Risks and benefits of vaccines in diabetes

We are approaching 4 years from the onset of the COVID-19 pandemic, and it is appropriate to review the topic of recommended vaccines for persons with diabetes. Infections are important causes of adverse outcome among persons with diabetes and are associated with morbidity and mortality substantially above rates among persons not having diabetes. Analysis of >500 000 UK Biobank participants showed that diabetes was associated with 1.5-fold and 1.8-fold greater COVID mortality in women and in men, respectively, but also with 2.2-fold and 1.9-fold greater influenza/pneumonia mortality in women and in men.¹ The 2014 US Medical Expenditure Panel Surveys, representing 24 million persons with and 218 million without diabetes, showed that diabetes was associated with nearly a doubling in the likelihood of pneumonia and with a 2.6-fold increase in hospitalization for pneumonia.² Diabetes was associated with a nearly 4-fold increase in likelihood of herpes zoster,³ with these infections in turn associated with increases in myocardial infarction (MI) and stroke.⁴ The response of “social media” to COVID, however, has been associated with considerable misinformation about many healthcare recommendations.⁵ Many of my patients now question the benefit of routinely recommended vaccination to prevent influenza, COVID-19, pneumonia, zoster, and the many other infections for which effective preventative measures are available. We will review some of these considerations and how they apply to the treatment of diabetes.

A number of vaccines are considered “must-haves” for people with diabetes: vaccines for influenza, COVID-19, hepatitis B, pneumococcal pneumonia, tetanus, and at age 50 or greater herpes zoster.⁶ New vaccines being developed may be appropriate, for example, that for respiratory syncytial virus (RSV).⁷

For influenza vaccination, manufacturer product information in studies of <2000 persons receiving vaccine versus placebo, most not having diabetes or other medical conditions, suggests moderate efficacy of 68% in reducing infection.⁸ There is evidence of benefit in at-risk persons such as those with diabetes. An epidemiologic study of vaccinated persons with diabetes showed a 46% lower mortality than among those not vaccinated and an 11% lower rate of hospitalization for pneumonia.⁹ In two studies from Taiwan, one of persons with chronic kidney disease (CKD) not known to have underlying (Cardiovascular disease (CVD) showed that those who had received one, two, three, and four influenza vaccines between 1997 and 2008 had likelihood of acute coronary syndrome reduced 38%, 61%, and 87% in comparison to those who had not received the vaccine,¹⁰ and another study of persons with CKD showed > 50% reduction in progression to requirement for dialysis among those who had had the vaccine.¹¹

A number of COVID-19 vaccine preparations are now available, with administration currently recommended for all persons with diabetes.¹² Although some evidence suggests greater likelihood of thrombotic and coronary events among persons having COVID with the Janssen and Oxford-AstraZeneca vaccines, the mRNA vaccines were associated with reduction in such events,¹³ and a study from Hong Kong comparing 3218 unvaccinated persons with 5248 who had been vaccinated showed reduction in post-COVID MI or stroke among persons with underlying CVD, with what appeared to be progressively greater reduction in such events with increasing number of doses either of an mRNA vaccine or a whole inactivated virus vaccine.¹⁴ There are population studies suggesting potential adverse events. In a study of events during the first 28 days after COVID-19 mRNA vaccine among >6 million persons >65 years old between December 2020 and July 2021, deep vein thrombosis or pulmonary embolism occurred in 0.5%, MI in 0.1%, stroke in 0.05%, thrombocytopenic purpura in 0.05%, facial nerve palsy in 0.04%, and myocarditis or pericarditis in 0.01%; encephalomyelitis, transverse myelitis, or Guillain-Barre syndrome occurred at a frequency of <1 per 100 000 vaccine recipients.¹⁵ The likelihood of such events in a comparator population not receiving the vaccine is uncertain, particularly for the very rare events, and for the more common CVD events the argument has been made that post-COVID myocarditis and pericarditis is 1.39-fold more common in unvaccinated than vaccinated persons, that MI was 26% less common in a UK study comparing vaccinated with unvaccinated persons age ≥ 70, and that MI and stroke risk were 52% and 60% lower, respectively, in a Korean study comparing vaccinated with unvaccinated persons.¹⁶

Other vaccines may be appropriate for persons with diabetes. The RSVpreF vaccine against RSV has been advocated based on the incidence of the infection of 3%–7%/year among older persons, leading in the United States to >150 000 hospitalizations annually, with a randomized controlled trial of >34 000 adults ≥age 60 showing 81% and 94% efficacy in reducing infection at ages 60–69 and 70–79, respectively, although with few infections requiring hospitalization.⁷ Although the initial report noted that three persons had had either Guillain–Barré syndrome or acute encephalitis,⁷ a recent update indicated that six cases of inflammatory neurologic events had occurred, leading the authors to caution, “RSV vaccination in older adults should be targeted to those who are at highest risk for severe RSV disease and therefore most likely to benefit.”¹⁷

The evidence for vaccines, then, is typically based on randomized controlled trials, which are usually underpowered to determine severe and uncommon adverse effects and even to determine true benefit in reducing hospitalization and mortality. Extrapolation from population studies may be flawed if the population receiving vaccine differs in unmeasured ways from the nonvaccinated population. Unless well-designed trials are carried out of specific populations, in particular, of persons with diabetes, we may well continue to ask which vaccines truly are beneficial.