The term “no-reflow” was first proposed by Majno et al. (1967) [1]. These investigators found that after ischemia of the rabbit brain lasting 15 min, complete restoration of brain blood flow does not occur. A few years later, Kloner et al. [2] could demonstrate that injury of cardiac microvascular vessels is involved in the pathogenesis of no-reflow of the canine heart. In 1985, the no-reflow phenomenon was found in patients with ST-segment elevation myocardial infarction (STEMI) [3]. Investigators reported that thrombolysis could not completely restore coronary blood flow (CBF) in these patients. The duration of chest pain was less than 3 h before admission [3]. Currently, researchers often use the term “microvascular obstruction” (MVO) or the term “the slow flow phenomenon” because complete no-reflow (thrombolysis in myocardial infarction (TIMI) = 0) in the infarct-related coronary artery was detected angiographically in only 5% of patients with acute myocardial infarction (AMI) and percutaneous coronary intervention (PCI), in other patients the incomplete restoration of CBF was observed [4], where TIMI is Thrombolysis In Myocardial Infarction. It should be noted that some investigators suggested that no-reflow could be distinguished as TIMI = 0–1 or TIMI = 0–2 [5, 6, 7, 8, 9]. In this case, the terms “no-reflow” or “MVO” are used as synonyms. There is a correlation between TIMI evaluated angiographically and the MVO size measured by magnetic resonance imaging (MRI) [10, 11]. In a study of patients with AMI and PCI it was found that the duration of cardiac ischemia and infarct size are major determinant of severe MVO [12, 13, 14, 15]. MVO was found in 59% of patients with STEMI + PCI and a 3-h duration of ischemia [13]. If the duration of ischemia was 4–6 h, MVO was found in 72% of patients with STEMI + PCI [13]. The MVO area was measured by MRI [12, 14]. Infarct size in patients with STEMI and MVO was 2-fold larger than in patients without MVO [14]. MVO peaked at 07:00 a.m. [14]. Consequently, infarct size and the duration of ischemia are predictors of the development of MVO. It was reported that the MVO area is 1.9–5.4% of left ventricular (LV) mass in patients with STEMI and PCI according to MRI [11, 16, 17, 18, 19] or 22% of the infarcted myocardium [20]. According to Zia et al. [21] the MVO area is 3.1% of the myocardium by MRI. Infarct size was 13–32% of LV mass in patients with STEMI and MVO after PCI by MRI [11, 20, 22]. The intra-myocardial hemorrhage (IMH) area was 3.8% of LV mass in patients with STEMI and MVO after PCI [11]. Currently, the assessment of MVO often uses both angiography and MRI (Table 1, Ref. [5, 7, 9, 10, 11, 12, 13, 14, 16, 17, 18, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70]). However, in recent years, investigators increasingly favor MRI as a more accurate method of assessing MVO. Angiographic parameters are more variable (Table 2, Ref. [5, 7, 9, 10, 11, 18, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 34, 35, 36, 38, 39, 43, 44, 71]). Sardu et al. [72] found that prediabetes promotes the disorder of acetylcholine-induced coronary vasodilation and major adverse cardiac events in patients with non-obstructive coronary stenosis. Treatment with metformin, an AMP-activated kinase activator, partially restored acetylcholine-induced coronary vasodilation and reduced the incidence of major adverse cardiac events [72]. These data indirectly demonstrated that diabetes can promote the development of MVO and metformin partially reversed this negative effect of diabetes,

Thus, the MVO area is correlated with infarct size and depends on the duration of ischemia.

Microvascular obstruction was detected by MRI in 25% of patients with STEMI [45]. Ndrepepa et al. [5] reported that MVO was angiographically observed in 29% of patients with STEMI and PCI. According to Mayr et al. [65], MVO was found in 56% of patients with STEMI and PCI by MRI. MVO was diagnosed by echocardiography in 50% of patients with STEMI and PCI [69]. It was reported that no-reflow (TIMI = 0–2) was documented in 25% of patients with STEMI and PCI [7]. Microvascular obstruction was found in 25% of patients by angiography (TIMI = 0 or 2) with STEMI and PCI [9]. According to our data, the incidence of MVO is 37% in patients with STEMI and PCI by MRI data [66].

Thus, the incidence of MVO is observed in 25%–56% of patients with STEMI and PCI.

According to Abbo et al. [43], the incidence of no-reflow was 66 of 566 (11.6%) patients with AMI and thrombolytic therapy or PCI. They reported that patients with AMI and no-reflow experienced a 10-fold higher incidence of in-hospital death compared to AMI without no-reflow [43]. Cardiovascular events 6 months after AMI in patients with MVO are observed more often than in patients without MVO [45]. The in-hospital mortality rate was 14% in patients with AMI and MVO and only 3% in patients with AMI and without MVO [73]. The no-reflow phenomenon was accompanied by increased mortality for 3 years after AMI [44]. The no-reflow phenomenon was detected angiographically [44]. Patients with STEMI and no-reflow had an increased incidence of in-hospital mortality than patients without no-reflow (TIMI = 0–1) [9]. The no-reflow phenomenon was evaluated angiographically [9]. The mortality rate in patients with STEMI and MVO was greater compared to patients without MVO [14]. Adverse cardiovascular events in patients with AMI and MVO for 2 years after AMI developed more often than in patients with AMI without MVO [45]. MVO is an independent predictor of adverse LV remodeling in patients with STEMI [74]. MVO was evaluated angiographically [74]. In patients with STEMI with PCI, no-reflow (TIMI = 0–1) is a strong independent predictor of the mortality rate for 5 years after AMI [5]. No-reflow was detected angiographically [5]. Microvascular obstruction was usually accompanied by increased myocardial infarct size, a decreased LV ejection fraction, and a high mortality rate for 5 years after AMI [67]. The MVO area was measured by MRI [67]. Microvascular obstruction was associated with adverse cardiac remodeling for 8 months after AMI [68]. The MVO area were measured by MRI [68]. Major adverse cardiac events (MACE) for 6 months after STEMI was documented more often in patients with AMI and MVO [75]. No-reflow was detected angiographically [4]. MVO is a predictor of MACE in patients with STEMI and PCI [22, 59, 60, 61, 64, 76].

In summary, MVO is a common manifestation of AMI, especially in patients with STEMI. Microvascular obstruction is accompanied by a high mortality rate and is associated with MACE.

The highly effective therapy and prevention of MVO are impossible without knowledge of the pathogenesis of this pathology.

3.1 Microembolization and Microthrombi

In a study performed in 2012, MVO was assessed by myocardial blush grade (MBG) in patients with STEMI and PCI [77]. No-reflow was detected angiographically (Fig. 1). Blood samples were drawn from coronary arteries and the aorta for the detection of microparticles. It was shown that the microparticles’ level in the coronary artery is accompanied by MVO. It was concluded that-microparticles could be involved in the development of MVO [77]. This evidence is questionable because these data were not confirmed before by other investigators over the last 10 years. Moreover, it was obtained data that microthrombi is not involved in MVO [10, 11, 18]. Placebo-controlled studies in patients with STEMI + PCI have been performed [10, 11, 18]. The control group received a placebo, whilst alteplase was injected into coronary arteries of patients of the treatment group [10, 11, 18]. The MVO area was measured by MRI [18]. It was found that alteplase did not alter the MVO size measured by MRI [10, 11, 18]. Alteplase had no effect on infarct size but promoted the development of intramyocardial hemorrhage in patients with TIMI flow grade 2 [11]. We found that tenecteplase had no effect on the incidence of MVO.

Fig. 1.

Microvascular obstruction. Gadolinium contrast-enhanced cardiac magnetic resonance (DE-CMR) imaging of the basal, middle and apical (a, b, c) short-axis slices. Hypointense areas correspond to no-reflow in the projection of the anterior-septum wall of the left ventricle (LV) of the heart (a, b, c) of the LV in the mode of delayed contrast (inversion recovery sequence). HAL, head anterior left; CA, contrast agent; RAH, right anterior head; FPR, foot posterior right; LPF, left posterior foot; BP, body position; ST, slice thickness.

Microembolization and microthrombi do not play a significant role in the pathogenesis of MVO.

3.2 Platelet Aggregation

It has been shown that the microvascular obstruction score is lower in patients with STEMI and PCI who received aspirin than in patients without aspirin [46]. The MVO area was measured by MRI [45]. Patients with STEMI received heparin in the pre-hospital stage following PCI [57]. No-reflow was detected angiographically. Pretreatment with heparin contributed to a decrease in the incidence of no-reflow (TIMI = 0–1) by 13% (p 0.001) [57].

A correlation between the incidence of MVO and ADP-induced platelet aggregation in patients with STEMI and PCI was demonstrated [47]. The MVO area was measured by MRI [47]. A correlation between the incidence of MVO and platelet-neutrophil aggregation was also shown. The incidence of MVO and platelet-monocyte aggregation is also correlated [47]. Consequently, platelets could be involved in MVO (Fig. 2) by releasing a strong vasoconstrictor - thromboxane A2 [78]. Chronic administration of aspirin, a non-selective inhibitor of thromboxane A2 synthesis, resulted in a decrease in the serum concentration of thromboxane B2, a stable metabolite of thromboxane A2 [78]. These findings demonstrate that thromboxane A2 could be involved in MVO. Microvascular obstruction was more frequently observed in patients with STEMI + PCI and high platelet reactivity than in patients with low platelet reactivity [79]. No-reflow was detected as ST-segment regression 50% observed 90 min after PCI [79]. Consequently, platelets could be involved in the development of MVO. The role of thromboxane A2 in the pathogenesis of MVO is required for a study.

Fig. 2.

Some hypothetical pathogenic factors of microvascular obstruction.

Microvascular obstruction is often accompanied by intra-myocardial hemorrhage. A combination of MVO and IMH was found in 35–51% of patients with STEMI and PCI, where the hemorrhage area was about 3% of the LV mass [13, 19, 22, 56, 58, 66]. MVO and IMH areas were measured by MRI mass [13, 56, 66]. IMH without MVO was observed in 15% of patients with STEMI and PCI mass [66]. It was reported that the IMH area reached a maximum 24 h after the restoration of coronary perfusion and was about 4% of the left ventricle in pigs with coronary artery occlusion (CAO, 40 min) and reperfusion (24 h), while the maximum MVO area peaked in these pigs was 120 min after the restoration of coronary perfusion [102]. MVO and IMH areas were evaluated by MRI [102]. The largest area of IMH was identified in rats subjected to a 90-minute CAO followed by 48 hours of reperfusion [106]. The IMH area was measured in rats by MRI [106]. The onset of microvascular obstruction precedes the destruction of microvessels and the subsequent emergence of IMH. The presence of IMH was linked to poorer outcomes and the development of unfavorable ventricular remodeling. In patients with AMI, a larger IMH area was associated with longer ischemic durations and delayed reperfusion events [107]. Intra-myocardial hemorrhage was developed in pigs after a 40–120-min CAO and followed reperfusion [102]. According to Ma et al. [13] IMH did not develop before successful reperfusion of the heart and IMH size was correlated with infarct size and the MVO area. Investigators used CMR for the measurement of MVO and IMH areas. Intra-myocardial hemorrhage often develops in STEMI patients who have wider and deeper Q waves [108]. Anticoagulant and antiplatelet therapy could contribute to the occurrence of IMH in patients with AMI and PCI [108]. It was reported that the use of alteplase in patients with STEMI and PCI promotes an appearance of IMH [11]. A combination of MVO and IMH was observed in 36%–44% of patients with STEMI and PCI according to MRI data [13, 56, 66]. The pathogenesis of IMH remains unclear. It was reported that patients with IMH had higher CRP, inteleukin-6, fibrinogen, and neutrophils levels compared to patients without IMH [58, 88]. We found that the appearance of IMH is accompanied by an increase in the plasma CRP level was 13-fold in patients with STEMI and MVO. The role of inflammation in the development of IMH requires further investigation.

In summary, IMH is a common manifestation of AMI, especially in patients with STEMI. It is possible that inflammation is involved in the pathogenesis of IMH.

In our opinion, microvascular obstruction could be a target for the treatment of reperfusion cardiac injury. In recent years, much attention has been paid to dual antiplatelet therapy (DAPT, three possible combinations: aspirin and clopidogrel; aspirin and prasugrel; aspirin and ticagrelor) for the prevention of MVO [76, 109]. Some investigators performed DAPT in 97% of patients with STEMI and PCI [76].

The incidence of MVO in patients with STEMI and PCI receiving the P2Y${}_{12}$ antagonist clopidogrel was 66%, and in patients receiving the P2Y${}_{12}$ antagonists prasugrel or ticagrelor the incidence of MVO was 49% [54]. The MVO area was measured by CMR [54]. The glycoprotein IIb/IIIa inhibitor tirofiban was administered intravenously or intracoronally to patients with STEMI and PCI [17]. All patients received clopidogrel and aspirin before PCI. Intracoronary administration of tirofiban reduced the incidence of MVO compared to intravenous injection of tirofiban [17]. Investigators used MRI to measure the MVO area by MRI [17]. The efficacy of DAPT depends on inhibition of platelet aggregation, therefore assessingf the effect of DAPT on platelet aggregation is required. Massalha et al. [109] reported that hyporesponsiveness to aspirin or P2Y${}_{12}$ receptor inhibitor agents and demonstrated in 29% of patients with STEMI and PCI. Decreased platelet response to DAPT was accompanied by a greater extent of MVO [109].

It has been reported above that intracoronary infusion of adenosine prevents the appearance of MVO in patients with AMI and PCI [30, 32, 71]. Nevertheless, Niccoli et al. [26] demonstrated that the intracoronary delivery of adenosine did not influence the occurrence of MVO in a cohort of patients (n = 160) with STEMI who underwent PCI.

Nazir et al. [50] also could not find an improvement in MVO in patients (n = 168) with STEMI and PCI. It should be noted that adenosine can aggravate ischemic/reperfusion cardiac injury in patients with AMI through triggering coronary steal [110]. Consequently, adenosine cannot be recommended for the treatment of AMI and MVO.

It was reported above that sodium nitroprusside, a NO donor, did not alter the MVO area [25, 26, 50]. However, nicorandil, a NO donor and K${}_{\text{ATP}}$ channel opener, prevents the appearance of MVO by 50% in patients (n = 81) with AMI [101]. Combined intracoronary infusion of adenosine and nicorandil reduced the incidence of no-reflow (TIMI = 0–1) by 40% [33]. No-reflow was evaluated angiographically [33]. Intracoronary infusion of nicorandil reduced the incidence of MVO (TIMI = 0–2) in patients (n = 170) with STEMI and PCI compared to placebo [8].

In dogs with intact coronary arteries and no myocardial hypoxia, $\beta$-adrenergic receptor agonists were found to enhance CBF [111]. Nonetheless, when norepinephrine was infused intracoronarily in dogs with coronary stenosis, it led to increased myocardial oxygen consumption and myocardial hypoxia [112]. Administering the $\beta$1- and $\beta$2-adrenergic receptor agonist isoproterenol (0.1 µg/kg/min) intravenously during coronary artery occlusion and reperfusion in rabbits led to an enlargement of infarct size [113]. This suggests that the use of $\beta$-adrenergic receptor agonists in patients with AMI might exacerbate heart I/R injury. It’s worth noting, however, that these findings didn’t rule out the potential of a clinical investigation into the effectiveness of epinephrine intracoronary infusion as a therapy for MVO [28]. It was reported that intracoronary infusion of epinephrine decreases the MVO area in patients with AMI and PCI compared to patients without epinephrine [28]. These data were confirmed by Darwish et al. [29]. It should be noted that both groups of investigators did not evaluate the effect of epinephrine on the serum troponin I or creatine kinase levels, thereby it is unclear whether epinephrine can aggravate reperfusion cardiac injury or prevent reperfusion damage. In addition, it should be noted that these studies were not double-blind or placebo controlled.

Consequently, epinephrine could be used for therapy of MVO, but it should be evaluated for its negative effects.

L-type Ca${}^{2+}$ channel blockers result in vasodilation of coronary arteries [80, 93]. Intracoronary infusion of verapamil after administration of nitroglycerin improved TIMI flow grade in 89% of patients with AMI and PCI [39]. Intracoronary administration of verapamil decreased the MVO area in patients with AMI and PCI [24, 34, 35]. Intracoronary infusion of the L-type Ca${}^{2+}$ channel blocker nicardipine improved TIMI flow grade in 71 of 72 patients with AMI and PCI [41]. Verapamil alleviated no-reflow (TIMI = 0–1) in STEMI + PCI patients [36]. The ability of Ca${}^{2+}$ channel blockers to improve TIMI flow grade in patients with STEMI and PCI was confirmed by other investigators who used MRI to evaluate MVO [46]. Co-administration of nicardipine, adenosine, and nitroglycerine reversed no-reflow in patients with AMI and PCI [42].

Consequently, L-type Ca${}^{2+}$ channel blockers can be used for the treatment and prevention of MVO. The aforementioned data were summarized in Table 3 (Ref. [8, 17, 24, 25, 26, 28, 29, 30, 32, 33, 34, 35, 36, 39, 41, 50, 54, 71, 101, 109]).

These data demonstrated that clopidogrel, ticagrelor, and nicorandil reduce the incidence of MVO (Table 3). Epinephrine, verapamil, and nicardipine decrease the MVO area (Table 3).

Thus, MVO could be the result of an imbalance between vasodilation and vasoconstriction. Microvascular obstruction could be the result of coronary microvascular injury and, above all, the result of endothelial cell damage. It could be an inflammation injury of coronary microvessels. However, anti-inflammatory agents, for example, glucocorticoids have not been used before for treatment of MVO. Therefore, we cannot evaluate a role for inflammation in microvascular injury in patients with AMI and PCI.

Many questions are still waiting for an answer. It is unclear whether there is a role for endothelial cell injury in the pathogenesis of MVO in patients with AMI and reperfusion of the heart. There is no indisputable evidence of the involvement of inflammation in the development of MVO. A role for ROS in the pathogenesis of MVO is also yet to be studied. The role of necroptosis and pyroptosis in the pathogenesis of MVO in patients with AMI and PCI is also not studied. The role of thromboxane A2, vasopressin, angiotensin II, disturbances of nitric oxide production and prostacyclin synthesis in the formation of MVO was not studied before. The role of neuropeptide Y and endothelin-1 in the development of MVO is required in further investigations. It is unclear of the role of coronary artery spasm in the formation of MVO. It was reported that sodium nitroprusside, a donor NO and endothelium-independent vasodilator, did not improve CBF in patients with AMI and PCI. However, verapamil, an L-type Ca${}^{2+}$-channel blocker and endothelium-independent vasodilator, mitigates MVO in patients with AMI and PCI. It is possible that smooth muscle cells in coronary arteries lost sensitivity to nitric oxide in patients with MVO. The vasodilator effect of nicorandil, a NO donor and K${}_{\text{ATP}}$ channel opener, could be mediated via K${}_{\text{ATP}}$ channel opening in smooth muscle cells in coronary arteries.

Platelets could be involved in the development of microvascular obstruction in patients with AMI and PCI. Ca${}^{2+}$ overload of vascular smooth muscle cells in coronary arteries also participates in the pathogenesis of MVO. The duration of ischemia and infarct size are predictors of the appearance of MVO. Intramyocardial hemorrhage was found in 41–50% of patients with STEMI and PCI. Indirect evidence for the involvement of inflammation in the formation of MVO has also been obtained. Increased blood viscosity could contribute to the appearance of MVO. No-reflow and slow flow promote the formation of adverse myocardial remodeling. Probably endogenous catecholamines did not participate in the development of MVO in patients with AMI and PCI. Adenosine and sodium nitroprusside are not able to prevent the appearance of MVO. Intracoronary infusion of nicorandil could be used for the therapy of MVO. Glycoprotein IIb/IIIa inhibitor tirofiban and P2Y${}_{12}$ antagonists are low effective in the prevention and treatment of MVO. L-type Ca${}^{2+}$ channel blockers could be used for the treatment of MVO. Dual antiplatelet therapy improves the efficacy of PCI in the prevention of MVO.

LNM, EVV and NVN had the idea for the paper, reviewed and edited it critically for important intellectual content, FF, JMP, ERD were responsible for curating data, IAD performed visualization, VVR and AAB performed the literature search, GZS, AVM, BKK, AVK, AEG, SVD, JOS substantially contributed to the conception of the paper, wrote the manuscript, designed the figures and critically revised the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Not applicable.

Not applicable.

This article was supported by state assignment 122020300042-4.

The authors declare no conflict of interest.