on 2,832 adults (mean age 63.4 years; 47.6% men) hospitalized with COVID-19 across five hospitals in the Henry Ford Health System between March and November 2020. In contrast to previous smaller studies, the large size of this cohort makes it possible to identify risk factors for VTE events in the posthospital setting, Li said.

The study confirms the low overall risk of thrombosis after discharge, with 1.3% of patients having a VTE event within 90 days and 0.5% having an arterial event (mostly ACS). The risk of venous events, but not arterial, declined over time.

On multivariate adjustment, there were three factors independently associated with a greater likelihood of postdischarge VTE events:

• History of VTE (OR 3.24; 95% CI 1.34-7.86)
• Peak D-dimer level > 3 μg/mL (OR 3.76; 95% CI 1.86-7.57)
• Predischarge CRP level > 10 mg/dL (OR 3.02; 95% CI 1.45-6.29)

Roughly one-quarter of the cohort received anticoagulation at discharge, including 6.6% at a prophylactic dose and 17.4% at a therapeutic dose. Postdischarge anticoagulation was associated with fewer VTE events (OR 0.14; 95% CI 0.03-0.58), a difference driven by significantly lower odds in patients who received therapeutic doses (OR 0.18; 95% CI 0.04-0.75).

Lingering Uncertainty

Zhao said because there haven’t been strict guidelines for how to handle extended VTE prophylaxis after discharge in patients hospitalized for COVID-19, practice has varied. The NIH guidelines suggest using the IMPROVE and modified IMPROVE VTE tools developed prior to the pandemic, as well as D-dimer levels, to identify higher-risk patients who might benefit from extended anticoagulation, although these scores have not been validated in COVID-19 populations.

“Reliance on the IMPROVE tool to guide postdischarge anticoagulation in patients with COVID-19 should be done cautiously because two of three risk factors identified in this study are not reflected in the IMPROVE tool,” Li et al write. “Furthermore, the weighted risk factors in the IMPROVE tool, such as thrombophilia, active cancer, ICU stay, and age, were not associated with postdischarge VTE in our cohort.”

The current study provides risk factors that are more specific to patients hospitalized with COVID-19, Zhao said, adding that he has not yet put these findings into practice because the data are new and require discussion within the medical community.

Wang pointed out that “a lot of the evidence on which postdischarge thromboprophylaxis was based was extrapolated from trials of hospitalized medically ill [patients] in the pre-COVID era, studies like APEX, MARINER, etc. Back then, the primary determinants of risk were related to degree of preexisting or hospital-imposed immobilization, as characterized by risk scores such as the IMPROVE VTE score.”

The MICHELLE trial, presented at the virtual European Society of Cardiology Congress 2021 at the end of the summer, provided some needed information in the setting of COVID-19. Trial investigators defined COVID-19 patients with high VTE risk as those with an IMPROVE VTE score ≥ 4 or an IMPROVE VTE score of 2 or 3 along with a D-dimer level > 500 ng/mL, showing that rivaroxaban (Xarelto; Bayer/Janssen) after discharge improved outcomes versus no intervention.

“The difference between treatment arms was quite striking,” Wang said, noting that “the majority of the patients in this trial qualified by the combination of a moderately high IMPROVE score with a D-dimer elevation.”

As for why the current study didn’t find an association between many of the components of the IMPROVE score and VTE risk, Wang said it “may in part be explained by the pathophysiology of COVID in and of itself.”

It wouldn’t be too difficult to integrate the high-risk factors identified by Li et al into clinical decision-making, Wang indicated, but she urged caution: “We need to be careful about generalizing the results too far given the limitations of the (albeit well-analyzed) study. In this study population, almost half of the patients had a CRP > 10 mg/dL, so if you really want to be selective about who to anticoagulate, you would need to take into consideration other factors.”

Wang is the principal investigator of the ongoing ACTIV-4c trial of posthospital thromboprophylaxis in COVID-19 patients, which will provide additional insights.

The trial “is still recruiting, and unfortunately for the world, there is an uptick in hospitalizations for COVID, so we’re busy recruiting right now, which will influence when first results will be available,” she said. “Also keep in mind that the ACTIV trials are adaptive trials, designed to leverage existing research resources to answer questions raised by a changing disease. This study [by Li et al] and others will certainly provide food for thought on any adaptations we might need to make to ensure relevant scientific and clinical insights from our trial.”