Stroke is currently the second leading cause of death (about 7 million) and the third leading cause of disability worldwide [1]. According to the WSO/Lancet Neurology Commission on Stroke, between 2020 and 2050, stroke mortality will increase by 50% (from 6.6 million per year to 9.7 million). Ischemic stroke (IS) accounts for 65.4%, intracerebral hemorrhage (ICH) for 28.8%, and subarachnoid hemorrhage (SAH) for 5.8% of all stroke cases [2]. The causes of ischemic stroke (IS) are stenosis, thromboembolism or thrombosis of cerebral vessels, while hemorrhagic stroke (HS) is caused by rupture of a blood vessel in the brain or under the meninges, as well as hemorrhagic transformation of IS (from small petechial diapedetic hemorrhages to intracerebral hematoma).

Ischemic stroke

About 50% of IS cases are caused by large vessel atherosclerosis and rupture of an atherosclerotic plaque, about 20% are caused by cardioembolism, and about 25% of IS present as lacunar infarcts due to small vessel disease and occlusion of deep perforating arteries [3]; vasculitis or extracranial artery dissection account for 5% [4].

Risk factors for IS include hypertension, diabetes mellitus, heart disease, smoking, hypercholesterolemia. Cardioembolic IS becomes the most common subtype with increasing age, while small vessel disease is the cause of IS in young people [5], but also increases with age – from 5% in people aged 50 to almost 100% in people over 90 [6]. In addition, there are differences in the distribution of IS subtypes among different ethnic groups [7]. Finally, not all IS cases have a known etiology, so it should be kept in mind that a certain number of IS cases have an unknown cause and are called cryptogenic strokes [8].

While atherosclerosis and cardioembolism often cause infarcts in both gray and white matter, lacunar infarcts are usually observed only in subcortical white or deep gray matter. Therefore, the main subtypes of IS differ significantly in blood RNA expression profiles [9].

As a result of ischemic brain injury, patients with IS show damage to neurons, glial cells, and blood vessels. Neurons in the center of the ischemic area (the infarct core) undergo colliquative necrosis, a process in which the neuronal body and axons disappear [10, 11]. Large neurons show edema, vacuolization of the neuroplasm, and disappearance of the nucleus and nucleolus. The shapes of smaller neurons become distorted, and the cell nucleus condenses. These symptoms indicate severe damage to organelles, including mitochondria, which no longer function properly and cannot produce energy for the cell; damage is also detected in glial cells, astrocytes, oligodendrocytes, and microglial cells [12].

The penumbra, which contains viable neurons surrounding the ischemic area, is also characterized by the presence of so-called “red neurons” or “ischemic neurons”. These neurons are characterized by several factors, such as acidophilic cytoplasm, changes in neuronal proteins, and disintegration of endoplasmic ribosomes and Nissl bodies [13].

Most cases of IS are caused by transient or permanent occlusion of a cerebral blood vessel, which leads to cerebral infarction. The final infarct size and neurological outcome depend on many factors, such as the etiology and location of the infarcts, the duration and severity of ischemia, the presence of collaterals and the level of systemic blood pressure, as well as the age, sex, comorbidities with corresponding drug therapy, and the patient’s genetic background [14]. Thus, IS is an extremely complex and heterogeneous disease.

The main approach in the treatment of IS resulting from cerebral vessel thrombosis is the use of thrombolytics (plasminogen tPA), which break down the thrombus and are the only FDA-approved therapeutic agent for the treatment of IS. Mechanical thrombus removal – thrombectomy – is also used. However, these treatments are limited to 3–4.5 hours from the onset of stroke symptoms. Recanalization (thrombolysis, thrombectomy) outside this therapeutic time window often leads to hemorrhagic transformation, which can cause additional brain damage, such as petechial hemorrhages along the borders of ischemia, confluent petechial hemorrhages in the ischemic zone, parenchymal hematomas, and artery dissection [15–17].

Reperfusion plays an important role in the outcomes of patients with IS because it helps restore normal cerebral blood flow and prevents progression of events in the infarct core. However, reoxygenation during reperfusion often exacerbates ischemic damage, primarily due to the induction of oxidative stress [18], thereby contributing to additional neuronal death [19]. Therefore, neuroprotective therapy with drugs possessing cerebrovascular and antioxidant properties becomes important [20–22].

To find new approaches and drugs for the treatment of IS, preclinical studies in adequate animal models are necessary. Modeling IS in animals serves as an indispensable tool for studying the mechanisms of ischemic brain damage and for developing new anti-ischemic drugs.

Most IS models are performed in rodents, and each model has its strengths and weaknesses. Reproducing all the manifestations of stroke observed in humans in a single animal model is impossible because IS is a heterogeneous disease. Experimental stroke models can only cover certain specific aspects of this multifaceted disease.

According to our understanding, a model of cerebral ischemia in animals should meet the following criteria: ischemic processes and pathophysiological responses should be relevant to human stroke, the size of the ischemic lesion should be reproducible, the modeling technique should be minimally invasive, physiological variables should be controllable and maintained within normal limits, and brain samples should be easily accessible for measuring outcomes such as histopathological, biochemical, and molecular biological assessment. These criteria for studying IS in experiments are also used in foreign IS studies [23, 24].

There are two main in vitro ways to induce cell ischemia – by oxygen and glucose deprivation, or by chemical or enzymatic blockade of cellular metabolism. The most commonly used in vitro model of cerebral ischemia is combined oxygen-glucose deprivation. For this purpose, the atmosphere in the incubator with cultured cells or brain slices is replaced. The usual O2/CO2‑equilibrated medium is replaced with an N2/CO2‑equilibrated medium in a hypoxic chamber that lacks glucose. The main cellular platforms are organotypic brain slices and primary cell cultures, respectively. The organotypic brain slice has the advantage of a functional system preserving neuronal morphology and the presence of glial cells and network connections, and allows separation of ischemic effects on neuronal tissue from effects caused by action on the cerebrovascular system.

For in vitro modeling of IS, models of the neurovascular unit (NVU)/blood‑brain barrier (BBB) are used, reproducing intercellular interactions and simulating blood flow and anatomical features of the brain. The advantage of the in vitro model lies in reflecting the main mechanisms and molecular pathways of cell death under ischemic conditions, as well as the possibility of working with human cells. Recent advances are observed in the field of 3D‑printed NVU models, which are expected to become a promising system for more reliable mechanistic studies and preclinical drug screening for the treatment of IS [25].

The main limitation of in vitro models of cerebral ischemia is the difference between the actual response of individual cell types that form part of brain tissue and the combined response of stroke‑affected tissue in in vivo rodent models.

All used in vivo IS models can be divided according to the nature of lesion spread into global ischemia models and focal ischemia models, and according to the time of vessel occlusion into permanent ischemia models and transient ischemia models. It should be noted that animal models mainly perform transient proximal occlusions, whereas patients more often have permanent proximal as well as permanent distal occlusions [26].

In experimental studies, the focal cerebral ischemia model, in which blood flow is reduced in a specific brain region, is widely used. Focal cerebral ischemia models are performed by mechanical vessel occlusion or by embolization methods [27]. Currently, this model is used in 5 variants: transcranial occlusion, endovascular filament occlusion of the middle cerebral artery (MCA), embolic occlusion, photothrombosis model, and endothelin-1 occlusion (an endogenous peptide with powerful and long‑lasting vasoconstrictor properties).

Since the MCA is affected most often in patients (almost in 50%), most focal ischemia models involve occlusion of this artery. MCA occlusion mainly leads to damage to the cortex and striatum, but the extent of infarction depends on the location and duration of occlusion, as well as the amount of collateral blood supply to the MCA. The most commonly used is the endovascular filament model, which can be used to model both permanent and transient focal ischemia [28, 29]. The size of the ischemic lesion varies considerably depending on the duration of ischemia. To obtain reproducible infarct volumes, 90–120 min of ischemia are required. The lesion caused by focal ischemia lasting more than 3 hours is irreversible.

Using a transient cerebral ischemia model with recanalization of the occluded artery, as occurs in transient ischemic attacks in humans or after therapeutic recanalization, the consequences of reperfusion in the ischemic territory (reperfusion injury) can be assessed [30, 31].

The advantages of focal cerebral ischemia models are a clear localization of the infarct (mainly the MCA), the presence of a penumbra, damage to the blood‑brain barrier, and inflammatory processes. Disadvantages of the model include: large infarcts mimicking malignant infarction [32], rapid reperfusion due to filament removal [33], and uneven reduction of cerebral blood flow [34].

Advantages of thromboembolism models are the ability to test thrombolytic drugs, assess ischemic lesions subjected to thrombolysis, and study combination therapy with thrombolytics and neuroprotectants.

Although a standardized animal model of thromboembolism is still lacking, data obtained from thromboembolic focal ischemia models clearly demonstrate the importance of studying the formation and composition of emboli used to induce sustained reduction of cerebral blood flow and reproducible lesions, while obstructive emboli should be located in the proximal segment of a large feeding artery [35, 36]. The characteristics of emboli (clots) can affect the efficacy of mechanical thrombectomy and thrombolysis.

Many compounds and artificial embolic materials are used to induce ischemia by injection into the common carotid or internal carotid artery, most often in rats, but also in larger animals, including primates. Viscous silicone [37], collagen [38], polyvinylsiloxane [39], and heterologous atheroemboli [40] are used to produce embolism.

Among embolization models without thrombus formation, microembolization induced by microspheres is a better studied model [41, 42]. The development of the lesion caused by microspheres occurs slowly, increasing in size over 24 hours after injection. This model can provide a wider therapeutic window for testing drugs in stroke, and also causes multifocal and heterogeneous lesions [43]. Advantages of the method include: presence of thromboembolic infarcts, permanent ischemia, penumbra, ischemic cell death, and inflammation. Disadvantages of this model include permanent ischemia without the possibility of reperfusion, multiple vessel occlusion (arterioles and capillaries), leading to redistribution of blood flow, disruption of the blood‑brain barrier, and vasogenic edema [44].

The photothrombosis model belongs to thromboembolic models, where a cortical infarct is induced by systemic injection of a photoactive dye (most often Bengal rose) combined with irradiation with a light beam of a specific wavelength. The formation of singlet oxygen leads to focal endothelial damage, platelet activation, and aggregation in both pial and intraparenchymal vessels within the irradiated area [32, 45–47]. The irradiation area can be defined to cause an ischemic lesion in any desired region of the cortex. The advantage of this model is that it addresses issues related to drug research and assessment of neuronal recovery. Other advantages of this model include minimal surgical intervention, high reproducibility of lesions, and low animal mortality.

However, the photothrombotic ischemic lesion lacks a penumbra because vasogenic edema and BBB disruption in the lesion occur within minutes. Another disadvantage of the photothrombosis model is that the damage induced by this method is occlusive in nature and resistant to therapy based on enhancing collateral perfusion.

Recent literature reports present a minimally invasive model of thrombotic stroke using magnetic particle delivery, which does not require craniotomy, is amenable to reperfusion therapy, can be combined with in vivo imaging techniques, and can be performed in awake mice [48]. The model is based on the delivery of PEGylated magnetic nanoparticles that are attracted to the middle cerebral artery (MCA) by a magnet placed on the skull. The model leads to reproducible cortical infarcts in the MCA territory with cytological and immune changes similar to those observed in more invasive models.

Endothelin-1 (ET-1) is a powerful vasoconstrictor produced by endothelial cells [49]. Artery occlusion is achieved by injecting ET-1 directly into or near the MCA, resulting in focal ischemia of the cortex and striatum. This model causes prolonged vasoconstriction leading to infarction. The infarct size depends on the concentration of ET-1 [50]. Advantages of the model: simplicity, minimal invasiveness, no need for craniotomy, possibility of conducting behavioral tests immediately after the intervention. The model is well suited for studying long‑term consequences of ischemia and recovery processes. Disadvantages: variability in infarct size, need for stereotaxic injection [51].

Permanent bilateral common carotid artery occlusion (2‑VO) leads to chronic cerebral hypoperfusion. This model is widely used to study vascular dementia and cognitive impairment in chronic cerebral ischemia. In rats after 2‑VO, cerebral blood flow is reduced by 40–60% in the cortex and hippocampus, and this persists for several months. The model is characterized by selective vulnerability of hippocampal neurons (CA1 field) and learning and memory impairments [52, 53]. It is important to note that 2‑VO alone may not cause obvious infarction, but leads to persistent changes in brain electrophysiological activity [54], neuropathological changes, and sustained oxidative stress [53]. Moreover, this model allows the study of ischemic white matter damage and ischemic eye disease.

The 2‑VO model in rats can be used to develop neuroprotective strategies not only for IS but also for other neurodegenerative diseases, including those associated with cognitive dysfunction in aging and Alzheimer’s disease [53, 55–58].

Consequences of 2‑VO include selective vulnerability of cortical neurons (death of pyramidal neurons in layers 3–4 of the neocortex, the granular layer of the neocortex (layer 2) and general disorganization of cortical neuronal layers), hippocampal neurons (death of pyramidal neurons in CA1 and CA3 fields and the dentate gyrus of the hippocampus), as well as vulnerability of small and medium neurons of the dorsoventral striatum [59]. In addition, expression levels of hnRNPA2/B1 and GABAAR‑α1 are significantly reduced in the hippocampus of rats with 2‑VO [60].

The 4‑VO model in rats involves electrocoagulation of the vertebral arteries between the first and second transverse processes through a dorsal incision with temporary bilateral occlusion of the common carotid arteries using microvascular clips 10, 20, or 30 minutes after vertebral artery occlusion [61]. However, non‑visual electrocoagulation of the vertebral arteries often leads to incomplete occlusion, which may hinder the creation of a successful model.

To study cerebral circulation disorders, global transient ischemia (GTI) is used by means of transient bilateral carotid artery occlusion combined with a reduction in mean arterial pressure to 50 mmHg, achieved by bloodletting followed by reperfusion and reinfusion, which represents an ischemic model of the forebrain. Autoradiographic measurements of local cerebral blood flow showed that under these conditions, cerebral blood flow decreases in the neocortex, hippocampus, and caudate nucleus to near‑zero values, while blood flow velocity in a number of subcortical regions is variable [62].

Using laser Doppler flowmetry, it has been shown that after global transient ischemia and reperfusion, blood flow in the rat cerebral cortex decreases by 30–40% of the baseline level and remains reduced for the entire observation period [63]. Under GTI conditions, an increase in glutamate content and a decrease in GABA levels are observed in the striatum [64], a decrease in catalase activity, increased free‑radical oxidation, as well as a significant decrease in nerve growth factor (NGF) content in the hippocampus, and an increase in stress protein HSP70 levels [65, 66]. Within this model, it was also shown that hypoperfusion impairs exploratory behavior of animals in the closed cross‑maze test [63].

One of the models close to the clinical manifestations of ischemic brain injury is the carotid artery stenosis (CAS) model, which involves partial ligation of the carotid arteries, leaving the vessel lumen within 30–50%, resulting in impaired motor and cognitive functions, morphological changes in brain tissue, and a decrease in GABA A R‑α1 levels in the hippocampus [60, 67].

The CAS model was used in our studies to examine both cerebral blood flow disorders and neurological behavioral and cognitive changes. Blood flow in the rat cerebral cortex under CAS conditions decreases by an average of 30%, and after one week, in these animals, against the background of neuroprotective treatment, cognitive functions improve and the GABAergic system is activated [68, 69].

One method of global cerebral ischemia is the gravitational overload model, which models global transient ischemia and subsequent reperfusion brain injury [70–73].

The advantage of the model is that experiments are performed on awake animals (no surgical interventions required) and it is possible to create both acute and chronic brain disorders depending on the magnitude of the overload. The model makes it possible to study the effects of different levels of hypergravity on brain neuronal activity, the immune system, physiological functions, and interactions between physiological systems [74, 75].

One important feature of human ischemic stroke that is ignored in most available animal models is the occurrence of spontaneous reperfusion, which is observed within the first hours after stroke onset [76]. In the early stage of IS, the infarct core is irreversibly damaged, and the penumbra expands to about two‑thirds of its maximum possible area within about 3 hours; therefore, early reperfusion can partially restore even the core tissue with secondary delayed damage developing up to 3 weeks after stroke onset, depending on the ischemic interval [33]. Peri‑infarct blood flow can also be maintained by collaterals via the circle of Willis and/or leptomeningeal anastomoses. Therefore, early restoration of blood circulation is of great importance.

According to the literature, the frequency of combination of myocardial infarction and cerebral stroke in clinical patients ranges from 1.3 to 12.8%, most often observed in the first 2 weeks of the disease [77–80]. According to medical registries (12 CVD registries) established under the supervision of the National Medical Research Center for Therapy and Preventive Medicine of the Ministry of Health of Russia, the proportion of persons who had a stroke against the background of myocardial infarction is about 20%. Over 10 years, the mortality rate of such hospitalized patients reaches 69%, compared to patients with IS (46.4%) and with myocardial infarction (47%) [81].

Given the prevalence of vascular comorbidity of infarction and stroke in the clinic, a model of combined vascular pathology of the brain has been developed. In rats, myocardial infarction is first reproduced by ligation of the descending coronary artery. Three days after the establishment of experimental infarction, which is confirmed by ECG recording in standard lead II, animals with confirmed infarction undergo global transient ischemia performed by occlusion of the carotid arteries for 10 minutes with a parallel reduction in blood pressure by withdrawing blood from the femoral artery to a pressure level of 50 mmHg. After removal of the clamps from the carotid arteries and reinfusion of blood, changes in cerebral blood flow and blood pressure are monitored. Under these conditions, when studying anti‑ischemic drugs, it turned out that their cerebrovascular activity changes: in some drugs the vascular effect is enhanced, while in others it disappears [82–85].

Subarachnoid hemorrhage (SAH) is bleeding into the space between the arachnoid and pia mater, which can occur spontaneously, usually due to rupture of an arterial aneurysm, or as a result of traumatic brain injury [86]. In SAH, intracranial pressure sharply increases, cerebral edema and ischemia develop. Consequences of hemorrhage can include depression of consciousness, paralysis, and even death. After SAH, almost 15% of patients die before hospitalization, and only less than 50% of survivors who receive adequate professional treatment can return to their former lives, while the remaining 50% suffer from neurological deficits, making them dependent on outside help for the rest of their lives [87]. A serious complication of SAH is cerebral vasospasm, leading to reduced blood flow. Symptomatic cerebral vasospasm (30% of patients) is the main cause of subsequent disability and mortality in patients with aneurysm rupture [88].

Several well‑studied animal models exist for investigating SAH pathophysiology, used over the last 40 years. Models using perforation with an intravascular filament of an intracranial artery (circle of Willis) or direct injection of blood into the cisterna magna or prechiasmatic cistern are the most commonly used procedures [89–91]. The advantage of the injection model is that the blood volume in SAH can be controlled [92]. However, this model, although it reflects a more “natural” course of the disease, does not accurately mimic acute pathophysiological changes and has a high mortality rate.

Intracerebral hemorrhage (ICH) is caused by rupture of pathologically altered walls of cerebral vessels or by diapedesis (the process of blood cells exiting through the walls of blood vessels into brain tissue as a result of inflammation of the tissues surrounding the vessels). The cause of intracerebral hemorrhage is most often hypertensive disease (80–85% of cases). Less frequently, hemorrhages are associated with atherosclerosis, blood diseases, inflammatory changes in cerebral vessels, intoxication, vitamin deficiencies, and other causes.

Brain injury after ICH can be primary and secondary. Primary brain injury occurs in the most acute stage of ICH and includes mechanical damage to the perihematoma tissue. The degree of primary injury depends on the location and volume of the hematoma. Most often, ICH is located in the basal ganglia, thalamus, and internal capsule, which are rich in white matter fibers and are easily damaged by mechanical stress from the hematoma [93]. Secondary brain injury after ICH is a complex process that includes neuroinflammation, oxidative stress, iron deposition (as hemosiderin), brain edema, and disruption of the blood‑brain barrier (BBB) [94]. White matter damage leads to neurological disorders such as sensory disturbances, motor dysfunction, cognitive impairment, and emotional disorders [95–97]. Functional impairments are also closely related to the location of the intracerebral hemorrhage and the expansion of the hematoma [98].

Various animal species are used to model ICH: rodents, rabbits, cats, dogs, pigs, primates. Experimental ICH models have been available since the 1960s and include intracerebral injection of autologous blood [99] or bacterial collagenase [100], balloon inflation [101, 102], or rupture of a cerebral blood vessel [103].

The most widely used method for creating a hematoma in the brain of an experimental animal is a single injection of autologous blood into the brain. Blood is taken from a superficial vessel and stereotaxically injected into the striatum to create a hematoma model. Rapid accumulation of intraparenchymal blood is important for intracerebral hemorrhage in patients [104, 105].

It has been established that under the ICH model conditions, cerebral blood flow decreases both around the hematoma and in the surrounding brain. This change depends significantly on the hematoma volume and is not accompanied by significant changes in cerebral perfusion pressure [106, 107]. The volume of injected blood varies in different studies and corresponds to the average hematoma size in humans [108]. For good reproducibility of hematoma volumes, slow injection of 50 μL of blood with a Hamilton syringe over 5 minutes is recommended [109]. Some researchers perform double or multiple injections, which cause persistent neurological deficit, brain edema, and cortical hypoperfusion [110, 111]. This technique has been adapted for mice [110]. The advantage of the method is high reproducibility, while the disadvantage is that the model does not reproduce the rupture of blood vessels and does not allow assessment of rebleeding.

The collagenase‑induced ICH model mimics spontaneous intraparenchymal hemorrhage that occurs in patients with intracerebral hemorrhage [112]. Collagenase is a protein that destroys collagen in the basal layer of the BBB, ultimately leading to microvascular rupture near the injection site. Hematoma expansion and vasogenic edema after ICH are thought to result from increased local concentration of collagenase released from damaged cells. Collagenase injection leads to continuous cerebral bleeding in a shorter time while avoiding surgical complications [94, 113]. However, collagenase damages many blood vessels, leading to extensive hemorrhage. In addition, degenerated erythrocytes and inflammatory cells are formed, leading to more severe inflammatory reactions [114].

The collagenase‑induced ICH model is used to study spontaneous intracerebral bleeding that develops over several hours [115, 116].

Makarenko A.N. et al. (2002) developed a combined ICH model with brain tissue damage using a special device (mandrel‑knife) stereotaxically inserted into the internal capsule or striatum, followed by injection of autologous blood into the damaged area [103, 117]. This model most fully reflects the clinical situation and reproduces the main diagnostic criteria of acute cerebrovascular accident. The model is convenient for reproducing lesions of different brain regions in neurophysiological and pharmacological studies and allows registration of cerebral blood flow [118]. In experimental studies on rats, it has been shown that the modified ICH model with internal capsule damage leads to long‑term axonal damage, neurological deficit, and histopathological and electrophysiological disturbances [98].

Our studies have shown that under this model, rats exhibit marked motor activity disorders, development of neurological and cognitive deficits, and paresis and paralysis of the limbs on the contralateral side of the lesion [119–122].

Morphological studies have shown that this method achieves a local autologous hemorrhagic lateral stroke in the region of the internal capsule without significant damage to overlying brain structures [123].

Barth A et al. (2007) proposed a modification of the ICH method in rats, according to which a cannula is stereotaxically inserted into the striatum of adult rats, then parenchymal damage is created using a rotating microcatheter inserted through the cannula, followed by slow infusion of 30 μL of autologous blood over 5 minutes. Hematoma volume and morphology are quantified, and animal behavior is analyzed using standardized tests [117].

The ICH model with brain tissue destruction is also used to study cerebral blood supply disorders. Using laser Doppler flowmetry, our studies have shown that in ICH, cerebral blood flow decreases by an average of 40% both in the damaged area and in the contralateral hemisphere; this decrease is reduced by the administration of nimodipine, which is used clinically in patients with HS [124, 125].

One of the side effects of thrombolytic treatment of stroke or thrombectomy is its transformation into hemorrhagic stroke [126, 127].

We have developed a model of combined cerebrovascular brain damage: a few days after carotid artery stenosis, hemorrhagic brain injury is induced in rats according to the method of A.N. Makarenko [103]. This model allows assessment of hemodynamic features in hemorrhagic transformation of ischemic stroke and development of new drug treatment approaches. It has been shown that cerebrovascular drugs that improve cerebral blood supply in IS and HS, within this model, increase cerebral blood flow to a greater extent than with each vascular pathology separately [125, 128].

Hypertension is one of the main risk factors for ICH. To study the features of ICH associated with hypertension, transgenic mice with increased expression of renin and angiotensinogen are used [79]. In this model, induction of ICH requires the addition of a special high‑salt diet and inhibition of nitric oxide synthase. To create a hypertension model in mice, subcutaneous administration of angiotensin II and inhibition of nitric oxide synthase, as well as acute injections of angiotensin to further increase blood pressure, are also used. Models reproducing cerebral amyloid angiopathy (CAA) in transgenic mice with increased expression of the amyloid precursor protein have also been developed. Mice with developing CAA exhibit spontaneous ICH [129].

The pathophysiology of brain stroke is complex and includes many interrelated processes, such as disruption of the integrity of the blood‑brain barrier, energy failure with subsequent development of energy deficit and oxidative stress, disruption of cellular ion homeostasis, increase in intracellular calcium levels with development of excitotoxicity and cytotoxicity, activation of astroglia and microglia, infiltration of the lesion by leukocytes, etc. These interrelated processes lead to cell death, primarily in the central part of the lesion – the zone where the most pronounced reduction in blood flow (close to “0”) is observed – and in the penumbra zone, which is functionally inactive due to reduced blood flow but remains metabolically active. In this regard, the main strategies for developing treatments for brain stroke are aimed at creating compounds with anti‑ischemic, neuroprotective action to restore neuronal function, primarily in the penumbra zone.

The main molecular, neurochemical, and biochemical cascades of ischemic brain damage development in humans and animals (rats) largely coincide. In particular, it has been shown that, as in humans with ischemic stroke, maximum brain edema in rats develops 24 hours after modeling ischemic stroke, and resolution of brain edema is observed by the end of the 3rd day. In the case of hemorrhagic stroke, both in humans and rats, a second wave of neurological complications is observed on days 5–6 after pathology creation, during the period of hemoglobin breakdown and hemosiderin deposition in brain tissues.

It can be assumed that experimental models of brain stroke are to some extent translational with human strokes. Understanding the advantages and disadvantages of experimental methods for reproducing cerebrovascular lesions allows modeling brain strokes with different etiologies and pathogenesis and assessing post‑stroke complications, including neurological, cognitive, and emotional disorders, and on this basis developing treatments for stroke and post‑stroke complications.