Insulin resistance and coronary microvascular dysfunction: a complex interplay

Abstract

Up to 40–60% of patients undergoing coronary angiography because of angina and/or evidence of inducible ischaemia on non-invasive stress testing are diagnosed with ischaemia and non-obstructive coronary artery disease (INOCA). The pathogenesis of this condition is primarily attributed to two mechanisms: coronary microvascular dysfunction (CMD) and coronary vasospasm. Notably, ischaemic heart disease is the leading cause of death in patients with type 2 diabetes mellitus (T2DM). Insulin resistance (IR), affecting 10–25% of the general adult population, plays a major role in the pathophysiology of T2DM, but can precede diabetes by years. IR is recognised as a major cardiovascular risk factor, involved in endothelial dysfunction and inflammation, two key processes leading to CMD and vasomotor dysfunction. Hyperinsulinaemia, dysglycaemia, and oxidative stress contribute to this complex relationship, yet the connection between IR, CMD and coronary vasospasm remains incompletely defined. Moreover, IR may represent a target for tailored therapies aimed at improving microvascular function and alleviating symptom burden. Although a few studies have investigated this relationship, the molecular mechanisms by which multiple pathways lead to different INOCA endotypes remain incompletely defined. The aim of this review is to summarise current evidence linking IR, CMD and coronary vasospasm, with emphasis on pathophysiological mechanisms and diagnostic approaches, and to highlight future research directions in this clinical setting.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12933-025-03068-x.

Research insight

What is currently known about this topic?

Microvascular dysfunction and microangiopathy are key features of diabetes and insulin resistance.

What is the key research question?

What is the impact of insulin resistance on coronary microvascular and vasomotor function?

What is new?

Insulin resistance is linked to distinct INOCA endotypes via multiple, interrelated mechanisms.

How might this study influence clinical practice?

Therapies targeting insulin resistance may improve CMD and alleviate symptoms in INOCA patients.

Introduction

Ischaemic heart disease is the leading cause of mortality in patients with type 1 and type 2 diabetes [1]. In these patients, myocardial ischaemia is often found even in the absence of obstructive coronary artery disease (CAD) [2, 3]. Coronary microvascular dysfunction (CMD) and coronary vasospasm play a major role in the pathophysiology of ischaemia with non-obstructive coronary artery disease (INOCA), and growing evidence suggests that a personalised diagnostic and therapeutic approach to patients with INOCA can improve quality of life and clinical outcomes [2–4].

Considering the established role of insulin resistance (IR) as a major cardiovascular risk factor and its detrimental effects on the endothelium, an association between IR, CMD and INOCA has often been postulated [2]. However, few studies have investigated the link between IR and CMD, with limitations including the use of surrogate measures of IR and a non-comprehensive evaluation of coronary microvascular function.

In this narrative review we summarise the evidence linking IR and coronary vascular dysfunction–including both CMD and (micro)vascular spasm–with a particular focus on the pathophysiological mechanisms. In addition, we also provide an overview of the available diagnostic techniques to detect IR and CMD. Finally, we aim to highlight current knowledge gaps and future research directions with the goal of developing personalised therapies for patients with IR and INOCA.

Insulin resistance and systemic endothelial dysfunction

Type 2 diabetes mellitus (T2DM) is a chronic metabolic disease, and its time-dependent deleterious effects on the microvasculature are well established [5, 6]. T2DM is defined by hyperglycaemia resulting from tissue insulin resistance and relative insulin deficiency [6]. IR is characterised by a reduced ability of insulin to mediate its metabolic actions on glucose uptake, glucose production, and lipolysis [7]. This condition plays a central role in the pathophysiology of T2DM and may already be present years before the clinical onset of the disease. Long before overt insulin deficiency develops, pancreatic β-cells compensate for reduced insulin sensitivity by increasing insulin secretion, leading to a state of chronic hyperinsulinaemia. This compensatory phase substantially contributes to the development of atherosclerosis and endothelial dysfunction [8]. Therefore, a significant proportion of apparently healthy individuals can be diagnosed with IR when appropriate assessments are performed [9]. The prevalence of IR in the general population ranges from 10 to 25% among adults, and it is even higher in specific populations, such as patients with polycystic ovary syndrome, metabolic syndrome, or obesity [10].

Under physiological conditions, insulin acts on multiple processes, stimulating the use of metabolic substrates in multiple tissues, including cardiomyocytes, and leading to an integrated signal that ensures the correct balance between nutrient supply and demand [11]. This metabolic regulation is tightly coupled with the control of tissue perfusion, highlighting the role of insulin as a key modulator of vascular function. Vascular tone is regulated by three types of endothelium-derived relaxing factors, including prostacyclin (PGI2), nitric oxide (NO), and endothelium-derived hyperpolarisation (EDH), as well as by the contractility of vascular smooth muscle cells (VSMCs) [12]. Notably, insulin contributes to the regulation of vascular function through all these mechanisms [13]. Cellular responses to insulin are mainly mediated by the activation of two molecular pathways: insulin receptor substrate 1 (IRS-1)/phosphoinositide 3-kinase (PI3K)/Akt signalling, which promotes vasodilation via endothelial nitric oxide synthase (eNOS), and the mitogen-activated protein kinase (MAPK) signalling, which promotes vasoconstriction through endothelin-1 (ET-1) secretion [14].

In an insulin-resistant state, the target cells fail to respond to ordinary levels of circulating insulin and higher concentrations of the hormone are required to achieve a given effect [15]. This condition leads to impaired glucose uptake in muscles and increased gluconeogenesis by the liver resulting in hyperglycaemia. IR impairs glycogen synthesis and protein catabolism in skeletal muscles and inhibits lipoprotein lipase activity in adipocytes, increasing the release of free fatty acids and inflammatory cytokines, such as interleukin (IL)-6, tumor necrosis factor a (TNFa), and leptin [12]. Simultaneously, the MAPK pathway is overstimulated by hyperinsulinaemia, promoting vasoconstriction, inflammation, mitogenic effects on the endothelium and increased water retention with consequent hypertension and atherosclerosis [16]. In the presence of IR, particularly when coexisting with obesity, mechanical stretch of cells leads to the upregulation of Rho and Rho-kinase, which selectively impairs PI3K-dependent signalling. This mechanism decreases the production of NO from endothelial cells, increases the release of pro-coagulant factors and shifts insulin actions towards hyperreactivity and hypertrophy of VSMCs (Fig. 1) [13, 17–22].

Fig. 1.

Integrated metabolic and vascular effects of insulin resistance and hyperinsulinaemia. Insulin resistance (IR) requires increased insulin production to achieve a given effect. IR and hyperinsulinaemia lead to dysglycaemia and activation of inflammatory pathways, which promote protein kinase C (PKC), Rho kinase, mitochondrial dysfunction, and oxidative stress. Simultaneously, IR also impairs endothelium-derived hyperpolarisation (EDH). Together, these alterations shift insulin vascular effects toward vasoconstriction. Specifically, hyperinsulinaemia overstimulates mitogen-activated protein kinase (MAPK) signalling, which promotes endothelin-1 (ET-1) secretion and mitogenic effects on the endothelium. Moreover, oxidative stress and the activation of several serine/threonine kinases impair the ability of insulin receptor substrate 1 (IRS-1) to bind and activate phosphoinositide 3-kinase (PI3K), resulting in a selective impairment of PI3K/Akt pathway, further weakening insulin vasodilation effects. AGEs: advanced glycation end products; EDH: endothelium-derived hyperpolarisation; ET-1: endothelin-1; IL: interleukin; MAPK: mitogen-activated protein kinase; NO: nitric oxide; PKC: protein kinase C; PI3K: phosphoinositide 3-kinase; TNFα: tumor necrosis factor α: VSMCs: vascular smooth muscle cells

Additionally, insulin has been shown to interact with EDH-mediated vasodilation. EDH acts on Ca²⁺ mobilisation and K⁺ channel activation and is mainly responsible for the modulation of vascular tone in small resistance arterioles. In an insulin-resistant state, EDH-dependent relaxation appears to be initially preserved or even enhanced to compensate for reduced NO bioavailability [23]. This early compensatory mechanism may partly explain the variability in clinical presentation and the heterogeneity in the development of vascular complications among patients with IR.

However, long-term metabolic disorders lead to EDH impairment through increased ROS production and K⁺ channel dysfunction [24].

Glucotoxicity and systemic endothelial dysfunction

Beyond IR and compensatory hyperinsulinaemia, dysglycaemia itself contributes to endothelial dysfunction and it is recognised as a relevant metabolic condition associated with increased cardiovascular risk. Prediabetes is defined by glucose levels higher than normal but still below the diagnostic thresholds for diabetes [25]. Subclinical vascular changes commonly occur in patients with prediabetes, with dysglycaemia promoting endothelial dysfunction through different mechanisms (Fig. 1) [26, 27]. First, altered intracellular glucose metabolism leads to increased glycolytic pathway and enhanced oxidative stress. Second, hyperglycaemia promotes the production of advanced glycation end products (AGEs), which damage endothelial cells by modifying both intracellular and extracellular matrix proteins. Third, the protein kinase C (PKC) pathway is activated, resulting in reduced eNOS activity and further amplification of oxidative stress [27]. Finally, glucotoxicity is also responsible for mitochondrial dysfunction leading to enhanced oxidative stress and apoptosis of endothelial cells [28]. Therefore, therapeutic strategies to maintain or restore mitochondrial function may hold promise for preserving microvascular function in patients with hyperglycaemia, even in the absence of overt diabetes [29, 30].

Assessment of insulin resistance

Identifying the insulin-resistant phenotype is clinically relevant [31]. Therefore, an accurate and reproducible method for diagnosis and quantification of IR is of paramount importance (Table 1).

Table 1.

Assessment of insulin resistance

Method	Formula	Advantage	Disadvantage
Hyperinsulinaemic euglycaemic glucose clamp [32]		Direct measure of insulin sensitivity under steady-state conditions	Laborious, time-consuming and expensive, infusion of insulin and glucose with risk of hypoglycaemia, frequent blood sampling and specialised expertise
ITT [31]		Relatively quick procedure. It allows direct measurement of insulin sensitivity	Risk of hypoglycaemia and unpleasant for the patient. Contraindicated in the presence of cardiovascular diseases because of its neurological and cardiovascular side effects, particularly in diabetic and elderly subjects with diffuse atherosclerotic disease
Minimal model derived from IVGTT [31]		Suitable for dynamic assessment of glucose metabolism	Laborious and time-consuming, requires frequent blood sampling and complex analysis
HOMA-IR [34]	fasting glucose (mg/dl) x fasting insulin (mU/l)/405	Simple, minimally invasive, widely used in clinical practice thanks to its good correlation with clamp method	Low sensitivity, no consensus on cut-off
QUICKI [35]	1/Log (fasting insulin, mU/L) + Log (fasting glucose, mg/dL)	Consistent, precise index of insulin sensitivity, minimally invasive, less affected by glucose variability	Significant inter-laboratory variations in insulin assay
Matsuda [36]	10,000/√[(Fasting glucose, mg/dL × fasting insulin, mU/L) × (mean glucose in OGTT × mean insulin in OGTT)]	Represents both hepatic and peripheral tissue sensitivity to insulin, it includes post-prandial measures	Its correlation is very weak in diabetic patients, more complex since it requires OGTT
TyG index [37]	Log[(triglycerides, mg/dl x fasting glucose, mg/dl)/2]	Simple and minimally invasive using routine clinical measures, precise index of insulin sensitivity in patients with different degrees of glucose tolerance, correlation with HF and CVD	Influenced by acute changes in triglycerides or glucose, less correlation in patients with normal values of triglycerides and fasting glucose or severe dyslipidemia, no universal cut-off

CVD: Cardiovascular disease; HF: Heart failure; HOMA-IR: Homeostasis model assessment of insulin resistance; ITT: Insulin-tolerance test; IVGTT: Intravenous glucose tolerance test; OGTT: Oral glucose tolerance test; QUICKI: Quantitative insulin sensitivity check index; TyG: Triglyceride-glucose index

The hyperinsulinaemic-euglycaemic clamp test is considered the gold standard for assessing insulin sensitivity in vivo. It offers a more reliable estimation of tissue sensitivity to insulin and significant advantages over the other commonly used techniques. During the test, exogenous insulin is administered as a priming dose followed by a constant infusion at a rate designed to maintain a pre-set hyperinsulinaemic plateau. The plasma glucose concentration is held constant at basal levels (around 5.0 mmol/L) by a variable glucose infusion. Under these steady-state conditions of hyperinsulinaemia/euglycaemia, the glucose infusion rate (M_value) equals glucose uptake by all the tissues in the body and is therefore a measure of tissue sensitivity to exogenous insulin [32]. Unfortunately, the clamp method is limited by its procedural complexity, substantial time requirements, the potential for hypoglycaemia, and the necessity for skilled operators [31].

Several alternative indices of IR, derived from plasma glucose and insulin levels either at fasting or after oral glucose load, are widely used in clinical research and are described in Table 1 and in Supplementary Material 1. However, the distribution of these surrogate indices varies according to the characteristics of the subjects, such as age, sex and race, making it difficult to estimate the optimal cut-off point. Thus, all surrogate indices should be validated in each specific setting before they can be appropriately used [33].

Endotypes of ischaemia with non-obstructive coronary artery disease

Coronary microvascular dysfunction. Up to 40–60% of all patients undergoing coronary angiography because of angina and/or inducible myocardial ischaemia at a non-invasive stress testing are classified as INOCA, and in nearly half of them CMD plays a significant role [38]. Moreover, CMD has been observed in several other clinical settings ranging from acute coronary syndromes to heart failure, cardiomyopathies, and valvular heart disease [39]. In all these clinical scenarios, it portends an increased risk of adverse cardiovascular events [39]. CMD is a heterogeneous entity with a complex pathophysiology and is characterised by functional and/or structural abnormalities in the microvasculature. The haemodynamic hallmarks of CMD are coronary flow reserve (CFR), which reflects the ratio of hyperaemic to resting coronary flow, and the index of microcirculatory resistance (IMR). Assessing these two indices helps differentiate functional and structural endotypes across the spectrum of CMD (Fig. 2).

Fig. 2.

Pathophysiology of coronary microvascular dysfunction. Functional coronary microvascular dysfunction (CMD) is defined by impaired coronary flow reserve (CFR) and preserved index of microcirculatory resistance (IMR). It is mainly due to the inadequate vasodilation in response to augmented myocardial oxygen demand; or augmented basal coronary flow. Structural CMD is defined in the presence of increased IMR. It depends on remodelling of the coronary microvasculature with intimal thickening and capillary rarefaction, which are associated with endothelial damage mediated by ROS, blebs and polymorphonuclear cell aggregates (PNAs). CFR: coronary flow reserve; IMR: index of microcirculatory resistance

Functional CMD. Functional CMD is defined in the presence of CFR < 2.5, and preserved IMR (≤ 25 units) [40, 41]. Functional dysregulation occurs mainly in medium- and large-sized arterioles, where flow-mediated vasodilation is the predominant regulatory mechanism. The causes of reduced CFR in functional CMD are twofold: the inadequate increase in coronary flow in response to augmented myocardial oxygen demand; or augmented basal coronary flow, a condition related to altered myocardial metabolism and increased lactate production at rest [40, 42, 43]. Decreased vasodilatory activity may be the result of endothelial injury, which leads to a reduced availability and effect of NO and PGI2 on VSMCs, as well as a primary impairment in the relaxation of VSMCs, not dependent on classic vasodilatory chemicals [44–46].

Structural CMD. Structural CMD is diagnosed in the presence of higher-than-normal microcirculatory resistance (IMR > 25 units) [40]. Structural CMD depends on remodelling of the coronary microvasculature, encompassing microvascular obstruction, luminal narrowing of the intramural arterioles, and capillary rarefaction. It is associated with a decrease in microcirculatory flow and impaired oxygen delivery capacity [47, 48]. Especially when ischaemic injury is present, endothelial protrusions and blebs may obstruct the capillary lumen, while leukocyte plugging promotes the formation of polymorphonuclear cell aggregates (PNAs), which further contribute to microvascular obstruction [49]. These pathological changes directly cause a reduction in the vasodilatory range of the coronary microcirculation, thereby impeding the delivery of maximal blood and oxygen to the myocardium.

Coronary vasomotor dysfunction. Up to three-quarters of INOCA patients exhibit identifiable disorders of coronary vasomotion, including epicardial or microcirculatory vasospasm [50]. Notably, coronary vasomotion disorders often coexist with both CMD and obstructive CAD [40]. These conditions are frequently misdiagnosed due to their transient nature. Therefore, a systematic diagnostic evaluation of coronary microcirculatory function including vasoreactivity testing is recommended, especially in the absence of obstructive CAD.

Epicardial coronary vasospasm. Epicardial vasospastic angina is a clinical condition characterised by enhanced reactivity of VSMCs to endothelium-derived vasoconstrictors. The evaluation of endothelium-dependent vasomotor function entails intracoronary vasoreactivity testing with incremental doses of acetylcholine (ACh) infusion. The neurotransmitter ACh plays a crucial role in flow autoregulation mediating the production of NO by the endothelium. Conversely, in the presence of endothelial dysfunction, ACh primarily binds to VSMCs, leading to paradoxical vasoconstriction [51]. Details of the vasoreactivity test are presented in Fig. 3 and in Supplementary Material 2.

Fig. 3.

Assessment of coronary vasomotor function and diagnosis of epicardial and microvascular spasm. The figure shows a typical case of epicardial spasm elicitated by incremental doses of ACh. In steps 2–3 of the vasoreactivity test, a significant spasm of the distal left anterior descending coronary artery is evident at coronary angiogram. ACh: acetylcholine; ECG: electrocardiogram

Coronary microvascular spasm. When compared with the epicardial compartment, the contribution of NO is comparatively less relevant in the microcirculation, where EDH-mediated vasodilation plays a major role [23]. Microvascular spasm seems to occur mainly in older adults, and it is characterised by an increase in microvascular resistance during ACh infusion [40].

Assessment of coronary microvascular function

Invasive diagnostic technique. The diagnosis of CMD is challenging because the microcirculation cannot be directly assessed using the most common diagnostic techniques [52]. Invasive techniques based on the use of coronary flow and pressure sensor wires are considered the reference standard for the assessment of CMD. These measurements are typically obtained in the left anterior descending artery (LAD), which supplies a substantial proportion of the myocardium [4]. Achieving steady-state hyperaemia is essential for an accurate evaluation of the coronary microcirculation and it is typically obtained by an intravenous infusion of adenosine (140 µg/kg/min), which produces endothelium-independent vasodilation through activation of vascular A2 receptors. The technical description of CMD invasive assessment in the catheterisation laboratory goes beyond the scope of this work and it has been reported in detail elsewhere [53]. The main indices are presented in Table 2 and in the Supplementary Material 3.

Table 2.

Assessment of coronary microvascular dysfunction

Index	Method	Description and significance	Normal values	Advantages	Disadvantages
Non-invasive indices
CFVR [57]	TTE	Ratio of hAPV to bAPV. Moderate correlation with microvascular obstruction at CMR	> 2.0	Non-invasive, low cost and widely available, combined with stress echocardiography	Dependent on acoustic windows, limited to LAD territory, highly operator dependent
MFR [54]	PET	Ratio of stress to rest MBF	> 2.0	Non-invasive reference standard technique	Low availability, high costs
MPR [44]	CMR	Ratio of stress to rest MBF	> 2.0	Non-invasive, combined with structural tissue characterisation	Technically demanding, high costs and time-consuming
Invasive indices
CFVR [58]	Intracoronary Doppler Wire		> 2.5	High temporal resolution (beat-to-beat flow velocity analysis), well validated	Technically demanding, lower signal stability compared to thermodilution methods, limited use in clinical practice
CFR [57]	Bolus or continuous thermodilution	Ratio of peak hyperaemic flow to resting flow, reflecting epicardial and microvascular vasodilatory ability	> 2.5*	Well correlated with clinical outcomes, easy to perform, good reproducibility	Influenced by epicardial disease, indirect measure of flow based on surrogate temperature changes
IMR [51]	Bolus thermodilution	Minimal microvascular resistance on hyperaemia	< 25	Specific of microvascular resistance, easy to perform in cathlab, less affected by haemodynamic conditions	Operator dependent, not validated for all clinical scenarios
MRR [59]	Bolus or continuous thermodilution	Vasodilatory reserve of coronary microcirculation	> 3	Specific of microvascular resistance, integrates both resting and hyperaemic data	Operator dependent (for bolus thermodilution), technical complexity (for continuous thermodilution)

*CFR cut-off values vary according to the diagnostic technique used. Using bolus thermodilution, a CFR < 2 is commonly considered impaired

bAPV: baseline average peak flow velocity; CFVR: coronary flow velocity reserve; CFR: coronary flow resistance; CMR: cardiac magnetic resonance; hAPV: 1⁄4 hyperaemic average peak flow velocity; IMR: index of microcirculatory resistance; LAD: left anterior descending artery; MBF: myocardial blood flow; MFR: myocardial flow reserve; MPR: myocardial perfusion reserve; MRR: microvascular resistance reserve; TTE: transthoracic echocardiography

Non-invasive imaging techniques. Several non-invasive imaging techniques allow the identification of CMD (Table 2). However, non-invasive assessment of CMD should only be considered after the exclusion of obstructive CAD with computed tomography angiography or invasive coronary angiography. Notably, non-invasive techniques can adequately identify flow-mediated CMD, but not vasomotor dysfunction.

Positron emission tomography (PET) allows the measurement of myocardial flow reserve.

(MFR), defined as the ratio of pharmacologically induced hyperaemic to resting MBF. PET is considered the reference standard for the non-invasive assessment of CMD. A value < 2.0 for MFR is generally considered abnormal [54]. Similar to PET, cardiac magnetic resonance (CMR) enables the quantification of MBF at rest and during maximal hyperaemia through the intravenous administration of an endothelium-independent vasodilator, to calculate the myocardial perfusion reserve index (MPRI) [44]. In addition, CMR is also commonly used to detect microvascular obstruction after STEMI, which is associated with adverse clinical outcomes in this clinical context [55].

Transthoracic Doppler echocardiography (TTE) with dipyridamole or adenosine infusion allows the measurement of the coronary flow velocity in the LAD both at rest and during maximal hyperaemia. Coronary flow velocity reserve (CFVR) is then obtained as the ratio between stress to rest diastolic peak flow velocity. A CFVR < 2.0 is commonly used to identify impaired coronary microvascular function [44, 45, 56, 57].

Clinical evidence linking insulin resistance to coronary microvascular dysfunction

Systemic microvascular dysfunction and microangiopathy are common features of T2DM and insulin-resistant states [60]. The coronary microvascular bed is particularly vulnerable to metabolic derangements, as it provides a larger endothelial surface area per gram of tissue than any other organ. Notably, coronary endothelial cells account for nearly one-third of all cardiac cells, highlighting the central role of endothelial homeostasis in myocardial perfusion [60].

This distinctive anatomical and functional profile may explain why individuals with IR and T2DM have a higher prevalence of CMD and an increased risk of heart failure and sudden cardiac death, even in the absence of obstructive CAD [61]. Although the association between IR and CMD has been increasingly explored, its underlying pathophysiological mechanisms remain incompletely defined.

Evidence consistently supports a link between IR and CMD across different clinical settings (Table 3). Among patients with CMD, Gallinoro et al. identified structural CMD as the predominant endotype in individuals with diabetes compared with those without diabetes (78% vs. 58%, p = 0.03) [62]. Conversely, Picchi et al. demonstrated a marked reduction in CFR in patients with T2DM compared with those without T2DM, with no significant differences in IMR, suggesting preferential impairment of flow-dependent mechanisms [63].

Table 3.

Current evidence on the correlation between insulin resistance and coronary microvascular disease

	Study Population	Methods	Main Findings
Dagres et al. [13]	18 non-diabetic patients without cardiovascular risk factors, structural heart disease or diagnosis of CAD	Hyperinsulinaemic-euglycaemic clamp test CFVR with intracoronary Doppler wire	Increased insulin sensitivity is associated with increased CFR (r = 0.76, p < 0.001)
Caliskan et al. [64]	95 non-diabetic patients with family history of T2DM and 34 without family history of T2DM	HOMA-IR CFVR with TTE	Increased risk of developing CMD in subjects with family history of T2DM. Inverse correlation between HOMA-IR and CFR (r = − 0.433, p < 0.05)
Takei et al. [65]	40 patients without diagnosis of CAD or T2DM	HOMA-IR CFVR with TTE before and after OGTT	Acute hyperglycaemia in response to an OGTT rapidly suppresses CFVR. HOMA-IR independently predict CFVR changes.
Trifunovic et al. [66]	104 non-diabetic patients with STEMI	HOMA-IR CFVR with TTE	Correlation between IR and CFR (r = − 0.331 to − 0.386, p < 0.01)
Picchi et al. [63]	13 diabetic and 17 non-diabetic patients	HOMA-IR CFR and IMR with continuous thermodilution	CFR, but not IMR, is impaired in diabetic patients (2.2 vs. 4.1, p < 0.02). Inverse relationship between HOMA-IR and CFR

CAD, cardiovascular artery disease; CFR, coronary flow reserve; CFVR, coronary flow velocity reserve; CMD, coronary microvascular dysfunction; HOMA-IR, homeostasis model assessment-insulin resistance; IR, insulin resistance; OGTT, oral glucose tolerance test; STEMI, ST-segment elevation myocardial infarction; T2DM, type 2 diabetes mellitus; TTE, transthoracic echocardiograph

Importantly, IR-related CMD has also been demonstrated in individuals without diabetes. In non-diabetic subjects with a family history of T2DM, Caliskan et al. reported an inverse correlation between HOMA-IR and CFVR, indicating early microvascular involvement before overt diabetes [64]. Dagres et al. showed a strong positive correlation between insulin sensitivity, assessed by the hyperinsulinaemic–euglycaemic clamp, and intracoronary Doppler CFR (r = 0.76, p < 0.001) [13]. Takei et al. further demonstrated that acute hyperglycaemia induced by an oral glucose tolerance test (OGTT) caused a rapid decline in CFVR, proportional to the severity of IR [65]. In acute coronary syndromes, higher HOMA-IR values were independently associated with lower CFR and impaired myocardial reperfusion, supporting a clinically relevant link between metabolic dysfunction and microvascular injury [66].

Despite these findings, few studies have used the hyperinsulinaemic–euglycaemic clamp, and even fewer have explored the relationship between IR and distinct CMD endotypes. Moreover, endothelium-dependent vasomotor function assessed by invasive vasoreactivity testing remains unexplored. The ongoing Insulin Resistance and Coronary Microvascular Dysfunction in Patients with Myocardial Ischaemia and Non-Obstructive Coronary Artery Disease (IRIN) study (NCT06597851) addresses these gaps by combining clamp-derived insulin sensitivity assessment with comprehensive invasive thermodilution-based evaluation of CMD endotypes and endothelial function in non-diabetic patients with INOCA. This study aims to identify a high-risk metabolic phenotype and support personalised preventive strategies.

Early and chronic effects of insulin resistance on the coronary microvasculature

The effects of IR on the coronary microcirculation can be broadly categorised into early functional alterations and chronic structural changes, which reflect distinct yet interconnected mechanisms.

Early functional alterations. In the early phases, CMD in insulin-resistant states is primarily driven by hyperglycaemia, hyperinsulinaemia, and oxidative stress. Acute glucose fluctuations increase ROS production, leading to reduced NO bioavailability through direct scavenging and eNOS uncoupling. This results in endothelium-dependent impaired vasodilation and enhanced VSMCs sensitivity to vasoconstrictors such as endothelin-1 (ET-1) [67]. At the molecular level, hyperinsulinaemia selectively overstimulates the MAPK pathway while impairing IRS-1/PI3K/Akt signalling, shifting insulin action toward vasoconstriction and inflammation (Fig. 1). These mechanisms may explain epicardial coronary vasospasm and functional CMD, as demonstrated by the acute reduction in CFVR observed after OGTT-induced hyperglycaemia [45, 65]. Reduced NO bioavailability and increased ROS also limit maximal vasodilatory responses to endothelium-independent stimuli such as adenosine, further promoting functional CMD [68].

Chronic structural alterations. Chronic exposure to IR and hyperinsulinaemia leads to persistent endothelial dysfunction and progressive microvascular remodelling. Sustained NO deficiency removes a key anti-inflammatory signal, facilitating activation of nuclear factor-κB (NF-κB) and transcription of pro-inflammatory cytokines, including IL-6, IL-1β, and TNF-α [69]. Concurrent activation of PKC and oxidative stress alters vascular endothelial growth factor (VEGF) and transforming growth factor-β (TGF-β) signalling, promoting endothelial apoptosis, capillary rarefaction, and perivascular fibrosis [70]. These processes characterise structural CMD, with reduced microvascular density, increased stiffness, and impaired vasodilatory reserve. Long-standing metabolic disturbances may also impair EDH-mediated relaxation, contributing to coronary microvascular spasm [23].

A vicious cycle between insulin resistance and microvascular dysfunction. IR and CMD are interconnected through a self-reinforcing vicious cycle. Endothelial dysfunction is associated with the activation of stress-related serine/threonine kinases, including PKC and IKK-β, which reduce GLUT4 translocation on the cell membrane, thereby worsening insulin sensitivity [71, 72]. Conversely, progressive IR perpetuates endothelial dysfunction through sustained oxidative stress and inflammation. Emerging evidence implicates microRNAs as additional regulators of this interplay. Dysregulated expression of microRNAs such as miR-126, miR-155, and miR-21, has been associated with impaired eNOS signalling, endothelial inflammation, and metabolic dysfunction in T2DM and obesity [73, 74]. These epigenetic mechanisms may contribute to the pathogenesis of INOCA and represent potential therapeutic targets.

Future perspectives and possible treatment strategies of coronary microvascular dysfunction in insulin-resistant state

The management of patients with CMD begins with the aggressive correction of cardiovascular risk factors, among which weight control represents a top priority because of the strong association between obesity and IR, where low-grade systemic inflammation is a key mediator [60]. Aerobic exercise training and intentional weight loss in obese patients have been shown to increase CFR in randomised trials [45, 75]. Exercise also improves endothelium-dependent vasodilation and cardiorespiratory fitness, and these benefits are maintained long-term when associated with a Mediterranean diet [76].

Recent advances in glucose-lowering medications have demonstrated to be effective in targeting cardiovascular disease and obesity, potentially serving as a tool to correct CMD.

Metformin is a cornerstone of T2DM treatment, and it is also currently used in patients with prediabetes, given its beneficial effects on IR. Beyond its glucose-lowering actions, metformin treatment has been associated with lower rates of rehospitalisation for angina in INOCA patients with hyperglycaemia. The suggested mechanism involves the attenuation of mitochondrial ROS production evoked by high glucose concentrations [77]. Vasculoprotective actions of pioglitazone are also well-established, and exert mainly on the microvasculature where it prevents capillary rarefaction and other structural alterations independently of improvements in glycaemic control. However, the specific molecular mechanisms remain unclear [78].

Glucagon-like peptide-1 receptor agonists (GLP1-RAs) exert multiple beneficial effects on the cardiovascular system and rank among the most effective glucose-lowering agents [15, 79]. Long-acting GLP1-RAs, such as semaglutide, are particularly effective, not only in lowering HbA1c levels but also in promoting weight loss. Semaglutide has shown to be crucial for improving insulin sensitivity and reducing visceral fat. Additionally, GLP1-RAs modestly reduce systolic blood pressure and positively affect lipid profiles. GLP1-RAs also exhibit anti-inflammatory effects, further contributing to their cardiovascular benefits [80–82].

Sodium glucose co-transporter-2 inhibitors (SGLT2i) provide cardiovascular benefits that extend beyond their primary function of lowering blood sugar levels [60]. SGLT2i increase the urinary glucose excretion at any given plasma glucose concentration, lowering plasma glucose in an insulin-independent manner [83]. Several preclinical studies have reported that SGLT2i may act on systemic inflammation and oxidative stress. The pathophysiological mechanisms are still debated and likely include multiple favourable actions on the vasculature, such as direct endothelium-independent vasodilation, reduction of vasoconstrictive mediators, improvement of endothelial dysfunction, and lowered arterial stiffness [60, 83, 84]. This hypothesis is further supported by the increase in MFR and stress MBF observed by Leccisotti et al. in diabetic patients treated with dapagliflozin [85]. Therefore, both GLP1-RAs and SGLT2i offer a comprehensive approach to protect against vascular injury by enhancing antioxidant defences and mitigating inflammation-associated mechanisms [79]. In addition, these therapeutic agents have been reported to improve mitochondrial function in diabetic patients by reducing ROS production and promoting mitochondrial biogenesis [86]. In particular, recent experimental data on empagliflozin have reinforced the idea that SGLT2i may exert beneficial effects on the coronary microvasculature through an improvement in cardiac energy status by increasing mitochondrial production of ATP [87].

Finally, therapies targeting pro-inflammatory cytokines, such as anakinra, an IL-1 receptor antagonist, and tocilizumab, an IL-6 receptor blocker, have been shown to improve vascular and cardiac function by reducing oxidative stress and inflammatory burden. They may therefore exert beneficial effects on CMD in patients with systemic inflammatory syndromes [88]. This hypothesis has to be confirmed in dedicated studies.

Conclusions

Multiple interrelated mechanisms – including hyperinsulinaemia, hyperglycaemia, oxidative stress and inflammatory pathways – contribute to the complex relationship between insulin resistance and coronary microvascular dysfunction. Despite growing evidence, this interplay remains only partially understood, underscoring the need for comprehensive, mechanistic investigations. Improved pathophysiological insight may facilitate the development of targeted, metabolism-oriented therapeutic strategies aimed at improving clinical outcomes and alleviating symptoms in patients with INOCA.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviation

ACh

Acetylcholine

CAD

Coronary artery disease

CFR

Coronary flow reserve

CMD

Coronary microvascular dysfunction

ET-1

Endothelin-1

GLP1-RAs

Glucagon-like Peptide-1 receptor agonists

IKK β

IκB Kinase complex β

IMR

Index of microcirculatory resistance

INOCA

Ischemia with non-obstructive coronary artery disease

IR

Insulin resistance

MAPK

Mitogen-activated protein kinase

NO

Nitric oxide

eNOS

endothelial Nitric Oxide Synthase

PI3K

Phosphoinositide 3-Kinase

PKC

Protein Kinase C

ROS

Reactive oxygen species

SGLT2i

Sodium Glucose co-transporter-2 inhibitors

T2DM

Type 2 Diabetes Mellitus

VSMCs

Vascular Smooth Muscle Cells

Author contributions

MM and LP conceived and wrote the initial draft of the manuscript. AT, EM, SP and MR contributed to writing and content development. AL prepared the figures. RS, FT, AM, FR, VK and RB critically revised and edited the manuscript. All authors approved the final version of the manuscript.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

RS reports research grant from Abbott Vascular and speakers fee from Abbott. FR reports research grant from Abbott Vascular and Philips. The other authors report no conflict of interest.