Introduction

Despite the undoubted successes achieved in the treatment of acute myocardial infarction (AMI) over the past 40 years, in-hospital mortality in this disease remains high. ^, Mortality is especially high in patients with cardiogenic shock, microvascular obstruction (MVO), and intramyocardial hemorrhage (IMH). The pathogenesis of MVO and IMH are interrelated. Endothelial cell injury is involved in the pathogenesis of both MVO and IMH. It was reported that MVO and IMH could precede the appearance of adverse postinfarction remodeling of the heart. Consequently, a study of the pathogenesis of MVO and IMH and the development of approaches to prevent the appearance of these adverse events is an important aim of cardiology.

The objective of the review article: to analyze clinical and experimental data on intra-myocardial hemorrhage and cardiac microvascular injury in ischemia/reperfusion of the heart.

Clinical data

In 1984, Yasuno et al. published a case report on a patient with ST-segment elevation myocardial infarction (STEMI) and thrombolytic therapy with urokinase. Autopsy showed severe transmural hemorrhage in the posterior wall of the left ventricle. Macroscopic hemorrhagic infarction was found in 14 autopsied patients with AMI and thrombolytic therapy with urokinase. IMH was associated with cardiac rupture in 4 patients. Forty-five patients with STEMI and percutaneous coronary intervention (PCI) underwent cardiac magnetic resonance (CMR) one week after AMI. All patients had MVO with with Thrombolysis In Myocardial Infarction (TIMI) (TIMI) flow grade 2 or 3 before PCI. IMH was found in 22 (49%) patients. Patients with IMH had larger infarct size, the larger MVO area after PCI, higher left ventricular end-diastolic volume (LVEDV), higher LV end-systolic volume (LVESV), and lower LV ejection fraction (LVEF). This study enrolled 108 patients with STEMI and PCI. Patients were divided into 3 groups: 30 (28%) patients with MVO(-)/IMH(-), 25 (23%) with MVO(+)/IMH(-), and 53 (49%) with MVO(+)/IMH(+). Two hundred and forty-five patients with STEMI and PCI were enrolled in a study. Patients were divided into 3 groups: 112 (45%) patients with MVO(-)/IMH(-), 32 (13%) with MVO(+)/IMH(-), and 101 (41%) with MVO(+)/IMH(+). IMH was found 2 days after AMI. It was accompanied by infarct size, a decrease in LVEF, an increase in LVESV, inflammation (P < 0.001, leukocyte cell count, neutrophil count, monocyte count), and adverse remodeling. However, investigators could not find an association between IMH and the C-reactive protein (CRP) or N-terminal prohormone of brain natriuretic peptide (NT-proBNP) levels in blood. It was found that IMH was more closely associated with adverse outcomes in patients than MVO. It was reported that isolated IMH without MVO is observed in 15% of patients with STEMI and PCI. A combination of IMH and MVO was found in 37% of patients with STEMI and PCI. We found that a combination of MVO and IMH in patients with STEMI and PCI is associated with an increase in the plasma CRP level but is not accompanied by leukocyte cell count [unpublished data]. These data demonstrated that inflammation could be involved in the pathogenesis of IMH. The appearance of a combination of MVO and IMH is associated with larger infarct size and contractile dysfunction. Consequently, MVO and IMH aggravate the course of AMI.

A study included 410 patients with STEMI and PCI. Cardiac magnetic resonance imaging (MRI) was carried out 5.5±1.8 days after PCI. IMH was found in 54% of patients. IMH was accompanied by larger infarct size, the larger MVO area, larger LVESV, and lower LVEF. The incidence of IMH reached a maximum at 7 days after PCI. It was reported that the MVO area correlated positively with the IMH size in patients with STEMI and PCI. IMH was found in 22 – 27% of patients with STEMI and PCI. ^, However, higher values are usually given. IMH was observed in 29% of females with STEMI + PCI and in 45% of males with STEMI + PCI (P = 0.032). It was reported that the incidence of IMH is 42 – 57% in patients with STEMI and PCI. IMH size is 0.8 – 8.3% of LV mass. ^{, ,} It is unclear why IMH size varies so much among different investigators. These data demonstrate that IMH is a common event in patients with STEMI and PCI observed in 42 – 57% of cases.

The appearance of IMH precedes the occurrence of adverse events. It was reported that IMH in patients with STEMI is associated with adverse remodeling of the heart.

A multicenter study enrolled 264 patients with STEMI and PCI. Major adverse cardiac events (MACE) occurred in 19 patients (7.2%) during the 12 months after AMI. MACE occurred more frequently in patients with IMH than in patients without IMH (P = 0.008).

What could be a trigger for IMH? IMH is associated with infarct size ^, and treatment with glycoprotein IIb/IIIa inhibitors. A multicenter study enrolled 421 patients with STEMI and PCI. Alteplase was administered intracoronary during PCI. Alteplase increased the incidence of MVO compared to placebo by 75% (P = 0.005). Alteplase increased the incidence of IMH compared to placebo by 2-fold (P = 0.001). According to our data using tenecteplase at the prehospital stage had no effect on the incidence of MVO or IMH in patients with STEMI and PCI [unpublished data]. According to Carrick’s data, IMH was accompanied by an increase in leukocyte cell count, neutrophil count, monocyte count in blood. However, investigators did not use treatment with anti-inflammatory drugs thereby it cannot be argued that inflammation is a trigger of MVO and IMH. Twenty patients with STEMI and PCI were enrolled in a study. Neutrophil count was increased during 4 – 24 h after admission in patients with IMH compared to patients without IMH. The plasma interleukin-6 (IL-6) level in patients with IMH was increased only at 24 h after admission. Plasma CRP concentration was increased 24 h after admission compared with patients without IMH. This increase was continued during the first week after PCI. Plasma fibronectin concentration was increased only one week after AMI. These data demonstrate that neutrophils, CRP, and IL-6 could be involved in the pathogenesis of IMH. An increase in the fibronectin level is most likely a consequence rather than a cause of IMH. The main disadvantage of this work is that the group of patients is too small (a total number of patients is 20). A study enrolled 170 patients with STEMI and PCI. Plasma IL-6 concentration was measured on day 2 after PCI. CMR was carried out on day 4 after admission. Patients with plasma IL-6 concentration ≥ 17 ng/L had lower LVEF, larger infarct size, the larger MVO area, and more frequent IMH (P < 0.001). Consequently, the high level of circulating IL-6 is accompanied by MVO and IMH.

Thus, IMH is observed in 42 – 57% of patients with STEMI and PCI. IMH promotes adverse post-infarction remodeling of the heart and MACE. The appearance of IMH is associated with larger infarct size, the larger MVO area, treatment with glycoprotein IIb/IIIa inhibitors, and intracoronary administration of alteplase. The occurrence of IMH is accompanied by inflammation. However, it remains unclear whether inflammation is an IMH trigger or a consequence of IMH.

Experimental studies

Dogs underwent coronary artery occlusion (CAO) for 1, 3, 5, and 7 h without reperfusion. Other dogs were subjected to CAO for 1, 3, 5, and 7 h with reperfusion (30 min). Endomyocardial IMH was found only in dogs with reperfusion after CAO (3 – 7 h). The quantity of blood in the epicardium was small. Consequently, long-term CAO and reperfusion are required for the development of IMH in dogs. IMH is located in the endocardium. However, it was demonstrated that IMH can develop in rats with long-term permanent CAO (24 h). Pigs were subjected to CAO (60 or 120 min) and reperfusion (75 min). Both a 60-min ischemia and a 120-min ischemia induced hemorrhagic myocardial necrosis. Pretreatment with the L-type Ca ²⁺ channel blocker diltiazem reduced IMH size by 50% if the duration of CAO was 60 min and had no effect if the duration of CAO was 120 min. Pigs (n = 23) underwent CAO (120 min) and reperfusion. Cardiac MRI was performed 1 and 5 weeks after reperfusion. MVO and IMH were identified in 17 pigs one week after reperfusion. Mice were subjected to CAO (60 min) and reperfusion (6 h and 7 days). Blood cell extravasation was found on day 7 after CAO. Pigs underwent CAO (75 min) and reperfusion. CMR was performed 7 days after CAO. MVO and IMH were found in 75% of pigs. The infarct size/area at risk (IS/AAR) ratio was 57%. The MVO size/infarcted area ratio was 35%. The IMH size/infarcted area ratio was 53%. Rats with CAO (60 min) and reperfusion (48) were included in a study. MRI was carried out 48 h after CAO. IMH was found in 70% of rats. Pigs underwent CAO (65 min) and reperfusion. CMR was performed 8 days after CAO. MVO was identified in all pigs. IMH was found in 46% of animals. Rats were subjected to CAO (60 min) and reperfusion. MRI was carried out 24 h, 48 h, 72 h, and 5 days after CAO. It was found that IMH size reached a maximum of 48 h after CAO. IMH size was 3.9% of LV, MVO was 3.8% of LV, infarct size was 26.7% of LV at 24 h (n = 5). IMH size was 5.2% of LV MVO was 3.1% of LV, infarct size was 25.7% of LV at 48 h (n = 5). However, these differences were not significant. In this study, infarct size was larger than IMH size. Rats underwent CAO (30 min or 90 min) and reperfusion (24 h). Larger IMH size was found by MRI only in rats with a 90-min ischemia. IMH size was 0.4% of LV in animals with a 30-min ischemia. IMH size was 9% of LV in animals with a 90-min ischemia. Consequently, if ischemia is prolonged then the IMH area is larger.

Pigs with CAO (45 min) and reperfusion were included in a study. CMR was carried out on day 1, week 1, and week 4 after CAO. It was reported that infarct size was larger in pigs with IMH than in animals without IMH. Contractile dysfunction was observed only in pigs with IMH. Cardiac fibrosis was developed primarily in animals with IMH. It was concluded that IMH is dependent upon infarct size. Investigators suggested that IMH promoted the appearance of contractile dysfunction and adverse remodeling.

What is the mechanism of development of IMH? It was shown that intracoronary administration of collagenase to pigs with CAO and reperfusion promoted the appearance of IMH. Collagenase is generally used for the disintegration of the myocardium and isolation of cells from it. Apparently, collagenase injured cardiac microvessels and contributed to IMH. These data indirectly demonstrated that microvascular injury promotes the occurrence of IMH. It was shown that infarct size is an independent predictor of IMH size in rats with CAO (60 min) and reperfusion (48 h).

Thus, the development of IMH is dependent upon the duration of ischemia and requires reperfusion of the heart. IMH is accompanied by contractile dysfunction and adverse remodeling of the heart. The most likely cause of IMH is microvascular injury.

Cardiac microvascular injury in ischemia/reperfusion of the heart

Rabbits with CAO (30 min) and reperfusion (3 h) were included in a study. Cardiac microvascular permeability (CMP) was measured by ¹²⁵ I-labelled albumin and ¹¹¹ In-labeled neutrophil accumulation in the area at risk. Ischemia/reperfusion (I/R) resulted in an increase in albumin (by about 20-fold) and neutrophil accumulation (by about 100-fold). Consequently, I/R of the heart induced an increase in CMP. Pigs underwent CAO (30, 60, and 90 min) and reperfusion. CMP increased progressively with an increase in the duration of ischemia and reperfusion. It was shown that CAO (45 min) and reperfusion caused glycocalyx injury in cardiac arterioles in mice.

The isolated rat heart was exposed to a 30-min ischemia, a 90-min ischemia, or a 30-min ischemia followed by a 60-min reperfusion. A 90-min ischemia induced minimal injury of coronary microvessels. In contrast, a 30-min ischemia followed by a 60-min reperfusion resulted in massive microvascular damage to I/R. Investigators suggested that reperfusion plays a key role in microvascular injury. These findings demonstrate that cardiac microvascular injury (CMI) can develop without the involvement of circulating leukocytes and proinflammatory cytokines in blood. Pigs underwent CAO (90 min) and reperfusion (3 h). Ischemic postconditioning and remote conditioning reduced infarct size and decreased CMI. I/R triggered injury of cardiac microcirculation endothelial cells in mice subjected to CAO (30 min) and reperfusion (2 h). ^, Investigators obtained evidence that I/R-induced injury of endothelial cells is a consequence of mitochondrial damage in these cells. Rats were subjected to CAO (60 min) and reperfusion (6 h). I/R-induced coronary microvessel permeability was detected by plasma albumin extravasation. Mice underwent CAO (45 min) and reperfusion. I/R caused cardiac microvessel permeability measured by Evans blue dye leakage in myocardial tissue. Cardiac ischemia (45 min) and reperfusion (120 min) resulted in CMI and erythrocyte extravasation in mice. Rats underwent CAO (45 min) and reperfusion (180 min). CMP was evaluated using fluorescent microspheres. I/R caused an increase in CMP, rupture of capillaries, and intramyocardial hemorrhage measured by erythrocyte extravasation.

Mice with CAO (1 h) and reperfusion (24 h) were included in a study. I/R (1/24 h) increased CMP by about 100-fold in the AAR. CMI was associated with an increase in the myocardial levels of proinflammatory cytokines: IL-1β, IL-6, tumor necrosis factor-α (TNF-α), and monocyte chemotactic protein-1 (MCP-1). Plasma IL-6 and TNF-α concentration is also increased. These data showed that CMI is associated with an increase in proinflammatory cytokine content in myocardial tissue and blood. Pretreatment with polypeptide relaxin (50 µg/kg) reduced infarct size, alleviated CMI, and reduced the myocardial and plasma proinflammatory cytokine levels. It was suggested that the cardioprotective effect of relaxin is accompanied by its anti-inflammatory effect. CAO (45 min) and reperfusion (24 h) induced cardiac microvascular damage, apoptosis of endothelial cells, and mitochondria injury in endothelial cells in mice. ^, This injury was accompanied by an increase in the IL-6, MCP-1, and TNF-α mRNA levels in myocardial tissue. ^, It is possible that microRNAs are also involved in I/R-induced CMI. It was reported that miR-499 isolated from plasma of patients with AMI can cause endothelial cell injury in coronary arteries of the isolated perfused rat heart.

Rats with CAO (30 min) and reperfusion (90 min) were included in a study. Cardiac microvessel permeability was measured by FITC-albumin leakage in myocardial tissue. I/R resulted in an increase in microvessel permeability and the downregulation of tight junction proteins (claudin-5, occludin, JAM-1, and VE-cadherin) in vascular endothelial cells.

These data demonstrate that reperfusion plays a key role in microvascular injury. Inflammation plays an important role in reperfusion cardiac injury, thereby it could be proposed that inflammation triggers CMI. CMI is associated with an increase in the myocardial and plasma levels of proinflammatory cytokines. It is possible that miR-499 is involved in the development of microvascular injury. An increase in cardiac microvessel permeability is accompanied by the downregulation of tight junction proteins. CMI can be developed without the involvement of circulating leukocytes and proinflammatory cytokines in blood.

Prospects for treatment of microvascular injury

As mentioned above, the L-type Ca ²⁺ channel blocker diltiazem reduced IMH size by 50% in rats if the duration of CAO was 60 min. However, diltiazem did not alter the IMH area if the duration of CAO was 120 min. It was reported that pretreatment with the adenosine A _2A agonist ATL-146e reduces reperfusion glycocalyx injury in mice. Melatonin protected cardiac microvasculature against I/R-induced injury via mitochondrial permeability transition pore closing in endothelial cells in mice. It was reported above that polypeptide relaxin, an agonist of relaxin receptors, alleviated CMI. Pretreatment with tanshinone IIA, a traditional Chinese drug, mitigated I/R-induced CMI in mice. Empagliflozin, an inhibitor of the sodium-glucose co-transporter-2 (SGLT2) in the proximal tubules of the kidneys, reduced CMI and erythrocyte extravasation in mice with CAO and reperfusion. ^, Dapagliflozin, an SGLT2 inhibitor, exhibited the same effect. Pretreatment with astragaloside IV reduced I/R-triggered CMP in rats with CAO (30 min) and reperfusion (90 min). Imatinib, a tyrosine kinase inhibitor and anticancer drug, alleviated I/R-induced CMP and IMH in rats with CAO and reperfusion.

These findings showed that pretreatment with ATL-146e, melatonin, tanshinone IIA, relaxin, empagliflozin, dapagliflozin, and astragaloside IV can mitigate I/R-induced CMI. L-type Ca ²⁺ channel blockers cannot be apparently used for treatment of CMI because diltiazem reduced IMH only in short-term CAO. Imatinib also cannot be used for treatment of CMI because most anticancer drugs exhibit toxic side effects.

Unresolved issues

It was mentioned above that CMI is associated with an increase in the myocardial and plasma levels of proinflammatory cytokines. However, there is no convincing evidence that proinflammatory cytokines trigger CMI. An increase in the proinflammatory cytokine and CMI could be two independent processes. Cardiac microvessel permeability is accompanied by the downregulation of tight junction proteins. However, the molecular mechanism of the downregulation of tight junction proteins is unknown. It is unknown whether remote postconditioning or adaptation to hypoxia can prevent the occurrence of CMI and IMH. It was reported that opioids, cannabinoids, and bradykinin can increase cardiac tolerance to I/R. However, it is unknown whether opioids, cannabinoids, and bradykinin prevent the development of MVO, CMI, and IMH. The ability of other cardioprotective compounds to reduce CMI and IMH has not been studied before.

Conclusion

A combination of MVO and IMH is a widespread pathology that is observed in 42 – 57% of patients with STEMI and PCI. The appearance of IMH is associated with inflammation. IMH is associated with infarct size, contractile dysfunction, and the duration of ischemia. IMH contributes to adverse postinfarction remodeling of the heart and MACE. Excessive thrombolytic therapy can promote the occurrence of IMH. Animal studies have shown that the appearance of IMH requires reperfusion and is dependent on the duration of ischemia and CMI. IMH promotes the occurrence of contractile dysfunction and adverse remodeling. IMH is accompanied by an increase in the myocardial and plasma proinflammatory cytokines. IMH is associated with a reduction in tight junction protein content in myocardial tissue. I/R-induced CMI can be formed without the involvement of circulating leukocytes and plasma proinflammatory cytokines. Pretreatment with ATL-146e, melatonin, tanshinone IIA, relaxin, empagliflozin, dapagliflozin, and astragaloside IV can mitigate I/R-induced CMI.

CRediT authorship contribution statement

Konstantin V. Zavadovsky: Conceptualization, Formal analysis, Writing – original draft. Vyacheslav V. Ryabov: Formal analysis, Validation. Evgeny V. Vyshlov: Data curation. Olga V. Mochula: Writing – review & editing. Maria Sirotina: Visualization, Writing – review & editing, Conceptualization. Artur Kan: Writing – review & editing. Alexander V. Mukhomedzyanov: Conceptualization, Supervision. Ivan A. Derkachev: Writing – review & editing. Nikita S. Voronkov: Writing – review & editing. Andrey V. Mochula: Writing – review & editing. Alexandra S. Maksimova: Formal analysis, Validation. Leonid N. Maslov: Conceptualization, Formal analysis, Supervision, Writing – review & editing.