Endothelial Dysfunction and Platelet Hyperactivation in Diabetic Complications Induced by Glycemic Variability.

The development and progression of the complications of chronic diabetes mellitus are attributed not only to increased blood glucose levels but also to glycemic variability. Therefore, a deeper understanding of the role of glycemic variability in the development of diabetic complications may provide more insight into targeted clinical treatment strategies in the future. Previously, the mechanisms implicated in glycemic variability-induced diabetic complications have been comprehensively discussed. However, endothelial dysfunction and platelet hyperactivation, which are two newly recognized critical pathogenic factors, have not been fully elucidated yet. In this review, we first evaluate the assessment of glycemic variability and then summarise the roles of endothelial dysfunction and platelet hyperactivation in glycemic variability-induced complications of diabetes, highlighting the molecular mechanisms involved and their interconnections. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial-License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Diabetes mellitus (DM) is the third most common chronic disease worldwide and a serious threat to human health and life 1 . DM-related macrovascular and microvascular complications, including coronary heart disease, cerebrovascular disease, heart failure, peripheral vascular disease, diabetic retinopathy (DR), neuropathy, and nephropathy, impair the quality of life and cause disability and premature death 2 .

Fasting plasma glucose (FPG), postprandial glucose excursions, and hemoglobin A1c (HbA1c), described as the “glucose triad,” are the main parameters used in monitoring patients with type 2 diabetes (T2D) 3 . Among elderly patients with T2D, all-cause mortality 4 , including cardiovascular disease mortality 5 , is mainly related to the variability or instability of fasting glycemia rather than its absolute values. In the Veteran Affairs Diabetes Trial, longitudinal variations in FPG were associated with all-cause mortality, even when accounting for standard measures of glucose control, as well as comorbidity and lifestyle factors 6 . Notably, many studies have found that HbA1c does not fully explain the risk of chronic complications of T2D. Therefore, other reliable and accurate monitoring parameters of diabetic complications need to be explored.

Good glucose control is one of the most effective means to prevent the complications of advanced DM 7 . Impaired glucose homeostasis is the main risk factor for cardiovascular disease 8 ; thus, glycemic variability (GV) might be as important as the glucose triad 9 . Additionally, several studies have confirmed that patients with DM and fluctuant hyperglycemia have higher risks of chronic vascular complications than those with persistent hyperglycemia 10 11 . Furthermore, greater degrees of glycemic fluctuation are associated with a higher incidence of complications and worse prognosis 12 . Thus, GV has become a research hotspot in the field of DM prevention and treatment.

Hyperglycemia fluctuation is an important factor that causes aggravation of DM-associated vascular complications. Vascular endothelial dysfunction is the initiating factor for the development of atherosclerosis and is an important pathophysiological basis for diabetic vascular disease. Platelet hyperactivation induced by thrombosis is also essential in the development of vascular events. Furthermore, both increased platelet reactivity and endothelial dysfunction are considered a “prothrombotic state” in DM 13 . This review aimed to provide a deeper understanding of the role and mechanism of GV-induced diabetic complications. Toward this goal, we focused on the relationship between endothelial dysfunction, platelet hyperactivation, and GV.

Indicators of GV

Glucose levels can be measured repeatedly in one day (within-day glucose variability) or during more days (between-day glucose variability) 14 . Another method is continual measurement of glucose levels using a continual glucose monitoring (CGM) system 15 .

Dysglycemia in diabetes can be classified into two mechanisms: sustained chronic hyperglycemia and acute fluctuant fluctuations over a daily period 16 17 . The former is integrated by HbA1c, which depends on both interprandial and postprandial hyperglycemia, and the percentage of each contributor is modulated by the degree of diabetic control 18 . There are many parameters for the clinical evaluation of GV 19 , with the most common measures being the following ( Table 1 ): (1) assessment of within-day blood glucose variability, including the mean amplitude of glycemic excursions (MAGE), largest amplitude glycemic excursion (LAGE), standard deviation (SD) of all blood glucose measurements, and high/low blood glucose index (HBGI/LBGI); (2) assessment of day-to-day blood glucose variability, including the FPG coefficient of variation (CV), mean of daily difference (MODD), and average daily risk range (ADRR); (3) assessment of postprandial blood glucose fluctuation, such as the mean indices of meal excursions (MIME); and (4) special assessment for islet transplantation of type 1 DM (T1D), such as the lability index (LI).

Table 1 Definition of the various indices used to assess glycemic variability (GV).

Measure [Ref]	Description	Advantages	Limits
Mean amplitude of glycemic excursions (MAGE) 20 21 22	Mean of glycemic excursions from nadir to peak blood glucose level and vice versa that are>1 SD of blood glucose mean	It is a diabetes-specific metric of the amplitude of glucose excursions.	It considers glycemic peaks and nadirs occurring daily but does not account for the total number of fluctuations; it depends on sampling frequency; it is ambiguous as to where peaks and nadirs begin and end.
Largest amplitude glycemic excursion (LAGE) 23	Maximal sensor glucose levels minus the minimal daily sensor glucose levels	It can reflect variations in the characteristics of within-day and day-to-day blood glucose.	It cannot reflect the frequency of fluctuations or full level of GV for a single day or several days.
Standard deviation (SD) 24 25	Variation around the mean blood glucose (intra-day or inter-day) 26	It is a simple, classical statistical method.	It combines information on variability from different sources; it does not address non-Gaussian skewed data.
Coefficient of variation (CV)=SD/mean	Magnitude of variability relative to mean blood glucose 27 28	It can be used to assign more importance to hypoglycemia than to hyperglycemia.	It is subject to the same limitations as SD. It fails to provide enough weight to hypoglycemic values.
Low blood glucose index (LBGI) 29	For glucose values<112.5 mg/dl, average of 27.695×{[log(glucose)] 1.084 – 5.381}	Heavier weights are assigned to severe hypoglycemic values.	The mathematical form is obscure.
High blood glucose index (HBGI) 30	For glucose values>112.5 mg/dl, average of 27.695×{[log(glucose)] 1.084 – 5.381}	Heavier weights are assigned to severe hyperglycemic values.	The mathematical form is obscure.
Mean of daily difference (MODD) 31	It is calculated as the average of the absolute difference between values on different days but at the same time for two consecutive days.	It can be used to assess the continuous variability of blood glucose at the same time between different days.	It may be affected by insulin injections.
Average daily risk range (ADRR) 32	Blood glucose is continuously monitored for 14–28 days at least four times a day. The results are converted to obtain ADRR, which is used to evaluate long-term GV.	It is the best predictor for variations of hypoglycemia and hyperglycemia, independent of the type of diabetes.	Patients are required to master self-monitoring of blood glucose. Because of the high monitoring frequency, long duration, and low patient compliance, this is less frequently applied in clinical practice.
Mean indices of meal excursions (MIME) 33	These include postprandial spike (PPGE), peak-reaching time, and percentage decrease in blood glucose 1 h after peaking (BR). PPGE is the difference between a postprandial spike and the corresponding preprandial glucose.	Dynamic changes in PPGE can be visually shown in detail.	It is related to mealtime, type of food, and eating style. Changes in postprandial levels during different days cannot be observed.
Lability index (LI) 34	It is calculated based on changes in glucose levels over time using 4-week glucose records and compared with a clinical assessment of glycemic lability.	It can be used as an indicator of patient prognosis.	It is only applicable to patients with type 1 diabetes mellitus having solitary islet transplantation with recurrent severe hypoglycemia and labile glucose control.
Continuous overall net glycemic action (CONGA-n) 35	It measures the intraday glycemic swings occurring over predetermined intervals.	It provides an accurate measure of intra-day glycemic variability.	It is difficult to calculate.

The development of CGM technology greatly expands the ability to assess glycemic control throughout the day and has enabled research on the influence of acute blood glucose fluctuations in real life 36 . The 2017 Advanced Technologies & Treatments for Diabetes consensus conference 37 identified “time in ranges” as a metric of glycemic control that provides more actionable information than HbA1c alone. The metric includes three key CGM measurements: percentage of readings and time per day within target glucose range [TIR, 70–180 mg/dl (3.9–10.0 mmol/l)], time below target glucose range [TBR, 70 mg/dl (3.9 mmol/l)], and time above target glucose range [TAR, 180 mg/dl (10.0 mmol/)] 38 .

These variability parameters are important in selecting the optimal treatment strategies and estimating the risk of chronic DM complications. GV is closely related to diabetic complications and monitoring it could help in controlling or reducing the risk of complications. Therefore, the parameters of both short- and long-term GV should be further explored. Recently, numerous studies have supported the hypothesis that GV acts as an important determinant in both the genesis and development of diabetic vascular complications 39 40 41 42 43 44 .

GV and vascular complications of DM

Diabetic vascular complications are conventionally classified as microvascular and macrovascular according to the size and location of the blood vessels involved 45 . Several studies have demonstrated that the presence of macro- and microvascular complications in patients with DM is related not only to chronic hyperglycemia represented by HbA1c but also to acute glycemic fluctuations 46 47 . There is substantial evidence supporting that GV has drawn a great attention for its role in macro- and microvascular complications in patients with T1D or T2D 48 .

GV and microvascular disease of DM

Microvascular disease, mainly including retinopathy, nephropathy, and neuropathy, is strongly associated with hyperglycemia. Recently, studies have shown a positive association between GV and microvascular complications of diabetes 49 . The present review suggests that increased levels of short-term glucose variability, particularly in FPG levels, may contribute to the development of microvascular complications in T2D, whereas the role of increased short-term glucose variability in the development of microvascular complications in T1D is less evident 50 .

Retinopathy

The risk of development and progression of retinopathy has been linked to glycemic exposure in many studies from the Diabetes Control and Complications Trial in 1995 51 . DR is currently the leading cause of blindness among working-aged persons in the developed world 52 . Many studies in the present literature indicate that short-term GV may contribute to the development or progression of DR in T2D, while long-term GV, represented by HbA1c, appears to play a more important role in retinopathy in patients with T1D or T2D 53 . One study of 415 DR patients with T1D in a tertiary referral centre suggested prevention and early detection of retinopathy in T1D patients, to take HbA1c variability into account when optimizing glycemic control 54 .

Nephropathy

In a cross-sectional study of patients with T1D, higher glucose variability, as estimated by SD, CV, and MAGE, was found significantly more often in those with elevated albuminuria than in those with normal albuminuria, although their mean HbA1c was comparable 50 . Moreover, in vivo studies have revealed that in patients with T2D, GV results in chronic kidney disease characterized by progressive proteinuria, which ultimately leads to end-stage renal failure 55 .

Neuropathy

Many forms of neuropathy can occur, including sensory, motor, and autonomic neuropathies, in the setting of diabetes after the exclusion of other causes. Several retrospective longitudinal studies on patients with T1D have demonstrated that glycemic fluctuations may contribute to diabetic peripheral neuropathy 56 and cardiovascular autonomic neuropathy 57 . Moreover, high levels of GV, assessed via CGM, appear to have even more deleterious effects than sustained hyperglycemia in the pathogenesis of cardiovascular autonomic neuropathy and other cardiovascular complications in patients with T2D 58 59 .

Both GV, as derived from visit-to-visit FPG measurements using CV, and HbA1c ≥53 mmol/mol were potent predictors of diabetic peripheral polyneuropathy (DPN) in patients with T2D. The associations among HbA1c, GV, and DPN suggest a linked pathophysiologic mechanism, which may play a crucial role in clinical risk assessments 60 .

GV and macrovascular disease of DM

There are three major manifestations of macrovascular disease: coronary artery disease, cerebrovascular disease, and peripheral artery disease. In patients with T1D and T2D, the role of increased short-term GV in the development of both micro- and macrovascular complications is less evident 50 .

Coronary artery disease

DM increases the risk of myocardial infarction (MI) more than any other risk factor aside from cigarette smoking. Coronary artery disease is the most common macrovascular complication recorded in patients with DM 61 . Fluctuant hyperglycemia has adverse effects on blood vessels that lead to cardiovascular and peripheral vascular diseases. Patients with T2D complicated by coronary heart disease have higher GV than those without coronary heart disease 62 . In addition, increased visit-to-visit GV has been shown to predict mortality in patients with T2D and acute MI 63 . Previous studies have already correlated an increased GV to the occurrence of adverse events in patients with acute coronary syndromes undergoing percutaneous coronary revascularization (PCI) 64 65 .

Cerebrovascular disease

Stroke incidence is elevated in the diabetic population, claiming a high cost in terms of morbidity and mortality 66 . More severe glycemic fluctuations in patients with T2D are associated with a greater risk of cerebral infarction and worse prognoses 67 .

Peripheral artery disease

The risk of developing peripheral artery disease (PAD) is increased two- to four-fold in patients with DM compared to those without 68 . One-fourth of patients with PAD demonstrate progression of symptoms over 5 years and a rate of amputation of around 4%. Diabetes not only affects large-calibre peripheral vessels but distal arterioles as well 66 . A recent large, retrospective cohort study reported that HbA1c and GV, as estimated using FPG-CV, were risk factors for PAD aside from other conventional risk factors in persons with T2D 69 .

Endothelial dysfunction in GV induces DM-associated complications

Endothelial dysfunction is defined as a disorder in the capacity of the endothelium to maintain vascular homeostasis 70 . It represents one of the most important determinants of coronary vascular disease 71 . Endothelial dysfunction could contribute to insulin resistance, potentially unifying the etiology of DM and coronary vascular disease 72 . Many studies have confirmed that fluctuant hyperglycemia is more likely to cause vascular endothelial dysfunction than persistent hyperglycemia 73 . Moreover, glucose fluctuation negatively influences endothelial function 74 .

Endothelial dysfunction arising from a chronic hyperglycemic state is the result of increased oxidative stress and overproduction of reactive oxygen species (ROS), reduced nitric oxide (NO), and increased expression of adhesion molecules and inflammatory reactions ( Fig. 1 ).

Fig. 1

Schematic representation of the main processes involved in the pathogenesis of endothelial dysfunction in patients with diabetes with glycemia variability (GV).

Oxidative stress plays a critical role in the pathogenesis of diabetic complications and several vascular diseases 75 . As stated in the unifying theory for DM complications proposed by Brownlee, oxidative stress response accompanied by sharp fluctuations in blood sugar is an important mechanism of vascular endothelial dysfunction 76 . The association between GV and oxidative stress has also been investigated using CGM. Monnier et al. 77 found that the average 24-hour urinary excretion rate of free 8-iso-prostaglandin (PG) F2α, a marker of oxidative stress, in 21 patients with T2D significantly correlated with MAGE, a marker of GV assessed via CGM, and with the area under the curve of the mean postprandial increment of glucose level above preprandial values. This suggests that glycemic fluctuations during postprandial periods are more likely to induce oxidative stress. In vitro studies and animal models have also substantiated these results.

Our study found that intermittent high glucose levels (5.56–25 mmol/every 24 h) induced oxidative stress injury with an increase in advanced oxidative protein products (AOPPs) and a decrease in total antioxidant capacity (T-AOC), which led to increased cellular apoptosis in human umbilical vein endothelial cells (HUVECs), compared with a constant high-glucose setting (25 mmol/l). Furthermore, the trend of these indices was verified in streptozotocin (STZ)-induced diabetic rats with fluctuating hyperglycemia treatment 78 . These effects were amplified during glucose fluctuations, consistent with previous observations 79 80 .

NO produced by nitric oxide synthases (NOS) is the smallest gaseous intercellular messenger involved in the modulation of several processes (e. g., blood flow and platelet aggregation control) and is essential for maintaining vascular homeostasis 81 . Aside from being a vasodilator, NO reduces vascular permeability and the synthesis of monocyte and lymphocyte adhesion molecules, which contribute to the reduction of tissue oxidation and inflammation, platelet aggregation, and thrombogenic factor activation. These processes, in turn, lead to the reduction of typical inflammatory processes induced by hyperglycemia 82 . Impaired endothelial NOS activity and enhanced ROS production in DM result in diminished NO bioavailability and vascular damage 83 . Therefore, NO is considered an important anti-atherogenic molecule that is necessary to contain diabetic endothelial alterations 84 85 .

Inflammation is widely considered a key etiological factor that plays a vital role in the development of diabetic complications 86 . Otsuka et al. injected glucose into the intraperitoneal space of Sprague-Dawley rats to cause a temporary increase in blood glucose. The results indicated that a transient increase in blood glucose could induce increased adhesion of monocytes and endothelial cells in the thoracic aorta 87 . Studies by Watada et al. 88 and Mita et al. 89 in rats have also shown that repetitive postprandial hyperglycemia could promote the adhesion of monocytes, macrophages, and aortic endothelial cells more than sustained hyperglycemia, thus increasing the surface area of arteriosclerotic injury. Additionally, studies in HUVECs have also yielded similar results, further confirming that fluctuant hyperglycemia can significantly increase the levels of inflammatory indicators (interleukin-6, tumor necrosis factor-α) and expression of adhesion molecules (ICAM-1, VCAM-1, and E-selectin) 90 91 92 . In human circulation, fluctuant hyperglycemia could induce tumour necrosis factor-α production in vivo 93 . A previous study showed that high glucose altered the expression profile of cytokines and chemokines via specific signalling pathways and can increase monocyte-endothelial adhesion in monocytes 94 . Acute glucose fluctuation may cause significant oxidative stress and inflammation in rat aortic endothelial cells, increase the adhesion of monocytes to rat aortic endothelial cells, and elevate endothelial cell apoptosis, resulting in severe cardiovascular injury 95 .

Platelet hyperactivation in GV-induced DM-associated complications

Increased coagulation, impaired fibrinolysis, and endothelial dysfunction result in a prothrombotic state for which platelet hyperreactivity is said to be an established contributing factor. Complications arise owing to these hyperactive platelets that play a vital role in the pathophysiology of thrombotic events 96 .

Platelets could both trigger thrombus formation and release oxidative, mitogenic, and vasoconstrictive substances that induce the development of local vascular lesions 97 . Studies have shown that platelet aggregation is significantly enhanced and is in a hyperactive state in patients with DM, with consequent increase in microcapillary embolization and accelerated local vascular lesions 98 99 . Increased platelet activities, such as altered morphology and function, may play a vital role in the development of diabetic vascular complications 100 . A study found that increased levels of the mean platelet volume, plateletcrit (PCT), and platelet-large cell ratio (P-LCR) are positively associated with the occurrence of increased HbA1c, retinopathy, nephropathy, and neuropathy individually in patients with DM 101 . Chronic hyperglycemia has been established as the cause of platelet activation and platelet hyperactivity in patients with DM 102 . For example, the profound increase in urinary excretion of 11-dehydro-TXB2 in patients with T2D suggests that acute hyperglycemia induces increased platelet activation from high shear-stress conditions 103 . T2D is associated with a greater production of 8-iso-PGF2α, a stable compound of non-cyclooxygenase peroxidation of arachidonic acid inducing vasoconstriction and platelet activation 104 . Basili et al. 105 found that acute glucose fluctuations strongly correlated with urinary excretion of 8-iso-PGF2α, but no relationship was observed when urinary 8-iso-PGF2α excretion rates were plotted against main markers of sustained hyperglycemia (HbA1c and mean daily glucose concentrations). In our previous study, we explored platelet aggregation in HUVECs exposed to different GV media and healthy platelets and found that endothelial cells intermittently incubated with high glucose showed a more relevant increase in maximum platelet aggregation rate than the increase observed in the stable high-glucose condition 106 .

The significant factors causing increased platelet reactivity in patients with DM are hyperglycemia and insulin resistance. Hyperglycemia can cause dysfunctional platelet adhesion, aggregation, and release through the following mechanisms ( Fig. 2 ) 107 108 . First, hyperglycemia is responsible for non-enzymatic glycation of proteins on the surface of the platelet, which decreases membrane fluidity and increases the propensity of platelets to become activated 109 . Next, the platelet activation signalling pathway is ultimately mediated by glycoprotein IIb/IIIa receptor (GPIIb/IIIa) platelet-fibrin interaction. Hyperglycemia leads to the release of larger platelets with more GPIb and GPIIb/IIIa receptors, thus increasing the aggregation baseline activation and thromboxane-forming capacity 110 . Third, both acute and chronic hyperglycemia induce increased protein kinase C (PKC) in vivo , a transduction pathway that triggers platelet activation 111 . Fourth, hyperglycemia promotes increased non-enzymatic glycation of circulating low-density lipoprotein, which may cause platelet dysfunction by increasing intracellular calcium concentration and platelet NO production 112 113 . As an important second messenger, calcium participates in the regulation of a series of platelet functions, including morphological changes, secretion, aggregation, and thromboxane synthesis. Lastly, hyperglycemia induces the coagulation mechanism by increasing the release of prothrombotic molecules (e. g., tissue factor and von Willebrand factor [vWF]) while inhibiting fibrinolysis by increasing plasminogen activator inhibitor-1 (PAL-1) concentration 114 115 . A high level of vWF in the circulation correlates with increased platelet activation, suggesting that acute, short-term hyperglycemia in T2D may precipitate vascular occlusions by facilitating platelet activation 116 . Additionally, mechanisms of platelet dysfunction in DM include upregulation of platelet P2Y purinoceptor 12 signalling, increased generation of thrombin, increased production of thromboxane A2 from arachidonic acid metabolism, and accelerated platelet turnover 117 .

Fig. 2

Schematic representation of the main processes involved in the pathogenesis of platelet hyperactivation in patients with diabetes with glycemia variability (GV). PKC: Protein kinase C; GP: Glycoprotein.

Nowadays, several studies have demonstrated that some antidiabetic agents also have antithrombotic effects. The potential benefit to platelets may be related to the normalization of glycemic control, but other additional direct antithrombotic and anti-inflammatory mechanisms may be involved. The modulation of platelet activation by antidiabetic drugs may mitigate the risk of thrombotic events and contribute to cardiovascular protection in patients with DM 118 . Metformin, recommended as first-line therapy for newly diagnosed T2D by the American Diabetes Association, is associated with decreased cardiovascular risk, such as significant reduction in cardiovascular endpoints (MI and stroke) or all-cause mortality 119 and decreased macrovascular complications (MI, stroke, peripheral vascular disease) in patients with DM already on insulin therapy 120 .

Interaction between endothelial dysfunction and platelet hyperactivation in vascular complications of DM

Several studies have demonstrated that hyperglycemia is the main mediator of endothelial dysfunction and platelet hyperaggregation in DM, which contributes to the development and progression of vascular complications. Interestingly, both endothelial dysfunction and platelet hyperactivation also regulate hyperglycemia-induced diabetic complications ( Fig. 3 ). However, their influence on this process remains unclear and needs further investigation.

Fig. 3

Schematic representation of the interaction between endothelial dysfunction and platelet hyperactivation in diabetic vascular complications with glycemia variability (GV). vWF: von Willebrand factor; AGE: Advanced glycation end-product.

Oxidative stress is the first link in the interaction between endothelial dysfunction and platelet hyperactivation in vascular complications of DM. Superoxide may increase platelet reactivity by enhancing intraplatelet activity to release calcium after activation 121 . In addition, oxidative stress impairs endothelial function and reduces the production of NO, thus increasing platelet reactivity 122 .

The second link of the interaction is vWF, which is a glycoprotein released into the circulation by secretion from endothelial cells. Studies have reported that an elevated level of vWF is associated with a higher risk of thrombotic cardiovascular events in patients with DM 123 124 . Hu et al. 125 showed that hyperglycemia-induced repression of microRNA-24 increases vWF expression and secretion in both patients with DM and diabetic mouse models. When the endothelial layer is disrupted, vWF binds to exposed collagen and then anchors platelets to the subendothelium. In addition, vWF is also able to bind to GPIb-IX and IIb-IIIa platelet receptors, promoting platelet aggregation and causing platelet plug formation 126 .

Advanced glycation end-products (AGEs) mediate the linkage between endothelial dysfunction and platelet hyperactivation. Previous data have indicated that increased AGE production under hyperglycemic conditions can induce decreased endothelial NOS expression and increased endothelin-1 expression and ROS through AGE-specific receptors, leading to endothelial dysfunction 127 128 . This further results in impaired vasodilatory response to NO 129 and enhancement of platelet aggregation in vivo and in vitro 130 . Another mechanism linking AGEs and platelet hyperreactivity is the increased expression of CD36, CD62, and CD63 on the platelet surface membrane 131 132 , which is associated with enhanced platelet reactivity in vitro as well as enhanced arterial thrombosis in vivo .

Conclusions

The long-term hyperglycemic state in patients with DM leads to the development and progression of microvascular and macrovascular complications of DM and increases their all-cause and cardiovascular mortality. Current research shows that GV leads to the development of vascular complications in patients with DM. Therefore, aside from FPG, postprandial blood glucose, and HbA1c, glycemic fluctuations should also be considered when formulating a blood glucose-lowering regimen. In addition, GV should be a target of clinical intervention. GV measurement has become more accurate with the development of new technologies (e. g., CGM systems) and provides directions for clinical and basic research. However, there is still no gold standard for measuring GV. Further, more studies on the effects of glycemic fluctuations on vascular complications in patients with DM should be conducted, with the reduction of glycemic fluctuations as a therapeutic target. This will help clarify the effect of GV on hard endpoints such as the development and progression of microvascular and cardiovascular events in patients. An increasing number of clinical and basic studies are investigating the molecular mechanism of vascular complications mediated by GV. It appears that endothelial dysfunction and platelet hyperactivation mediated by excessive oxidative stress play important roles, and in-depth research on molecular mechanisms related to GV is needed to obtain data for developing novel treatment regimens. Due to two mechanisms of blood glucose fluctuations mediating diabetes complications in our paper, the literature cited were mainly basic studies. So clinical studies were not mentioned, which was our limitation. However, our previous study focused on the relationship between blood glucose fluctuation and the prognosis of diseases, especially in critically ill patients 133 . The correlation between GV and macro- or micro-vascular complications of T1D or T2D as well as other critical diseases remains to be further explored.

Author Contributions

Conceptualization and funding acquisition: YH and JW; Writing of the original draft: YH and LY; Writing, review, and editing: JQ, MG, JW; Supervision: SL. All authors read and approved the final manuscript.