SARS-CoV-2 infection was associated with an increased risk for diabetes among youth, whereas other acute respiratory infections were not, new data from the U.S. Centers for Disease Control and Prevention indicate.
The results from two large U.S. health claims databases were published in an early release in the CDC’s Morbidity and Mortality Weekly Report by Catherine E. Barrett, PhD, and colleagues of the CDC’s COVID-19 Emergency Response Team and Division of Diabetes Translation.
Clinicians should monitor individuals younger than 18 years in the months following a SARS-CoV-2 infection for new diabetes onset, they advise.
The findings, which are supported by independent studies in adults, “underscore the importance of COVID-19 prevention among all age groups, including vaccination for all eligible children and adolescents, and chronic disease prevention and treatment,” Dr. Barrett and colleagues say.
Diabetes type couldn’t be reliably distinguished from the databases, which is noted as an important study limitation.
“SARS-CoV-2 infection might lead to type 1 or type 2 diabetes through complex and differing mechanisms,” they say.
Emerging evidence began to suggest, in mid-2020, that COVID-19 may trigger the onset of diabetes in healthy people. A new global registry was subsequently established to collect data on patients with COVID-19–related diabetes, called the CoviDiab registry.
Not clear if diabetes after COVID-19 is transient or permanent
From one of the databases used in the new study, known as IQVIA, 80,893 individuals aged younger than 18 years diagnosed with COVID-19 during March 2020 to February 26, 2021, were compared with age- and sex-matched people during that period who did not have COVID-19 and to prepandemic groups with and without a diagnosis of acute respiratory illness during March 1, 2017, to February 26, 2018.
From the second database, HealthVerity, 439,439 youth diagnosed with COVID-19 during March 1, 2020, to June 28, 2021, were compared with age- and sex-matched youth without COVID-19. Here, there was no prepandemic comparison group.
Diabetes diagnoses were coded in 0.08% with COVID-19 vs. 0.03% without COVID-19 in IQVIA and in 0.25% vs. 0.19% in HealthVerity.
Thus, new diabetes diagnoses were 166% and 31% more likely to occur in those with COVID-19 in IQVIA and HealthVerity, respectively. And in IQVIA, those with COVID-19 were 116% more likely to develop diabetes than were those with prepandemic acute respiratory illnesses. Those differences were all significant, whereas non–SARS-CoV-2 respiratory infections were not associated with diabetes, Dr. Barrett and colleagues say.
In both databases, diabetic ketoacidosis (DKA) was more common at diabetes onset among those with, vs. without, COVID-19: 48.5% vs. 13.6% in IQVIA and 40.2% vs. 29.7% in HealthVerity. In IQVIA, 22.0% with prepandemic acute respiratory illness presented with DKA.
Dr. Barrett and colleagues offer several potential explanations for the observed association between COVID-19 and diabetes, including a direct attack on pancreatic beta cells expressing angiotensin-converting enzyme 2 receptors, or via stress hyperglycemia resulting from cytokine storm and alterations in glucose metabolism.
Another possibility is the precipitation to diabetes from prediabetes; the latter is a condition present in one in five U.S. adolescents.
Steroid treatment during hospitalization might have led to transient hyperglycemia, but only 1.5% to 2.2% of diabetes codes were for drug- or chemical-induced diabetes. The majority were for type 1 or 2.
Alternatively, pandemic-associated weight gain might have also contributed to risks for both severe COVID-19 and type 2 diabetes.
“Although this study can provide information on the risk for diabetes following SARS-CoV-2 infection, additional data are needed to understand underlying pathogenic mechanisms, either those caused by SARS-CoV-2 infection itself or resulting from treatments, and whether a COVID-19–associated diabetes diagnosis is transient or leads to a chronic condition,” Dr. Barrett and colleagues conclude.
A version of this article first appeared on Medscape.com.