1. Introduction

The post-ischemic no-reflow phenomenon is the absence of complete tissue perfusion at the microcirculatory level and occurs after removal of the cause of artery occlusion [1–5]. The incidence of angiographic no-reflow (TIMI 0–1 blood flow on angiography) ranges from 10 to 40% in subjects undergoing primary percutaneous coronary intervention [6]. Cardiac magnetic resonance imaging with gadolinium-based contrast, detects no-reflow in 54.9% of subjects, which is defined by a contrast-enhanced infarct core [4]. The occurrence of no-reflow reduces the infarct-limiting efficacy of early revascularization, worsens prognosis, increases incidence of early myocardial infarction complications and postinfarction scar size [4,5,7,8]. Two main mechanisms for the development of no-reflow are common: ischemia-reperfusion injury (IRI) and distal embolization by thrombus and atherosclerotic plaque fragments [9]. Markers such as carbon black particles, microspheres, and thioflavin S (ThS) are mainly used to visualize no-reflow in experimental studies with IRI [4]. Vital fluorescent dye ThS has been used since 1974 and remains the main fluorescent dye for estimating the size of the no-reflow zone [1]. It stains the entire endothelium and allows visualization of no-reflow zones or ThS negative zones, in which circulation is absent or severely reduced due to microvascular obstruction (MVO). ThS distribution in the organ is observed on its sections under ultraviolet excitation, where hypofluorescent or nonfluorescent dark areas can be found on the background of the fluorescent surface of section [1,4,10].

The inconvenience of using ThS is that this dye is not stable in the light or at room temperature, therefore dyed tissue samples should be stored in the dark at 4°C. In addition, ThS is an irritant, so it is necessary to wear appropriate personal protective equipment when working with it. In this context, the search for other fluorophores for no-reflow visualization is a relevant issue. There are conditions that allow us to consider the use of indocyanine green (ICG) for this purpose. Visualization in the near-infrared range with ICG has several advantages over fluorescence in the visible range, namely: greater depth of penetration, weak dependence on ambient light, higher contrast due to the absence of tissue autofluorescence [11]. In addition, ICG is FDA approved for clinical use and this method is not harmful to either the animal or the operator.

The method of no-reflow visualization in the experiment with ICG is based on the ability of the fluorophore to accumulate in the area of tissue injury and necrosis [11–13]. We have previously shown that ICG injected in the first minutes of reperfusion (“early ICG”) after 30 min of ischemia in rat myocardium remains in the infarct area and its fluorescence can be observed in vivo, on the surface of the heart for two hours, and ex vivo in heart sections at the end of 2 h of reperfusion [13]. Early ICG fluorescence in rat heart sections was visible over the entire surface of the necrosis zone, but the no-reflow zone was not visualized under these conditions. The absence of no-reflow zones (immediate no-reflow) in the first minutes of reperfusion after brief ischemia (30 minutes) has also been demonstrated with ThS in dogs [2] and rabbits [10]. This may be explained by the fact that no-reflow zones in rat’s myocardium after 30 min of ischemia do not appear until tens of minutes after the onset of reperfusion and expand with time (delayed no-reflow). We hypothesized that delayed administration of ICG (“late ICG”) injected after 90 min of reperfusion would allow simultaneous visualization of no-reflow zones and areas of increased vascular permeability throughout the myocardium. It is expected that the intensity of ICG fluorescence will depend on the degree of vascular permeability, i.e. the area with the brightest ICG fluorescence will indicate the presence of the highest vascular permeability. The contrast between areas of normal and increased vascular permeability and ICG accumulation can be seen in 20–30 min from the moment of intravenous fluorophore injection, which is necessary for clearance of ICG from the microcirculatory bed by the liver and its retention in the interstitial space of the injured tissue.

To test the feasibility of usage of ICG for visualization of the no-reflow, the current study was designed to answer the following questions: 1) do the no-reflow zones detected by ICG correspond to those visualized by ThS, which is the standard fluorophore for this purpose; 2) how does the zone of ICG fluorescence and its intensity correlate with the zone of necrosis; 3) is it possible to detect ICG fluorescence by delaying intravenous ICG administration for 24 hours?

Therefore, the aim of the present study was to investigate the imaging characteristics of myocardial infarction using ICG for estimation of the severity of no-reflow phenomenon in comparision with ThS.

2. Materials and methods

The experiments were performed on male Wistar rats weighing 250–300 g. The animals were fed standard laboratory rodent chow and given water ad libitum. All experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals and approved by the local ethics committee.

An in vivo rat model of regional myocardial IRI was used in all experiments. Figure 1 shows the experimental study protocol.

 

Fig. 1. The experimental study protocols.

Download Full Size | PDF

2.1 Preparation of ICG solutions

ICG (Pulsion medical systems, AG, Germany) was dissolved in distilled water at a final concentration of ICG 0.25 mg/ml. NaCl was added to the ICG solution to get concentration of NaCl 0.9%.

2.2 In vivo rat model of myocardial IRI

Male Wistar rats were anesthetized with isoflurane 2–3%. Body temperature was maintained at 37.0 ± 0.5 °C (ATC1000-220, World Precision Instruments, Inc., USA). The rats were tracheostomized and mechanically ventilated (CWE-SAR-830/AP, World Precision Instruments, Inc, USA) with 60% oxygen (respiratory rate – 60 / min, tidal volume of 3 mL / 100 g body weight). Ventilation was adjusted by repeated arterial blood gas analyses throughout the experiment (ABL80FLEX, Radiometer, Denmark). Femoral vein catheterization was performed for infusions of the dyes. The right common carotid artery was cannulated with polyethylene tube (PE-50, Intramedic, USA) for blood sample withdrawal, measurement of arterial blood pressure (BP) and heart rate (HR). The arterial cannula was connected to a pressure transducer (Baxter, USA). During the experiments, animals had continuous monitoring of hemodynamic parameters using software PhysExp (LLC “Kardioprotekt”, Russia). For induction of regional IRI the chest was opened by incision at the fourth intercostal space and the ribs were spread to expose the heart. The pericardium was opened, and a 6-0 polypropylene non-traumatic suture was passed around the major branch of the left coronary artery, about 2 mm of its origin [14]. The ends of suture were passed through an occluder – small polyethylene tube ∼ 6–7 cm (PE-90, Intramedic, USA) and exteriorized. After the end of surgical procedures and 30 min stabilizing period, myocardial ischemia was initiated. The reversible myocardial ischemia was produced by shifting occluder down along ligatures and placing a surgical clamp on the occluder to prevent its shifting back. Registration of hemodynamic parameters was performed just before the 30 min occlusion, after 5 and 15 min after the occlusion, at the beginning of reperfusion (in the 5th min) and then every 30 min until the end of the experiment.

2.3 Ex vivo visualization of the IRI

At the end of the experiment (at 120 min or 24 h of reperfusion) the left coronary artery was reoccluded, followed by administration of 2.5 mL of 2.5% Evans blue (Merck, USA) via the femoral vein for identification of the area at risk (AAR). The hearts were excised for photographing of the outer surface followed by the registration of the fluorescence area and intensity of ICG and ThS in areas of IRI and the surrounding undamaged myocardium. Then hearts were cut into five 2 mm thick sections parallel to the atrioventricular groove. The basal surface of each section was photographed, using a digital camera. The sections were immersed in a 1% solution of 2,3,5-triphenyltetrazolium chloride (TTC; ICN Pharmaceuticals, USA) at 37 °C (pH 7.4) for 15 min and photographed again for identification of infarct area and registration of fluorescence area and intensity of ICG and ThS. The images were analyzed using ImageJ software (Bethesda, MD, USA). The AAR (Evans-negative sites) was expressed as a percentage of the whole section, and the infarct area (TTC-negative sites) was expressed as a percentage of the AAR. Values of AAR and infarct area for each heart were obtained by summarizing data for the sections and calculating mean values.

2.4 Methods of fluorescence registration

Semiconductor laser “Diolan” (NPP VOLO, Russia) with a wavelength of 808 nm, power 0.5-5 W and quartz fiber optic light guide was used to excite ICG fluorescence [15]. A lamp illuminator was used for ThS, in which an HXP 120VIS short-arc mercury lamp Osram, 120 W was used as a light source, radiation selection is carried out by switchable light filters (in this case, a bandpass filter with a central wavelength of 390 nm, width 40 nm, FF01-390/ 40-25) and a liquid lightguide for radiation delivery [16]. To improve the uniformity of illumination, the light guides are equipped with collector lens (illumination unevenness ±5%). The power density of the exciting radiation was about 15 mW/cm² (390 nm) and 50 mW/cm² (808 nm). At this excitation intensity, no damage to the samples was observed. Image registration was carried out using a multispectral imaging system is based on a high-sensitivity RGB-television array (ICX285AQ single-array progressive-scanning CCD detector of 2/3-inch format (SONY), maximum frame frequency is 14 Hz with 1280 × 1024 resolution elements, intrinsic noise equals 10 e-). In front of the camera, which is equipped with a Computar M1614-MP2 megapixel lens (Japan) (f = 16 mm, F/1.4), detector filters BLP01-442R-25 and NF03-808E-25 (Semrock, USA) were installed to block exciting radiation 390 nm and 808 nm, respectively. The dimensions of the field of view in these studies were 25 × 25 mm, which made it possible to simultaneously record all 5 sections of the myocardium, which were in a Petri dish filled with a buffer solution. To stabilize the position of the sections and eliminate glare, the sections were covered with a microscope slide. The exposure time did not exceed 70 ms, the s/n ratio was approximately 100.

2.5 Comparison of ICG and ThS fluorescence intensity in the area of myocardial IRI

The obtained images with ICG and ThS fluorescence were evaluated using Image-Pro Plus software (Rockville, MD, USA). To study the nature of ICG and ThS fluorescence intensity, a grid was drawn on the images of the AAR of the second and third sections from the apex of the heart, with an equal number of sectors and grid cells and limited by the boundaries of the AAR.The grid formed in this way was then transferred to images of the same sections stained with 2,3,5-triphenyltetrazolium chloride (TTC) and with ICG and ThS fluorescence.

The grid allows comparison of ICG and ThS fluorescence intensities in the same section, as well as localization and relation of borders of increased vascular permeability with borders of necrosis and no-reflow zones. The average fluorescence intensity within selected region of interest (ROI) was then calculated and expressed as a.u. The average fluorescence intensity of ROI in any grid cell within the AAR was compared with the average fluorescence intensity in the corresponding cell in the reference sector in the interventricular septum equidistant from the AAR borders.

Quantitative comparison was performed using the contrast parameter calculated by the formula:

Additionally, to compare the fluorescence intensity of ICG and ThS in different layers of the left ventricular myocardium, a graphical tool was used – the line intensity scan function in Image Pro Plus software (Media Cybernetics, USA). Graph of ICG and ThS fluorescence intensity along the scan line and the reference line were plotted using the line profile tool.

2.6 Comparison of no-reflow area sizes from ICG and ThS fluorescence images

To calculate the AAR, necrosis, and no-reflow zones, planimetric imaging of cardiac sections was performed in the ImageJ software. The boundaries of the no-reflow zones in the images of ICG fluorescence were considered to be the dark edge of the ICG peak fluorescence stripe along which the boundary of the no-reflow area was delineated. Also, the borders of nonfluorescent areas in the images of ThS fluorescence were traced for calculation of the no-reflow zone.

2.7 Myocardial histology

Heart samples were excised and fixed in buffered 10% formaldehyde and histology slides were prepared by Mallory’s trichrome-stain method. Morphometry of 5-µm sections was used to quantify the areas of stasis (the areas occupied by erythrocytes). The areas of stasis (the target area) in the intramural layer of the area at risk (total area 3 mm²) were detected using the ImageJ threshold tool. The percentage of the target area was calculated using the formula: (target area per section/total area) × 100.

2.8 Statistical analysis

Statistical analysis was performed using the nonparametric Mann-Whitney U test, with P values less than 0.05 considered to be statistically significant.

3. Results

No differences in hemodynamic parameters were observed between the three groups. No significant hemodynamic changes were observed (data not shown).

3.1 Effect of myocardial TTC staining on ThS fluorescence intensity

To examine the relation of nonfluorescent areas (no-reflow) with the borders of the risk zone and necrosis zone in hearts stained with ICG and ThS, additional double staining with Evans blue (EB) and TTC was performed. Figure 2(c) shows that after additional double staining (EB and TTC), ThS fluorescence intensity decreased compared to TTC(-) sections (Fig. 2(c)). However, this decrease allowed us to see the ThS peak fluorescence stripe along the border of the no-reflow zone (Fig. 2(c), 3(d), 4(a)). This finding showing the presence of increased vascular permeability at the border of the no-reflow zone. The same peak fluorescence stripe of ICG was seen along the border of the no-reflow zone in the near-infrared light. These observations suggest similar mechanisms of accumulation of a portion of ThS with ICG at the border of the no-reflow zone.

 

Fig. 2. Effect of myocardial EB and TTC staining on ThS fluorescence intensity. (a) Selection of rat heart sections for measurement of ICG and ThS fluorescence intensity in the no-reflow zone. Numbers from 1 to 5 are the number of transverse sections from the apex to the base of the rat heart. Arrows indicate 2 (apical) and 3 (middle) sections. (b) Fluorescence of ThS and ICG, respectively. Numbers in circles indicate the second and third sections used to measure ICG and ThS fluorescence intensities. (c) Comparison of ThS fluorescence intensity in TTC-stained (TTC+) and non-TTC-stained (TTC-) representative sections of rat hearts. (d) Comparison of ThS fluorescence intensity in representative sections of the interventricular septum (reference zone) – red scan lines and the left ventricular free wall – orange scan lines, where no-reflow zones are visible. Vertical blue arrows show the change of ThS fluorescence intensity in the reference zone of the section compared to the background (blue lines). Scale bars: 1 mm.

Download Full Size | PDF

 

Fig. 3. Evaluation of the initial stage of no-reflow expansion via double ICG and ThS staining. Imaging of ICG and ThS fluorescence of the same section was performed after double histochemical staining (EB and TTC). (a) The area between the dashed lines delineating the anatomical AAR is divided into 6 equal sectors: two border sectors (BZ-1 and BZ-2) and four inner sectors (S1–S4). Each sector is divided into three grid cells: 1) subepicardial (Subep.), 2) intramural (Intr.), and 3) subendocardial (Suben.). Reference – red reference line in the reference sector plotted equidistant from the AAR. (b) Image of the same section stained with TTC. (c) Section in near-infrared light. * – the site of maximum ICG fluorescence intensity. Scan line 3 and ref. are highlighted in red to show ICG and ThS fluorescence intensity in plot (e). (d) Section in UV-light. * – the point corresponding to the maximum ICG fluorescence intensity. (e) Graph of ICG and ThS fluorescence intensity along the scan line 3 and the reference line (red) plotted. The gray “White” line was drawn from the image of the TTC-stained section (plot (b)). Scale bars: 1 mm.

Download Full Size | PDF

 

Fig. 4. Comparison of sizes of anatomic no reflow detected with both ICG and ThS in the same heart. There was increase in size of no-reflow area between 90’ICG and 120’ThS (n = 7). (a) Representative ICG (ICG'90) and ThS fluorescence (ThS'120) images of a single section. Dotted lines are the borders of the no-reflow zone. (b) Comparison of the area of the no-reflow obtained by planimetric analysis of fluorescence images in cross-sections of rat hearts. (c) Positive correlation of no-reflow area sizes obtained by ICG and ThS fluorescence imaging (p = 0.0341). Scale bars: 1 mm.

Download Full Size | PDF

3.2 Imaging of the early stage of no-reflow expansion by double ICG and ThS staining

Figure 3 shows that injection of ICG, which has a short plasma half-life, in the first minutes of reperfusion allowed to capture the initial stage of no-reflow development, when there is increased vascular permeability in the center of the AAR (Fig. 3(a)–3(c)) and there is no microvascular obstruction (MVO) (Fig. 3(c)). In this moment there is the greatest intensity of ICG accumulation in the subepicardial and intramural layers of the inner sectors (S2 and S3) where drastic increase in vascular permeability occurs. From the Fig. 3(e), it is evident that boundaries of ICG retention zone are wider than borders of necrosis zone on 0.5 - 1.0 mm, however where the ICG retention zone overlaps the necrosis zone, the intensity of ICG fluorescence increases greatly (Fig. 3(b), 3(c), 3(e)).

ICG and ThS fluorescence images (Fig. 3(c), 3d) show that the sites of increased vascular permeability at the beginning of reperfusion became the MVO zones after 2 hours (Fig. 3(d)), as indicated by the presence of the ThS(-) zone in the corresponding grid cells (S2 and S3 sectors). Figure 3(e) shows that the maximum ICG fluorescence intensity overlaps by the no-reflow zone. Thus, double ICG and ThS staining with early ICG injection can be useful as it allows evaluation not only the initial stage of no-reflow expansion, but also visualization of MVO including track the correlation between MVO severity at the end of reperfusion and the severity of vascular permeability at its beginning.

3.3 Comparison of the size of no-reflow zones obtained with two fluorophores: ICG and ThS after 2 h of reperfusion

The main objective of the study was to evaluate the possibility of using ICG to visualize no-reflow zones and to compare the size of no-reflow zones in the same animal obtained with two fluorophores: ICG and ThS. For this purpose, rats were injected intravenously with ICG solution at the 90th min of reperfusion and with ThS solution 30 min later (120 min after the onset of reperfusion).

ICG signal was detectable macroscopically within the zones of myocardial necrosis (Fig. 4(a), 4(b)), within which ICG-negative zones were detected. The sizes of ICG-negative zones were significantly smaller by 8.7% [3.16–12.98] than the sizes of the no-reflow zones detected at 120 min with ThS (Fig. 4(b)). The boundaries of the ICG-negative zones lie within the boundaries of the ThS-negative zones, and the sizes of the no-reflow zones obtained with ICG correlate with the sizes of the no-reflow zones detected with ThS (Fig. 4(c)). These observations suggest that the ICG-negative zones are the area of the no-reflow zone that became larger 30 min after ICG injection (90’ICG) and was detected with ThS (120’ThS).

3.4 Comparison of no-reflow zone sizes obtained with ICG and ThS after 24 h of reperfusion

To assess the ability to visualize no-reflow after 24 h of reperfusion with ICG, we performed 30 min of regional ischemia and 24 h of reperfusion (I-30’/P-24 h). Animals were reanesthetized 24 h after IRI to visualize no-reflow and measure infarct size. Figure 5(a)–5(c) show images of double fluorescence (ICG and ThS) and TTC stained section one day after the ischemic episode. The images 5(a) and 5(b) show ICG and ThS fluorescence within the zone of myocardial necrosis, within which ICG-negative zones are identified, the size of which is not significantly different from the size of the no-reflow zones identified with ThS, indicating that the ICG-negative zones are no-reflow zones (Fig. 5(d)).

 

Fig. 5. Comparison of the size of no-reflow zones after 30 min of ischemia and 24 h of reperfusion in the same heart using both ICG and ThS staining in the same heart. (a) and (b) are ICG and ThS fluorescence images. (c) The same section in white light. The free wall of the left ventricle shows a TTC-negative zone, which is a zone of necrosis. (d) Sizes of no-reflow zones detected by ICG and ThS (n = 5). Scale bars: 1 mm.

Download Full Size | PDF

3.5 Comparison of ICG and ThS fluorescence intensity in the area of myocardial infarction

ICG and ThS fluorescence intensity was compared in the intramural layer of the myocardium of apical and middle cross sections of the heart (Fig. 6(a), 6(b)), in the central sectors of which (S2 and S3) the MVO zone was often observed. To compare the fluorescence intensity of two fluorophores, we used the parameter “contrast”, which reflects the ratio between the fluorescence in the examined (injured) myocardial zone and the reference zone (nonischemic), calculated by the formula after subtraction of the values of background fluorescence (see 2.5). The contrast value is directly proportional to the fluorescence intensity, and its negative values occur when the fluorescence intensity in ROI decreases below the fluorescence intensity of the myocardial reference zones. Comparison of the contrast between ICG and ThS fluorescence in all sectors showed a significant difference between the fluorophores, with ICG fluorescence contrast being higher than ThS fluorescence contrast in the risk zone (Fig. 6).

 

Fig. 6. Comparison of contrast expression between sectors of the intramural layer of the left ventricular wall in the risk zone and the remote zone (interventricular septum) after 2 h of reperfusion and after 24 h. BZ-1 and BZ-2 are border sectors of the risk zone, and S1, S2, S3, and S4 are inner sectors. *,** – statistically significant difference between the contrast of the inner sectors and the next border sector of the intramural layer.

Download Full Size | PDF

Negative contrast values were observed in the negative ThS fluorescence zone corresponding to the MVO zone. This was observed at both 2 h (Fig. 6(a), 6(b)) and 24 h of reperfusion (Fig. 6(c), 6(d)). Contrary to the central sectors, contrast was positive in the border sectors. Significant differences in ThS fluorescence in the two sections after 2 h of reperfusion were obtained between the contrasts of the border and central sectors, with the negative ThS fluorescence contrast being more pronounced in the middle section than in the apical section. This pattern of ThS fluorescence persisted after 24 h. Contrary to ThS fluorescence, the intensity of ICG fluorescence in the risk zone was higher than in the reference zone in all sectors, including the central sectors; therefore, the contrast in the central sectors of the intramural layer was mostly positive at 2 h of reperfusion and at 24 h. The distribution of zones with high and low intensity of ICG fluorescence corresponded to the character of ThS fluorescence, i.e., ICG fluorescence intensity was high in the border sectors and relatively low in the central sectors. This character of ICG fluorescence intensity distribution allows us to delineate the no-reflow zone and measure its area.

3.6 Comparison of ICG fluorescence intensity with severity of blood stasis in the area of myocardial necrosis

Marked red blood cell stasis was found in ICG and ThS-negative areas. The magnitude of the negative ICG fluorescence contrast depended on the severity of blood stasis in the infarct zone after 30 min of myocardial ischemia and 2 h and 24 h of reperfusion (Fig. 7). We quantified the severity of red blood cell stasis in the intramural layers of rat hearts after 2 h of reperfusion by counting the area occupied by erythrocytes in Mallory-stained myocardial sections and found an inverse relationship between the value of contrast and the severity of blood stasis. The median percentage of area occupied by erythrocytes in the intramural sectors (S1-S4) of apical and middle sections was 3.4 [2.3–8.7]. Figure 7(a) shows the relation between blood stasis and a delayed flow impairment: the higher the percentage of area occupied by erythrocytes per unit section area, the lower ICG fluorescence contrast value in this myocardial zone. This observation was also evident in 24 h reperfusion experiments (Fig. 7(b)).

 

Fig. 7. Dependence of negative contrast values on the severity of blood stasis in the area of myocardial infarction. (a) Two representative images of cross-sections of rat hearts of group I-30’/R-2 h, which differs in value of negative contrast, to show relation between ICG fluorescence contrast and severity of blood stasis (BS). (b) A representative cross-section of a rat heart from group I-30’/R-24 h, in which marked negative ICG fluorescence contrast corresponds to severe blood stasis. Scale bars are 1 mm for all images of rat whole-heart cross section, 50 µm for images ×20.

Download Full Size | PDF

4. Discussion

We report for the first time the benefits of usage of ICG for estimating the size of the no-reflow zone in myocardial ischemia/reperfusion injury in a rat model. We compared ICG with frequently used ThS by dual-fluorescence staining of one heart that applies direct macroscopic examination using a fluorescence camera. Additionally, risk zone and infarct size were determined by blue dye and triphenyltetrazolium chloride (four stains in one heart). Comparative analysis of macroscopic images (magnification ×1) of ICG and ThS fluorescence intensity using a grid dividing the IRI zone into cells (squares) showed that ICG and ThS fluorescence have a similar character of fluorescence intensity distribution, indicating similar mechanisms of fluorophore accumulation in the IRI zone. This is supported by correlations in the size of the no-reflow areas. In contrast to ThS, ICG-stained myocardial sections can be stored at room temperature, and ICG fluorescence intensity does not decrease when sections are incubated in TTC solution. This property of ICG allows us to use it instead of ThS to measure the no-reflow zone and to compare areas of different fluorescence intensities with areas of necrosis, blood stasis and other histological images of the same heart.

The ability of fluorophores to bind to plasma proteins, including albumin, in the blood flow allows their use as a marker of increased vascular permeability, since intact vascular endothelium is “impermeable” to albumin [17,18]. Extravasation of proteins in inflammation-induced tissue edema occurs by paracellular transport [17] and appears to be the primary mechanism for tissue retention of any dye that has been bound to albumin in the blood flow or preconjugated prior to intravenous administration (enhanced permeability and retention effect). For example, Evans blue used in the Miles test [18,19] is carried by albumin in the blood flow. FITC (FITC-albumin) covalently bound to albumin is used for the same purpose [20]. ICG has been used as a marker of increased vascular permeability in tissue injury and inflammation in several experimental [13,21–23] and clinical studies [24–26]. ICG is transported in plasma with high and low density lipoproteins and albumin, allowing it to be used for fluorescence-based localization of areas of increased vascular permeability and its quantification [24–26]. Apparently, ThS is able to penetrate the damaged myocardium by the same mechanism as ICG. This assumption can be made based on the coincidence of ThS fluorescence peaks with ICG fluorescence peaks at the edges of the no-reflow zone. Another thioflavin, thioflavin T, is known to bind to albumin [27]. The structure of the ThS molecule contains an acidic residue $\mathrm{SO}_3^{-}$ [28], the presence of which suggests the ability of ThS to form ionic bonds with plasma proteins.

The ability of ICG to persist in tissues with increased vascular permeability allows us to simultaneously assess both the borders of no-reflow and the degree of increased vascular permeability in the IRI zone in a single image. In ThS, this property is evident when its fluorescence intensity is “quenched” after incubation in TTC solution. It should be noted that the interval from the time of intravenous injection to the time of heart excision was different for the two fluorophores. For ICG, this interval was 30 min to obtain almost maximal reduction of its background fluorescence in the intact myocardium, and for ThS, this interval was 15 sec before the onset of the microcirculatory phase and staining of the whole heart. There is a direct dependence of the degree of accumulation of the substance in tissues on the duration of its circulation in the blood flow [29]. The longer interval of circulation time in blood and the absence of ICG retention in the reference zone can explain the higher ICG fluorescence contrast throughout the zone of IRI.

With the use of ICG, injured areas of myocardium that perfused by blood appear bright and areas not perfused appear dark (ICG-negative), especially in the central sectors and the intramural layer, which is a diagnostic sign of no-reflow and is explained by the presence of MVO. The complex mechanism of MVO has been well studied [3–5] and includes the gradual appearance of blood stasis in the no-reflow zone, which is visible in native myocardial sections and in myocardial staining with Mallory trichrome or hematoxylin-eosin [30]. In the present study, we used for the first time a technique that allows us to observe the dynamics of no-reflow development due to early ICG and late ThS injection with simultaneous registration of the fluorescence of two fluorophores. Using this technique, we confirmed the literature data on the absence of visible MVO in the first minutes of reperfusion after 30 min of ischemia at the beginning of reperfusion [2,10] and showed the possibilities of dual fluorescence staining of the heart with ICG and ThS in comparing the early manifestations of increased vascular permeability with the late manifestations of no-reflow. Figure 3 shows that during early ICG administration (at the beginning of reperfusion), bright ICG fluorescence was observed in the central sectors, indicating increased vascular permeability and absence of blood stasis. After 2 and 24 hours, a decrease in ICG and ThS fluorescence intensity was observed in the intramural layer. Our data suggest that blood stasis is the main cause of negative contrast ICG fluorescence after 30 min of ischemia in rats. It is known that with increasing ischemia in rats, its consequences appear earlier and in the first minutes of reperfusion there are zones where blood flow is practically not restored, so stasis cannot develop in such zones due to the absence of blood flow during the entire period of reperfusion [10]. A similar observation has been made in dogs after 90 min of ischemia, in which “true” or “immediate” no-reflow developed in the central zones of myocardial ischemia that had no collateral blood flow during coronary artery occlusion [1,2]. Consequently, the presence of blood stasis in the IRI zone is a sign of temporary restoration of blood flow after relatively brief ischemia with gradual development of MVO.

Previously, using the 1 h myocardial ischemia mouse model with double histochemical fluorescence staining with Evans blue and ThS, it was shown that increased vascular permeability in the no-reflow zone persists for up to 3 days [31]. The authors made the interesting observation that 1 day after 1 h of ischemia, intravenously injected Evans blue accumulated in the center of the no-reflow zone (intramural layer). Of note, Evans blue was administered intravenously 3 h before the mice were euthanized, allowing the dye to accumulate in significant amounts, indicating that little blood or plasma flow remains in the MVO zone. Our data on ICG accumulation in the no-reflow zone after 30 min of ischemia complement the authors’ findings by showing the presence of increased vascular permeability at the border of the no-reflow zone and in the no-reflow zone itself at 24 h, as detected by ThS. The penetration of ICG into the no-reflow zone is evident when analyzing the contrast of ICG and ThS fluorescence, which shows the presence of positive contrast with ICG in sectors where ThS has negative contrast. However, in the case of zones of maximum blood stasis, they correspond to zones of negative contrast of ICG fluorescence, indicating extremely low residual blood or plasma flow in the no-reflow zone.

Contrary to acute experiments with 2 h reperfusion, the intensity of ICG accumulation decreases after 24 h of reperfusion, which can be explained by a gradual decrease in the intensity of increased vascular permeability. These data do not agree with those obtained using Evans blue in mice [31], where the authors showed an increase in vascular permeability at 24 hours of reperfusion after 60 min of ischemia. These differences may be explained by differences in dyes used, duration of ischemia, methodological approaches, and species differences. Data from the present study and data from Gao et al. (2017) confirm the presence of increased vascular permeability at 24 h [31], which allowed us to visualize no-reflow at 24 h using ICG. Sufficient contrast was observed between the boundaries of the no-reflow zone, the MVO zone, and the reference zone, allowing us to clearly distinguish the boundaries of the no-reflow zone on the cardiac section as well as the IRI zone within healthy tissue on the epicardial surface of the heart. Previously, in the same in vivo model, in acute experiments on rats, we demonstrated the possibility of intraoperative IRI visualization from the epicardial surface of the heart [13]. The phenomenon of ICG fluorescence in the IRI area detected in the present study after 24 h of reperfusion can be used for intraoperative visualization of ischemia-reperfusion myocardial damage that occurred one day ago.

5. Conclusion

Thus, ICG can be used for simultaneous visualization of no-reflow and assessment of the severity of the processes involved in its formation: increased vascular permeability and MVO. At 30 min after a single intravenous injection of ICG at a dose of 1 mg/kg, the no-reflow zones become visible in NIR light due to absence of ICG retention and visualized microscopically on myocardial sections. In the surrounding myocardium, where blood flow is preserved, ICG accumulation occurs due to increased vascular permeability, creating a stripe of bright ICG fluorescence (ICG-positive zones) around of the no-reflow zone. Contrary to ThS, ICG-stained myocardial sections can be stored at room temperature and ICG fluorescence intensity is not reduced by incubation of the sections in TTC solution. This property of ICG allows us to use it instead of ThS to measure the no-reflow zone and to compare areas of ICG fluorescence with areas of necrosis, blood stasis, and different histological images of the same heart. Double fluorescence staining of ICG and ThS with early ICG injection and late ThS injection makes it possible to observe the dynamics of the transition from processes of increased vascular permeability with ICG to processes of MVO formation with ThS.