A new study in Cell Metabolism has suggested that patients with COVID-19 and Type 2 diabetes that received insulin treatments were significantly more likely to die than those that did not receive insulin. And among survivors, those that had been treated with insulin had significantly longer hospital stays as well as higher rates of inflammation and adverse secondary outcomes, such as acute kidney injury and invasive mechanical ventilation.
The authors called the finding “unexpected” and urged caution with the use of insulin to treat patients with both COVID-19 and T2D. The suggestion that insulin therapy might actually be harmful to patients with COVID-19 is at odds with some of our current understanding of the topic; several sources have recommended aggressive glucose control, including the use of insulin, even in patients without diabetes.
The present study investigated the outcomes of 689 patients with both COVID-19 and T2D from Wuhan, China, the original site of the coronavirus pandemic. An astonishing 27.7% of these patients that were administered insulin in the hospital died, versus a mere 3.5% of those patients that were not given insulin. (Only about one-quarter of those that were administered insulin in the hospital were known to use insulin before they had been treated for COVID-19, suggesting that the majority of insulin used by the patient group was administered in an effort to fight hyperglycemia caused by the viral disease.)
While it might be natural to assume that patients that had been prescribed insulin to treat their diabetes were less healthy to begin, in fact the two groups had very similar baseline characteristics. The insulin-reliant patients were not significantly more likely to have most other major underlying diseases or co-morbidities: they had similar rates of high blood pressure, kidney disease and heart disease, and were of the same age. Nor did they display a greater rate of symptoms of illness upon admission to the hospital.
The two sets of patients did differ, however, in at least one predictable way: those that were prescribed insulin had significantly higher fasting blood glucose and HbA1c upon admission. Even so, glycemic control prior to infection could not account for the differences in outcomes. When researchers examined only “well-controlled” patients (HbA1c <6.5%), for example, they still found that the risk of death hugely increased for patients that received insulin therapy.
Even when the sample was further restricted to only patients who were “critically ill,” so as to control for disease severity, those receiving insulin were nearly three times as likely to die.
The researchers offer little speculation on the mechanism behind these associations, except to note that insulin appeared to cause inflammation and aggravate organ injury. Some of the deaths could also be attributable to hypoglycemia, a known danger associated with aggressive insulin usage.
It is not immediately clear how the new study fits into the existing research. The preponderance of evidence suggests that good blood sugar control, in people both with and without diabetes, is a huge factor in preventing bad outcomes from COVID-19. And insulin is indisputably the most powerful way to quickly lower blood glucose. The authors of the Cell Metabolism study still agree that “controlling blood glucose is very important for the prognosis of patients with COVID-19.” Nevertheless, they are able to point to several other studies showing “that glucose control by insulin treatment increases the mortality in critical patients, likely through promoting inflammation.”
Is insulin really harmful to people with diabetes and COVID-19, despite how effective it can be at moderating hyperglycemia? One study isn’t enough to know. And it’s possible that we will have a widely distributed vaccine before a consensus forms around the question.
In the meantime, some may take this recent study as further evidence that insulin can be a problematic therapy for patients with Type 2 diabetes. The large amounts of insulin that are needed to surmount insulin resistance are indeed known to cause inflammation. Our contributor and medical advisor Dr. Mariela Glandt has been one of many voices calling for a reevaluation of insulin’s appropriateness for use in Type 2 diabetes: Should We Reconsider Treating Type 2 Diabetes with Insulin?
How does type 1 diabetes fit into this? I’ve been taking insulin for 37 years.