Evidence
Microvascular disease is common, its treatment depends on which endotype the patient has, naming the endotype measurably improves outcomes, and the anatomical tests most of these patients receive cannot see it.
A substantial share of patients sent for invasive coronary angiography with angina turn out to have no obstructive disease. The imbalance falls heavily on women: 50–70% of women referred for angiography have non-obstructive coronary arteries, compared with 30–50% of men.1
The profile is distinct from obstructive disease, which is part of why it gets missed. Symptomatic women are more likely than men to have ischaemia with non-obstructive coronary arteries.2 Among patients referred for comprehensive invasive physiology testing for suspected INOCA, women were younger than men, with a mean age of 57.0 years against 62.1.3 And the risk-factor profile differs rather than simply being milder: compared with obstructive CAD, INOCA patients had hypercholesterolaemia less often, were less often active smokers, had a lower mean BMI, and more often reported a family history of coronary disease, while also being prescribed antithrombotics, cholesterol-lowering drugs, and beta-blockers less often.4 The absence of obstructive disease is read as low risk, and these patients are often falsely reassured.2
This is not a rare phenotype at the margins of coronary disease. The 2024 ESC chronic coronary syndromes guidelines describe microvascular dysfunction as an increasingly acknowledged factor characterizing the entire spectrum of chronic coronary syndromes, capable of causing angina and ischaemia even where the large and medium vessels are unobstructed.5
Microvascular and vasospastic disease are not one condition with one drug. They separate into endotypes distinguishable by mechanism and by response to therapy, and discriminating epicardial vasospasm from microvascular dysfunction permits specific and distinct treatment.6 The therapies point in different directions, which is why guessing is costly.
| Endotype | Mechanism | First-line therapy |
|---|---|---|
| Microvascular angina | Impaired microvascular vasodilation or microvascular spasm; raised microvascular resistance | Non-selective beta-blocker, for example carvedilol 6.25 mg twice daily or nebivolol 2.5 mg, as first-line antianginal7 |
| Vasospastic angina | Abnormal vasoconstriction of the epicardial arteries causing dynamic obstruction | Calcium-channel antagonist, with or without nitrates5 |
| Mixed | Both microvascular dysfunction and epicardial spasm present together | Combined, guided by the dominant contributor6 |
| Non-cardiac | Chest pain not attributable to a coronary vasomotion disorder | Withdraw antianginal therapy and redirect the workup6 |
Baseline risk-factor therapy runs alongside this. Patients with hyperlipidaemia or hypertension were considered for aspirin, a statin, and an ACE inhibitor.7 And the mixed row is not a footnote: microvascular dysfunction and vasospastic angina frequently occur together, which makes both diagnosis and treatment harder.1
That the assignment holds up in practice is visible in the prescribing. At twelve months, patients with vasospastic angina in the intervention arm were more likely to be taking calcium-channel antagonists than controls, while those with microvascular angina were more likely to be on beta-blockers and ACE inhibitors.8
CorMicA randomized 151 patients with angina and no obstructive disease to therapy stratified by coronary function testing, against a blinded control arm receiving standard care and a sham procedure.6
The effect grew rather than washed out. At one year the difference was 13.6 units, a 27% higher overall angina score, corresponding to roughly one grade of the Canadian Cardiovascular Society classification and exceeding the minimum clinically important difference for the SAQ domains.8 Quality of life improved in parallel, with an EQ-5D index difference of 0.11 units.8
Two details are worth keeping. The microvascular angina group showed the most statistically significant treatment effect, while there was no significant effect in the non-cardiac chest pain group — which is the point of ruling patients out as much as ruling them in.8 And the one-year MACE rate of 12% is a reminder that microvascular and vasospastic angina without obstructive epicardial disease are not benign diagnoses.9
Most of these patients are worked up with coronary CT angiography, and increasingly with CT-derived fractional flow reserve. Both answer a question about the epicardial vessels. Coronary CT angiography is an anatomical method for ruling out obstructive disease, and it does not currently provide information on coronary microvascular dysfunction.10 FFR-CT extends this to the haemodynamic significance of a stenosis; it does not measure microvascular resistance.11
The consequence is a systematic miss. Vessels with under 50% stenosis on CT or angiography, or FFR above 0.8, are treated as low risk and often not investigated further.11
PET addresses the gap directly, because it quantifies myocardial blood flow noninvasively in absolute terms, in millilitres per minute per gram of tissue, rather than comparing one region of myocardium to another.12 It is not a complete answer: PET, Doppler echocardiography, and CMR provide surrogates of flow, and perfusion assessment alone does not separate the epicardial and microvascular contributions to a reduced flow measurement.10 The invasive procedure remains the reference standard. But a patient who leaves a CT scanner labelled low risk has been measured on the wrong axis entirely.