The Coronary Microcirculation Re-explored: Pathophysiological Insights and Clinical Implications

The human heart is a highly demanding organ because of its substantial need for oxygen and nutrients. It relies on a specialised and intricate network of blood vessels known as the coronary circulation to meet these demands. The coronary microvessels are those elements of this circulation that are less than 500 µm in diameter and are critical in transporting vital elements necessary for myocardial metabolism. The coronary circulation has unique properties that are adaptable to metabolic demands at any given time, facilitating high blood flow and a short oxygen diffusion distance. The potential for high metabolic demands requires a system that can modulate perfusion momentarily.

Coronary microvascular dysfunction (CMD) is a pathological phenomenon that leads to a variety of anatomical and physiological changes which manifest in impaired coronary blood flow (CBF). Subsequently, these pathological alterations result in myocardial ischaemia. The critical role of CMD is increasingly recognised in patients with angina without obstructive coronary artery disease (CAD). Evidence shows a contributing role for CMD in patients with obstructive CAD, takotsubo syndrome, heart failure and cardiomyopathies.

This review explores coronary microvascular anatomy, CMD pathophysiology, diagnostic methods and current and emerging treatments.

Overview of Coronary Microcirculation: Structure and Function

The coronary microcirculation is composed of vessels <500 µm in diameter, including pre-arterioles, arterioles, capillaries and venules and it is specially structured to regulate myocardial blood flow.¹ Among the microvascular components, arterioles play a central role in controlling vascular tone. They function as resistance vessels, surrounded by vascular smooth muscle and lined internally with endothelial cells, similar to the lining found in many capillaries.² The coronary microcirculation controls the distribution of blood flow to meet myocardial metabolic needs and regulates coronary vascular resistance. In normal epicardial arteries, coronary arterioles adapt to increased myocardial metabolic needs by progressively dilating, often allowing CBF to increase by as much as five times the resting level.¹ Myocardial blood flow (MBF) amounts to approximately 1 ml of blood per gram of myocardium being continuously supplied every minute of an individual’s life by the coronary vessels under resting conditions.³ Given that the myocardium extracts nearly all available oxygen at rest, any increased demand must be met by elevated blood flow – a process that is entirely dependent on an intact coronary microcirculation.⁴

Arterioles

Arterioles are small-calibre branches that emerge from intramural vessels and course alongside myocardial fibres. Their morphology is not uniform throughout; the proximal and mid-segments display features akin to those of small arteries, including a tunica media composed of multiple concentric layers of vascular smooth muscle cells (VSMCs). In comparison, the terminal portions exhibit a thinner tunica media, often limited to one or two layers of VSMCs – or entirely without them – with pericytes present instead (Figure 1 ).⁵ Furthermore, the internal elastic membrane may be absent distally and the tunica adventitia is usually thinner.⁶

Arteries with larger diameters exhibit a distinct, variable layer of vascular smooth muscle, unlike capillaries that do not possess smooth muscle and instead have pericytes (Figure 1 ). The capillary neurovascular unit thus comprises endothelial cells, a basement membrane and pericytes, all enveloped by neuronal endings. Additionally, fibroblasts and collateral branches form part of the vascular system.⁷

Functionally, the arterioles are characterised by a noticeable gradual decline in pressure along their length and they contribute to blood flow resistance through differing mechanisms. Relaxation of the larger proximal arterioles is mediated by endothelium-dependent relaxation factors, such as adenosine and nitric oxide (NO). Relaxation in medium-sized arterioles is predominated by the myogenic smooth muscle response, where the vessel constricts against high perfusion pressures and relaxes when the pressure is low, thus protecting the thinner vessels downstream from being pressure-damaged. The pre-capillary arterioles are more controlled by metabolic factors, such as pH and endothelium-dependent hyperpolarising factor (EDHF). Ultimately, a defect in any of these mechanisms can lead to a loss of correct control of perfusion and cause CMD.^8,9

Capillaries

Capillaries are microscopic vessels (<10 μm diameter, averaging 5.7 μm) connecting arterial and venous networks through numerous anastomotic loops (Figure 1) and they are essential for nutrient and oxygen exchange between blood and myocardium.² Healthy myocardium typically has capillary densities up to 3,500/mm², with higher densities in sub-endocardial regions due to increased oxygen demand compared to sub-epicardial areas.^9–11 Structurally, capillaries differ from larger vessels by having walls composed primarily of an endothelial layer and its basal lamina, featuring small intercellular clefts or larger intercellular gaps (Figure 1).

Capillaries can be morphologically categorised into: non-fenestrated (continuous); fenestrated; and sinusoidal (discontinuous) types.² Continuous microvessels predominant within the coronary circulation are common in muscles, lungs and the central nervous system and contain numerous pinocytotic vesicles without fenestrations.⁶ In fenestrated subtypes, the endothelial lining exhibits small pores (typically 80–100 nm), facilitating efficient molecular exchange in tissues such as intestines and endocrine glands.⁶ Discontinuous capillaries (sinusoids), characterised by larger diameters and irregular shapes, are present in tissues such as the liver.⁶

Pericytes, small contractile cells situated within the capillary basal membrane, vary morphologically and functionally across vascular beds (Figure 1).¹² These cells extend along capillaries and are essential for angiogenesis, endothelial regulation and regional blood flow, particularly within the central nervous system and kidneys.^12,13 Pericytes exhibit contractile features mediated by proteins such as α-smooth muscle actin and tropomyosin, although their contractile mechanism differs from VSMCs due to the absence of specific muscle machinery.¹² Despite an incomplete understanding of pericyte functions in coronary physiology, their abundance and similarities to central nervous system pericytes suggest critical roles in regulating capillary diameter and permeability.¹³ Additionally, other structures, such as capillary sinuses may, act as micropumps within the network.⁵

The two main factors determining coronary flow are arterial pressure and the oxygen requirement of the myocardium.¹⁴ Additionally, autoregulation facilitates oxygen supply irrespective of arterial pressure. The tight control of coronary flow maintains a coronary venous oxygen partial pressure of 20 mmHg despite varying physiological demands and conditions.¹⁵ Coronary flow has the capacity to increase fivefold during exercise to ensure adequate delivery of oxygen and nutrients. The coronary flow is controlled, at least partially, by a feed-forward system. The metabolic factors that contribute to the system include reactive oxygen species and carbon dioxide; however, during exercise, the addition of neural and endothelial control is required to ensure adequate modulation of coronary flow.

Pathophysiology of Coronary Microvascular Dysfunction

CMD results from alterations at the coronary microvascular level that lead to impaired coronary flow, with subsequent ischaemia. CMD pathogenesis involves multiple mechanisms and is primarily linked to structural and functional disruptions in the coronary microcirculation (Figures 2 and 3).

The combination of these factors impairs coronary perfusion and contributes to ischaemia, which is often regional even when no obstructive CAD is present. Additionally, these alterations are driven by endothelial-dependent and endothelial-independent factors. The endothelial-dependent factors react to changes in coronary flow and myocardial metabolites, balancing vasoconstricting factors, such as endothelin-1 (ET-1), against vasodilators (prostaglandins, NO and EDHFs).¹⁶ Endothelium-independent mechanisms are based on myocyte tone.¹⁷

Functional Factors in CMD

Functional causes of CMD are the product of pathological alterations that lead to impaired vasodilation and increased constriction of the microvasculature. Although the mechanisms underlying impaired vasodilatation may involve endothelium-independent vasodilators, the endothelium still contributes significantly to the vasomotor activity of the microvasculature through regulatory (vasoactive) factors released by the endothelial cells.¹⁸ Vascular tone is influenced by a balance of mediators that either promote vessel relaxation, including NO, prostaglandins and EDHFs, or induce constriction, as is the case with ET-1.¹⁹ Additionally, the roles of these individual vasoactive mediators may vary depending on the segment of coronary microvasculature. For instance, NO facilitates vasodilation mainly at the epicardial coronary artery and arteriolar level, while EDHFs primarily mediate endothelial-dependent vasodilation of microvessels.¹⁷

Although not fully understood, endothelial-dependent mechanisms may be disrupted due to reduced ability of VSMCs to relax, heightened sensitivity to vasoconstrictors and autonomic stimuli, and elevated production of vasoconstrictive agents such as ET-1.¹ Several studies have elucidated the critical role of ET-1 in CMD. ET-1 is found in endothelial cells and is produced by smooth muscle cells and cardiomyocytes. The peptide exhibits both vasoconstrictive and mitogenic properties, promoting the release of growth factors such as vascular endothelial growth factor and basic fibroblast growth factor. It also amplifies the activity of transforming growth factors. Prolonged ET-1 activation contributes to pathological changes, including myocardial and vascular fibrosis, cardiac hypertrophy and extracellular matrix expansion.²⁰ Various physicochemical stimuli, including pulsatile stretch, shear stress, pH and blood flow, modulate the release of ET-1. For example, acute hypoxic conditions specifically enhance ET-1 gene expression and peptide synthesis within the pulmonary vascular system.¹

There are two distinct G protein-coupled receptors that ET-1 stimulates: ET-A and ET-B. ET-A receptors are present in the medial smooth muscle layer of the atria, ventricular myocardium and blood vessels. The stimulation of this receptor leads to vasoconstriction and cellular proliferation due to an increase in intracellular calcium. ET-B receptors are localised in the endothelial cells, smooth muscles and macrophages. Activating this receptor leads to the release of NO and prostacyclin and prevents apoptosis. In normal physiological states, ET-A and ET-B balance these regulatory processes. In pathological states, however, the ET-B receptors have a similar function to the ET-A receptors, leading to more profound vasoconstriction while propagating the mitogenic effects of ET-1.²¹ ET-1 promotes microvascular constriction in individuals experiencing non-obstructive ischaemia, significantly influencing coronary microvascular dysfunction. Additionally, ET-1 can enhance proliferation and endothelial inflammatory processes. Therefore, ET-1 holds promise as a novel treatment target for individuals diagnosed with CMD.²¹ While arterioles and larger vessels contribute significantly in modulating vascular resistance, the role of microvessels in regulating blood flow is increasingly recognised in many pathological manifestations. The endothelium is critical in regulating the epicardial coronary artery vasomotor tone. However, endothelial capacity to mediate vasodilatory signals, such as acetylcholine, can be hindered even in the absence of atherosclerosis in those with diabetes, hypercholesterolaemia, hypertension and those who smoke. Furthermore, oxidative stress has been demonstrated to initiate a constrictor response in the endothelium.²² In conditions leading to tissue hypoxia, such as vessel stenosis, anaemia or certain physiological states such as high-altitude exposure, the resultant reduced oxygen delivery has adverse effects, particularly during exertion. Typically, in these high-demand conditions, oxygen-sensitive K channels serve as molecular integrators between cardiac electrical activity and energy homeostasis. In contrast, calcium channels are closed to prevent intracellular overload.¹⁹ This mechanism is important to avoid mismatches between myocyte energy demand and production, and any disruptions to this process will exacerbate mismatches and impair myocardial oxygenation.²³

Structural Factors in CMD

In CMD, structural changes such as inward remodelling – often seen in hypertrophic cardiomyopathy and hypertension – affect the left ventricle, driven by smooth muscle growth, the buildup of collagen and intimal thickening, ultimately impairing coronary flow.²⁴

Separate from these two key factors, some underlying molecular mechanisms are crucial in the pathophysiology that impair coronary flow. Elevated levels of reactive oxygen species (ROS) and associated inflammation contribute significantly to both the initiation and advancement of CMD. ROS are mainly generated through the activation of nicotinamide adenine dinucleotide phosphate oxidases (Nox enzymes). Such activation by Nox leads to the phosphorylation of the p66shc protein and translocation within the mitochondria. This process leads to changes to the mitochondrial bio-energetic properties that stimulate further Nox activity, accelerating apoptosis.²⁵

Moreover, the buildup of oxygen free radicals – especially superoxide – promotes conversion of NO to peroxynitrite, while disrupting endothelial NO synthase (eNOS) function. This shift causes eNOS to produce ROS instead of NO, thereby reducing NO-dependent vasodilation.^26,27

Clinical Classification of CMD

The classification is based on the clinical circumstances in which they occur. They are as follows:²⁸

• CMD without underlying myocardial or epicardial coronary artery disease is linked to traditional risk factors such as hypertension, smoking and diabetes and is identified by coronary flow reserve assessment.
• CMD in the setting of myocardial disease typically presents in genetic (primary) cardiomyopathies, including hypertrophic and enlarged forms, as well as secondary cardiomyopathies, such as hypertensive and valvular types. It results from microvascular remodelling and fibrosis, contributing to ischaemia.
• CMD in obstructive CAD occurs in stable CAD or acute coronary syndromes, where diagnosis may require both invasive and non-invasive assessment.
• Iatrogenic CMD develops post-revascularisation due to distal embolisation or vasospasm and is often transient and reversible.

Diagnosing Microvascular Dysfunction

A historical dependence on the epicardial lumenogram as the gold standard in diagnosing coronary insufficiency and a lack of technologies that can accurately measure changes in microvascular blood flow have long hampered efforts to diagnose CMD. However, some newer modalities have helped shed light on aspects of the coronary microvasculature.²⁹ Coronary microvascular and endothelial function can be measured using invasive and non-invasive techniques. These techniques mainly depend on CBF and have been well-validated in multiple studies. Imaging modalities are summarised in Supplementary Table 1.

Non-invasive Techniques

Exercise treadmill testing (ETT) has gained renewed recognition for its diagnostic value in evaluating CMD among angina and non-obstructive coronary artery disease (ANOCA) patients. A 2024 single-centre study involving 102 patients with ANOCA validated ETT against invasive coronary physiological assessments, including adenosine-derived coronary flow reserve (CFR) and acetylcholine flow reserve (AChFR), which assess non-endothelial and endothelial contributions to microvascular regulation, respectively.³⁰ The study demonstrated that ischaemic ECG changes during ETT, defined as ≥0.1 mV horizontal or down-sloping ST-segment depression, were 100% specific for CMD. All patients with ischaemic ETT results were found to have CMD on invasive testing, with endothelium-dependent dysfunction (AChFR ≤1.5) being the strongest independent predictor of ischaemia. Notably, 97% of patients with ischaemic ETT results had endothelium-dependent CMD, while 63% exhibited endothelium-independent dysfunction.³⁰ These findings highlight the physiological relevance of acetylcholine testing and its ability to link ischaemic ETT results to impaired coronary vasodilation. This study underscores the importance of ETT as a simple, low-cost, non-invasive strategy for identifying CMD in ANOCA. The findings also challenge the historical misconception of ‘false positive’ ETT results in ANOCA patients, redefining these outcomes as markers of microvascular dysfunction rather than diagnostic inaccuracies. However, given the study’s single-centre design and limited sample size, further multi-centre trials are required to validate these results in larger, more diverse populations. Additional studies are warranted to better define the prognostic significance of ischaemic ETT results and to assess the integration of ETT with other diagnostic modalities in CMD evaluation.

Doppler-derived CFR assessed through pharmacologic stress echocardiography using adenosine offers a non-invasive approach to assess coronary microvascular function. It calculates the ratio between blood flow under stress and at rest using pulsed-wave Doppler velocity (CFV) within the mid/distal-segment of the left anterior descending artery (LAD).³¹ Studies showed a modest to good correlation between thermodilution-derived CFR and CFV reserve (CFVR), while showing a stronger correlation with PET imaging.^32–37 While PET provides a global assessment of myocardial perfusion, Doppler CFR offers a more targeted evaluation limited to the LAD. A CFVR value below 2.0 is a strong indicator of CMD.³⁸ However, an optimal CFVR cut-off is between 2.0 and 2.5 depending on the study and patient population.³⁹ Advantages include its cost-effectiveness, accessibility, and non-invasive nature, making it suitable for routine use. However, it is operator-dependent, limited to single-vessel evaluation and it is less accurate in patients with poor acoustic windows. Despite these challenges, Doppler CFR remains a valuable diagnostic option, especially where PET or invasive techniques are less available.

Cardiac magnetic resonance (CMR) stress perfusion imaging is a reliable technique for assessing CMD. Comparing myocardial perfusion during pharmacologically induced stress (typically with adenosine) with rest employs visual, semi-quantitative and quantitative approaches, each offering progressively greater precision. CMR can identify transmural maldistribution of myocardial blood flow, a key feature of CMD.⁴⁰ Semi-quantitative analysis refines this evaluation by calculating the myocardial perfusion reserve index, which compares signal intensity during stress to that at rest.^40–44 The most accurate method is quantitative analysis, which quantifies MBF in ml/min/g of tissue using advanced methods such as dual-bolus or dual-sequence imaging to enhance measurement precision and myocardial flow reserve (MFR).^45–47 These quantitative methods have been validated against PET and invasive coronary physiology, demonstrating strong diagnostic performance.^48,49

CT perfusion (CTP) imaging is a valuable tool for assessing CMD. Its high spatial resolution enables differentiation between the endocardium and epicardium, with decreased endocardial-to-epicardial contrast ratios and attenuation abnormalities serving as key indicators of CMD. Reduced MFR values and regional perfusion deficits are consistent with CMD and help distinguish it from diffuse coronary artery disease. CTP has been benchmarked against PET and fractional flow reserve (FFR), demonstrating high diagnostic performance in evaluating ischaemic myocardial conditions.^50,51 By integrating anatomical and functional data, CTP provides a non-invasive and reliable method to detect early microvascular dysfunction, supporting its clinical value in CMD detection and decision-making.^52–54 CT-derived FFR has also been shown to detect physiologic abnormalities in patients with non-obstructive CAD, revealing discordance between anatomical and functional assessments that may reflect underlying microvascular dysfunction. Furthermore, the myocardial blood volume-to-mass ratio (vol/mass) derived from CTP has emerged as a potential marker for evaluating coronary microvascular health, offering additional insights into perfusion abnormalities in the absence of obstructive epicardial disease.^55,56

PET is widely recognised as the reference standard for quantifying myocardial blood supply and flow reserve, calculated as the proportion of MBF during stress to that during rest due to its ability to provide absolute quantification in ml/min/g. Numerous studies have validated PET’s accuracy and reliability, demonstrating its superiority over other imaging modalities. PET consistently delivers precise, reproducible and quantitative assessments of coronary circulation making it the preferred technique for evaluating CMD.⁵⁷

PET radiotracers differ in availability, image quality, extraction fraction and MBF accuracy due to extraction linearity at high flow rates. Rubidium-82 (⁸²Rb) is generator-produced, allowing rapid imaging without a cyclotron, but has non-linear extraction above ~2.5 ml/min/g and a short 75-second half-life. Nitrogen-13 ammonia (¹³N-NH₃) offers better image quality and MBF accuracy (moderate nonlinearity at ~3.0 ml/min/g) but requires a cyclotron and has a 10-minute half-life. Oxygen-15 Water (¹⁵O-H₂O) provides linear extraction across all flow ranges, making it the most accurate for MBF, but its 2-minute half-life requires an on-site cyclotron and immediate imaging. Flurpiridaz F-18, a next-generation tracer, has high, nearly linear extraction up to ~4.0 ml/min/g, superior image resolution and a long 110-minute half-life, enabling off-site cyclotron production and transport, making it a promising alternative for MBF and MFR assessment.⁵⁸

Most clinical centres define CMD using a hyperaemic MBF threshold of <1.7–2.3 ml/min/g and an MFR cut-off of <2.0, which helps mitigate methodological errors. While MFR <2.0 is widely used in invasive and non-invasive imaging, studies suggest that stricter thresholds of <1.7 or <1.5 are linked to a higher risk of major adverse cardiovascular events (MACE), offering greater prognostic value.⁵⁷ When obstructive CAD is excluded, MBF and MFR can be used to categorise coronary microvascular function into four distinct types (Supplementary Table 2).⁵⁷

Invasive Techniques

A needle-probe video microscope has been used to gain in vivo images of the vasculature over several centimetres of myocardial thickness. The probe is inserted into the myocardium or near the endocardial surface, often within a doughnut-shaped device that creates a blood-free zone. This imaging modality facilitates the assessment of regulatory mechanisms within the vascular diameter of the subendocardial and subepicardial microcirculation, but its use is mainly reserved for research purposes.⁵⁹

Intracoronary pressure-temperature wire studies (PTW) allow simultaneous measurement of FFR, CFR and the index of myocardial resistance (IMR). FFR is the ratio of distal to proximal (aortic) pressure, determined during pharmacologically induced hyperaemia.⁶⁰ CFR is the ratio of maximum hyperaemic flow to resting flow, reflecting the myocardial reserve vasodilator capacity.⁶⁰ IMR is the ratio of distal coronary pressure to the inverse of the mean transit time during maximal hyperaemia.⁶¹

A recent study has shown that PTW-CFR mapping provides a good prediction of coronary flow impairment that correlates with MACE. IMR measurement is also a well-validated invasive technique that uses the thermodilution principle to measure coronary flow.⁶² According to this principle, the transit time of a bolus of cold saline (TmnHyp) is inversely proportional to flow and the intraluminal pressure in the coronary artery distal to the lesion (Pd) can be measured with a pressure wire. When both measurements are recorded during maximal hyperaemia, microvascular resistance can be calculated using IMR = Pd×TmnHyp.⁶³ This is quite distinct from the CFR and allows the use of the IMR in those with epicardial obstruction.

PTW overcomes the limitations of measuring CFR alone (as the latter is often affected by baseline epicardial stenosis and hyperaemia), providing a comprehensive assessment of coronary flow characteristics. On the other hand, compared to CFR, IMR is unaffected by resting haemodynamics and is more reproducible after haemodynamic perturbations. Thus, it is considered superior to CFR and is regarded as more specific for microcirculation assessment.⁶³ More recently, continuous thermodilution methods, such as the assessment of CFR and absolute CBF (quantification of absolute CBF by perfusion imaging [Q]), have been introduced to evaluate microcirculatory function by directly measuring mean transit time at baseline and during hyperaemia. The continuous infusion of saline via a dedicated catheter allows for a steady-state assessment, improving reproducibility and mitigating variability related to manual injections.⁶⁴ These methods have demonstrated potential in detecting microvascular dysfunction and ischaemia more precisely, complementing traditional indices like IMR.

Ischaemia with non-obstructive coronary arteries (INOCA) consists of different endotypes based on specific diagnostic markers summarised in Supplementary Table 3.

Epicardial and microvascular vasospasm result from hyperreactivity of smooth muscle or endothelial dysfunction. Epicardial spasm is defined by a reduction in coronary diameter exceeding 90% accompanied by clinical symptoms and ischaemic ECG abnormalities during higher dose acetylcholine testing, while microvascular spasm occurs with chest pain and ECG changes indicative of ischaemia but with <90% luminal narrowing in the epicardial vessels. Adenosine is used to evaluate endothelium-independent function by inducing smooth muscle vasodilation, while acetylcholine (at lower dose than that used for spasm provocation) tests endothelium-dependent function, triggering NO-mediated vasodilation in healthy endothelium or vasoconstriction where there is dysfunction.⁶⁵ In the CorMicA trial, which looked at Invasive coronary function testing, 17% of participants demonstrated isolated vasospastic angina and approximately 21% showed overlapping features of microvascular and vasospastic angina, highlighting the limitations of non-invasive tests in detecting endothelial dysfunction and vasospasm, which require intracoronary acetylcholine provocation. By combining adenosine and acetylcholine testing, clinicians can assess both pathways of CMD, providing critical insights into disease mechanisms and guiding tailored management.⁶⁶ Figure 4 offers a streamlined overview of the diagnostic pathway for suspected INOCA, moving from initial clinical assessment and non-invasive testing through to invasive evaluation, functional assessment and final classification based on defined diagnostic endotypes.

Management of Microvascular Dysfunction: Current Practice

The mechanisms of medical therapies vary by CMD endotype, targeting vascular tone, endothelial function or myocardial energetics. The 2020 Expert Consensus Document on Ischaemia with Non-Obstructive Coronary Arteries (INOCA), developed by the European Association of Percutaneous Cardiovascular Interventions in collaboration with the European Society of Cardiology Working Group on Coronary Pathophysiology and Microcirculation and endorsed by the Coronary Vasomotor Disorders International Study Group, advises adopting a systematic strategy for managing INOCA. It emphasises modifications to lifestyle, including healthy nutrition, consistent exercise, stress reduction and smoking cessation.⁶⁷

Risk factor management should focus on controlling hypertension with angiotensin-converting enzyme inhibitors (ACEi) or angiotensin receptor blockers, which improve endothelial function and reduce microvascular resistance, managing dyslipidaemia with statins and optimising diabetes care. Antianginal therapy needs to be individualised according to the specific pathophysiological process of ischaemia. For microvascular angina, treatment includes ß-blockers, which lower myocardial oxygen demand; calcium channel antagonists, which reduce vasospasm; ranolazine, which improves diastolic relaxation; nicorandil, a potassium-adenosine triphosphate channel opener that improves microvascular perfusion; trimetazidine, which enhances metabolic efficiency via shifting myocardial energy use from fatty acid to glucose oxidation; and ACEi or angiotensin receptor blockers (ARBs). Vasospastic angina is primarily managed with a combination of calcium channel antagonists, nicorandil, renin–angiotensin system inhibitors and lipid-lowering agents. In cases where symptoms remain unresponsive to pharmacological intervention, enhanced external counterpulsation may be considered.⁶⁸ In the iPOWER study, among patients with coronary microvascular dysfunction (CFVR <2.0; n=241), 51% were on aspirin therapy, 35% were managed with ß-blockers, and 25% received calcium channel antagonists. Renin–angiotensin system blockers were prescribed in 15%, while both statins and ARBs were administered in 25%.⁶⁹

The CorMicA trial randomised 151 individuals presenting with angina in the absence of obstructive coronary lesions to invasive coronary function testing with stratified therapy (n=75) or standard care (n=76). The invasive group underwent microvascular and vasospastic dysfunction testing, guiding targeted treatment. All received initial pharmacotherapy, such as antiplatelet agents, statins, and renin–angiotensin system modulators when appropriate. The microvascular angina group (57%) received ß-blockers, calcium channel antagonists, or nicorandil based on clinical need. The vasospastic angina group (16%) was managed with calcium channel antagonists with adjunctive nitrates or nicorandil, while ß-blockers were avoided. Mixed cases (19%) received calcium channel antagonists and nicorandil, while therapy was discontinued in those without coronary dysfunction (8%). At 6 months, the intervention group had greater symptom relief, improved health-related quality of life, and greater usage of calcium channel antagonists, renin–angiotensin inhibitors and lipid-lowering agents. These results endorse invasive assessment of coronary function to guide optimal treatment in patients experiencing angina without evidence of obstructive epicardial disease.⁶⁶ Ranolazine has shown modest benefits in women with CMD, offering slight improvements in symptoms and myocardial blood flow compared to placebo.⁷⁰ Similarly, Vilano et al. reported that in patients with microvascular angina (MVA), ranolazine was more effective than ivabradine in enhancing exercise performance and alleviating anginal symptoms.⁷¹ A 2024 meta-analysis of five randomised controlled trials with 209 patients evaluated ACEis for coronary microvascular dysfunction. ACEis significantly improved CFR but showed no significant effect on chest pain episodes or systolic blood pressure. While ACEis may enhance microvascular function, larger trials are needed to confirm their clinical benefits.⁷² Calcium channel antagonists are known to reduce microvascular tone, which relieves spasms and subsequently improves coronary flow. Additionally, verapamil, nifedipine and lidoflazine are associated with improved stress test parameters.⁷³ However, while lidoflazine has shown reduced coronary resistance, it is linked with increased rates of fatal arrhythmia.⁷⁴

Statins are key in managing CMD due to their anti-atherosclerotic and anti-inflammatory effects. Pravastatin has been shown to improve CFR with the administration of an acetylcholine provocation test. While Zhang et al. and Eshtegardi et al. have shown an increase in CFR with fluvastatin and atorvastatin, respectively, the effects were not clinically significant.^75,76 However, patients with a poorly controlled lipid profile had worse CFR and its associated symptoms.

The role of oestrogen has also been investigated in post-menopausal women due to increased rates of CMD in this population. However, no therapeutic benefits were identified.⁷⁷

Recent studies show that empagliflozin does not directly improve coronary microvascular dysfunction in type 2 diabetes. A 13-week study in 90 patients used ⁸²Rb-PET/CT to assess MFR and found no improvement.⁷⁸ A 4-week crossover study in 13 patients used ¹⁵O-H₂O PET/CT and showed reduced myocardial glucose uptake and resting blood flow but no effect on MFR.⁷⁹ A 12-week crossover trial in 19 patients used CFVR and also found no improvement.⁸⁰

Alongside pharmacological agents, the role of device therapy for CMD has been explored. For instance, enhanced external counterpulsation (EECP) is a device that modulates pain-related signals and potentially increases MBF through its effects on sympathetic tone.⁸¹ In addition, 4 weeks of neurostimulation via transcutaneous electrical nerve stimulation was found to improve angina pectoris with a concomitant improvement of myocardial perfusion.⁶⁸ Figure 5 provides a simplified visual summary of current INOCA treatment recommendations, highlighting lifestyle modification and tailored anti-anginal therapy while also outlining future directions such as precision medicine, stratified approaches and emerging therapies.

Potential Novel Treatment of Microvascular Dysfunction (Role of Endothelin Receptor Blockers)

ET-1, a strong endogenous vasoconstrictor, plays a central part in CMD. Early studies revealed that ET-1 concentrations were notably increased in individuals with endothelial dysfunction, particularly those exhibiting acetylcholine-induced vasoconstriction.⁸² Similarly, individuals with microvascular angina and unobstructed coronary arteries on angiography had higher plasma ET-1 concentrations, particularly those with a positive exercise test for ischaemia, reinforcing endothelin’s role in microvascular pathology.⁸³ Building on these findings, later studies explored the therapeutic implications of modulating the endothelin pathway. In a randomised trial conducted in 2016, patients with non-ST elevation acute coronary syndromes who underwent percutaneous coronary intervention had their microvascular coronary function evaluated using Doppler-derived CFR, both before and after receiving BQ-123, a selective ET-A receptor blocker. The notable improvement in CFR following treatment pointed to ET-1 as a contributor to post-PCI microvascular dysfunction.⁸⁴ Ford et al. analysed 185 angina patients without obstructive CAD, identifying the rs9349379-G allele as a risk factor for CMD, linked to higher ET-1 levels, impaired CFR, increased IMR, reduced myocardial perfusion on stress CMR and lower exercise tolerance based on the Duke Treadmill Score. Ex vivo studies showed that zibotentan, a selective ET-A receptor antagonist, reversed ET-1-mediated coronary vasoconstriction, supporting its potential as a targeted therapy for CMD.⁸⁵ Naya et al. studied 49 CAD patients with one- or two-vessel disease and found that 18 (37%) had CMD, identified by CFR <2.0 in non-obstructed regions on 15O-water PET/CT. These patients exhibited significantly elevated ET-1 levels compared to those without CMD.

ET-1 remained a strong predictor of CMD, even after adjusting for diabetes.⁸⁶ Recently, Morrow et al. published a randomised, placebo-controlled, crossover trial to evaluate the effects of zibotentan (ET-A receptor selective antagonist 10 mg daily for 12 weeks) in 118 patients with microvascular angina. The trial did not demonstrate a meaningful increase in exercise capacity among patients receiving zibotentan compared to placebo. Furthermore, the higher incidence of adverse effects associated with zibotentan has prompted caution regarding its long-term tolerability and clinical safety profile.⁸⁷ This lack of efficacy may reflect limitations of selective ET-A receptor blockade. Under pathological conditions, ET-B receptors can become vasoconstrictive and pro-inflammatory.⁸⁸ ET-A antagonism also raises circulating endothelin-1, further activating ET-B pathways.⁸⁹ These factors, combined with CMD heterogeneity and the short 12-week treatment duration likely explain the neutral exercise outcomes in the PRIZE trial.⁹⁰

Several emerging therapies are under investigation for CMD. While sodium-glucose co-transporter 2 inhibitors offer systemic benefits, they do not improve CFR.⁹¹ Soluble guanylate cyclase stimulators like vericiguat, studied in heart failure with preserved ejection fraction, may enhance microvascular function. Anti-inflammatory agents, such as colchicine and interleukin-1 blockers, show potential but lack definitive evidence.^92–94 Finally, since ET-B receptors may also mediate vasoconstriction in disease, dual ET-A/ET-B blockade may offer advantages over ET-A-selective agents.^95–97

Conclusion

The regulation of myocardial blood flow is a finely tuned and intricate process, historically attributed primarily to arterioles within the microvasculature. However, modern insights emphasise that the regulation extends beyond arterioles to involve the entire microvascular network, including capillaries and venules. The endothelium plays an essential regulatory role by coordinating the balance between vasodilatory and vasoconstrictive factors. Among these, ET-1 is a particularly potent vasoconstrictor, extensively associated with the pathogenesis of microvascular angina. Its role in microvascular dysfunction has positioned ET-1 as a key pharmacological target for therapeutic intervention in individuals diagnosed with microvascular angina.

Click here to view Supplementary Material.