Patchy Anticoagulation Use During COVID Pandemic Raises Questions

Despite recommendations advising prophylactic anticoagulation in all hospitalized patients with SARS-CoV-2 infection, a significant proportion received none during some of the worst periods of the pandemic, according to data from the American Heart Association’s COVID-19 Cardiovascular Disease Registry.

The look back at how patients were treated between 2020 and 2022 also found wide-ranging variability in use and dosing of prophylactic anticoagulation across US hospitals.

“What these data show and why they're important is that despite the debate shifting from whether or not someone should be prophylactically anticoagulated to what dose they should be on, there was still a large proportion of patients—one in six—that received nothing,” said Mathew S. Lopes, MD (Brigham and Women’s Hospital, Boston, MA).

“That decision seemed to be driven not by your risk coming in, but really by where you were and what hospital you went to,” he added.

In response to early reports, like a New York study of hospitalized patients finding thrombotic events in 16% and others noting VTE in nearly 70% of those requiring intensive care, by 2021 the recommendation was for prophylactic anticoagulation in all patients hospitalized with SARS-CoV-2 infection.

The study was published this week online ahead of print in the Journal of the American Heart Association.

Large Variation in Prescribing and Dosing

Lopes and colleagues used the registry to examine data on 26,775 individuals (median age 60 years; 48% women) hospitalized for COVID-19. Of those, 28% spent time in the ICU during the hospitalization period.

Anticoagulation was not given in 16%. Of those who did receive it, the most commonly used agents were unfractionated heparin and low-molecular-weight heparin. A low dose was given in 58%, an intermediate dose in 11%, and a full dose in 15%.

Compared with patients who received anticoagulation, those who did not were younger, more often white and women, and tended to have fewer comorbidities as well as lower levels of C-reactive protein and ferritin. Patients who received full-dose versus low-dose anticoagulants were more often men, had higher rates of diabetes and hypertension, and had higher levels of natriuretic peptide, C-reactive protein, and ferritin.

Rates of anticoagulation among the hospitals in the registry ranged from 0 to 98%, with a similar pattern seen for full-dose anticoagulation.

The possibilities that a patient would receive any versus no anticoagulation or a full versus low dose were 2.8-and threefold higher from one hospital to another even after accounting for identical covariates of age, sex, race, body mass index (BMI), corticosteroid use, ICU admission, mechanical ventilation, and dichotomized admission date. These odds did not change over the study period.

In multivariable analysis, patient-level and hospital-level factors associated with an increased odds of anticoagulation included older age, male sex, nonwhite race, higher BMI, higher platelet count, corticosteroid use, and ICU admission.

Similarly, patient- and hospital-level factors associated with an increased odds of receiving full-dose anticoagulation included older age, male sex, higher BMI, higher platelet count, corticosteroid use, ICU admission, and need for mechanical ventilation.

By region, patients at hospitals in the Midwest had the lowest likelihood of receiving any anticoagulation, with wide variation seen among the other regions.

To TCTMD, Lopes said a surprising and disappointing finding was that there was an increase in the proportion of patients who received no anticoagulation over the study period, up to and including February 2022. Much of this upward swing was seen in patients who were not admitted to the ICU (P for trend < 0.0001). This was accompanied by a downturn in use of low-dose anticoagulation (P for trend = 0.02).

“There was an evolution of strains throughout the course of the COVID pandemic, so it's not entirely clear whether thrombotic risk changed as the pandemic went on. But, I think there may have been a false assumption that has yet to be crystallized as to whether subsequent variants of COVID carry less thrombotic risk,” he noted.

Another potential possibility is that clinical inertia was at play.

“Despite the recommendations, some physicians and regions may have been still going about asking themselves when a patient [came] in using common risk scores whether or not they should be anticoagulated,” Lopes added.

The ability to look back now on what happened is a luxury that clinicians at the height of the pandemic could only have hoped to have, but Lopes said it’s important to study what happened, and when, so that future management of hospitalized COVID patients can be applied with clarity.

“What I would want folks to take away is that [these data] should be used as justification for more implementation science from randomized trials to guidelines, to then implementing those guidelines [while] trying to minimize regional variation and focus less on care that's dictated by where a person presents, and more on the severity of illness in the patient in front of you,” he said.