2. Pathophysiology of MB and Its Association with MINOCA
The main feature associated with MB is a dynamic compression of the tunnelled artery during the systole. As coronary blood flow occurs primarily during diastole, only a small percentage (≃15%) of coronary blood flow should theoretically be compromised by the presence of a bridge; accordingly, MB has been long considered a benign condition. However, recent evidence has demonstrated that MB could lead to clinically evident myocardial ischemia through different mechanisms, including the interplay between a high sympathetic tone and a delayed early diastolic coronary artery relaxation
[15], the occurrence of a Venturi-like effect along the bridged artery
[16], an MB-mediated increased risk of epicardial or microvascular coronary spasm
[13[13][17],
17], atherosclerotic plaque disruption (i.e., plaque rupture or plaque erosion)
[18], spontaneous coronary artery dissection (SCAD)
[19] or their combination. Each of these mechanisms associated with MB may potentially be the underlying aetiology of MINOCA.
The presence of MB is associated with a delayed early diastolic coronary artery relaxation within the intramyocardial segment and a hindered rapid early diastolic hyperaemia, leading to a reduced coronary blood flow. This mechanism may be particularly relevant in the sub-endocardium, as it is more susceptible to ischemia, and it may be accentuated by an increased sympathetic tone
[20,21][20][21]. Indeed, the latter increases the heart rate, decreases the diastolic perfusion time, increases the strength of contraction across the MB and delays the relaxation beyond systole into the early diastolic phase, thus impairing coronary blood flow. Moreover, the presence of MB is associated with the occurrence of a Venturi-like effect at the ostium of the side branches, in particular the septal branches, also known as “branch steal”. Indeed, flowing through the tunnelled artery in the end systole/early diastole, the blood passes through a constricted segment leading to an increase in fluid velocity and a decrease in the perfusion pressure at the ostium of the side branches
[16]. The key aspects for the risk of myocardial ischemia associated with MB are the depth and the length of the tunnelled segment. Indeed, the depth of MB (<2 mm superficial, ≥2 mm deep, ≥5 very deep) is one of the main determinants of the degree of systolic compression and course of the affected artery. The length of MB is related not only to the amount of affected arteries but also to the number of side branches affected
[8].
Of importance is the fact that, as demonstrated by a recent study from researchers' group enrolling patients with myocardial ischemia and non-obstructive coronary arteries undergoing intracoronary provocative testing with acetylcholine (ACh), the prevalence of epicardial or microvascular coronary vasospasm is high among patients with MB (up to 79%, mainly epicardial spasm rather than microvascular spasm). Moreover, coronary spasm coexists more frequently in patients with MINOCA than in stable patients with no obstructive CAD (21.3% vs. 7%,
p = 0.002). Of note is the fact that the presence of MB was a predictor of MINOCA only in patients with a positive ACh test, and patients with MB and a positive ACh test had a lower major adverse cardiac-event-free survival, representing the group with the worst prognosis
[14]. Another study by Nam et al. demonstrated that severe MB (defined as bridge narrowing >90%) was significantly associated with a high risk of coronary spasm, and patients with both conditions were more likely to have recurrent angina compared with MB patients without spasm
[17].
Epicardial coronary vasospasm is defined as the vasoconstriction of an epicardial coronary artery caused by vascular smooth muscle hyper-reactivity due to increased sympathetic stimulation, local or systemic inflammation or environmental factors (i.e., smoking, alcohol consumption, environmental exposures)
[22]. Microvascular spasm results from an increased constriction of coronary microvessels due to a reduced production and/or enhanced degradation of nitric oxide and other endothelial-derived relaxation factors
[23,24][23][24]. Mechanistically, the compression–relaxation effect of MB on the coronary arteries may result in injury to the intima and endothelium, leading to impaired endothelium-dependent vasodilatation and enhanced local vascular reactivity to systemic vasoconstrictor stimuli
[25]. Moreover, a reduced expression of endothelial nitric oxide synthase has been demonstrated in the MB segment compared to the proximal and distal segments
[26]. These alterations also involve coronary microcirculation, as the presence of MB has been associated with alterations in vasoactive agents such as nitric oxide synthase and endothelin-1 and local variations in shear stress
[27]. The anatomical features of MB, such as the length and percentage of systolic compression, may further promote the occurrence of coronary vasospasm
[28,29][28][29].
Importantly, the presence of MB might favour the progression and destabilization of atherosclerotic plaques in the segment proximal to the tunnelled one
[18]. Plaque rupture is caused by a fibrous cap discontinuity, which exposes the plaque core to the coronary lumen, leading to the activation of a pro-thrombotic cascade and a consequent thrombus formation
[30,31][30][31]. Plaque erosion is due to the apoptosis of endothelial cells leading to thrombus formation without fibrous cap discontinuity
[31,32][31][32]. Both plaque rupture and plaque erosion may not determine an angiographically evident flow-limiting stenosis and present as MINOCA through different mechanisms, including a transient thrombosis with spontaneous thrombolysis, distal embolization, superimposed vasospasm or a combination of these processes
[33]. In patients with MB, the longstanding compression–relaxation effect of the intramural segment can cause changes in the wall shear stress, leading to the formation of an area of reduced wall shear stress proximal to the MB and the release of inflammatory mediators and endothelial vasoactive agents
[25]. Conversely, the wall shear stress is increased along the tunnelled artery, thus sparing this segment from atherosclerosis. Another mechanism that could promote atherosclerotic plaque formation and progression in the segment proximal to the MB is the abnormal blood flow profiles in this area caused by the collision of anterograde coronary flow with the retrograde flow due to the compression of the MB during the systole. Such alterations can determine endothelial injury, thus promoting plaque formation and disruption
[34].
The presence of MB might increase the risk of SCAD, a frequently overlooked cause of MINOCA, especially in young women. SCAD is caused by the separation of the layers of an epicardial coronary artery wall by intramural haemorrhage, with or without an intimal tear, not associated with atherosclerosis, iatrogenic injury or trauma, leading to the formation of a false lumen wherein the axial propagation of blood flow may result in a narrowing of the true lumen. The exact pathophysiology of SCAD is complex and likely involves multiple predisposing and precipitating factors, including genetic predisposition, hormonal influences, vascular and systemic inflammation, intense exercise, emotional stress and drug abuse
[35,36][35][36]. Different case reports showed that SCAD might occur in patients with MB even in the absence of other predisposing or precipitating factors and tends to localize within the intramyocardial segment or distal to the MB
[37,38][37][38]. Several mechanisms may be involved, including the occurrence of coronary vasospasm in the bridged segment that could increase its susceptibility to dissection or the presence of MB-induced endothelial dysfunction
[19,39][19][39]. Moreover, the flow disturbance caused by the pressure gradient between the bridged and non-bridged segment can lead to a state of chronic coronary pressure overload, which may cause intimal trauma, which eventually leads to intimal tearing and coronary dissection
[40].
Finally, previous studies have reported that MINOCA might be associated with a low but still not neglectable rate of complications during the acute phase, mainly represented by intraventricular septal rupture, free wall rupture with pericardial effusion or cardiac tamponade and ventricular arrhythmias
[41]. A case of ventricular septal rupture caused by MINOCA has recently been reported, in which a septal branch steal phenomenon might have contributed to its occurrence
[42]. However, to date, there are no data reporting the risk of mechanical complications in MB-related MINOCA, likely because this association is still frequently underdiagnosed in clinical practice.
3. Invasive Diagnostic Approach to Patients with MINOCA and MB
Coronary angiography is the cornerstone for the diagnosis of MINOCA by excluding the presence of stenosis of more than 50% in any major epicardial coronary artery. The use of intracoronary vasodilators during diagnostic coronary angiography can increase the systolic “milking” of the tunnelled artery induced by the systolic compression of the intramural artery and facilitate the detection of MB
[43,44][43][44]. A significant “milking effect” is present when there is a visual >70% reduction in the minimal luminal diameter during systole and a persistent >35% reduction in the minimal luminal diameter during mid-to-late diastole. Once the diagnosis of MINOCA and MB has been established or suspected, the use of advanced diagnostic techniques might be performed at the time of coronary angiography, aiming at confirming the presence of MB and identifying the presence of one of the above-described mechanisms of MINOCA associated with MB (
Figure 1).
Figure 1. Invasive diagnostic algorithm in patients with MINOCA associated with myocardial bridging. Abbreviations: CAD: coronary artery disease; DFFR: diastolic fractional flow reserve; iFR: instantaneous wave-free ratio; IVUS: intravascular ultrasound; MINOCA: myocardial infarction with non-obstructed coronary arteries; OCT: optical coherence tomography; SCAD: spontaneous coro-nary artery dissection.
Intracoronary imaging techniques (i.e., intravascular ultrasound (IVUS) or optical coherence tomography (OCT)) are extremely helpful in this setting. IVUS is an ultrasound-based technology that allows the detection of the systolic compression of the intramyocardial segment due to MB and of an echo lucent area between the intramyocardial tract of the coronary artery and the epicardium known as a “half-moon” sign, which represents the muscle band overlying the intramyocardial segment
[45]. Moreover, IVUS may allow the detection of atherosclerotic plaque rupture or SCAD. OCT is a light-based technique that provides high in vivo resolution (~10 µm) images of the coronary artery that are approximately ten times greater than those of IVUS
[46]. Therefore, OCT has a higher diagnostic accuracy for the presence of plaque rupture, plaque erosion or SCAD, but it has a lower diagnostic accuracy for the detection of MB, mainly because of its limited penetration and rapid OCT pullback and image acquisition (20 mm/s vs. 0.5 mm/s in IVUS). However, the presence of a heterogeneous fusiform band with sharp borders and a low-/intermediate-intensity signal surrounding the vessel adventitia has been described in the presence of MB (similar to the “half-moon” sign in IVUS), but further research is needed for its validation
[47,48][47][48]. IVUS and OCT show plaque rupture as a discontinuity of the fibrous cap overlying a lipid-rich core. A ruptured fibrous cap is associated with a vessel wall cavity without an IVUS or OCT signal
[49,50][49][50]. A thrombus often overlies the ruptured segment but may be absent in case of an old plaque rupture or with a recent rupture treated with antithrombotic therapies. Conversely, in the case of plaque erosion, OCT shows the absence of fibrous cap disruption in a lesion that is frequently composed of fibrous tissue with an overlying luminal white thrombus
[49]. IVUS and OCT can also be used to facilitate the diagnosis of SCAD, showing the intramural haematoma with or without the intimal flap
[49].
Intracoronary provocative tests using pharmacologic vasoactive agents can elicit a vasoconstrictive response at both the epicardial and microvascular level and, therefore, are fundamental for the diagnosis of coronary vasomotor disorders. ACh is usually preferred over ergonovine as a provocative agent due to its lower rate of complications. Moreover, performing an ACh-provocative test in MINOCA is safe with a relatively low risk of transient complications, which are mainly represented by transient bradyarrhythmia and supraventricular tachycardia
[51,52,53][51][52][53]. Given that ACh, by binding muscarinic 3 receptors, stimulates nitric oxide release from the endothelium to vasodilate and directly induce vascular smooth muscle cells (VSMCs) contractility, the net effect of intracoronary ACh administration is either vasodilatation (healthy endothelium) or coronary spasm (endothelial dysfunction or increased VSMCs reactivity)
[54,55,56][54][55][56]. This test is considered positive for epicardial coronary spasm in the presence of: (1) focal or diffuse epicardial coronary diameter reduction ≥90% in comparison to the relaxed state following intracoronary nitroglycerine administration given to relieve the spasm; (2) the reproduction of the patient’s symptoms; and (3) ischemic ECG shifts
[57]. On the other hand, microvascular spasm is diagnosed when typical ischemic ST-segment changes and angina develop in the absence of epicardial coronary constriction (<90% diameter reduction)
[58].
If other aetiologies have been excluded, the dynamic stenosis caused by MB may be responsible for myocardial ischemia and troponin release in particular circumstances (hypertensive crisis, tachyarrhythmias, etc.). To assess the hemodynamic significance of MB-induced stenosis, the use of several techniques has been suggested. Fractional flow reserve (FFR) is a pressure-wire-based index that is routinely used to assess the hemodynamic relevance of intermediate coronary stenosis. FFR is obtained through the calculation of the ratio of mean blood pressure distal to coronary stenosis to the mean aortic pressure during maximal hyperaemia achieved by the intravenous administration of adenosine (cut-off value of ≤0.80)
[59]. However, FFR may not be accurate for the assessment of MB as it is based on the assumption that the difference between the mean and diastolic pressure gradient values across the lesion is neglectable at maximal hyperaemia. Indeed, the presence of MB reduces systolic pressure gradients because of distal pressure overshooting during myocardial contraction, with the consequent overestimation of the mean pressure and, consequently, the underestimation of the hemodynamic impact of MB
[60]. Partially overcoming these limitations, diastolic FFR (dFFR), which is only calculated from diastolic pressures and is obtained during the maximal hyperaemia induced by intracoronary adenosine, has been demonstrated to be more sensitive than FFR for the functional assessment of MB. Escaned et al. found that, among twelve patients with symptomatic MB, dFFR identified a hemodynamic relevance in five patients, whereas mean FFR did so in only one patient, and the discrepancy between mean FFR and dFFR increased with dobutamine challenge
[61]. Indeed, as the dynamic obstruction caused by MB is deeply influenced by the degree of extravascular compression and intramyocardial tension that is increased during exercise or stress, its evaluation at rest may underestimate the hemodynamic relevance. For this reason, inotropic stimulation with dobutamine may enhance the diagnostic sensitivity of dFFR, and a cut-off value of ≤0.76 for dFFR has been proposed during dobutamine provocation to identify stress-induced myocardial ischemia in MB patients
[61]. However, dFFR is still not routinely performed because it is time-consuming, mainly due to the need for hyperaemic stimulus with adenosine, inotropic stimulation with dobutamine and data post-processing for its calculation
[62]. Additionally, non-hyperaemic pressure indices not requiring the administration of adenosine have been developed, including the instantaneous wave-free ratio (iFR), a diastolic-specific flow index. Tarantini et al., in a prospective study including 20 patients with angina and an abnormal non-invasive test suggestive of ischemia, demonstrated that iFR was associated with a higher proportion of positive obstructive bridges compared to FFR
[63]. Of note is the fact that iFR has only been formally validated in patients without MB with a cut-off value considered to be <0.89, while the measurement iFR in patients with MB in both rest and stress states (i.e., during inotropic infusion) requires further validation studies
[8].
Finally, the non-invasive functional evaluation at rest of MB with computed tomography (CT)-derived FFR has been investigated (cut-off value 0.75, grey zone 0.75–0.80)
[64]. However, further studies are needed for its validation, and, moreover, this approach is affected by the same limitations that apply to invasive FFR, especially in MINOCA patients
[8]. Indeed, during the acute phase, microvascular and endothelial dysfunction may be present, raising concerns regarding the achievement of maximal hyperaemia and, consequently, the reliability of FFR that can lead to the underestimation of the hemodynamic significance of MB
[56]. Similarly, cardiac magnetic resonance imaging could be used to assess for segmental myocardial perfusion defects and allow the physiological assessment of the functional effects of MB, but its routine use is limited by technical challenges and a lack of spatial resolution
[65]. In the context of MINOCA, cardiac magnetic resonance, thanks to its high accuracy in discriminating between ischaemic or non-ischaemic aetiologies, may help in the differential diagnosis between myocardial infarction, inflammatory cardiac diseases and Takotsubo syndrome
[50].
4. Therapeutic Implications
The management of MINOCA still has a limited number of evidence-based studies, as only a few registries have addressed this issue, and there is a strong need for randomized controlled trials in this setting
[66,67][66][67]. Adding complexity, specific therapeutic considerations are likely needed in patients presenting with MINOCA in which a specific pathogenetic mechanism associated with MB is diagnosed (
Table 1). However, there is a lack of evidence supporting any specific intervention as well as reporting the incremental prognosis of MINOCA associated with a MB. Accordingly, there is a strong need for properly designed randomized clinical trials as well as for data from observational studies or registry to support recommendations in future clinical consensus or guidelines in these patients.
Table 1. Therapeutic implications according to the underlying pathogenetic mechanism of MINOCA associated with a MB.