An Observational Study of Hemostatic Profile during Different Stages of Liver Transplant Surgery Using Laboratory-Based Tests and Thromboelastography.

Background: Liver produces most of the blood coagulation factors, so it is not surprising to see a deranged coagulation profile in patients receiving liver transplants. Besides standard laboratory methods to evaluate coagulation profile, point-of-care assays are being used regularly since their results are rapidly available. However, sparse information is available on the comparability of point-of-care coagulation assays with laboratory coagulation assays in this special setting. In this study, our aim is to observe the changing hemostatic profile during different stages of liver transplant surgery using laboratory-based tests and thromboelastography (TEG).
Methods: Fifty patients undergoing living donor liver transplantation surgery were selected. Coagulation tests (prothrombin time [PT], activated partial thromboplastin time [APTT], platelet count, and fibrinogen) and TEG were performed at various intervals during liver transplant surgeries - before induction of anesthesia, 2 h into dissection phase, 30 min into anhepatic phase, 30 min after reperfusion of homograft, postoperative - at closure of surgery, 12 h postoperative, and 24 h postoperative. Statistical analysis and Pearson correlation were performed between laboratory-based coagulation tests and TEG, and their pattern through various stages of the surgery analyzed.
Results: Platelet count and fibrinogen have a significant positive correlation with TEG in almost all phases of liver transplant. PT and APTT have a positive correlation with TEG until uptake of new liver and predominantly negative correlation after that. However, this correlation is significant only before induction of anesthesia and anhepatic phase. Conclusions: TEG can be used to estimate platelet count and fibrinogen concentrations in all phases but PT and APTT only before induction and anhepatic phase of liver transplant surgery. The decision regarding transfusion of blood products should be based on a combination of the clinical assessment of surgeon and anesthesia personnel combined with results from laboratory and TEG. Copyright:

INTRODUCTION

Liver transplantation, which began as a speculative surgery 20 years ago, is now accepted as the modality of treatment for patients with end-stage liver disease.[1] However, massive blood loss during liver transplantation is a major concern. The liver produces most of the blood coagulation factors, so very low levels of these factors and prolonged prothrombin time (PT) and activated partial thromboplastin time (aPTT) are predictable in many patients receiving liver transplants.[2] In liver cirrhosis, the synthesis of procoagulant factors as well as anticoagulant factors and fibrinolytic proteins is reduced. Similarly, the decreased platelet count may be counterbalanced by increased platelet aggregability. The coagulation system is, therefore, conceived to be rebalanced.[3]

The bleeding complication in liver transplantation is caused by difficult dissection through friable and dilated collateral vessels in the presence of portal hypertension and coagulopathy associated with reduced coagulation factors and increased activation of coagulation and fibrinolysis.[24] This becomes a major perioperative challenge as both bleeding episodes and thrombotic events can be expedited. Hence, monitoring of the coagulation system during liver transplantation becomes indispensable.[56]

Besides standard laboratory methods, point-of-care assays are getting established since their results are rapid.[7891011]

Conventional coagulation tests (CCTs) cannot detect fibrinolysis or give an indication of clot stability, nor can they generally detect hypercoagulability.[1213] Thromboelastography (TEG) can discriminate between different phases of the coagulation system. As such, TEG can be used for identifying clotting abnormalities and guiding the administration of blood products, thereby decreasing intraoperative blood transfusions during orthotopic liver transplant.[14] Less literature is available correlating point-of-care coagulation assays with laboratory coagulation assays in liver transplant surgery to determine the coagulability of blood in such patients and also, whether any of these can be taken as the prime method to decide regarding blood and blood products transfusion during the surgery. The purpose of this study is to observe the changing hemostatic profile during different stages of liver transplant surgery by correlating laboratory-based tests and TEG.

METHODS

Study site

This study was conducted in the Department of Anesthesia, Artemis Hospitals, Gurugram, Haryana.

Study population

Eighteen–sixty years old patients of either sex scheduled to undergo orthotopic live donor liver transplantation under general anesthesia were enrolled in this study.

Study design and clinical trial registry of India registration

It was a prospective, observational study which was registered prospectively with the Clinical Trial Registry of India (CTRI) (CTRI/2018/06/014568).

Sample size

Fifty patients were included in this study.

For tests of association using bivariate correlations, a good correlation between laboratory-based coagulation tests – PT, aPTT, platelet count, fibrinogen at various intervals, and thromboelastometry in liver transplantation – was considered meaningful. To detect at least a good correlation (r = 0.60),[15] it was found that a sample of 50 analyzable subjects would provide 80% power to discover that the correlation is significantly different from there being no correlation at the 0.05 level (two sided).

Formula used

N = [(Zα + Zβ)/C]2+3

Where

The standard normal deviate for α = Zα =1.96

The standard normal deviate for β = Zβ =0.842

C = 0.5 × ln ([1 + r]/[1 − r])

Time frame

July 1, 2018–December 30, 2019.

Inclusion criteria

Recipients of liver transplant surgery

Patients aged 18–60 years.

Exclusion criteria

Clinically significant renal disease – combined hepatorenal transplant

All exclusion criteria for liver transplant surgery like:

Extrahepatic malignancy

Active substance abuse

Active extrahepatic infections.

Methodology

The study was conducted after the institutional ethics committee and scientific committee approval. Written informed consent was obtained from all patients for participation in the study and use of the patient data for research and educational purposes.

Procedure

Coagulation values of patients undergoing liver transplantation between July 1, 2018, and December 30, 2019, were observed for the study.

Induction of anesthesia was performed using propofol, fentanyl and atracurium via 20 G peripheral venous cannula and trachea intubated with appropriate sized cuffed orotracheal tube. Maintainence of anesthesia was done with oxygen, air, inhalational agent, fentanyl, and atracurium infusions.

All patients had one indwelling 8.5-French (4 lumen) catheter for volume infusion in internal jugular vein. In addition, two intravenous cannula (one 20 G and other 18G or 16G cannula) were placed in the peripheral vein. Two 20G arterial cannula were placed in the right and left radial artery each. Left radial artery cannula was used for arterial blood pressure and stroke volume variability (SVV) measurement and right radial artery cannula was used for blood sampling.

Temperature and urine output monitoring was done at regular intervals.

Normal saline solution without heparin was used for continuous flushing of the pressure monitoring catheters to avoid anticoagulant effects. A rapid infusion system (H-1200/H-1000 fluid warmer set up) was used, which was able to deliver warmed, premixed fluid at a rate of up to 950 ml. min − 1.

All blood products were preserved with citrate phosphate and dextrose with adenine-1. Intraoperative volume infusion was guided by systolic blood pressure, heart rate, central venous pressure, and SVV. Arterial blood gas tensions and serum electrolytes, glucose, and ionized calcium levels were monitored as and when indicated.

Coagulation tests (PT, aPTT, platelet count, and fibrinogen) and TEG [images of apparatus used shown in Images 1 and 2] were performed simultaneously at below mentioned stages of the surgery:

Image 1

Thromboelastography analyzer

Image 2

Thromboelastography cuvette, pipette, and kaolin activated vial

Before induction of anesthesia

Two hour into the Pre– anhepatic phase/dissection phase

Thirty minutes into the anhepatic phase

Thirty minutes after the reperfusion of the homograft

Postoperative – at the closure of surgery

Twelve hour postoperative

Twenty-four hour postoperative.

The following variables were measured in TEG: reaction time (R, min); maximum amplitude (MA, mm); coagulation time (R + k, min); and clot formation rate (α°).

Results of both the methods were used as adjunct to clinical judgement for deciding about blood and blood product transfusion.

Statistical methods

The continuous data were shown as mean +/‒standard deviation and categorical data were represented as absolute numbers and percentages. For continuous data, the Kolmogorov–Smirnov tests were performed to assess normality and where appropriate, the data were analyzed with required statistical tests and descriptive statistics. Parametric data were analyzed with student's t-test/Z-test. Nonparametric data were analyzed with Kruskal–Wallis test and further paired comparisons were done using Mann–Whitney U-test. Nominal categorical data between the groups were compared using the Chi-square test or Fisher's exact test as appropriate and Pearson correlation coefficient was used to observe the linear relationship. All major data analysis packages as well as spreadsheets, such as Microsoft Excel, were used as per requirement. For all statistical tests, P < 0.05 was taken to indicate a significant difference.

RESULTS

Fifty patients aged 18–60 years undergoing orthotopic live donor liver transplant surgery were included in this study. TEG and conventional laboratory tests (PT, APTT, platelet count, and fibrinogen) values were observed at specified intervals in each patient. The results were evaluated and correlated statistically. The following were the profiles of patients studied:

Age

The mean age of the patients included was 43.70 years (range 26–60 years).

Body mass index and duration of surgery

The mean body mass index of patients included in the study was 26.53 (Range 20.4–38.7).

The mean duration of anesthesia exposure time (from wheeling the patient into the odds ratio to shifting to intensive care unit postoperatively) was 12.75 h (Range 9–18 h).

Diagnosis

Out of 50 patients, 40% of them (20 patients) came with a diagnosis of hepatitis B virus (HBV), whereas 30% (15 patients) came with HCV-Hepatocellular carcinoma (HCC). One patient had HBV–HCC. Five patients (10%) came with a diagnosis of alcoholic steatohepatitis [Table 1].

Table 1

Distribution of diagnosis of patients

Diagnosis	n (%)
HBV	20 (40)
HCV and HCC	15 (30)
Alcoholic	5 (10)
HBV-HDV	3 (6)
HBV-HCC	3 (6)
CLD	2 (4)
Others	2 (4)
Total	50 (100)

HCC=Hepatocellular carcinoma, HDV=Hepatitis D virus, HBV=Hepatitis B virus, HCV=Hepatitis C virus, CLD=Chronic liver disease

We analyzed the results of 350 sets of both laboratory and TEG values obtained during orthotopic liver transplantations.

Correlation coefficient (r) had values −1 to +1. Zero means no correlation, >0 to +1 is positive correlation, and − 1 to <0 is negative correlation. The correlation was considered strong for r ≥ 0.6, moderate for 0.6> r ≥0.3, and weak for 0.3> r ≥0 [Table 2].

Table 2

Correlation coefficient values (R) between conventional coagulation tests and thromboelastography parameters in various phases of liver transplant surgery

	Before induction	2 h into the preanhepatic phase/dissection phase	30 min into the anhepatic phase	30 min after the reperfusion of the homograft	Postoperative - at the closure of surgery	12 h postoperative	24 h postoperative
Correlation of PT
R (min)	0.694	0.019	0.718	−0.019	−0.243	−0.078	−0.065
MA (mm)	−0.144	−0.428	−0.475	−0.280	−0.037	−0.098	−0.059
R + K (min)	0.471	−0.020	0.734	−0.007	−0.140	−0.138	0.189
α	0.169	−0.289	−0.690	−0.063	−0.013	−0.125	−0.157
Correlation of aPTT
R (min)	0.514	0.077	0.609	−0.040	0.046	−0.030	0.061
MA (mm)	−0.140	−0.302	−0.586	−0.196	−0.528	−0.426	−0.371
R + K (min)	0.290	−0.023	0.708	0.003	0.097	0.127	0.259
α	0.147	−0.161	−0.633	0.144	−0.247	−0.135	−0.390
Correlation of PLT count
R (min)	−0.193	−0.343	−0.127	−0.262	−0.537	−0.161	−0.028
MA (mm)	0.671	0.506	0.502	0.643	0.718	0.441	0.552
R + K (min)	−0.376	−0.514	−0.231	−0.394	−0.588	−0.236	−0.099
α	0.417	0.667	0.183	0.316	0.640	0.385	0.315
Correlation of fibrinogen
R (min)	−0.329	−0.397	0.007	0.005	−0.274	−0.111	−0.457
MA (mm)	0.571	0.565	0.405	0.576	0.460	0.491	0.500
R + K (min)	−0.317	−0.370	−0.310	−0.144	−0.355	−0.135	−0.663
α	0.307	0.444	0.398	0.314	0.420	0.228	0.597

PT=Prothrombin time, aPTT=Activated partial thromboplastin time, PLT=Platelet, MA=Maximum amplitude

MA and alpha angle parameters of TEG were found to be most positively and moderately correlating with both platelet count [Figures 1 and 2] and fibrinogen values [Figures 3 and 4], and these correlations were stronger than those of other parameters.

Figure 1

Bland–Altman plots between platelet count and maximum amplitude during various stages of liver transplant surgery. Data are correlation coefficient values (r). (a) Comprehensive analysis. (b) Before induction of anesthesia. (c) Preanhepatic phase. (d) Thirty minutes into the anhepatic phase. (e) Thirty minutes after reperfusion. (f) Immediate postoperative period. (g) Twelve hour postoperative. (h) Twenty-four hour postoperative

Figure 2

Bland–Altman plots between platelet count and alpha angle during various stages of liver transplant surgery. Data are correlation coefficient values (r). (a) Comprehensive analysis. (b) Before induction of anesthesia. (c) Preanhepatic phase. (d) Thirty minutes into the anhepatic phase. (e) Thirty minutes after reperfusion. (f) Immediate postoperative period. (g) Twelve hour postoperative. (h) Twelve hour postoperative

Figure 3

Bland–Altman plots between Fibrinogen and maximum amplitude during various stages of liver transplant surgery. Data are correlation coefficient values (r). (a) Comprehensive analysis. (b) Before induction of anesthesia. (c) Preanhepatic phase. (d) Thirty minutes into the anhepatic phase. (e) Thirty minutes after reperfusion. (f) Immediate postoperative period. (g) Twelve hour postoperative. (h) Twelve hour postoperative

Figure 4

Bland–Altman plots between Fibrinogen and alpha angle during various stages of liver transplant surgery. Data are correlation coefficient values (r). (a) Comprehensive analysis. (b) Before induction of anesthesia. (c) Preanhepatic phase. (d) Thirty minutes into the anhepatic phase. (e) Thirty minutes after reperfusion. (f) Immediate postoperative period. (g) Twelve hour postoperative. (h) Twelve hour postoperative

We also found that PT and aPTT had poor correlation with TEG parameters (R, R + k) [Figures 5-8].

Figure 5

Bland–Altman plots between prothrombin time and R during various stages of liver transplant surgery. Data are correlation coefficient values (r). (a) Comprehensive analysis. (b) Before induction of anesthesia. (c) Preanhepatic phase. (d) Thirty minutes into the anhepatic phase. (e) Thirty minutes after reperfusion. (f) Immediate postoperative period. (g) Twelve hour postoperative. (h) Twenty-four hour postoperative

Figure 8

Bland–Altman plots between activated partial thromboplastin time and R + k during various stages of liver transplant surgery. Data are correlation coefficient values (r). (a) Comprehensive analysis. (b) Before induction of anesthesia. (c) Preanhepatic phase. (d) Thirty minutes into the anhepatic phase. (e) Thirty minutes after reperfusion. (f) Immediate postoperative period. (g) Twelve hour postoperative. (h) Twenty-four hour postoperative

Bland–Altman plots between prothrombin time and R + k during various stages of liver transplant surgery. Data are correlation coefficient values (r). (a) Comprehensive analysis. (b) Before induction of anesthesia. (c) Preanhepatic phase. (d) Thirty minutes into the anhepatic phase. (e) Thirty minutes after reperfusion. (f) Immediate postoperative period. (g) Twelve hour postoperative. (h) Twenty-four hour postoperative

Bland–Altman plots between activated partial thromboplastin time and R during various stages of liver transplant surgery. Data are correlation coefficient values (r). (a) Comprehensive analysis. (b) Before induction of anesthesia. (c) Preanhepatic phase. (d) Thirty minutes into the anhepatic phase. (e) Thirty minutes after reperfusion. (f) Immediate postoperative period. (g) Twelve hour postoperative. (h) Twenty-four hour postoperative

DISCUSSION

In this study, the correlations of CCT and TEG were assessed prospectively in 50 cases of adult orthotopic liver transplant.

MA and alpha angle parameters of TEG were found to be most positively and moderately correlating with both platelet count and fibrinogen values, and these correlations were stronger than those of other parameters.

We also found that PT and aPTT had a poor correlation with TEG parameters (R, R + k).

Patients with end-stage chronic liver disease, i.e., cirrhosis patients, exhibit profound changes in hemostatic profile. They show a decrease in coagulation factors as well as a decrease in inhibitors and regulatory proteins. This leads to a rebalanced state of hemostasis. They also show a decrease in the number of platelets as well as platelet dysfunction. Various studies have demonstrated an increase in the platelet activity, similar to the Von Willibrand factor, derived from the endothelium. This has been shown to compensate for the decreasing platelet count at the end-stage liver disease patients. MA of TEG is reflective of clot strength, which results from interactions between platelets and fibrinogen. This result corresponds with the research of Herbstreit et al., which demonstrated a good correlation between platelet count, fibrinogen, and TEG parameters.[15]

Before induction of anesthesia

Reduced coagulation factor activities can commonly be observed in end-stage liver disease because the liver synthesizes nearly all coagulation factors including factors II, V, VII, VIII, IX, X, XI, XII, and XIII. There is also a documented decrease in the inhibitors and regulatory proteins (AT III, alpha 1-antitrypsin, C1 inhibitor, and alpha 2-macroglobulin). Hence, there is a hemostatic rebalance in these patients. This explains moderate-to-strong positive correlation between PT, aPTT, and TEG parameters in the preoperative period. This is similar to the results of Kang et al. who demonstrated a good correlation between aPTT and TEG preoperatively.[16]

Preanhepatic and anhepatic phases

In the preanhepatic and anhepatic phases, levels of all coagulation factors and platelet count decrease, as preexisting coagulopathy is compounded by surgical bleeding.

Furthermore, fibrinolysis in patients with severe hepatocellular disease or in those requiring massive transfusion and activation of coagulation reduces procoagulant level even with an infusion of blood products rich in coagulation factors. Hypothermia and ionized hypocalcemia also impair coagulation.[17]

In our study, the fibrinogen values gradually decreased from 144 before induction to 107 in preanhepatic phase and further 86 in anhepatic phase.

The platelet values almost remained constant in these phases, as in the preinduction phase.

The PT and aPTT values increased grossly in most cases whereas TEG parameters did not show much change.

Hence, we witnessed no or weak correlation between PT, aPTT, and TEG.

The increasing fibrinolysis and platelet dysfunction continued, and hence, a moderate positive correlation was seen between platelet, fibrinogen, and TEG parameters.

Very few patients received massive transfusions in our study (only 5 out of 50) and these transfusion decisions were based on clinical parameters, laboratory values, and TEG parameters taken into consideration altogether. Most of the transfusions were observed in the pre anhepatic phase, which probably resulted in improvement in coagulation factors by the time anhepatic phase started. This explains the strong positive correlation between PT, aPTT and TEG parameters which is in agreement with the correlation obtained before induction of anesthesia. However, the fibrinolysis and platelet dysfunction continued to proceed in anhepatic phase, similar to that in preanhepatic phase/dissection phase. Thus, the correlation remained moderately positive between platelet, fibrinogen, and TEG parameters.

Postreperfusion

A severe coagulopathy, a component of the postreperfusion syndrome, occurs on reperfusion of the grafted liver; more specifically, a decrease in coagulation factor levels, a sudden increase in tissue plasminogen activator (tPA), thrombocytopenia, and fibrinolysis, and a moderate increase in fibrin degradation products (FDP) and thrombin antithrombin (TAT) levels are seen.[4]

In our study, we observed similar changes; PT and aPTT values increased and fibrinogen levels further decreased. TEG parameters did not show any major changes. This resulted in a poor correlation among PT, aPTT, and TEG parameters.

However, an increase in platelet count was noticed in most patients. A strong positive correlation persisted between platelet count and TEG, mostly dependent on the functionality of platelets as well as the number.

Despite a decrease in fibrinogen, a moderate correlation persisted between fibrinogen and TEG. This is probably due to the increased FDP and fibrin monomers in this phase.

The cause of the postreperfusion coagulopathy is multifactorial. The release of endogenous heparin from the donor liver results in moderate-to-severe heparin effect in approximately 30% of patients and lasts for 60–120 min. Fibrinolysis occurs in approximately 80% of patients, although severe fibrinolysis occurs in approximately 40% of patients.[18] It is caused by the release of tPA from the grafted liver, congested viscera, and lower extremities, together with a reduction of plasminogen activator inhibitor (PAI) activity. Other causes include contact activation of fibrinolysis and activation of protein C or urokinase-type plasminogen activator.[19]

Excessive activation of coagulation also occurs during this phase and is evidenced by increased levels of TAT complex, FDP, and fibrin monomers and decreased levels of AT-III and PAI. This appears to be caused by tPA-induced activation of platelet aggregation or by extracellular release of lysosomal proteinases from macrophages (Cathepsin B) and granulocytes (elastase).[20] Excessive activation of coagulation, which may result in consumption coagulopathy and thromboembolism, is receiving more attention by clinicians.

De Wolf et al.[21] reported six patients who developed clinical pulmonary embolism on reperfusion, and Suriani et al.[22] reported a significant echocardiographic pulmonary thromboembolism in 59% of patients without venovenous bypass and 11% of patients with venovenous bypass. This excessive activation of coagulation may be caused by anoxic damage of endothelium, release of lysosomal proteinases from activated macrophages and platelets, or low AT-III levels as postulated by Kang et al.[17]

Postoperative period

Flute et al.[23] performed a study in 1969 and documented the hematological changes in the postoperative period using CCTs. They found that within the first 48 h, the one-stage coagulation tests had returned to normal, platelet count, fibrinogen, and prothrombin had begun to rise.

Thrombocytopenia occurs on reperfusion and postoperatively with a transhepatic decrease in platelet count of as much as 55%. Although platelet function parallels platelet count, platelet function can be impaired by the loss of granulation and decreased platelet aggregation.[24]

Fibrinolysis and the heparin effect dissipate gradually, and the procoagulant level increases toward baseline as the grafted liver begins to function. However, bleeding or oozing may persist in some patients.

Our study showed a drop in fibrinogen values at closure; however, it was on an improving trend in the first 24 h postoperative.

Platelet count showed a continuous decreasing trend.

PT and aPTT did not show much change. Furthermore, TEG parameters did not show much difference in the first 24 h postoperative.

A moderate positive correlation was observed between platelet, fibrinogen, and TEG parameters.

However, a weak negative correlation was found between PT, aPTT, and TEG parameters after the uptake of new liver.

CONCLUSIONS

Our results show that PT, aPTT, and TEG have a positive correlation UPTILthe uptake of new liver and predominantly negative correlation after that. However, this correlation is good and significant only in anhepatic phase and before induction.

Platelet count and fibrinogen concentration are important determinants for the mechanical properties of a clot. TEG is capable to measure the MA directly and correlation with platelet count and fibrinogen concentration was found to be significantly positive and good in almost all phases of live donor liver transplant surgery.

The above findings suggest specific uses of point-of-care testing. TEG (ma and alpha angle ) is a good guide of platelet and fibrinogen levels.TEG cannot, however, be recommended as a substitute for laboratory PT and aPTT.

In conclusion, our results suggest that though TEG more closely correlates with the coagulation profile of the patient, it cannot replace the CCTs in total. We could not establish either of them as a prime method to decide blood and blood product transfusion.

We suggest that the decision regarding transfusion of blood products should be based on a combination of the clinical assessment of surgeon and anesthesia personnel combined with results from laboratory and TEG.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.