COVID-19 not linked to increased long-term risk for CVD or diabetes
Click Here to Manage Email Alerts
People infected with COVID-19 do not have increased long-term risks for developing cardiovascular disease or diabetes, according to a study published in PLOS Medicine.
In findings from a population-based cohort study conducted in the U.K., the increased risk for CVD in people who contract COVID-19 begins to decline 5 weeks after infection, whereas the risk for both type 1 and type 2 diabetes remains elevated up to 23 weeks after infection before dropping back to preinfection levels.
“COVID-19 patients are most at risk of developing CVD and diabetes during the first 4 weeks since initial diagnosis,” Emma Rezel-Potts, PhD, postdoctoral researcher in epidemiology and public health at King’s College London, told Healio. “After this, the risk of CVD begins to decline; however, the risk of diabetes remains elevated by 27% from 4 to 12 weeks after infection.”
Rezel-Potts and colleagues analyzed data from registered patients in the Clinical Practice Research Datalink Aurum, a comprehensive medical record database of patients from 1,356 general family practices in England. Researchers extracted data from 428,650 patients with confirmed or suspected COVID-19 from the database’s February 2021 release. The COVID-19 cohort was matched by year of birth, sex, and family practice with a control cohort of people with no recorded COVID-19 diagnosis. The control cohort was randomly selected from the database’s March 2021 release. First recorded CVD and diabetes diagnoses were the primary outcomes.
Diabetes, CVD risks up 4 weeks after infection
Before contracting COVID-19, the infection group had a slightly higher diabetes incidence rate of 15.81 per 100,000 patient-weeks compared with 11.32 per 100,000 patient-weeks for controls. The infection cohort also had a higher baseline CVD incidence rate of 14.07 per 100,000 patient-weeks compared with 7.58 per 100,000 patient-weeks for controls.
Four weeks after COVID-19 infection, the diabetes and CVD incidence rates increased to 23.79 per 100,000 patient-weeks and 76.92 per 100,000 patient-weeks, respectively, in the infection group. The CVD incidence rate declined to 22.06 per 100,000 patient-weeks at 5 to 12 weeks and to 12.63 per 100,000 patient-weeks at 13 to 52 weeks after infection. Diabetes incidence remained slightly elevated at 19.54 per 100,000 patient-weeks 5 to 12 weeks and 19.57 per 100,000 patient-weeks 13 to 52 weeks after infection.
After adjusting for covariates and allowing for underlying differences between infected patients and controls, COVID-19 infection was associated with an increase in CVD events (adjusted incidence rate ratio [aIRR] = 5.82; 95% CI, 4.82-7.03) and diabetes diagnoses (aIRR = 1.81; 95% CI, 1.51-2.19). At 5 to 12 weeks after infection, there was still an increased likelihood for CVD (aIRR = 1.49; 95% CI, 1.28-1.73) and diabetes (aIRR = 1.27; 95% CI, 1.11-1.46). At 13 to 52 weeks, there was no difference in diabetes incidence between the infection and control groups, and those with a COVID-19 infection had a lower incidence rate of CVD events compared with controls (aIRR = 0.8; 95% CI, 0.73-0.88).
“As the COVID-19 pandemic has continued, evidence from electronic health records has accumulated and enabled us to look at longer-term outcomes,” Rezel-Potts said. “Some of the literature emerging when we began the study was suggestive of heightened cardiometabolic disease risk from 6 months to 1 year postinfection. We may have been able to evaluate the period of elevated CVD and diabetes risk after COVID-19 more precisely because we used statistical approaches that enabled us to include our cohort’s baseline risk for CVD and diabetes diagnoses when predicting their risk after infection.”
Increased diabetes risk persists up to 23 weeks
When IRRs were estimated by 4-week period, there was no evidence that CVD events are increased beyond 8 to 11 weeks after COVID-19 infection, and there were fewer CVD events in the infection group compared with controls from week 20 onward. There was an increased risk for diabetes incidence up to 20 to 23 weeks after a COVID-19 infection (aIRR = 1.29; 95% CI, 1.08-1.55; P = .005), but not beyond 24 weeks.
“It is important to note that COVID-19 patients face an increased risk of diabetes for at least 3 months after infection,” Rezel-Potts said. “We suggest that clinical and public health interventions to reduce diabetes risk are targeted toward patients recovering from COVID-19, such as advice on healthy eating and exercise.”
For more information:
Emma Rezel-Potts, PhD, can be reached at emma.rezel-potts@kcl.ac.uk.
Editor’s note: On July 22, 2022, this article was updated to clarify that both type 1 and type 2 diabetes mellitus were included in the study’s outcome measures.
Perspective
Back to Top
Epidemiological evidence from large cohort studies has linked SARS-CoV-2 infection with a higher incidence of cardiometabolic conditions, including cardiovascular disease, venous thromboembolic disease and diabetes, up to 12 months after infection, but the time period after infection when these risks are highest is unknown.
Rezel-Potts and colleagues used electronic health records data from primary care clinics in the U.K. to look at associations of SARS-CoV-2 infection with incident cardiometabolic conditions. They followed 430,000 people with documented SARS-CoV-2 infection and an equal number of controls without a positive test starting 52 weeks before their COVID diagnosis (or a representative index date for controls) until 52 weeks after diagnosis. Then they looked at incidence of cardiometabolic conditions at 4 weeks, 5 to 12 weeks, and 13 to 52 weeks. They adjusted for factors thought to be associated both with risk of getting COVID and with cardiometabolic diseases.
The cohort averaged aged 35 years, more than 50% female, more than 50% white, with a low burden of comorbidity. There was a substantial amount of missing data on BMI, smoking and blood pressure. Interestingly, even prior to infection, the two groups were not identical in terms of cardiometabolic risk, with higher incidence in the group that went on to get SARS-CoV-2. Up to 4 weeks after the index date, risk of conditions including diabetes, myocardial infarction, heart failure, atrial arrhythmias and venous thromboembolic disease were higher in the SARS-CoV-2 positive group compared to controls. Notably, the RR of pulmonary embolism was 11.5, atrial arrhythmias 6.4 and venous thromboembolic disease 5.4. Five to 12 weeks after the index date, the risk of diabetes was still higher, but risks of CVD and venous thromboembolic disease were not significantly higher in the SARS-CoV-2-positive group compared to controls. Thirteen to 52 weeks after the index date, SARS-CoV-2 infection was not associated with a higher risk of any cardiometabolic condition. In fact, for several conditions the risk was lower in the SARS-CoV-2-positive group during this time period.
These findings are consistent with previous studies which have shown a higher risk of cardiometabolic conditions after SARS-CoV-2 infection, including our own work. In addition, this study provides new information about the time period during which higher risk is seen. For CVD and venous thromboembolic disease, the period of higher risk was not long after infection. This is consistent with existing evidence that suggests that risk of thrombotic events is elevated in acute COVID, and with well-known risks of CV conditions in acute illness in general. For diabetes, risk was higher in the SARS-CoV-2-positive group for up to 12 weeks after diagnosis. They assessed for the presence of surveillance bias by using consultation frequency, and findings were similar, which lends strength to the results. Missing values for covariates were represented by indicator variables, which can bias the results of a study in unpredictable ways.
In all, the study adds to current epidemiological evidence suggesting that SARS-CoV-2 infection is associated with a higher risk of diabetes, CVD and venous thromboembolic disease. The public health impacts of these associations remain uncertain. As the pandemic wears on, epidemiological studies are urgently needed to examine variant-specific associations as well as effects of recurrent infections and vaccination. In parallel, laboratory research should characterize contributing mechanisms, and detailed phenotypic characterization should be undertaken in longitudinal human cohorts.
Collapse
Click Here to Manage Email Alerts