COVID-19 in Diabetes: What Mediates Risk?

Certain clinical factors were predictive of severe COVID-19 in people with diabetes, a researcher reported.

In a retrospective study of nearly 2,000 patients with diabetes hospitalized with COVID-19, those with out-of-range blood glucose levels were more likely to have severe infection requiring admission to the intensive care unit (ICU), leading to death or fitting the clinical criteria indicating respiratory decompensation, reported Samarth Virmani, MBBS, of the University of Central Florida College of Medicine in Orlando.

Specifically, 44% of diabetes patients in the study with severe COVID-19 had hyperglycemia (blood glucose over 180 mg/dL) versus only 38% of those with a non-severe case of COVID-19, he explained during a presentation at the virtual meeting of the American Association of Clinical Endocrinology (AACE).

On the other hand, just slightly more patients with diabetes with hypoglycemia (below 70 mg/dL) had a severe case of COVID-19 versus those with non-severe infection (9.75% vs 8.90%).

Patients with type 2 diabetes who were on an SGLT-2 inhibitor prior to hospitalization — such as canagliflozin (Invokana), empagliflozin (Jardiance), dapagliflozin (Farxiga), or ertugliflozin (Steglatro) — had an 85% greater chance of having a severe COVID-19 infection (OR 1.85), a finding Virmani called “interesting.” Similarly, individuals who had pre-hospitalization exposure to steroids also had an elevated chance of having severe COVID-19 (OR 1.49).

However, those who were taking an ACE inhibitor (OR 0.75) or statin (OR 0.66) prior to being hospitalized had a lower chance of severe infection.

In line with prior findings, a significantly increased risk of severe COVID-19 was observed in patients who were older (1% increased risk with each year), male, or who had obesity (body mass index [BMI] over 35 versus under 25), nephropathy, or cardiovascular disease.

Virmani and co-authors also assessed the risk associated with other clinical factors that were not significantly tied to severe COVID infection, including having hypertension, hyperlipidemia, fatty liver disease, use of insulin GLP-1 receptor agonists, sulfonylurea, DPP-4 inhibitors, thiazolidinediones, biguanides, or having type 1 versus type 2 diabetes.

He explained that his group selected these variables because prior COVID-19 research coming out of Italy and China had indicated that these factors played the biggest roles in infection outcomes.

“Our study was able to outline a risk profile for severe infection in diabetic patients,” said Virmani. “These findings will help us determine which diabetics would be at a higher risk of severe COVID-19.”

For the multicenter analysis, the researchers analyzed data on 1,796 patients with diabetes, representing 1,818 separate hospital admissions from 128 hospitals over 7 months, from January through July 2020. “Severe” COVID-19 was defined as a composite based on ICU admissions, ICD-10 codes that specified acute respiratory distress syndrome due to COVID-19, certain respiratory markers like a respiratory rate greater than 30, a P/F ratio less than 300, or a pulse oximeter reading less than 93%.

The average age of patients with diabetes was 61, and they had a mean BMI of 32.1 and a mean HbA1c of 8.49.

One of the limitations of the retrospective study, Virmani noted, was a lack of detailed data on pre-hospitalization medications, such as the duration of steroid and statin use and patient compliance. Future studies, he said, should therefore compile more data on these patients to help pinpoint if certain patients with diabetes could benefit from earlier and more intensive COVID-19 care.