August 18, 2020

expert reaction to study linking COVID-19 to increase in type 1 diabetes in children

A study, published in Diabetes Care*, has looked at a potential link between COVID-19 and an increase in type 1 diabetes cases in children.

Profs Paul Zimmet (Monash University) and Francesco Rubino (King’s College London), Co-principal Investigators of the Global CoviDIAB Registry, said:

“The findings of the multi-centre regional study from North West London published in Diabetes Care add to our concerns on a potential link between COVID-19 and diabetes. Early in the course of the pandemic we had become aware of similar reports of new cases of both Type 1 and Type 2 diabetes as well as other atypical diabetes manifestations in people with SARS-CoV-2 (COVID-19) infection.

“As a result, in a report in the New England Journal of Medicine (Rubino F, Amiel SA, Zimmet P, et al. New-Onset Diabetes in COVID-19. N Engl J Med 2020) we expressed our concerns about a potential diabetogenic effect of COVID-19. This has a biological rationale since the new coronavirus can bind to receptors that are highly prevalent in cells of crucial metabolic organs such as the pancreas, the adipose tissue, the liver and the intestine, which may explain also the fact that both type 1 and type 2 diabetes have been reported in association with COVID-19.

“However, all studies on the subject so far are limited by generally small sample sizes and by the fact that a diagnosis of COVID-19 could not confirmed in all suspected cases – two issues that also apply to the new Imperial study. This is inevitable due to the short duration of human contact with the new coronavirus and the low accuracy of COVID-19 testing in the early phases of the pandemic.

“To address these limitations an international group of leading diabetes researchers have joined forces to design a global registry of COVID-19–related diabetes with the aim to establish the extent and nature of new-onset diabetes and the outcomes of diabetes in the context of COVID-19. The COVIDIAB Global Registry http:// COVIDiab.e-dendrite.com/# is already accepting data on new cases from major centres around the word including from countries where the pandemic is at its height as Latin America, Brazil, India and Africa. We agree with the authors of the latest study that it is extremely important to improve our understanding of the relationship between SARS-Cov2 and diabetes as we face inevitable new outbreaks of COVID-19 in the next few months. For this reason we call on the international medical community to lodge information on COVID-19-related diabetes on the Global COVIDiab Registry to help understand how COVID-19–related diabetes develops, its natural history, and appropriate management. This knowledge will inform clinical practices and public health strategies during the course of this pandemic and beyond.”

Prof Deborah Dunn-Walters, Chair of the British Society for Immunology taskforce on COVID-19 and Immunology, and Professor of Immunology at the University of Surrey, said:

“Type 1 diabetes is an autoimmune disease in which the body’s immune system attacks its own tissues in the pancreas, meaning that the body can no longer produce insulin, the hormone that regulates blood glucose levels. This paper reports an increase in the number of type 1 diabetes cases in children in part of London during the months of April and May compared with previous years. As not all children in the study were tested for COVID-19, the findings do not show that this increase was linked to the COVID-19 pandemic – in fact, it is currently unclear what was behind this rise in cases.

“COVID-19 is caused by the virus, SARS-CoV-2, and we know that other viral diseases can act as a trigger for some autoimmune diseases, such as Guillain-Barre syndrome. As SARS-CoV-2 is such a new virus, there is still much we need to learn about how it interacts with our immune systems and the long-term effects that it might have on us. Currently, there have been no comprehensive studies published linking COVID-19 to the development of any autoimmune disease, including type 1 diabetes. However, we are still in the early days of finding out about the longer-term effects of COVID-19 and follow-up studies in this area will be important. As the British Society for Immunology recent report recommended, large scale cohort studies and research programmes to track long-term health consequences in COVID-19 patients from all demographics over a number of years are now urgently needed so that we can properly assess any longer-term health risks due to COVID-19 infection.”

Prof Julian Hamilton-Shield, Professor of Diabetes and Metabolic Endocrinology at the University of Bristol, said:

“Unsworth et al’s paper is interesting as the high percentage of new cases of Type 1 diabetes presenting with severe diabetic ketoacidosis mirrors anecdotal observations from across England. Any suggestion of an increase in actual incidence should be treated cautiously, as only some units contributing had greater than expected presentations across a really short time span.

“The yearly Royal College of Paediatrics and Child Health (RCPCH) Paediatric Diabetes Audit should provide a true estimate of any increased incidence around the COVID-19 period. Unfortunately, the data in this current report does not include presence of Type 1 diabetes-associated autoimmunity in these children.

“Type 1 diabetes presentation in childhood has classically displayed seasonal variation probably due to viral illness precipitating metabolic decompensation in those in whom autoimmunity has been gradually degrading pancreatic insulin producing capacity for years before. SARS-CoV-2 could simply be acting as another viral infection causing this decompensation.

“However, it is possible that SARS-CoV-2 may be causing direct damage to the islet cells in some cases and thus immune markers might prove different in these children. It should still be possible to examine this in the future on stored samples from these cases. The suggestion made recently in the New England Journal of Medicine by Rubino et al for a registry of new diabetes cases during COVID-19 again seems eminently sensible given this current report. The report of severe hypokalaemia (low potassium levels that can trigger heart arrhythmias) is also notable. Again, as pointed out by the authors, hypokalaemia may simply reflect treatment effects in those with severe ketoacidosis but SARS-CoV-2’s influence on the renin-angiotensin-aldosterone system requires further scrutiny.”

Prof Paul Hunter, Professor in Medicine at the University of East Anglia, said:

“This paper raises a concern that does need further studies to either confirm or refute its observations. Diabetic ketoacidosis can be triggered in susceptible individuals by a range factors and infection is one of the most common. There is also a plausible mechanism for associating diabetes with COVID-19 in particular, in that the binding site for SARS-CoV-2 is the ACE-2 receptor which is strongly expressed in pancreatic endocrine cells, i.e. the cells that produce insulin.

“However, the epidemiological evidence presented in this paper is weak and cannot be used as evidence of proof of association. The main weakness of this paper is the identification of an increase in only two of five inpatient units. It is not clear what triggered this study. If it was because someone had noticed an increase in one of other of the units and then decided to do the study, it is difficult to prove cause and effect. Across the UK at any time cases of new ketoacidosis will vary from one year to the next and it is often tempting but can be wrong to assume that whatever infection is common in the community is the cause of an increase in diagnoses that could have arisen just by chance. During any epidemic like the current one, people will be looking out for any change in the incidence of disease in their own practice that could be related to the epidemic pathogen. Sometimes any such association will be causal and at other times the association will be random.

“Nevertheless, this is an association that does need further analysis across the whole country. Any association between diabetic ketoacidosis and COVID-19 may simply represent the association between ketoacidosis and any infection or in may represent something particular to COVID-19. It is important that this observation is followed up.”

* ‘New-onset Type 1 Diabetes in Children during COVID-19: Multi-centre Regional Findings in the UK’ by Rebecca Unsworth et al. was published in Diabetes Care on Tuesday 18 August 2020.

Declared interests

Prof Hamilton-Shield: No conflicts declared

Prof Dunn-Walters: “I am a Trustee of the Dunhill Medical Trust and of BSI. I hold various positions on strategy and grant panels at BBSRC and oversight committees at MRC.”

No others received.