Diabetes Appears With Long COVID. Can the Thyroid and Adrenals Be Far Behind?

Richard M. Plotzker, MD


July 13, 2022

My edition of Harrison's Principles of Internal Medicine, the one with the azure blue cover, the source of lectures and exam questions during my third year clerkship in 1976, didn't have a chapter on AIDS. There was no AIDS. It did discuss pneumoencephalography to follow up a J-shaped sella on a lateral skull radiograph, and the chapter on peptic ulcer disease included the indications for antrectomy and vagotomy. Our gynecology textbook had a chapter on surreptitious septic abortions. I don't know if the current edition still does. Some conditions just become rare, and not always because of changes in medicine. A chapter on Legionella pneumonia didn't appear until two editions later.

By now, acute SARS-CoV-2 infections are finding their way into the textbooks of medicine, infectious diseases, pulmonary disorders, and critical care. It's an acute disease with its epidemiology, socioeconomic inequities, lab tests, treatments, preventions, and decision branch points. As with septic abortions, whose prevalence depends more on legal or social realities than medical advancements, how people fare with COVID-19 — a new disease that needs experience and innovation — has its competing political influences.

Two years into a devastating public experience, our learning curve has enhanced survival, containment, and prevention. But we also find ourselves at a transition — one we have seen with other diseases, from diabetes a century ago to AIDS in our generation — that transforms a disease from something life-threatening to something chronic. Long COVID, formally designated post-acute sequelae of SARS-CoV-2 (PASC), sits at the beginning of that experience. After clinical recovery, some of those who were infected do not return entirely to premorbid wellness or after a quiescent period become ill again, but in a less dramatic way.

Wide-Ranging Manifestations

Much of the descriptive elements of PASC fail to distinguish this into a single category. Neurologic symptoms such as headache, dizziness, or mood changes seem to predominate. However, some reported symptoms identify organ systems, such as rash, limb pain, bowel dysfunction, and chest symptoms. And although the classic endocrinology diseases do not yet seem very prominent, everyone graduates from fellowship with the belief that your internal chemicals define you and what happens to you; therefore, the search for the place of endocrinology in chronic COVID sequelae has begun with a few early findings reported, both within recent endocrinology literature and at annual meeting presentations.

Because diabetes dominates endocrine practice and extensive databases related to diabetes exist, the first foray into understanding the endocrine effects of post-COVID syndromes have begun here. The Lancet published an extensively intricate exploration of data collected by the US Veterans Administration (VA). The investigators asked whether people who recovered from COVID infection were more likely to acquire new diabetes in the first year than did veterans seeking care through that system who never had COVID.

They divided the cohort into three groups: those treated for COVID at the VA who recovered; those treated for other things at the VA but not COVID during the time of the pandemic; and a baseline of veterans receiving care at the VA the year before COVID-19 arrived in the United States. They also subdivided their patients by severity of illness to include those treated as outpatients, those hospitalized in general wards, and those requiring intensive care. Defining diabetes either as newly elevated A1c or new prescription for hypoglycemic agents, they found a relative risk increase of about 40% for those who had recovered from COVID. Among those hospitalized, the risk of acquiring new hyperglycemia during the next year approached threefold. Among those who recovered, the presence of traditional risk factors before infection, such as weight, age, prediabetes, and African-American ethnicity, seemed to mirror who would be pushed past the glycemic edge after recovery.

Although it is tempting to generalize this because of the thoroughness with which the data were analyzed, the investigators cautioned that the VA population differs somewhat from the composite US population, being older, largely male, and often less prosperous — groups that have been most devastated by acute COVID. The data seem conclusive, however, that some elements of health, including glycemic control for many, remain precarious even in recovery, particularly when the acute infection and its cytokine storm was particularly severe.

What About COVID and Established Diabetes?

Could those who already had diabetes when their acute COVID illness presented be more subject to acquiring a more protean PASC syndrome after recovery from the acute phase? An oral presentation at the most recent American Diabetes Association annual meeting offers some insight, though with inconclusive data from limited studies.

From a review of seven studies published within the past few months, the presenter found that about half the identified studies showed that patients with preexisting diabetes had up to four times the risk of developing PASC compared with those who did not have diabetes; the rest of the studies showed no appreciable correlation. Those most severely affected by the acute phase were more likely to experience adverse symptoms in their postacute phase. There was also a trend toward PASC in those with poorer glycemic control. Unfortunately, meta-analyses involve studies with very different methods and populations, but the diversity of conclusions, including some with worrisome results, suggest the need for a large targeted multi-site investigation with less heterogeneity of populations and clearer definition of symptoms.

Thyroid, adrenal, and gonadal disorders remain largely unexplored in studies of PASC.

Growing Challenges

Needless to say, as this plays out, there will be challenges in our exam rooms and for health planners. At present, we must assume that persons with post-COVID diabetes will be subject to the same end-organ risks as others with diabetes. Nonspecific complaints that defy firm conclusions have always been the bane of clinician history-taking, though the convenience of a COVID diagnosis may compete with investigations of other plausible possibilities that may better address symptoms. The National Institutes of Health in its RECOVER initiative has offered funding opportunities to enable more exploration on this emerging though poorly elucidated condition.

How big a problem is this for our health system? Expansion of the prevalence of diabetes from any cause adds to the resources needed for management. Opinions range from a small blip upward in the need for medical care to a much more burdensome prediction of many more chronically ill people needing a broad spectrum of medical care. The Annals of Internal Medicine held a symposium, available on podcast, dedicated to the future directions that long COVID may take. The Atlantic also published a detailed perspective written by a physician.

Considering the vast numbers infected, a few percent with chronic symptoms brings a lot of people to their doctors seeking care, yet the mass disability event that some have feared has not yet happened. Individual impairment of function or new presentations of chronic diseases such as diabetes, which the Lancet survey seems to unmask, will challenge endocrinologists and our colleagues who want our patients to function their best but are still grappling with a fundamentally unfamiliar, poorly defined condition. Eventually data will become synthesized in a more applicable way.

For more diabetes and endocrinology news, follow us on Twitter and Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.