Optical assessment of alterations of microrheologic and microcirculation parameters in cardiovascular diseases

: In this work, we compare the blood aggregation parameters measured in vitro by laser aggregometry and optical trapping techniques in blood samples with the parameters of blood rheology measured in vivo by digital capillaroscopy in the nail bed capillaries of patients suffering from the hypertension and coronary heart disease. We show that the alterations of the parameters measured in vivo and in vitro for patients with different stages of these diseases are interrelated. Good agreement between the results obtained with different techniques, and their applicability for the diagnostics of abnormalities of rheological properties of blood are demonstrated.

Much of the research in clinical and experimental cardiology focuses on the problems associated with myocardial contractility and perfusion deterioration due to lesions of the coronary arteries. However, quite recently, an increasing interest of researchers was directed to rheologic and microcirculatory disorders in heart diseases. This is due to the growing awareness that the delivery of oxygen and nutrients to the heart tissue mainly depends on the coronary microcirculation and rheological properties of blood [8][9][10].
The study of RBC aggregation parameters is also important for modern cardiology because in the last decade, high-performance oral anticoagulants have appeared in the arsenal of cardiological drugs that have a powerful effect on the coagulation properties of blood [11]. Along with antiplatelet agents, these drugs, in addition to their benefits, have a high risk of inducing bleeding. Administration of the antiplatelet and anticoagulants agents into blood is based on general assumptions about the average doses of the drug, which often do not fully comply with the requirements of the personalized medicine. As a rule, monitoring the effectiveness of the treatment carried out is not sufficient, especially when using the so-called New Oral Anticoagulants (NOAC), for which the necessary laboratory tests have not been developed yet. However, overdose and uncontrolled intake of anticoagulants and antiplatelet agents can cause serious hemorrhagic complications [12,13]. On the other hand, insufficiently effective antiplatelet therapy leads to complications of different kinds, associated with the risk of increased thrombosis.
One of the techniques for in vivo studying the rheological properties of blood is digital capillaroscopy -an optical in vivo microscopy technique allowing for direct visualization of the superficial skin microvessels and tissue surrounding them. Original image-processing software provides non-invasive quantitative assessment of static and dynamic parameters of microcirculation [14].
It also allows for estimating the density of the capillary network, the degree of perivascular tissues oedema, the diameters of the capillary sections. In addition, this technique provides an opportunity to assess the capillary blood velocity, as well as visualize the presence of blood aggregates. The RBC aggregates formation usually occurs against the background of the capillary blood flow velocity reduction and stasis [15,16].
To measure the aggregation parameters on a large number of RBCs the diffuse light scattering method based on registration and subsequent analysis of intensity of light scattered from the layer of whole blood is used. This method is implemented in several commercial systems: laser-assisted optical rotational cell analyzer LORCA (Mechatronics, Netherlands) [17], laser aggregometer and red blood cell deformometer LADE (Rheomedlab, Russia) [18,19], and fully automatic erythrocyte aggregometer FAEA (Myrenne, Germany) [20]. During the last two decades, a new device RheoScan (RheoMeditech, Seoul, Korea) for studying RBC aggregation in whole blood samples in vitro was developed [21]. Another optical approach for studying heart failure severity in humans suggested was based on imaging of the viable epidermis around the nailfold capillaries [22]. To measure RBC aggregation on the level of individual cells laser tweezers (laser traps) are used [23].
The aim of this study is to analyze the relationships of the results obtained in vivo with capillaroscopy and in vitro with RBC aggregometry and laser tweezers in order to assess the pathologic alterations of the microrheologic properties of blood in patients with such widespread cardiovascular diseases as arterial hypertension (AH) and established coronary heart disease (CHD).

Materials and methods
In this work, we used three optical techniques for assessing microrheologic and microcirculatory parameters of blood: laser aggregometry, optical trapping and manipulation of individual red cells, and vital digital capillaroscopy.
In vitro measurements of microrheologic properties of blood that characterize the aggregation of erythrocytes were carried out on large ensembles of cells. Measurements were performed on whole blood samples (macrolevel), and also on the singe cell interaction level (cellular microlevel) in highly diluted suspension of RBCs in blood plasma. Detection of diffuse light scattering from a layer of whole blood (the method of laser aggregometry) allows us to get the parameters of RBC aggregation that characterize the average time of aggregate formation as well as determine the relative amount of aggregated cells. Using the method of optical trapping, we can manipulate individual cells without mechanical contact, measure the forces of RBC interaction during their aggregation and the aggregation time. In vivo visualization of blood microcirculation with digital capillaroscopy technique allows us to determine visually the presence of blood aggregates and stasis in blood flow, and calculate the capillary blood velocity (CBV).
Comparison of the results obtained using these three in vitro and in vivo techniques can clarify the effects of microrheological parameters on capillary blood flow, and also shed light on the possibility of using in vitro measurements to characterize native blood microcirculation.

Laser aggregometry technique
In order to assess the aggregation parameters on whole blood samples we used laser aggregometer RheoScan which utilizes the method of diffuse laser light scattering (laser aggregometry) [19,21].
Disposable microcuvettes that were used during the experiment consist of a small flat reservoir for whole blood with a thickness of 300 μm and diameter of 5 mm. Inside the reservoir there is a thin magnetic stirrer bar that can rotate when an electrical field is present inside the microcuvette thus stirring its contents. Whole blood samples of 8 μl were administered into the reservoir using a micro-doser for each measurement.
The process of measurement begins after the microcuvette is put inside the device. It records the time dependence of the intensity of laser light (wavelength of 633 nm) scattered forward by the blood sample, regarded as the RBC aggregation kinetics (Fig. 1). First, the bar rotates at a high speed, thus creating an external shear stress that destroys all the aggregates that have already appeared in the sample. Then, the rotation is abruptly stopped (time t = 0) and the process of spontaneous aggregation of erythrocytes at rest (without any external shear stress) in our sample begins to take place. Laser beam shines on the blood reservoir and is scattered by the individual cells and the newly forming aggregates. After some time the intensity of light scattered in forward direction increases because of the increase of the average size of the scattering centers due to the formation of aggregates. This process of spontaneous aggregation is accomplished after approximately 2 minutes when the scattered light intensity reaches its maximum value indicating that almost all the cells have aggregated (Fig. 1).
Most of aggregation parameters measured with the techniques based on light scattering are significantly influenced by the hematocrit of the blood samples due to the dependence of scattered light intensity on the volumetric concentration of scattering centers -red blood cells [21]. In this work, all measurements by laser aggregometry were implemented in the RheoScan device and were performed on whole blood samples after normalizing the hematocrit to the value of 40% by autologous blood plasma.   he lasers, der to get the needed intensity gradient in the focal area of the beams (the positions of laser traps are shown with arrows in Fig. 2). Upon passing the lens system the beam is split by the dichroic mirror, the larger part of the intensity being directed to the aperture of the camera lens Olympus (х100, N.A. = 1, water immersion), while the smaller part -to the photodetector to measure its power. The power of each beam can be changed using half-wave plates that are connected to electric motors with small steps of rotation. One beam is always stationary, while the second one can be moved by rotatable mirror. This way we are using 2 areas of trapping: one is stationary, one can move inside our sample. A lens and CMOS-camera are installed vertically and are illuminated by white light from under the sample. The cuvette is placed on a motorized platform. It consists of an object and a cover glass plates with the distance between them equal to 100 μm created by double layer of adhesive tape on the sides. The suspension of erythrocytes is diluted in plasma (1:1000 proportion) and is placed between the glass plates using a micro-doser. After that it is sealed off by Vaseline to prevent unwanted currents and evaporation.
Before any measurements can take place, a calibration procedure must be performed to find one-to-one relation between the trapping force F trap and the beam power. It was made by matching F trap with the force of viscous friction acting on the trapped cell when the platform was moved with a controlled speed relative to the trap. Calibration procedure is described in details in [24]. The maximum trapping force achieved was 15 pN with the laser power being 30 mW in the focal area. The aggregation force F A is defined as the trapping force F trap which is minimally sufficient for preventing the two cells from overlapping during the aggregation process.
The procedure of measuring F A and aggregation time consisted of several steps (Fig. 3). During the first step two RBCs were trapped with the beams and were aligned parallel to each other by moving the platform. So, one cell was trapped by the stationary beam, the other one by the movable trap. In the second step the erythrocytes were moved closer together until their interaction area reaches approximately 40% of the area of each cell. In the final step the movable beam power was slowly decreased until the trapping force (black arrow in Fig. 3) was no longer sufficient to resist the spontaneous aggregation force (white arrow) between the two cells. After that the cell can escape the trap and complete the aggregation (due to F trap ≤ F А ). At this moment the laser power is recorded and F A is calculated. Also, the aggregation time needed for full overlapping of the cells is measured.
For each sample the measurements of F A and aggregation time were carried out on at least 15 pairs of cells. The values were averaged to get the final results. All experiments with optical traps were conducted at room temperature in a temperature-controlled room.

Vital digital capillaroscopy (VDC)
Following a minimum 15 minutes seated rest, the in vivo microvascular measurements were conducted between 9 and 11 am in a quiet temperature-controlled room (the temperature was maintained between 22 and 23.5 °C), with the subject in the seated position and the left hand at the heart level. All participants were required to refrain from smoking and caffeinated drinks one day before the examination. Capillary blood flow velocity (CBV) was measured in the eponychium of the fourth or third finger of the left hand.
Skin temperature was measured at the dorsal middle phalangeal area of the tested finger of the left hand by medical precision thermometry; the mean skin temperature was 33.2 ± 1.7 °C with no significant differences in the studied group.
Nail fold capillaries were visualized using the digital capillaroscope Kapillaroskan-1 (AET, Russia) equipped with high speed CCD-camera (1/3'' monochrome progressive scan IT CCD sensor, resolution 640 x 480 px, frame rate 200 fps full frame), TM-6740GE (JAI, Japan). The nail bed illumination was achieved with a LED-based illuminating system. Two ranges of the total magnification (125x) and (400x) were used to visualize the nail bed capillaries. 125x magnification was used for obtaining the panoramic images of the capillaries, while more detailed imaging of single capillaries was performed at 400x total magnification and included the measurement of static parameters (capillary length and diameters in different parts) and CBV in different parts of the capillary. We also recorded the presence or absence of blood aggregates in the capillaries (Fig. 4) from the analysis of obtained digital microscopy images and video movies. For determining of the CBV after recording the video fragment the program stabilizes the dynamic images of the capillaries and then processes the images in the specified region of interest in the offline regime. The tracks of specific spots (RBCs) differ in the level of light intensity. The program marks them and then recognizes in the next frame. The program determines the average velocity along the axis of the capillary over 5 seconds long time intervals (500 frames). The CBV is estimated in 6 capillaries and the results are averaged. We estimated CBV only in the capillaries of the first line, where the capillaries are located within one layer. Thus, the obtained values of CBV are not affected by the movement of blood in the vessels lying above and below the investigated capillary. Detailed procedure of CVB measurements is described in [25]. Usually, at rest in healthy people not taking caffeinecontaining substances and drinks on the eve of the study, the average CBV varies in the range from 800 to 1500 µm/s.
In this paper, we make a decision about the two states of the presence or absence of RBC aggregates in the capillaries based on visual processing of the nail bed images and videos obtained with VDC technique in vivo. The criteria of distinguishing RBC aggregates presence in the capillary blood flow can be formulated as follows: 1. Blood aggregates are clusters of blood cells that form autonomous conglomerates separated by plasma gaps.
Thereby, the determination of the presence of RBC aggregates in the capillary blood flow was performed by visual assessment of nail bed capillaries images and videos according to the scale: the absence of aggregates in the capillaries -0, the presence of aggregates in the capillaries -1. In case of visual detection of at least several distinguished clusters of RBC in the flow we considered the state as Aggregates = 1, otherwise the Aggregates state was indicated by 0 (Fig. 4).
In this work, registration the CBV dependence on time was used to reveal the presence of a stasis, as a sign of a pronounced deterioration in the rheological properties of blood. The minimum duration of the blood flow halting in the capillary, which in our study is classified as stasis, is 0.25 sec. Generally, the duration of stasis can reach several seconds.  Figure 5 demonstrates the dependence of CBV on time during 30 seconds of measurements in a patient with CHD. In the center of the graph, one can see a drop in CBV to zero. The duration of this stasis was 1.16 sec. In our study, the presence of a stasis in the capillaries was assessed as "1" if CBV remained about zero more than 0.25 sec in at least one visualized capillary; otherwise the absence of a stasis was evaluated as "0". Accordingly, the higher CBV, the lower probability of the stasis.

Blood samples
All in vitro measurements were carried out using samples of whole human blood and erythrocyte suspensions. In our experiments, we have taken into account the latest recommendations for hemorheoligical laboratories made by an international expert team to standardize hemorheological methods [26]. The samples were kept at 37°C during the experiment and were used only during the first 3 hours after blood draw from the cubital veins of donors on an empty stomach. All blood samples were collected into containers (Vacuette EDTA K3E tube, 1.8 mg/ 1 ml of blood, total volume 4.5 ml) with EDTA as an anti-coagulant agent.
Laser aggregometry measurements was performed with whole blood. A highly diluted erythrocyte suspension was used for measuring the time and force of aggregation by the laser tweezers. The dilution was made with autologous poor platelets plasma with the final erythrocytes concentration of about 0.1%. Plasma was acquired by centrifuging in a following manner: firstly, centrifuging for 10 minutes at 170g, then centrifuging twice for 10 minutes at 3000g.

Experimental groups and statistics
Overall, 88 patients with arterial hypertension were enrolled in the study. They were divided into two groups. First group enrolled only patients with hypertension without cardiovascular complications (n = 48) and second group included patients (n = 40) with established coronary heart disease (CHD). Comparison of clinical backgrounds between hypertensive patients with and without coronary heart disease (CHD) is presented in Table 1.
The average age of all patients was 62.8 ranging from 24 to 87 years, 51.7% of them were males. In the group of hypertensive patients without CHD, females prevailed -28 persons (58.4%). In the group of patients with CHD males prevailed (63.4%). The body mass index (BMI) for the groups was significantly higher than normal value, but not statistically significantly different. The mean age was also higher in the group of patients with CHD. The left ventricle ejection fraction in the group with CHD was statistically lower in comparison with group of hypertensive patients without CHD. Standard Student T-test was used to analyze statistical difference between sets of values measured from compared groups. The difference was considered statistically significant when p < 0.1.
The study was approved by the ethics committees of the Medical Research and Education Center of Lomonosov Moscow State University. All donors were informed on the purpose of the study and agreed to participate in this study.

Results and discussion
Blood samples from every patient were investigated by methods of laser aggregometry and optical trapping. Microrheologic parameters of RBC aggregation (aggregation index, characteristic time of aggregate formation, aggregation force and time) were measured in vitro. Before blood was drawn from the patients all of them underwent the microcirculation measurements in vivo by means of vital digital capillaroscopy (VDC) technique.

Conclusion
In this work, a series of in vitro measurements of RBC aggregation parameters with blood samples obtained from the patients suffering from hypertension and coronary heart disease were performed. Aggregation index and characteristic time of aggregates formation were measured by laser aggregometry technique in whole blood samples comprising large populations of RBCs. In addition, forces and time of aggregation on individual cells level were obtained with an optical trap. The capillary blood velocity, presence or absence of RBC aggregates and stasis in blood flow were assessed in vivo with vital digital capillaroscopy. The relationships between the parameters measured in vitro demonstrate good agreement of the results for the patients distinguished into subgroups in accordance with VDC data obtained in vivo. In particular, impairing of capillary blood flow in case of AH or CHD leads to unconditional deterioration of aggregation parameters of blood cell.
Our results also clearly demonstrate the possibility of using laser aggregometry and optical trapping for estimating the alterations of microrheologic and, consequently, microcirculation parameters. Alterations of RBC aggregation parameters measured in vitro can be used to evaluate the alterations of vital capillary blood flow parameters in human body.