Thromboelastographic description of intraoperative coagulopathy in patients undergoing liver transplant surgery: A single-center observational study

Abstract

BACKGROUND

Patients undergoing liver transplantation often develop multiple hemostatic abnormalities requiring dynamic transfusion and antifibrinolytic support; however, how hemostatic status changes over time during liver transplant surgery is not well defined, limiting clinicians’ ability to anticipate and manage hemostatic instability.

AIM

To describe the distribution of intraoperative coagulation abnormalities across the surgical phases of liver transplantation.

METHODS

We conducted an observational study of adult patients who underwent liver transplantation at Henry Ford Health between January 2017 to April 2022. Intraoperative coagulation data from conventional coagulation tests and 2 thromboelastography (TEG) platforms (TEG 5000 and TEG 6s) were collected. Data were stratified across 4 surgical phases: Pre-anhepatic, anhepatic, post-reperfusion, and end-of-surgery.

RESULTS

Of 364 liver transplant recipients, coagulation abnormalities occurred during all surgical phases, with the highest frequency in the anhepatic and post-reperfusion phases. Platelet dysfunction was the most common abnormality, peaking during post-reperfusion, while hyperfibrinolysis was less frequent and resolved in most patients by the end of surgery. TEG 5000 and TEG 6s showed slightly different result patterns for several parameters, particularly for clot initiation time and fibrinogen activity.

CONCLUSION

TEG was more informative than conventional tests, showing peak coagulation derangements during anhepatic and post-reperfusion phases of liver transplantation, likely driven by platelet dysfunction and warranting increased vigilance for coagulopathy.

Key Words: Liver transplant; Coagulation abnormalities; Hemostatic; Thromboelstography; Coagulation tests; Hyperfibrinolysis; Anti-fibrinolytic; Reperfusion

Core Tip: This study describes intraoperative coagulation abnormalities across different phases of liver transplantation using thromboelastography. Findings highlight that coagulopathy peaks during the anhepatic and post-reperfusion phases, primarily driven by platelet dysfunction, underscoring the importance of targeted transfusion and antifibrinolytic strategies to optimize hemostatic management during these critical stages.

INTRODUCTION

End-stage liver disease and other indications for orthotopic liver transplantation (OLT) are associated with complex dysregulation of the hemostatic system[1], and during transplant surgery, patients can develop profound intraoperative coagulopathy due to trauma, ischemia, surgical bleeding, native liver removal, and liver graft reperfusion[2]. While transfusion support is crucial for managing coagulopathy during OLT surgery, perioperative transfusion is associated with adverse outcomes, including prolonged length of hospital stay, increased risk of thrombosis, and decreased survival[3,4]. Crucially, a careful balancing of transfusion and antifibrinolytic therapy during OLT surgery is often needed to meet the patient’s rapidly changing coagulation profile. However, to maintain this balance successfully, a thorough and real-time understanding of the patient’s intraoperative coagulation status throughout the various phases of OLT is critically necessary.

Although still used clinically, conventional coagulation tests, including platelet counts, partial thromboplastin time, international normalized ratio, and fibrinogen measurements, are incomplete and imperfect measures that provide delayed results and are poor predictors of bleeding risk in patients undergoing OLT[5,6]. Recently, viscoelastic hemostatic diagnostic assays, such as thromboelastography (TEG) and rotational TEG, have become widespread methods for monitoring coagulation status during OLT procedures[7,8]. These assays have several key advantages over conventional coagulation tests, such as performing near real-time and simultaneous evaluation of clotting and fibrinolytic function, which provides a more comprehensive representation of hemostatic balance[7,9]. However, few studies have considered the temporal changes in coagulation status during OLT as measured by TEG, and existing reports have typically not addressed fibrinolytic function or fibrinogen levels[10,11].

Overall, an understanding of coagulopathy and clotting dynamics during OLT is essential to enhance transfusion safety and guide therapeutic interventions. Therefore, we conducted a retrospective study characterizing coagulation derangements in patients undergoing OLT monitored with TEG across 4 surgical phases: Pre-anhepatic, anhepatic, post-reperfusion, and end-of-surgery. Our goal was to describe intraoperative hemostatic changes measured by TEG to provide insights for optimizing transfusion and antifibrinolytic management during OLT.

MATERIALS AND METHODS

Study design and setting

We conducted an observational study of all patients who underwent OLT and were monitored with TEG during transplantation at Henry Ford Health in Detroit (MI, United States). The study was approved by the Institutional Review Board of Henry Ford Health (Approved No. 15528-01) and conformed to both the Declaration of Helsinki and Declaration of Istanbul ethical guidelines. The need for informed consent was waived due to the retrospective nature of the study.

Data and patients

Data were collected from a prospective electronic liver transplant database. We included all adult patients who underwent OLT at Henry Ford Health between January 1, 2017, when TEG was introduced at Henry Ford Health, and April 30, 2022, and had ≥ 1 intraoperative TEG report in the medical record. Patients who were pregnant, were admitted to the intensive care unit (ICU) before surgery, and those who had long-term preoperative anticoagulation therapy, fulminant liver failure, transfusion support, or tranexamic acid 24 hours before surgery were excluded. These patients reflect a different disease state that might have different TEG patterns (overestimation or underestimation of coagulopathy), our intentions were to describe a typical liver transplant patient at our institutions, hence our exclusion criteria.

Variables

This report was developed according to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines[12]. The following demographic and clinical variables were collected from the medical record: Age at OLT, sex, race (Asian, Black, White, and Other), body mass index, blood type, pre-surgical model for end-stage liver disease sodium score, comorbidities (chronic pulmonary disease, diabetes, end-stage renal disease, and hypertension), and cause of liver failure (alcohol-related, hepatitis C, hepatitis B, metabolic dysfunction-associated steatohepatitis, primary sclerosing cholangitis, primary biliary cholangitis, autoimmune hepatitis, hepatocellular carcinoma, cholangiocarcinoma, cryptogenic, and other).

Surgical and postoperative variables were also extracted, including the TEG assay platform used (TEG 5000 or TEG 6s), type of transplant (deceased donor liver transplant or living donor liver transplant), donor type (donation after brain death, donation after circulatory death, or living donor), tranexamic acid use (total, pre-anhepatic, anhepatic, or post-reperfusion), intraoperative blood products (packed red blood cells, cell saver units, fresh frozen plasma, cryoprecipitate, and platelets), intraoperative continuous veno-venous hemofiltration, ICU length of stay, intubation status upon ICU admission, mechanical ventilation duration, ICU transfusion within 72 hours, need for re-operation within 7 days, and 30-day mortality.

The outcomes of interest were hemostatic data across the 4 surgical phases, including conventional coagulation tests (partial thromboplastin time, international normalized ratio, fibrinogen level, and platelet count) and TEG parameters (TEG 5000 and TEG 6s). The 4 surgical phases of OLT were defined as follows: Pre-anhepatic (between incision and inferior vena cava clamp), anhepatic (between inferior vena cava clamp and reperfusion), post-reperfusion (between reperfusion and the end of surgery), and end of surgery. Viscoelastic parameters from TEG 5000 included reaction time (R), kinetics time (K), α-angle, maximum amplitude (MA), and percent clot lysis at 30 minutes post-MA (LY30). Viscoelastic parameters from TEG 6s included citrated kaolin (CK) R, citrated rapid TEG MA (CRT MA), citrated functional fibrinogen (CFF) MA, CFF functional fibrinogen level (FLEV), and CK LY30. Reference ranges were based on manufacturer guidelines and are summarized in Supplementary Table 1.

Secondary outcomes included hemostatic data across the 4 surgical phases stratified based on donor type and whether the patients received tranexamic acid or not. In addition, to postoperative outcome differences based on TEG values at the end of surgery.

Statistical analysis

Continuous variables were described as the mean ± SD, and categorical variables were represented as n (%). Univariate analyses, data cleaning, and validation were performed in IBM SPSS Version 30.0.0.0 (Armonk, NY, United States). All patients who met the inclusion criteria were included; as such, no sample size calculations were performed. Secondary analysis was completed in StataNow/BE 19.5 (revision January 28, 2026). Comparisons of means were done using t-tests with Satterthwaite’s approximations for unequal variances.

RESULTS

Baseline characteristics

We screened 478 transplant recipient records, of which 364 were included in the study (Figure 1). The mean patient age was 56.7 (10.6) years, one-third of the patients were female (123/364; 33.8%), two-thirds were male (241/364; 66.2%), and most patients were White (271/364; 74.5%). Patients had a mean body mass index of 29.1 (5.9) kg/m² and a high mean end-stage liver disease sodium score before surgery of 22.2 (8.7). The most common comorbidity was hypertension (47.0%), followed by diabetes (28.0%). The most common cause of liver failure was alcohol-related (46.6%), whereas metabolic dysfunction-associated steatohepatitis and hepatocellular carcinoma contributed similarly (23.9% and 22.3%). Other clinical and demographic characteristics are described in Table 1.

Figure 1 Study group selection flowchart. TEG: Thromboelastography; ICU: Intensive care unit.

Table 1 Demographic, clinical characteristics and postoperative variables of patients undergoing liver transplant surgery, mean ± SD/n (%).

Characteristic	Result (n = 364)
Age at OLT, years	56.7 ± 10.6
Sex
Female	123 (33.8)
Male	241 (66.2)
Race
Asian	5 (1.4)
Black	24 (6.6)
White	271 (74.5)
Other	64 (17.5)
Body mass index, kg/m2	29.1 ± 5.9
MELD-Na score pre-surgery	22.2 ± 8.7
Blood type
A	147 (40.4)
AB	22 (6.0)
B	49 (13.5)
O	144 (39.6)
Unknown	2 (0.5)
Comorbidities
Chronic pulmonary disease	32 (8.8)
Diabetes	102 (28.0)
End-stage renal disease	10 (2.7)
Hypertension	171(47.0)
Cause of liver failure
Autoimmune hepatitis	16 (4.4)
Alcohol-related	166 (46.6)
Cholangiocarcinoma	5 (1.4)
Cryptogenic	14 (3.8)
Hepatocellular carcinoma	81 (22.3)
Hepatitis C	59 (16.2)
Hepatitis B	8 (2.2)
MASH	87 (23.9)
Primary sclerosing cholangitis	33 (9.1)
Primary biliary cholangitis	17 (4.7)
Other	12 (3.3)
TEG platform
TEG 5000	246 (67.6)
TEG 6s	118 (32.4)
Type of transplant
Deceased donor liver transplant	329 (90.4)
Living donor liver transplant	35 (9.6)
Donor type
Donation after brain death	253 (69.5)
Donation after circulatory death	76 (20.9)
Living donor	35 (9.6)
Tranexamic acid used
Pre-anhepatic	62 (17.0)
Anhepatic	11 (3.0)
Post-reperfusion	108 (29.7)
Quantity of intraoperative blood product units
Packed red blood cells	4.4 ± 6.4
Cell saver units	4.0 ± 4.7
Fresh frozen plasma	5.8 ± 6.7
Cryoprecipitate	1.6 ± 1.9
Platelets	1.0 ± 1.3
Intraoperative CVVH	23 (6.3)
ICU length of stay, days	4.8 ± 15.6
Intubated upon ICU admission	143 (39.3)
Mechanical ventilation duration, hours	22.1 ± 23.2
ICU transfusion within 72 hours
Packed red blood cells	147 (40.3)
Fresh frozen plasma	135 (37.08)
Cryoprecipitate	116 (31.8)
Platelets	83 (22.8)
Need for re-operation within 7 days	37 (10.2)
Mortality at 30 days	3 (0.8)

OLT: Orthotopic liver transplantation; MELD-Na: Model end-stage liver disease sodium; MASH: Metabolic dysfunction–associated steatohepatitis; TEG: Thromboelastography; CVVH: Continuous veno-venous hemofiltration; ICU: Intensive care unit.

Most patients (246/364; 67.6%) had coagulation monitored during surgery with TEG 5000, and the remaining 118 patients (32.4%) were monitored with TEG 6s. Most liver grafts came from deceased donors (90.4%) and were donated after brain death (69.5%). About half of the patients received intraoperative tranexamic acid: 29.7% during the post-reperfusion phase, 17% during the pre-anhepatic phase, and 3% during the anhepatic phase. The mean amount of blood products used included 4.4 (6.4) units of packed red blood cells, 4.0 (4.7) units of cell saver, 5.8 (6.7) units of fresh frozen plasma, 1.6 (1.9) units of cryoprecipitate, and 1.0 (1.3) units of platelets. Intraoperative continuous veno-venous hemofiltration was required in 23 (6.3%) patients. The mean ICU stay was 4.8 (15.6) days, and 143 (39.3%) required intubation with a mean mechanical ventilation time of 22.1 (23.2) hours. Within 72 hours of ICU admission, 40.3% of patients required packed red blood cell transfusion, 37.1% received fresh frozen plasma, 31.8% received cryoprecipitate, and 22.8% received platelets. Re-operation within 7 days was required in 37 (10.2%) patients, and 3 patients (0.8%) died within 30 days of transplantation (Table 1).

Coagulation parameters by phase of OLT

Figure 2 illustrates the proportions of patients with TEG parameters indicating hypocoagulability and normal hemostatic states across the 4 phases of OLT surgery. Overall, the graphs reveal that the 2 middle phases of surgery (anhepatic and post-reperfusion) represent the times at which increasing proportions of patients had TEG parameters suggesting hypocoagulability, particularly the anhepatic phase. Conversely, the earliest and latest transplant phases generally had higher rates of patients with normal TEG measures.

Figure 2 Distribution of patients with hypocoagulability and normal hemostatic parameters across 4 phases of liver transplant surgery. Hemodynamic parameters were monitored during transplant surgery with either the thromboelastography (TEG) 6s or TEG 5000 platforms. Hemodynamic trends suggest higher rates of hypocoagulability may occur during the anhepatic and post-reperfusion phases. A: The show proportions of patients with TEG results indicating hypocoagulability; B: The show proportions of patients with normal hemostatic TEG results over time. Denominators for percentages vary per parameter and result range (Table 2). Reference ranges for all parameters are in Supplementary Table 1. TEG: Thromboelastography; CRT: Citrated rapid thromboelastography; MA: Maximum amplitude; CFF: Citrated functional fibrinogen; CK: Citrated kaolin; R: Reaction time; LY30: Percent clot lysis at 30 minutes post-maximum amplitude; K: Kinetics time.

Notably, the 2 TEG platforms measure coagulation parameters in slightly different ways. Although various results were observed at similar rates across OLT surgery with each platform, the 2 instruments did not reveal identical trends. Figure 3 shows the proportions of patients with normal and coagulopathic TEG parameters across the 4 phases of OLT. Each panel shows a side-by-side comparison of the temporal trends obtained from the TEG 6s and TEG 5000. In particular, clot initiation and platelet function in the anhepatic phase were 2 areas in which the platforms varied slightly. However, patterns of fibrinolysis over time aligned very closely.

Figure 3 Head-to-head comparison of basic hemodynamic parameters from thromboelastography 6s and thromboelastography 5000. Hemodynamic parameters were monitored during 4 phases of transplant surgery with either the thromboelastography (TEG) 6s or TEG 5000 platforms. Hemodynamic trends suggest that coagulability is the least concerning during the pre-anhepatic and end-of-surgery phases, with coagulation stability less prevalent during the middle phases of transplant. A: Proportions of patients with normal results indicating coagulopathy over time; B: Proportions of patients with results indicating coagulopathy over time. Denominators for percentages vary per parameter and result range (Table 2). Reference ranges for all parameters are in Supplementary Table 1. TEG: Thromboelastography; CK: Citrated kaolin; R: Reaction time; CFF: Citrated functional fibrinogen; MA: Maximum amplitude; K: Kinetics time; CRT: Citrated rapid thromboelastography; LY30: Percent clot lysis at 30 minutes post-maximum amplitude.

The different TEG parameters for the 2 platforms measure specific characteristics of clot dynamics, which are indirectly related to the conventional hemodynamic measurements. Table 2 shows the proportions of all patients who had low/shortened, normal, and high/prolonged TEG results for all 4 stages of liver transplant surgery. Table 3 shows the distribution of results for conventional hemodynamic tests taken during transplant surgery. Not all patients had all parameters measured during each phase of surgery.

Table 2 Distribution of coagulation measures by thromboelastography across 4 phases of liver transplant surgery, n (%).

TEG parameters and levels	Distribution of TEG levels across liver transplant surgical phases
	Pre-anhepatic	Anhepatic	Post-reperfusion	End of surgery
Clot initiation time
CK R (TEG 6s)	n = 116	n = 100	n = 116	n = 113
Shortened/Low	19 (16.4)	22 (22.0)	22 (19.0)	12 (10.6)
Normal	87 (75.0)	46 (46.0)	54 (55.2)	85 (75.2)
Prolonged/high	10 (8.6)	32 (32.0)	30 (25.9)	16 (14.2)
R (TEG 5000)	n = 214	n = 175	n = 206	n = 207
Shortened/Low	148 (69.2)	150 (85.7)	93 (45.1)	116 (56.0)
Normal	64 (29.9)	23 (13.1)	94 (45.6)	84 (40.6)
Prolonged/high	2 (0.9)	2 (1.1)	19 (9.2)	7 (3.4)
Clot formation
CFF MA (TEG 6s)	n = 112	n = 94	n = 116	n = 113
Shortened/Low	40 (35.7)	49 (52.1)	53 (45.7)	55 (48.7)
Normal	62 (55.4)	38 (40.4)	60 (51.7)	57 (50.4)
Prolonged/high	10 (8.9)	7 (7.4)	3 (2.6)	1 (0.9)
CFF FLEV (TEG 6s)	n = 46	n = 46	n = 76	n = 62
Shortened/Low	18 (39.1)	25 (54.3)	39 (51.3)	36 (58.1)
Normal	26 (56.5)	20 (43.5)	35 (46.1)	26 (41.9)
Prolonged/high	2 (4.3)	1 (2.2)	2 (2.6)	ND
K (TEG 5000)	n = 212	n = 175	n = 199	n = 203
Shortened/Low	147 (69.3)	123 (70.3)	80 (40.2)	105 (51.7)
Normal	60 (28.3)	44 (25.1)	80 (40.2)	84 (41.4)
Prolonged/high	5 (2.4)	8 (4.6)	39 (19.6)	14 (6.9)
α-angle (TEG 5000)	n = 214	n = 175	n = 205	n = 205
Shortened/Low	4 (1.9)	5 (2.9)	35 (17.1)	10 (4.9)
Normal	37 (17.3)	28 (16.0)	48 (23.4)	43 (21.0)
Prolonged/high	173 (80.8)	142 (81.1)	122 (59.5)	152 (74.1)
Fibrinolysis
CK LY30 (TEG 6s)	n = 78	n = 58	n = 78	n = 74
Normal	69 (88.5)	43 (74.1)	67 (85.9)	73 (98.6)
Prolonged/high	9 (11.5)	15 (25.9)	11 (14.1)	1 (1.4)
LY30 (TEG 5000)	n = 213	n = 175	n = 204	n = 204
Normal	176 (82.6)	118 (67.4)	161 (78.9)	185 (90.7)
Prolonged/high	37 (17.4)	57 (32.6)	43 (21.1)	19 (9.3)
Platelet contribution to clot strength
CRT MA (TEG 6s)	n = 113	n = 93	n = 114	n = 104
Shortened/Low	61 (54.0)	60 (64.5)	80 (70.2)	73 (70.2)
Normal	51 (45.1)	33 (35.5)	34 (29.8)	31 (29.8)
Prolonged/high	1 (0.9)	ND	ND	ND
MA (TEG 5000)	n = 213	n = 175	n = 204	n = 204
Shortened/Low	92 (43.2)	65 (37.1)	138 (67.6)	109 (53.4)
Normal	106 (49.8)	96 (54.9)	64 (31.4)	92 (45.1)
Prolonged/high	15 (7.0)	14 (8.0)	2 (1.0)	3 (1.5)

TEG: Thromboelastography; CK: Citrated kaolin; R: Reaction time; CFF: Citrated functional fibrinogen; MA: Maximum amplitude; FLEV: Functional fibrinogen level; K: Kinetics time; LY30: Percent clot lysis at 30 minutes post-maximum amplitude; CRT: Citrated rapid thromboelastography; ND: No data.

Table 3 Distribution of conventional coagulopathy test results over 4 phases of liver transplant surgery in patients concomitantly monitored with thromboelastography, n (%).

Conventional test and level	Distribution of intraoperative conventional coagulopathy test results
	Patients monitored by TEG 5000				Patients monitored by TEG 6s
	Pre-anhepatic	Anhepatic	Post-reperfusion	End of surgery	Pre-anhepatic	Anhepatic	Post-reperfusion	End of surgery
International normalized ratio	n = 156	n = 194	n = 234	n = 234	n = 50	n = 70	n = 98	n = 93
Low/shortened	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A
Normal	2 (1.3)	1 (0.5)	ND	ND	2 (4.0)	4 (5.7)	ND	ND
High/prolonged	154 (98.7)	193 (99.5)	234 (100)	234 (100)	48 (96.0)	66 (94.3)	98 (100)	93 (100)
Fibrinogen	n = 142	n = 185	n = 222	n = 221	n = 50	n = 62	n = 93	n = 89
Low	74 (52.1)	111 (60.0)	166 (74.8)	149 (67.4)	25 (50.0)	43 (69.4)	76 (81.7)	58 (65.2)
Normal	63 (44.4)	71 (38.4)	53 (23.9)	69 (31.2)	23 (46.0)	19 (30.6)	17 (18.3)	31 (34.8)
High	5 (3.5)	3 (1.6)	3 (1.4)	3 (1.4)	2 (4.0)	ND	ND	ND
Partial thromboplastin time	n = 155	n = 194	n = 234	n = 234	n = 51	n = 70	n = 99	n = 93
Shortened	ND	ND	ND	ND	1 (2.0)	ND	ND	1 (1.1)
Normal	55 (35.5)	59 (30.4)	12 (5.1)	19 (8.1)	7 (13.7)	12 (17.1)	6 (6.1)	11 (11.8)
Prolonged	100 (64.5)	135 (69.6)	222 (94.9)	215 (91.9)	43 (84.3)	58 (82.9)	93 (93.9)	81 (87.1)
Platelets	n = 155	n = 192	n = 231	n = 231	n = 58	n = 65	n = 96	n = 94
Low	136 (87.7)	170 (88.5)	213 (92.2)	214 (92.6)	52 (89.7)	58 (89.2)	87 (90.6)	88 (93.6)
Normal	18 (11.6)	22 (11.5)	18 (7.8)	16 (6.9)	6 (10.3)	7 (10.8)	9 (9.4)	6 (6.4)
High	1 (0.6)	ND	ND	1 (0.4)	ND	ND	ND	ND

TEG: Thromboelastography; N/A: Not applicable; ND: No data.

Clot initiation time and coagulation factor activity: CK R and R reflect clot initiation time, and prolonged results suggest hypocoagulability due to coagulation factor deficit. For patients monitored with TEG 6s, around 75% of patients had normal CK R results during the pre-anhepatic and end-of-surgery stages, with lower rates during the middle anhepatic (46.0%) and post-reperfusion phases (55.2%). Prolonged CK R rates mirrored the normal rates, being lowest in the first and last surgery stages (8.6% and 14.2%) and highest in the middle 2 phases (32.0% and 25.9%). For patients monitored with the TEG 5000, prolonged R rates were generally low, ranging from around 1% to 9%, but the pattern of the highest rates occurring in the middle 2 phases mirrored that of CK R (Table 2). However, different from the TEG results, almost all patients (94%-100%) showed prolonged international normalized ratio results indicative of hypocoagulability during all phases of surgery (Table 3).

Clot formation and fibrinogen: The TEG parameters indicating clot formation dynamics and fibrinogen contribution to clot strength include CFF MA and CFF FLEV for the TEG 6s and K and α-angle for TEG 5000. Prolonged K and low CFF MA, CFF FLEV, and α-angle suggest hypocoagulability due to mostly a fibrinogen deficit. About a third of patients (35.7%) had low CFF MA in the pre-anhepatic stage; the proportion rose to over half (52.1%) during the anhepatic stage, decreased slightly to 45.7% post-reperfusion, and was at 48.7% by the end of surgery. Similarly, 39.1% of patients had low CFF FLEV in the pre-anhepatic phase, and the proportion was 54.3% in the anhepatic phase, 51.3% post-reperfusion, and 58.1% by the end of surgery. Few patients had prolonged K during the pre-anhepatic and anhepatic phases (2.4% and 4.6%), but the proportion was higher at post-reperfusion (19.6%) and closer to baseline by the end of surgery (6.9%; Table 2). Low α-angle followed a similar distribution, observed in 1.9%, 2.9%, 17.1%, and 4.9% of patients across the phases of OLT. Most patients who had fibrinogen levels assessed, a greater proportion of patients had low fibrinogen than normal levels, and the low fibrinogen levels percentages were highest in the post-reperfusion phase (74.8% and 81.7% for TEG 5000 and TEG6s, respectively; Table 3).

Fibrinolysis: CK LY30 and LY30 are the TEG parameters that measure decline in clot strength, reflecting fibrinolysis, with high levels being part of the hypocoagulable TEG profile. Hyperfibrinolysis peaked during the anhepatic phase in patients monitored with both TEGs (CK LY30 25.9% and LY30 32.6%) and was the lowest at post-reperfusion (14.1% and 21.1%). By the end of surgery, only 1 patient had high CK LY30 (1.4%), while 9.3% had high LY30 (Table 2).

Platelet activity: The TEG measures that reflect the platelet contribution to clot strength are the CRT MA for TEG 6s and MA for TEG 5000. Low results for these parameters are part of the hypocoagulability TEG profile. Over half of the patients monitored for CRT MA had low results in the pre-anhepatic and anhepatic phases (54.0% and 64.5%), and the proportion increased to 70.2% post-reperfusion, where it remained at the end of surgery. Low MA was observed in 43.2% of patients in the pre-anhepatic phase and 37.1% in the anhepatic phase. However, the proportion rose to 67.6% post-reperfusion, and at the end of surgery, it was 53.4% (Table 2). The conventional platelet count measure was consistently low across all phases of surgery in at least 88% of patients throughout all phases of OLT (Table 3).

Outcomes analysis: We have split the sample according to platelet dysfunction at the end of the surgery (the most prevalent problem we encountered and thus clinically meaningful) and compared differences in postoperative outcomes (Supplementary Table 2). Out of all outcomes, transfusion within 72 hours of ICU admission was increased and statistically significant in patients who had platelet dysfunction at the end of surgery in both TEG platforms. Intubation upon ICU admission was increased and statistically significant in patients who had platelet dysfunction at the end of surgery according to TEG 6s only. The rest of the outcomes were not statistically significant.

Stratified analysis: We performed a stratified analysis based on type of donor and whether the patient received tranexamic acid or not (Supplementary Figure 1). Parameters trends suggest that hypocoagulability is mostly increased during the anhepatic and post-reperfusion phases, in all different stratified groups, like the general pattern observed. Patients who received a liver after circulatory death had the highest percentage of platelet dysfunction and hyperfibrinolysis compared to other donor types. In contrast, patients who received a liver from a living donor or after brain death had the highest percentage of coagulation deficiency and fibrinogen deficiency respectively. Patients who received tranexamic acid had a higher percentage of hyperfibrinolysis in the pre-anhepatic, anhepatic and post-reperfusion phases compared to those who did not receive tranexamic acid. However, by the end of surgery patients who received tranexamic acid had a lower percentage of hyperfibrinolysis compared to patients who did not receive tranexamic acid.

DISCUSSION

In this study, we observed that intraoperative hypocoagulopathy peaked during the anhepatic and post-reperfusion phases of liver transplant surgery in patients monitored with TEG. By characterizing these temporal transitions, our findings may help clinicians better anticipate the need for specific blood products and hemostatic agents during different stages of OLT surgery.

Platelet and clotting factor activity across OLT phases

Our findings were consistent with a previous study that showed that patients undergoing OLT often exhibit hypocoagulability, particularly during the anhepatic and post-reperfusion phases of surgery[10]. In our patients, this pattern was primarily driven by decreased platelet activity, while coagulation factor activity was mostly normal or increased. Thrombocytopenia is also observed in patients undergoing OLT, which is often secondary to portal hypertension, splenic sequestration of platelets, and increased endothelial nitric oxide production[13]. Despite this, low levels of procoagulant factors may be counterbalanced by similarly reduced anticoagulant factors, producing a perceived prothrombotic state[14]. In our study, this was reflected by shortened R peaking during the anhepatic phase in most patients. A previous study of 124 patients reported a similar R distribution and reduced platelet activity that peaked during reperfusion[11]. These findings highlight the value of monitoring in uncovering hemostatic complexity during OLT.

Platelet activity and fibrinogen deficits during the pre-anhepatic phase

Pre-existing coagulopathy in patients undergoing OLT can be compounded in the pre-anhepatic phase by surgical bleeding, large-volume fluid shifts from ascites drainage, and the hemodynamic effects of portal hypertension[13]. In our patients, this was reflected by hypofibrinogenemia (low CFF FLEV and CFF MA) and decreased platelet function (low CRT MA and MA) as the most common abnormalities. This finding highlights the importance of fibrinogen and platelet dysfunction in the anhepatic phase and may guide clinician’s transfusion practices to reduce excessive bleeding during this phase.

Clotting factor activity and fibrinolysis during the anhepatic phase

Clamping of the liver’s blood supply during the anhepatic phase of OLT prevents the synthesis and clearance of coagulation factors and anti-coagulation factors[13,15]. Additionally, increased endothelial release of tissue plasminogen activator along with minimal changes in plasminogen activator-1 can promote hyperfibrinolysis and subsequent hypofibrinogenemia[13,15]. These events are reflected in our TEG findings, in which we observed the highest incidence of hyperfibrinolysis (elevated LY30 and CK LY30) and low fibrinogen contribution to clot strength (low CFF MA) in the anhepatic phase. Clotting factor activity showed a mixed pattern. Most patients had shortened R, suggesting hypercaogulopathy; however, of over half of the patients with abnormal CK R, 22% were shortened and 32% were prolonged. This reflects the complexity of clotting factor activity during liver dysfunction, highlighting the balance between low levels of procoagulant factors counterbalanced by similarly reduced anticoagulant factors.

Platelet activity and fibrinolysis post-reperfusion

Reperfusion of the donor liver is associated with platelet entrapment within the grafted liver, the release of heparin-like substances, and hemodilution from the cold preservative solution[13,15,16]. Post-reperfusion, we observed the highest prevalence of decreased platelet function, as evidenced by low CRT MA and MA values, which correlates with these physiological changes. We also observed a substantial decline in hyperfibrinolysis by the end of surgery, as indicated by the low number of patients with high LY30 and CK LY30 at this time. This correlates with the simultaneous clearance of tissue plasminogen activator, upregulation of plasminogen activator inhibitors, and administration of antifibrinolytic agents after implantation of the donor liver[13,15,16]. This shows that fibrinolysis activity is most readily corrected post-reperfusion, while platelet’s function might be harder to correct. This highlights the need for perhaps a more attentive platelet transfusion strategy during this phase.

Phase-specific discrepancies between teg platforms and conventional tests

Discrepancies between TEG 6s and TEG 5000 results during surgery have previously been reported especially with increasing coagulopathy[17,18]. In our study, platelet function and hyperfibrinolysis had comparable results across the TEG platforms; however, clot initiation time varied, with a higher prevalence of shortened TEG 5000 R times, and TEG 6s CK R values were prolonged more often. A similar discrepancy was observed in fibrinogen activity; TEG 5000 frequently indicated increased activity, with high α-angles and shortened K, while TEG 6s showed decreased activity more often, with low CFF MA values. Nonetheless, in our sample, both TEG6S and TEG5000 identified a significant amount of qualitative coagulopathy, showing a similar pattern of increased coagulopathy in the anhepatic and post-reperfusion phases. Hence, both tests can be used to describe patterns in coagulopathy during different phases of liver transplant surgery.

In contrast to TEG platforms, conventional coagulation tests showed consistently abnormal results throughout all surgical phases with minor changes. This supports previous observations that conventional tests may not capture the dynamic hemostatic changes during OLT and that TEG platforms may provide more precise monitoring[5,6].

Secondary analysis

Our outcome analysis showed expectedly that patients who had abnormal TEG 6s and TEG 5000 were more likely to received blood transfusions in the ICU, as the providers use TEG to guide transfusion therapy in the ICU. Also, stratified analysis showed interestingly that patients who ended up receiving tranexamic acid, had a higher percentage of hyperfibrinolysis early during the surgery compared to those who did not receive tranexamic acid. This finding implies that patients received tranexamic acid using TEG guidance mostly, rather than prophylactically or based on clinical judgment.

Strengths and limitations

The strength of our study was the relatively large sample size compared with prior reports on OLT-related coagulopathy. By describing coagulopathy as a primary endpoint, we provided detailed assessment across coagulation factor activity, platelet function, fibrinogen function, and fibrinolysis. Inclusion of two different TEG platforms also enabled descriptive comparison of patterns between two widely used instruments.

Limitations include cautious use of platelet transfusion, given its association with adverse outcomes in liver transplant recipients[19,20], which may have contributed to the high prevalence of reduced platelet activity. Not all coagulation tests and TEG parameters were performed in every phase, and sampling times varied. We did not limit analysis to patients with complete data, as this could bias results toward those requiring closer monitoring. Transfusion practices also reflect clinicians’ judgment beyond TEG results, so our findings may capture local practice as well as underlying physiology. Finally, being single-center, results may reflect institutional protocols and limit generalizability.

CONCLUSION

Among adults undergoing OLT, coagulopathy was most commonly observed during the anhepatic and post-reperfusion phases as measured by intraoperative TEG, with platelet dysfunction being the most common abnormality. Also, hyperfibrinolysis was the derangement most likely to resolve by the end of the surgery. Multi-center studies are needed to explore the postoperative implications of these temporal changes and to optimize intraoperative blood transfusion and antifibrinolytic strategies.

ACKNOWLEDGEMENTS

We thank Dr. Karla D Passalacqua for providing editorial assistance and Stephanie Stebens for helping with manuscript formatting.