This research aims to inform patient management of venous thromboembolism (VTE) during the present Coronavirus disease 2019 (COVID-19) pandemic. VTE refers to the blood clots in the veins usually formed in deep veins (deep vein thrombosis (DVT)) in the lower leg, thigh, or pelvis. When the clots break off and travel up to the lung, blood flow to part of the lung is blocked, resulting in pulmonary embolism (PE). The main reasons for the blood clots are the damage of the vessel’s wall, the change of blood flow and the thickness of the blood. This has raised concerns that lockdown measures affecting people’s daily physical activities might have increased the risk of DVT and thus PE. Autopsy findings of Covid-19 related mortality have identified severe vessel injury and widespread thrombosis in the lungs. This has implied that Covid-19 virus may increase the risk of PE as it mainly enters the body through the lungs and may cause thrombosis directly in pulmonary vessels. We are conducting two studies to test these hypotheses. To estimate whether lockdown measures have increased the risk of VTE incidence, we will map the chronological trends of incident VTE from 2015 to date to see if lockdown measures during pandemic time contributes to VTE incidence among Covid and non-Covid patients. To estimate whether Covid-19 associated with the risk of VTE, we will compare the risk for VTE patients according to Covid-19 diagnosis meanwhile allowing for other differences including demographic characteristics, underlying health conditions, and certain drug prescriptions.
The SARS-CoV-2 virus, which causes Covid-19, enters the human body mainly through the respiratory system. Covid19 increases the risk of thromboembolic events by including an inflammatory and high hypercoagulable state, as has been reported among hospitalized Covid-19 patients. This has raised concerns that Covid-19 induced immune thrombosis may lead to increasing risks of venous thromboembolism (VTE). Additionally, given that the lockdown measures have adversely impacted people’s physical activity levels, and immobility is a proven risk factor for deep vein thrombosis (DVT), which in itself is a common precursor for pulmonary embolism (PE), it has been suspected that VTE incidence might be increasing among at-risk populations during the pandemic. We propose two related analyses with primary care electronic health records to test these hypotheses. We will conduct a cohort study using GP health records to map the chronological changes of VTE incidence from 1st January 2015 to 30th April 2021. An interrupted time series analysis will be conducted by fitting hierarchical Poisson regression models to estimate changes in VTE incidence after the first lockdown in March 2020 and during the first and second wave of Covid-19, adjusting for age group, sex, ethnicity and general practices. A matched cohort study will be conducted, comparing Covid-19 patients with controls, matched for age, sex and general practice, to compare the incidence of VTE cases in relation to Covid-19 exposure. The second study is nested within the longer period being from 29th January 2020 onwards. Because 29th January 2020 was the official date of the UK's first confirmed COVID-19 case. We will estimate the incidence rate ratio of VTE associated with Covid-19 diagnosis in the preceding 28 and 84 days, and compare Covid-19 patients with non-Covid-19 patients, matched for general practice and index date using Poisson regression models to quantify the relative rates of VTE.
Thank you for the helpful comments.
Read/SNOMED-CT codes for venous thromboembolic disease have been added as appendixes. We also request HES APC for case identification. ICD-10 code lists for VTE are added in appendixes.
Given that patient with a confirmed incident PE would necessarily receive anticoagulant treatments, we further restrict incident PE cases to the ones with relevant clinical diagnoses receiving at least one prescription of anticoagulant. In order to differentiate between PE and DVT being recorded as VTE, information on diagnostic assessments to establish PE diagnoses 90 days before and after the index date will be sought after from the Observation file of CPRD Aurum. Incident VTE cases with clinical tests documented for PE within 90 days receiving at least one prescription of anticoagulant will be classified as incident PE. The remaining incident VTE cases with investigatory tests for DVT within 90 days will then be categorised into DVT incidence group.
We now revise the outcomes to be measured into:
Clinical diagnoses of incident PE receiving at least one prescription for anticoagulant in CPRD Aurum dataset
Clinical diagnoses of incident VTE with diagnostic tests for PE within 90 days of index dates in CPRD Aurum dataset
Clinical diagnoses of incident VTE with diagnostic tests for DVT within 90 days of index dates in CPRD Aurum dataset
Clinical diagnoses of incident DVT in CPRD Aurum dataset
Record of concurrent DVT and PE in CPRD Aurum dataset
Clinical diagnoses of incident PE in HES APC
Clinical diagnoses of incident DVT in HES APC
Record of concurrent DVT and PE in HES APC
Xiaohui Sun - Chief Investigator - King's College London (KCL)
Xiaohui Sun - Corresponding Applicant - King's College London (KCL)
Alex Dregan - Collaborator - King's College London (KCL)
David Morris - Collaborator - King's College London (KCL)
Martin Gulliford - Collaborator - King's College London (KCL)
HES Admitted Patient Care;Practice Level Index of Multiple Deprivation