In early 2020, the COVID-19 pandemic changed everything about life as we know it, with widespread shutdowns across the globe. The U.S. health care system quickly adapted, pivoting to telehealth visits when able and proactively managing outpatient conditions to prevent overwhelming hospital resources and utilization. Meanwhile, at my practice, the typical rate of about one new-onset pediatric type 1 diabetes (T1D) case per week increased to about two per week.
Dr. Jessica Sparks Lilley
Dr. Jessica Sparks Lilley
However, the new diabetes cases continued to accumulate, and I saw more patients being diagnosed who did not have a known family history of autoimmunity. I began to ask friends at other centers whether they were noticing the same trend.
One colleague documented a 36% increase in her large center compared with the previous year. Another noted a 40% rise at his children’s hospital. We observed that there was often a respiratory illness reported several weeks before presenting with T1D. Sometimes the child was known to be COVID-positive. Sometimes the child had not been tested. Sometimes we suspected that COVID had been a preceding illness and then found negative SARS-CoV-2 antibodies – but we were not certain whether the result was meaningful given the time lapsed since infection.
Soon, reports emerged of large increases in severe diabetic ketoacidosis (DKA) and hyperglycemic hyperosmolar state at initial presentation, a trend reported in other countries.
Is COVID-19 a trigger for T1D?
There is known precedent for increased risk for T1D after viral infections in patients who are already genetically susceptible. Mechanisms of immune-mediated islet cell failure would make sense following SARS-CoV-2 infection; direct islet toxicity was noted with SARS-CoV-1 and has been suspected with SARS-CoV-2 but not proven. Some have suggested that hypercoagulability with COVID-19 may lead to ischemic damage to the pancreas.
With multiple potential pathways for islet damage, increases in insulin-dependent diabetes would logically follow. Still, whether this is the case remains unclear. There is not yet definitive evidence that there is uptake of SARS-CoV-2 via receptors in the pancreatic beta cells.
Our current understanding of T1D pathogenesis is that susceptible individuals develop autoimmunity in response to an environmental trigger, with beta-cell failure developing over months to years. Perhaps vulnerable patients with genetic risk for pancreatic autoimmunity were stressed by SARS-CoV-2 infection and were diagnosed earlier than they might have been, showing some lead-time bias. Adult patients with COVID-19 demonstrated hyperglycemia that has been reversible in some cases, like the stress hyperglycemia seen with other infections and surgery in response to proinflammatory states.
The true question seems to be whether there is a unique type of diabetes related to direct viral toxicity. Do newly diagnosed patients have measurable traditional antibodies, like anti-glutamic acid decarboxylase or anti-islet cell antibodies? Is there proof of preceding SARS-CoV-2 infection? In the new cases that I thought were unusual at first glance, I found typical pancreatic autoimmunity and negative SARS-CoV-2 antibodies. The small cohorts reported thus far have had similar findings.
A stronger case can be made for the risk of developing diabetes (types 1 and 2) with rapid weight gain. Another marked pattern that pediatric endocrinologists have observed has been increased weight gain in children with closed schools, decreased activity, and more social isolation. I have seen weight change as great as 100 lb in a teen over the past year; 30- to 50-lb weight increases over the course of the pandemic have been common. Considering the “accelerator hypothesis” of faster onset of type 2 diabetes with rapid weight gain, implications for hastening of T1D with weight gain have also been considered. The full impact of these dramatic weight changes will take time to understand.
The true story may not emerge for years
Anecdotes and theoretical concerns may give us pause, but they are far from scientific truth. Efforts are underway to explore this perceived trend with international registries, including the CoviDIAB Registry as well as T1D Exchange. The true story may not emerge until years have passed to see the cumulative fallout of COVID-19. Regardless, these troubling observations should be considered as pandemic safeguards continue to loosen.
While pediatric mortality from COVID-19 has been relatively low (though sadly not zero), some have placed too little focus on possible morbidity. Long-term effects like long COVID and neuropsychiatric sequelae are becoming evident in all populations, including children. If a lifelong illness like diabetes can be directly linked to COVID-19, protecting children from infection with measures like masks becomes all the more crucial until vaccines are more readily available. Despite our rapid progress with understanding COVID-19 disease, there is still much left to learn.
A version of this article first appeared on Medscape.com.