Diabetes, COVID‐19, and questions unsolved

Recent evidence suggests a role for Diabetes Mellitus in adverse outcomes from COVID‐19 infection; yet the underlying mechanisms are not clear. Moreover, attention has turned to prophylactic vaccination to protect the population from COVID‐19‐related illness and mortality. We performed a comprehensive peer‐reviewed literature search on an array of key terms concerning diabetes and COVID‐19 seeking to address the following questions: 1. What role does diabetes play as an accelerator for adverse outcomes in COVID‐19?; 2. What mechanisms underlie the differences in outcomes seen in people with diabetes?; 3. Are vaccines against COVID‐19 efficacious in people with diabetes? The current literature demonstrates that diabetes is associated with an increased risk of adverse outcomes from COVID‐19 infection, and post‐COVID sequelae. Potential mechanisms include dysregulation of Angiotensin Converting Enzyme 2, Furin, CD147, and impaired immune cell responses. Hyperglycaemia is a key exacerbator of these mechanisms. Limited studies are available on COVID‐19 vaccination in people with diabetes; however, the current literature suggests that vaccination is protective against adverse outcomes for this population. In summary, people with diabetes are a high‐risk group that should be prioritised in vaccination efforts. Glycaemic optimisation is paramount to protecting this group from COVID‐19‐associated risk. Unsolved questions remain as to the molecular mechanisms underlying the adverse outcomes seen in people with diabetes; the functional impact of post‐COVID symptoms on people with diabetes, their persistence, and management; how long‐term vaccine efficacy is affected by diabetes, and the antibody levels that confer protection from adverse outcomes in COVID‐19.


| Background
The SARS-CoV-2 virus, which causes COVID-19, is a single-stranded enveloped RNA beta-coronavirus, first identified in Wuhan, China, in January 2020. 1 This virus presents a significant global burden. To date, COVID-19 has infected over 500 million people and has resulted in more than 6 million deaths. 2 An increasing body of evidence suggests that those with diabetes mellitus, a disease estimated to have a worldwide prevalence of at least 422 million, have worse prognoses when infected with COVID- 19. Indeed, this has been the case in previous coronavirus outbreaks, such as Middle East Respiratory Syndrome and Severe Acute Respiratory Syndrome virus. 3,4 The COVID-19 pandemic placed a further detrimental impact on people with diabetes as lockdowns restricted access to outpatient healthcare. That said, some recent evidence suggests that, at least for people with diabetes reliant on multiple insulin injections, metabolic control did not worsen during the lockdown period. 5 More recently, a post-acute sequelae of symptoms has been described. This refers to a new onset or persisting effects of COVID-19 after the acute phase of infection and represents the period between microbiological and clinical recovery. Frequently reported symptoms include fatigue, shortness of breath, cognitive dysfunction, post-exertional malaise. 6 As many as 70% of people infected with COVID-19 experience this syndrome. 6 This is likely to impose a significant burden on health systems and result in lost productivity. Limited work has been done to summarise the state of knowledge surrounding the effects of this syndrome on people with diabetes.
Focus has now turned to global vaccination efforts as a means to mitigate the risk of severe disease and mortality from COVID-19, so far with good efficacy in the general population. 7 However, historical evidence suggests that people with diabetes may mount an inferior immune response to some vaccines. 8 Moreover, COVID-19 vaccines represent the first mRNA vaccines authorised for widespread use in history. It is therefore paramount to establish the efficacy of these vaccinations in people with diabetes.

| Aims
This review aims to investigate and summarise the current state of knowledge concerning the below questions and to establish key questions that remain unsolved.
� What role does diabetes play as an accelerator factor for adverse outcomes in COVID- 19? � What mechanisms underlie the different outcomes seen in people with diabetes?
� Are vaccines efficacious in diabetes?

| Methodology
To answer these questions, a comprehensive search of peerreviewed literature was performed between the periods of March 2022 and April 2023 based on a wide array of key terms, such as Diabetes, Covid-19, Coronavirus, Post-COVID syndrome, and COVID vaccination. Two databases were searched: PubMed and Google Scholar. Articles were selected based on their relevance to the topic, identified by the abstract. The reference lists of the articles found were also searched to identify further relevant literature.

| DIABETES AS AN ACCELERATOR FACTOR FOR ADVERSE OUTCOMES IN COVID-19
In early 2020, the first reports of confirmed COVID-19 cases hinted at the possible involvement of underlying disease. Huang et al. 9 described the clinical characteristics of 41 confirmed cases in the epicentre city Wuhan, China, observing that one third of the patients had at least one comorbid condition. Zhou et al. 10 later found diabetes to be the second most prevalent comorbidity (19%), after hypertension (30%), among patients hospitalised with COVID-19. As this novel coronavirus spread and captured scientific attention, the number of observational studies increased, covering a broader geographic scope. In a case series 11  Given that inter-institutional differences in guidelines and polices for admission to ICU may influence such statistics, it is useful to examine the data retrieved by single-centre studies. A Kuwaiti study 22 found that diabetes was an independent predictor of transfer to ICU (OR = 9.38). However, this was not observed for pre-diabetes.
A centre in the United States observed associations between diabetes and increased odds of ICU admission and a mechanical ventilation requirement 14 days from initial admission (ORs 1.59 and 1.97 respectively). 23 Moreover, a Chinese centre observed that COVID-19 patients with diabetes were more likely to require transfer to ICU compared to those without (20.8% vs. 8.3%, p = 0.044). 24

| Diabetes and COVID-19 mortality
Diabetes confers an increased risk of mortality in COVID-19 infection. To date, several meta-analyses have observed the risk of death to be at least doubled for patients with diabetes compared to those without. 9,15,17,18,25 27 the prevalence of diabetes was the only comorbidity that influenced the mortality rate with statistical significance. In this study, a 1% increase in diabetes prevalence correlated with a 1.5% increase in the absolute mortality rate.
However, the increased mortality risk associated with diabetes may not affect all patients equally. Despite older age being a prominent risk factor for mortality in COVID-19 infection, there has been some suggestion that younger people with diabetes may be more greatly impacted. 17 Corona et al. 26 observed that diabetes-associated mortality risk was independent of sex, but attenuated by older age. A large cohort study 28 36 and Zein et al. 37 both observed reduced risk of COVID-19-associated mortality for people on statin therapy, but that these associations were not affected by diabetes. As such, further studies are required to explore potential risk and benefits for this population.

| Type 1 versus type 2
Despite differing aetiologies, only few studies distinguish between Type 1 diabetes mellitus (T1DM) and Type 2 diabetes mellitus (T2DM). The authors of one meta-analysis 38 reported that only 14.5% of the 45,775 included patients had the type of diabetes specified on their records. One reason could be the lower prevalence of T1DM relative to T2DM; people with T2DM represent 90%-95% of people with diabetes. 3 However, the consequence of this is reliance on extrapolated results that are inaccurate and do not reflect the true picture. It is therefore vital to examine the risks conferred by these two conditions separately.
Owing to the limited number of studies that differentiate between T1DM and T2DM, the existing literature is inconclusive as to how excess risk compares between these conditions. Barrett 44 found a greater in-hospital mortality risk for T1DM (OR = 3.51) compared with T2DM (OR = 2.03). These odds were adjusted for age, sex, deprivation, ethnicity, and geographical region, but not for established confounders, such as BMI, smoking, and hypertension.

| Post-COVID sequelae
The post-acute sequelae of COVID-19 infection is well documented in the literature. Commonly reported symptoms include headache, anosmia, fatigue, shortness of breath, cognitive dysfunction, anxiety, depression, and post-exertional malaise. 6,45 These symptoms can be new-onset or exacerbated, and may persist from the acute phase or occur with delayed onset. 46 A study from the United Kingdom found that 70% of subjects experienced symptoms 4 months after infection. 6 About 30% of these patients may have pre-existing comorbidities, and within this population, the prevalence of diabetes has been reported to be as high as 26%. 47 To date, there is no universally agreed term for these symptoms. A WHO-led Delphi survey opted for the term 'post-COVID-19 condition' to describe symptoms that occur more than 3 months after initial COVID-19 infection, persist for at least 2 months, and cannot be explained by an alternative diagnosis. 48 However, other health authorities draw a temporal distinction, defining signs and symptoms that persist between 4 and 12 weeks post-acute infection as 'ongoing symptomatic COVID-19', and those that occur 44% of studies reported that diabetes was a significant risk factor for the development of long COVID, and 56% did not. Notably, 38% of the studies that did find a significant association reported odds ratios >4.
Undoubtedly, the heterogeneity of symptoms included in these studies poses a considerable challenge. Long COVID symptoms reported in patients with diabetes are, as is the case for the general population, highly variable. Nesan et al.'s study 56 of patients hospitalised with COVID-19 3 months post-discharge found that diabetes was associated with an increased risk of ongoing cardio-respiratory and neurological symptoms, but not abdominal or psychological symptoms. Mechi et al. 57 observed that people with diabetes (compared to without) reported higher occurrences of fatigue, dyspnoea, and chest pain 9 months after infection. However, these comparisons did not account for common confounders such as hypertension and age.
Muscle weakness has also been described. Raveendran 55 12 out of the 14 included studies (86%) observed a significant association between COVID-19 infection and incident diabetes. In another metaanalysis that included 3700 hospitalised patients with COVID-19, the pooled proportion of newly diagnosed diabetes was 14.4%. 60 Concerningly, people with new onset diabetes appear to be at an increased risk of mortality and adverse events compared with people with known diabetes and without diabetes. 61 The underlying pathophysiology is yet to be firmly established, but there is some suggestion that the physiologic stress response may play a role, especially given that a considerable proportion (40%) of people regress to normoglycaemia within a year. 62 However, when compared with other conditions that provoke a stress response, patients with severe COVID-19 demonstrate significantly and disproportionately higher insulin requirements, alluding to a potential diabetogenic effect. 63,64 Current theories for this include COVID-19-induced autoimmune and inflammatory insult to pancreatic beta cells, and insulin resistance as the consequence of mACE2-virus binding. 65

| COVID-19: WHY DO PATIENTS WITH DIABETES DO WORSE?
With the evidence pointing towards people with diabetes experiencing a more severe course of infection and increased susceptibility to mortality from COVID-19, the question of 'why?' becomes poignant. The below information outlines the current state of knowledge regarding these findings and is further summarised in Table 1.

| Angiotensin Converting Enzyme 2 and COVID-19: What is the link?
Before examining the link between diabetes and poor outcomes in COVID-19, it is useful to consider how SARS-CoV-2 enters human cells in the first place. Coronaviruses have a transmembrane 'Spike' protein that determines diversity and host tropism. In humans, Angiotensin Converting Enzyme 2 (ACE2) has been identified as a functional receptor for SARS-CoV-2. 66 Angiotensin Converting Enzyme 2 has both membrane-bound (mACE2) and soluble (sACE2) forms. 66 mACE2 is expressed in the small intestine, testes, kidneys, heart, lung, and adipose tissues and sheds to form sACE2 in response to inflammation. 66 Spike. 68 sACE2-virus complexes can also interact with vasopressin to gain entry via the AVPR1B receptor, which may explain why the virus appears to affect tissues that do not express mACE2 highly. 66,69 The clinical relevance of ACE2 lies in its role in vascular homoeostasis, which is regulated by the Renin-Angiotensin-Aldosterone-System (RAAS). Angiotensin II (AngII), a key component of this system, has vasoconstrictive, pro-inflammatory, prooxidant, and pro-fibrotic effects. 66 ACE2 acts as a negative feedback mechanism by metabolising AngII and exerting the opposite effect (vasodilation, anti-inflammation, and attenuated fibrosis). 66 Viral entry of SARS-CoV-2 into human host cells via the ACE2 receptor leads to the loss of mACE2 via receptor internalisation and shedding. 70 It is thought that this decrease shifts the balance in favour of the RAAS axis and AngII levels rise, resulting in a sequela of pulmonary oedema, decreased pulmonary ventilation, insufficient oxygenation, and eventually Acute Respiratory Distress Syndrome (ARDS). 69

| Altered Angiotensin Converting Enzyme 2 expression in diabetes
Angiotensin Converting Enzyme 2 expression appears to be altered in diabetes. However, there is limited consensus over the direction and extent to which this dysregulation occurs. Both increased and decreased ACE2 expression has been reported in the literature, and it appears this dysregulation is influenced by disease progression. 71 Mouse model studies demonstrate the upregulation of ACE2 in the early stages of diabetes and the subsequent down regulation later on. 72,73 Corroborating this, biopsies from people with diabetic nephropathy, a feature of later stage disease, have also revealed decreased ACE2 expression. 74 Glycaemic control also appears to play a role in ACE2 dysregulation. For example, the increased expression of ACE2 and higher viral loads were observed in Garreta et al.'s study 75 of kidney organoids treated with a high glucose diabetic milieu. Moreover, autopsy sections of myocardial tissue demonstrate that total ACE2, the ratio of glycosylated ACE2, and TMPRSS2 expression are all significantly higher in the tissues from people with diabetes. 68 Notably, increased TMPRSS2 expression is associated with an oscillatory virus loads, which have been linked to a more severe disease course. 76,77 The significance of this dysregulated ACE2 expression is uncertain and may even have paradoxical effects. Increased ACE2 expression may facilitate viral entry and permit higher viral loads, which has been linked to a more severe disease course. 78 However, people with diabetes do not appear to be any more susceptible to acquiring COVID-19 infection than those without the condition. 79 On the other hand, ACE2 has a clear protective role in regulating AngIImediated damage. 66 Given the pro-inflammatory, pro-fibrotic effects of AngII, the decreased expression of ACE2 observed in the later stages of diabetes may offer insight as to why people with diabetes do worse. 72,73 However, further research is required to better T A B L E 1 Why do patients with diabetes do worse in COVID-19 infection? The key hypotheses.

Dysregulation of ACE2
� ACE2 expression is increased in the early stages of diabetes and decreased in the later stages � Increased ACE2 expression may facilitate viral entry, resulting in a higher viral load � Decreased ACE2 expression may favour AngII-driven inflammation and fibrosis Autoantibodies against ACE2 � Higher levels of circulating inhibitory autoantibodies against ACE2 favours AngIImediated inflammation and fibrosis  understand how ACE2 expression changes in diabetes and to elucidate the relationship between ACE2 and COVID-19 severity.

| Angiotensin Converting Enzyme 2 autoantibodies
Autoantibodies against ACE2 (AA-ACE2) have been detected in people infected with COVID-19 -likely enhanced by the presence of circulating sACE2-virus complexes. Rodriguez-Perez et al. 80 found that people with diabetes had significantly higher levels of AA-ACE2 relative to the controls. These antibodies have an inhibitory effect on ACE2 and so may shift the balance to favour the pro-inflammatory actions of AngII. 81,82 It is not clear why people with diabetes seem to be more susceptible to forming AA-ACE2. However, genetic studies have associated diabetes with polymorphisms of ACE2 that enhance conformational binding, increasing the probability of ACE2 being included in the viral part presented to the immune system, favouring autoantibody formation. 83

| Elevated furin levels in diabetes
Furin is a ubiquitously expressed proprotein convertase that acts on a diverse range of proproteins. It is found in membrane-bound and secreted forms and has numerous roles in normal physiology. 84 Sequencing of the SARS-CoV-2 genome has now identified the insertion of a Furin cleavage site in the virus' spike protein, assisting viral entry into host cells. 85 Type 2 Diabetes, Obesity, and Hypertension are all conditions associated with increased Furin levels and are frequently regarded as the most common comorbid conditions in severe cases of COVID-19. [86][87][88] People with diabetes also have elevated levels of the pro-inflammatory glycoprotein osteopontin, which has also been shown to upregulate Furin expression. 84 One of Furin's many functions is to promote cytokine maturation, notably the potent pro-inflammatory cytokine TNFa. 84 People with diabetes may therefore experience a chronic state of metabolic inflammation, predisposing them to the cytokine storm seen in severe COVID-19.

| CD147: An alternative mechanism for viral entry?
As knowledge about the entry of COVID-19 into human host cells continues to accrue, other receptors of interest have been identified. 66 One example is CD147, also referred to as Basigin, or extracellular matrix metalloproteinase inducer. Wang et al. 89 infected ACE2-deficient T cells with SARS-CoV-2 and found that humanised anti-CD147 antibodies attenuated infection. Whilst this suggests an alternative mechanism of entry for COVID-19, it may also explain the lymphopenia seen in severe COVID-19 infection. 90 CD147 has a particular relevance to T2DM because of its response to the hyperglycaemic environment. Hyperglycaemia and Advanced Glycation End products upregulate CD147 expression, suggesting that people with poor glycaemic control may be at increased risk. 91

| Diabetes and hyperglycaemia
Glycaemic control appears to influence prognosis in COVID-19 infection in diabetes. People with poorly controlled diabetes are twice as likely to require hospitalisation for COVID-19 compared to those with good glycaemic control, and five times more likely to require admission to an ICU. 24,105 Conversely, those with well controlled diabetes appear to benefit from reduced risk of COVID-19-associated mortality. 106,107 Recent studies suggest that hyperglycaemia is an independent risk factor for COVID-19 mortality regardless of diabetes status.
Morse et al. 108  The mechanisms underlying these observations are yet to be fully elucidated. Since individuals with poorly controlled diabetes are more likely to suffer diabetes-related vascular disease, they may be more vulnerable to the pro-inflammatory, pro-thrombotic effects of COVID-19. As for the independent risk associated with hyperglycaemia, the aforementioned associations between hyperglycaemia and dysregulated expression of ACE2, TMPRSS2 and CD147 (Sections 3.2 and 3.5), and immune cell dysfunction (Section 3.6) may offer some explanation.

| COVID-19 VACCINATION AND DIABETES
In December 2020, the first COVID-19 vaccine (Pfizer BioNTech BNT162b2) was granted emergency authorisation by the United Kingdom's Medicines and Healthcare products Regulatory Agency, making it the first mRNA vaccine to be approved for widespread use in history. Since then, several other vaccines based on mRNA and viral vector technologies have been developed and approved worldwide. Evidence continues to mount that these vaccines are efficacious in the general population. 7 Yet, it is also increasingly apparent that people with comorbid conditions suffer worse outcomes in COVID-19 infection. 29 In light of these findings, numerous public health authorities have prioritised people with comorbid conditions, including diabetes, in their vaccination efforts. The question therefore arises: how efficacious are these vaccines for people with diabetes mellitus?
Prior to the authorisation of COVID-19 vaccines, studies were conducted to examine the immune response in diabetes to natural COVID-19 infection. In a study of 150 patients hospitalised with COVID-19 pneumonia, serum from patients with diabetes produced similar neutralising antibody responses to those without diabetes. 110 This response was unaffected by blood glucose levels, and the presence of neutralising antibodies was found to be protective against COVID-19-associated mortality, regardless of diabetes status. 110  positivity for antibodies against SARS-CoV-2 is associated with increased likelihood of survival. [111][112][113] It is therefore useful to examine seroconversion rates for people with diabetes in response to COVID-19 vaccination. To date, many studies have observed no statistical difference in seroconversion rates between people with and without diabetes in response to COVID-19 vaccination. [114][115][116] Whitaker et al.'s large study 117 from England observed lower seroconversion rates in people with diabetes after receiving the first dose; however, after the second dose, 100% of people with diabetes had detectable anti-spike antibodies.
There is little consensus over what quantitative antibody levels confer protection from COVID-19, so caution should be exercised when interpreting differences in antibody levels in the absence of clinical outcomes. 118 Whitaker et al. 117 observed no statistically significant difference between the anti-spike antibody levels produced by people with and without diabetes who received the AstraZeneca vaccine (a viral vector based vaccine). However, of the people who received the Pfizer BioNTech vaccine, those with diabetes produced a median antibody titre that was 22% lower compared to those without diabetes.

| What role does diabetes play as an accelerator for adverse outcomes in COVID-19?
Diabetes is associated with adverse outcomes from COVID-19 infection. To date, several meta-analyses describe increased risks of severe illness in people with diabetes compared with the general population. [15][16][17][18][19]30 This is seen across a variety of different outcomes, such as the requirement for mechanical ventilation, ICU admission, and composite scores, that consider a vast array of clinical information. 12,20,21 Moreover, these patients are at an increased risk of experiencing complications, such as ARDS, acute cardiac injury, sepsis, and mortality. 51,52,101 Few studies differentiate between type 1 and type 2 diabetes though the available literature is suggestive that both types increase the risk of adverse outcomes. 39,43 The burden of diabetes may extend beyond the acute phase of infection. Though there is limited research focused specifically on interplay between diabetes and post-COVID-19 sequelae, the available evidence suggests that people with diabetes are more likely to suffer from fatigue, cardiorespiratory, and neurological symptoms several months after infection. 57,58 Furthermore, COVID-19 may exacerbate pre-existing sarcopenia in people with type 2 diabetes. 58,59 Questions have been raised as to how long these will persist, adding to the COVID-19-associated burden suffered by people with diabetes. 51,52,101 Further research is required to establish this risk.

| What mechanisms underly the differences of outcomes seen in people with diabetes?
To date, several mechanisms have been suggested to explain these phenomena though many rely on theoretical application of prepandemic knowledge about immunity in diabetes mellitus. Some theories focus on diabetes causing an overexpression of proteins that facilitate viral entry, such as ACE2, TMPRSS2, Furin, and CD147, whilst others centre around the dysfunctional immune response that is known to occur in diabetes. 71,78,84 Examples include impaired CD8 + T cell, NK cell, and phagocytic responses, which are likely to impede the clearance of virus-infected cells. 93,99,103 The current literature emphasises the role of hyperglycaemia in exacerbating the aforementioned mechanisms. Given that diabetes is a condition of chronic hyperglycaemia, it is highly probable that glycaemic control plays a role in worsening outcomes in COVID-19 infection. Indeed, people with poorly controlled diabetes are at a greater risk of hospitalisation and severe illness than those with well controlled diabetes. 24,105,108

| Are vaccines efficacious in diabetes?
Overall, the current evidence suggests that people with diabetes can mount effective immune responses against both natural infection and vaccination. Seroconversion rates do not differ significantly between people with and without diabetes which, in theory, should be protective against mortality. [114][115][116][117] Clinical studies describe a relative difference in vaccine effectiveness at preventing breakthrough infection, symptomatic illness, and hospitalisation in people with diabetes compared to without. Nevertheless, vaccination is still highly effective in this population, and no differences in mortality prevention have been observed. 119 It has come to light that poor glycaemic control may impair the immune response to vaccination. 115,121 Further studies are required to confirm this association; however, considering the existing evidence for poor glycaemic control compounding adverse outcomes COVID-19, careful monitoring and optimisation of glycaemic control is advised.

| Implications
People with diabetes are at increased risk of severe illness and death, and are likely to suffer considerable burden from post-COVID sequelae. As such, public health authorities should continue to recognise this population as an 'at-risk' group. Research into the mechanisms underpinning this excess risk highlights the role of hyperglycaemia in promoting dysfunctional immune responses and a pro-inflammatory state. Healthcare providers should encourage patients with diabetes to diligently monitor their blood glucose levels, routinely monitor longer-term glycaemic control, and use appropriate therapeutics to optimise accordingly. With evidence of COVID-19 exacerbating sarcopenia in people with diabetes, efforts should be made to support these individuals; recent literature highlights the utility of physiotherapy in post-COVID sarcopenia. 122 Vaccination is likely to be efficacious in preventing severe illness and mortality in people with diabetes. In light of the increased risk this population face from COVID-19, the findings of this review support the decisions of many public health bodies to prioritise this group for COVID-19 vaccination. Patients should be encouraged to take the vaccinations when offered. Table 2 provides a summary of these key findings.

| Questions unsolved
This review has revealed many knowledge gaps that require further research to better understand the intersection between COVID-19 and diabetes. These are further summarised in Table 3.
First, the proposed mechanisms underpinning our understanding of why people with diabetes suffer worse outcomes in COVID-19 are still highly theoretical. Improved knowledge of the molecular mechanisms driving these differences might shed some insight as to possible pharmacological targets for this patient population. The persistence of these symptoms in the long term is yet to be established. A recent study 125 indicates that the resolution of long COVID symptoms occurs in the general population. It is therefore pertinent to establish the persistence of symptoms affecting people with diabetes. A prevailing barrier to this is the high degree of heterogeneity of the literature, which prevents the acquisition of higherlevel evidence, such as meta-analyses. Nonetheless, this information T A B L E 2 COVID-19, type 1, and type 2 diabetes: summary findings.

Outcomes in COVID-19 and Diabetes
� Diabetes is associated with an increased likelihood of severe disease course and mortality risk from COVID-19. � Though few studies differentiate between type 1 and type 2 diabetes, the current literature suggests that the increased risk of adverse outcomes occurs in both types of diabetes.
Post-COVID sequelae � People with diabetes are more likely to experience post-COVID sequelae. � Commonly described symptoms in people with diabetes include fatigue, dyspnoea, chest pain, and muscle weakness. � Further longer-term studies are required to assess the persisting effects of COVID-19 on people with diabetes.
Mechanisms underpinning different outcomes � Several mechanisms have been suggested to explain the increased risk of adverse outcomes seen in people with diabetes. � The current evidence poises ACE2 dysregulation, ACE2 autoantibodies, furin, CD147, and immune cell dysfunction as playing a key role in this.
COVID-19 vaccination � The available vaccines appear to be efficacious in preventing severe illness and mortality from COVID-19 in people with diabetes but may be less effective at preventing infection. � Hyperglycaemia appears to impair the immune response to vaccination, highlighting the importance of optimising glycaemic control in people with diabetes.
T A B L E 3 Summary of the key questions that remain unsolved.
Question 1 � What molecular mechanisms underpin the discrepancy in outcomes between people with diabetes and without?
Question 2 � What is the functional impact of post-COVID sequelae on people with diabetes?
Question 3 � What are the common post-COVID symptoms experienced by people with diabetes, how long do they persist, and how are they best managed? Question 4 � How does the long-term efficacy of COVID-19 vaccinations compare between people with diabetes and without?
Question 5 � What is the quantitative level of antibodies that confers protection against COVID-19 from infection, severe illness, and death? D'SOUZA ET AL.
-9 of 14 will allow for healthcare providers to target vulnerable populations with appropriate follow up care and treatment.
Finally, there is still limited research comparing the long-term efficacy of vaccination in people with diabetes to those without.
Given that the dysfunctional immune response seen in people with diabetes is already well established, this presents a crucial gap in the literature. Moreover, future studies should aim to identify the antibody titres that confer protection against infection, severity, and mortality in people with diabetes. This information would allow antibody levels to be monitored and additional booster doses to be administered if required.

AUTHOR CONTRIBUTIONS
F.D. was the main writer of this manuscript. R.B. and P.P. provided substantial contribution to the structuring and content design of this review, as well as providing critical feedback for revisions. All authors have read and approved this manuscript for submission.