Remote history of VTE is associated with severe COVID‐19 in middle and older age: UK Biobank cohort study

Background Venous thromboembolism (VTE) is a common, life‐threatening complication of COVID‐19 infection. COVID‐19 risk‐prediction models include a history of VTE. However, it is unclear whether remote history (>9 years previously) of VTE also confers increased risk of COVID‐19. Objectives To investigate possible association between VTE and COVID‐19 severity, independent of other risk factors. Methods Cohort study of UK Biobank participants recruited between 2006 and 2010. Baseline data, including history of VTE, were linked to COVID‐19 test results, COVID‐19‐related hospital admissions, and COVID‐19 deaths. The risk of COVID‐19 hospitalization or death was compared for participants with a remote history VTE versus without. Poisson regression models were run univariately then adjusted stepwise for sociodemographic, lifestyle, and comorbid covariates. Results After adjustment for sociodemographic and lifestyle confounders and comorbid conditions, remote history of VTE was associated with nonfatal community (RR 1.61, 95% CI 1.02–2.54, p= .039), nonfatal hospitalized (RR 1.52, 95% CI 1.06–2.17, p= .024) and severe (hospitalized or fatal) (RR 1.40, 95% CI 1.04–1.89, p= .025) COVID‐19. Associations with remote history of VTE were stronger among men (severe COVID‐19: RR 1.68, 95% CI 1.14–2.42, p= .009) than for women (severe COVID‐19: RR 1.07, 95% CI 0.66–1.74, p= .786). Conclusion Our findings support inclusion of remote history of VTE in COVID‐19 risk‐prediction scores, and consideration of sex‐specific risk scores.


| INTRODUC TI ON
Venous thromboembolism (VTE), which includes pulmonary embolism and deep vein thrombosis, is a common and a serious complication of severe COVID-19 infection. 1 Despite anticoagulant thromboprophylaxis being recommended and empirically used for all hospitalized COVID-19 patients, 2 larger studies suggest that 7% to 8% still develop VTE, as do 46% of those admitted to intensive care units. 3 Many randomized controlled trials are now evaluating antithrombotic agents and intensities of anticoagulation for patients hospitalized with COVID-19. 4 Recent publications cast uncertainty about the best course of action but potentially patients who are less sick might benefit from more intense anticoagulation but for patients with severe disease it may be too late. 5 The potential benefits of more intense anticoagulation must be weighed against the risk of bleeding, which is also increased with COVID-19 infection. 6,7 Following discharge from hospital, there is an elevated risk of both thrombosis and hemorrhage, so universal postdischarge anticoagulation or antiplatelet therapy needs further evaluation, too. 8 Risk stratification for VTE complications in COVID-19 patients is needed to improve the risk-benefit ratio of targeted anticoagulation strategies. 9 In the most recent QCovid predictive score (https:// qcovid.org/), 10 history of VTE was identified, and included as a risk factor for both COVID-19 severity and VTE complications after COVID-19 infection, but no differentiation was made between remote and recent history of VTE. It is uncertain whether a remote history of VTE is a risk factor for severe COVID-19 infection. We used UK Biobank data, collected >9 years before the COVID-19 pandemic, to address this question.  16 March 2020 and 22 July 2020. COVID-19 infection was defined as more than one COVID-19-positive polymerase chain reaction test or COVID-19 stated on death certificate and categorized into: nonfatal community, nonfatal hospitalized, and died from COVID-19 (in hospital or in community). Severe COVID-19 was defined as hospitalization for or death from COVID-19 and ascertained by combining the latter two categories. These data were linked to the UK Biobank participants' data collected at enrollment. smoking (never, past. or current smoker); physical activity, derived from the questions about moderate and vigorous physical activity in the International Physical Activity Questionnaire, converted into metabolic equivalents (METs) and dichotomized into inactive (<600 MET min/week) and active (>600 MET min/week) 13 ; and frequency of alcohol consumption (never, special occasions only, one to three times a month, once or twice a week, three or four times a week, and daily or almost daily). Self-report of physician diagnoses included cardiovascular disease (myocardial infarction, heart failure, COVID-19. Associations with remote history of VTE were stronger among men (severe COVID-19: RR 1.68, 95% CI 1.14-2.42, p = .009) than for women (severe COVID-19: RR 1.07, 95% CI 0.66-1.74, p = .786).

Conclusion:
Our findings support inclusion of remote history of VTE in COVID-19 risk-prediction scores, and consideration of sex-specific risk scores. • COVID-19 prediction scores include history of VTE but it is unknown if remote history is relevant.
• UK Biobank data show an independent association between remote history of VTE and severe COVID-19.
• Our findings support inclusion of remote history of VTE in COVID-19 risk-prediction scores.
angina, stroke, transient ischemic attack, atrial fibrillation/flutter, valve disease) and diabetes mellitus (type 1 and type 2). Women were categorized as currently receiving exogenous estrogens (combined oral contraceptive pill or hormone replacement therapy) or not. Medication included any antiplatelet and anticoagulant medication taken regularly at the time of recruitment.

| Statistical analyses
The characteristics of the study population, broken down by history of VTE or not, were summarized using the median and interquartile range for continuous variables and frequencies and propor-  Participants who reported a history of VTE at baseline were older, more deprived, and more likely to be female and white (Table 1). They were less physically active, more likely to be obese and current or past smokers, more likely to have comorbid diabetes and cardiovascular disease and be on regular antiplatelet or anticoagulant therapy, but were less likely to consume alcohol weekly or daily (Table 1)  Also, patients with a remote history of VTE treated with antiplatelet agents or anticoagulants appeared to be protected from developing more severe COVID-19. This is interesting, because activation of hemostatic pathways, in the micro-or macro-circulation, may be an important mechanism determining the severity of COVID-19. Antithrombotic prophylaxis with, for instance, a low dose of a novel oral anticoagulant and/or aspirin might reduce the risks of developing severe COVID-19. 15 In the COMPASS trial, Finally, ascertainment of COVID-19 was dependent on the testing strategy in operation at the time. In our report, 70% of cases were severe, which is a much higher proportion than expected. It is likely that participants with mild or asymptomatic infections did not seek medical attention and were therefore not tested and, con-

| CON CLUS IONS
Our findings support inclusion of any history of VTE, including remote, in the QCovid risk score and other risk scores. They show that the risk was specific to men. The lack of an association in women needs to be corroborated in other studies but sex-specific risk scores should be considered. These findings also support testing anticoagulants and antiplatelet agents in the prevention of more severe COVID-19 outcomes. Individuals with a history of VTE should be considered a potential priority group for vaccination.

ACK N OWLED G M ENTS
We thank the UK Biobank participants. This research was conducted using the UK Biobank Resource under Application Number 7155.

CO N FLI C T O F I NTE R E S T
There is no conflict of interest to report.

AUTH O R CO NTR I B UTI O N S
Jill P. Pell, Naveed Sattar, and John G.