With new research on Long Covid emerging every day, it is becoming increasingly clear that Covid-19 infection impacts our health beyond the acute stage of the illness. A study demonstrates that infection with Covid-19 impacts the risk of cardiovascular events up to 12 months post-infection, regardless of age, race, sex, and other cardiovascular risk factors. This study emphasizes the recognition of and the need for more effective strategies to address the long term effects of Covid-19.
A group of researchers from the Clinical Epidemiology Center in Saint Louis, Missouri investigated the risk and excess burden of cardiovascular disease following the acute phase of Covid-19. In the study, over 150,000 veterans who had recovered from infection were compared with non-infected peers, in addition to a pre-pandemic control group. Xie and colleagues followed these three groups for twelve months and conducted a thorough analysis to estimate the risk and associated burden of cardiovascular outcomes.
Who and What Factors were Considered?
Xie and colleagues obtained information from databases managed by the United States Department of Veteran Affairs to construct their three cohorts. Of the 6,241,346 veterans who encountered the department of Veteran Affairs in 2019, 162,690 had a positive Covid-19 test between March 1st 2020 and January 15th 2021. 153,760 of such individuals were alive 30 days after their positive test date and selected into the Covid-19 test group. Researchers based the average date of the positive test for the cohort, T0, based on the distribution of the positive Covid-19 test dates.
5,960,737 veterans who encountered the department of Veteran Affairs in 2019 and were alive by March 1st, 2020; 5,806,977 of such individuals were not part of the Covid-19 group and were selected into the contemporary control group. Xie et al randomly chose the average enrollment date for the cohort, T0, which would allow for the participant distribution to be identical to the distribution of the Covid-19 group. In doing so, the contemporary control group and the Covid-19 cohort had similar follow-up times.
Xie et al selected individuals who encountered the department of Veteran Affairs in 2017 to be part of the pre-pandemic cohort. Of the 6,461,205 veterans, 6,150,594 of them were alive by March 1st 2018. 6,008,499 were not included in the Covid-19 cohort and were further chosen to be included in the pre-pandemic group. Researchers randomly selected the average enrollment date for the group, T0, which would allow for the participant distribution to be the same as the distribution of the Covid-19 group. This action similarly ensured that the historical control group and Covid-19 group had identical follow-up times.
The cardiovascular outcomes assessed were based on Xie and colleagues previous work regarding Long Covid. These outcomes include cerebrovascular disorders, dysrhythmia, inflammatory heart disease, ischemic heart disease, thrombotic disorders, and other cardiovascular diseases. Researchers conducted a follow-up for each cardiovascular outcome that the participant had no prior history with one year before their enrollment date. The follow up period began 30 days after the average enrollment date, T0, and ended by October 31st 2021.
Xie et al considered pre-defined and algorithmically selected variables to account for any baseline differences between the cohorts. Previous studies have shown that race, sex, body mass index, Area Deprivation index, smoking status, frequency of hospitalization, and use of long term care can influence risk and associated burden of cardiovascular outcomes. Xie and colleagues identified
cancer, chronic kidney disease, diabetes, and several other comorbidities as variables to also consider. In addition, they adopted an algorithm that determined the top 100 variables with the highest risk relative to the cardiovascular outcome and cohort. The program factored the diagnoses, medications, and lab abnormalities common in at least 100 members of the cohort. Researchers tested each cardiovascular outcome within each cohort independently so the algorithm could be applied.
Recovered from Infection Vs. Never Infected Vs. Pre-Pandemic: What Happened?
In the study, Xie et al estimated the risk, burden, and excess burden up to 12 months post infection of cardiovascular outcomes through inverse probability weighting, a common method used to estimate the probability of exposure observed for a particular person and using the value as a weighting factor in further analyses.
In this case, researchers calculated a propensity score, which describes the chance of being selected to the target population, for individuals who had no prior history (up to one year before enrollment) for each specific cardiovascular outcome. These scores estimated the probability of being a veteran who encountered the department of Veteran Affairs in 2019 based on the pre-defined and algorithmically selected variables; they were then used to calculate the inverse probability weight of being part of the sample who interacted with the department in 2019. Additionally, researchers applied the weight scores in hazard ratio models for each cardiovascular outcome. These models estimated the risk of each cardiovascular complication caused by Covid-19, using death as the other, competing risk.
Among the pre-specified cardiovascular outcomes, those who recovered from the acute phase of infection had higher risk of all pre-specified cardiovascular outcomes in comparison to those who were not infected. The composite scores for each group of cardiovascular outcome were above one, indicating increased risk: cerebrovascular disorders scored 1.53±0.16 , dysrhythmias scored 1.69± 0.11, inflammatory heart disease scored 2.02 ± 0.53, ischemic heart disease scored 1.66± 0.28, thromboembolic disorder scored 2.39± 0.24, and other cardiovascular disorders scored 1.72± 0.14 . In particular, veterans who survived the first 30 days following a positive test result exhibited higher risk of myocarditis (5.38 ± 3.79), cardiac arrest (2.45 ±0.81), cardiogenic shock (2.43 ± 1.30), and pulmonary embolism (2.93 ± 0.42).
Xie et al estimated the associated burden of cardiovascular outcomes caused by Covid-19 per 1,000 people at twelve months based on the differences between the estimated rate of Covid-caused cardiovascular outcomes caused in the Covid-19 cohort and the contemporary control cohort. They found similarly high and excess burdens with all cardiovascular outcomes. The associated excess burden for heart failure (11.61 ± 2.78) and atrial fibrillation (10.74 ± 2.3) were especially high.
Taking care setting into consideration, risks and associated burdens persisted among those who were not hospitalized for infection, which gradually increased with severity of infection. Participants who were not hospitalized, representing the majority of the US general population, had higher risk and excess burden for cardiovascular disease than those who were not infected by Covid-19. Those who were hospitalized for infection had higher risk and associated burdens than those who did not, and those who were admitted to intensive care had the highest risk of cardiovascular disease and excess burden.
Relative to the pre-pandemic control group, participants who recovered from Covid-19 infection had higher risk and excess burden of any pre-specified cardiovascular disease per 1,000 individuals in one year. The results of each assessment were consistent to what was found when comparing the covid-19 cohort with the contemporary control group.
A few studies suggested a potential association of some Covid vaccines and a very rare risk of heart or pericardium inflammation; Xie et al conducted two analyses to diminish any potential impact vaccine exposure may have had on cardiovascular outcomes. In their first test, researchers excluded participants who received their first dose of a Covid-19 vaccine. In the second assessment, they considered vaccination status as a time-sensitive covariate. In both analyses, Covid-19 was associated with higher risk of both heart and pericardium inflammation. This set of findings reinforces the importance of getting vaccinated.
Xie and colleagues validated their analytical approach by testing variables with expected outcomes. They tested the association between Covid-19 and the signature risk of fatigue; Covid-19 increased the risk of experiencing fatigue, as expected. Researchers tested the association between receiving a flu shot on even-numbered versus odd-numbered calendar days and the pre-specified cardiovascular outcomes. Following the same analytical approach and resources as the study, they found no significant association between the even versus odd-numbered calendar day of the influenza shot and pre-specified cardiovascular outcomes.
The cardiovascular disease risk associated with Covid-19 infection further highlights how we need a coordinated global response strategy to urgently address the challenges of dealing with the long-term health effects of Covid-19. Physicians should also be adjusting their screening questions to include past infection with Covid-19 and assess for all Long Covid symptoms including cardiovascular. Early identification, diagnosis, and treatment of heart disease are essential to lessen the risk of adverse health impacts.