Initial TEG 6s parameters for risk stratification of hemorrhage-related outcomes in massively transfused trauma patients

Abstract

Purpose

While thromboelastography (TEG) and rotational thromboelastometry (ROTEM) allow rapid coagulation assessment, the prognostic value of arrival TEG 6s remains unclear. We examined the association between initial TEG 6s findings and in-hospital mortality and early transfusion requirements in trauma patients undergoing massive transfusion, excluding severe traumatic brain injury to better assess hemorrhage-related outcomes.

Methods

This retrospective study included trauma patients who received massive transfusion and TEG 6s testing upon arrival at a Level I trauma center in South Korea. Patients with transfers, incomplete records, or head/neck AIS ≥ 3 were excluded. TEG 6s parameters were classified using published thresholds. Multivariable logistic and linear regression analyses evaluated associations with in-hospital mortality and RBC transfusion volumes at 4 and 24 h.

Results

Among 194 patients, 54.1% had abnormal TEG 6s results. These patients received higher RBC transfusion volumes at 4 and 24 h, and had higher in-hospital mortality. Abnormal TEG 6s findings were associated with higher in-hospital mortality in the multivariable analysis (adjusted OR 3.505, p = 0.050) and increased transfusion volume at 4 h (B = 1.686, p = 0.023) and 24 h (B = 2.313, p = 0.042).

Conclusion

Initial TEG 6s abnormalities were associated with in-hospital mortality and early RBC transfusion in trauma patients undergoing massive transfusion. In massively transfused patients without severe traumatic brain injury, these findings suggest that arrival TEG 6s parameters may aid early risk stratification for hemorrhage-related outcomes.

Introduction

Severe hemorrhage is a common occurrence in patients with major trauma and is recognized as one of the leading causes of early post-traumatic mortality. In addition to direct blood loss, severe hemorrhage triggers systemic pathophysiological responses, including tissue hypoperfusion, endothelial dysfunction, inflammatory activation, and dysregulated coagulation, which collectively contribute to the development of trauma-induced coagulopathy (TIC). Although severe hemorrhage and TIC are distinct clinical entities, they frequently coexist in severely injured patients and interact to exacerbate bleeding, organ dysfunction, and mortality. TIC is a well-recognized complication of severe trauma [1–3] and has been reported to occur in approximately 30–40% of patients with major injury in observational studies [4–6]. TIC is associated with worse clinical outcomes and increased mortality, particularly in severe hemorrhage [7, 8], with mortality rates approaching 50% reported in a large observational study [9]. However, considerable heterogeneity exists in the reported incidence and prognostic impact of TIC across studies, likely due to differences in injury severity, definitions of coagulopathy, and resuscitation strategies.

In recent years, viscoelastic analyzers (VEAs), including thromboelastography (TEG) and rotational thromboelastometry (ROTEM), which are used to perform viscoelastic hemostatic assays (VHAs), have gained increasing attention for the rapid diagnosis of TIC and the development of individualized resuscitation strategies. Compared to conventional coagulation tests (CCTs) such as platelet count, prothrombin time (PT), and activated partial thromboplastin time (aPTT), VHAs offer the advantage of real-time, quantitative, and dynamic assessment of overall coagulation function, thereby facilitating more precise transfusion management [10, 11]. While numerous studies have reported associations between VHA parameters and clinical outcomes in trauma patients, evidence supporting their use to guide early admission-based therapeutic interventions remains heterogeneous and inconsistent [12–15]. In particular, the prognostic significance of VHA parameters obtained upon hospital arrival, before substantial resuscitative interventions, remains unclear, as most existing studies have evaluated various TEG systems and are limited by heterogeneity in analytic methods, patient populations, and outcome definitions [16–20]. As a result, the prognostic value of TEG 6s measurements obtained upon arrival for hemorrhage-related outcomes in trauma patients has not yet been clearly established.

Accordingly, this study aimed to evaluate the association between TEG 6s parameters obtained upon arrival and hemorrhage-related outcomes, including in-hospital mortality and early red blood cell transfusion requirements, in trauma patients undergoing massive transfusion. Patients with severe traumatic brain injury were deliberately excluded because coagulopathy is known to occur early after brain injury—even in the absence of major systemic hemorrhage—and has been reported in approximately 20% of isolated traumatic brain injury cases, with associated mortality rates approaching 50% [21]. These findings suggest that outcomes in this population are predominantly driven by neurologically mediated pathophysiology rather than systemic bleeding. Exclusion of this subgroup therefore minimized neurologically driven confounding and enabled a more focused assessment of outcomes primarily influenced by hemorrhage severity and systemic coagulation derangement.

Methods

Patients

This retrospective study was conducted at a single regional trauma center functioning as a Level I trauma center in South Korea, between January 2023 and December 2024. The study included trauma patients who underwent massive transfusion and had a TEG 6s test performed immediately upon arrival. Massive transfusion was defined as the administration of ≥ 5 units of RBCs within 4 h or ≥ 10 within 24 h. The following patients were excluded from the analysis: those transferred from other hospitals, those with incomplete medical records, and those with an Abbreviated Injury Scale score ≥ 3 in the head and neck region (Fig. 1). These exclusions were applied to reduce the influence of mortality primarily driven by severe neurological injury and to allow a more accurate assessment of hemorrhage-related outcomes.

Fig. 1.

Flowchart of patient selection

Variables and data collection

Data were collected on sex, age, mechanism of injury, presence of underlying comorbidities, use of antithrombotic agents (including anticoagulants and antiplatelet agents), vital signs on arrival, Glasgow Coma Scale (GCS) score, initial lactic acid level, TEG 6s parameters, Injury Severity Score (ISS), hemostatic intervention, in-hospital complications, in-hospital mortality, and the volume of RBC transfusion within 4 and 24 h. Hemostatic intervention was defined as any definitive hemorrhage control procedure performed during the index hospitalization, including emergency surgical procedures such as laparotomy or thoracotomy, as well as endovascular interventions including transcatheter arterial embolization. Causes of in-hospital death were determined by retrospective review of clinical records, operative findings, and treating physician documentation. Hemorrhagic mortality was defined as death primarily attributed to uncontrolled bleeding or its direct physiological consequences. Among patients who died during hospitalization, the minimum time from emergency department arrival to death was approximately 90 min.

Initial coagulation status was assessed using the TEG 6s system (HAEMONETICS^®, Braintree, MA, USA), an automated viscoelastic hemostatic assay, performed immediately upon patient arrival at the emergency department (trauma bay). Blood sampling for the initial TEG 6s analysis was integrated into the initial trauma assessment workflow and obtained during establishment of vascular access. Whole blood samples were used for testing, utilizing Citrated Kaolin (CK) cartridges. The CK cartridge is the standard assay used in our institutional trauma workflow and was applied consistently throughout the study period to ensure standardized measurements. A tissue factor–based rapid TEG assay, which differs from the kaolin-based CK assay, was not available in our institutional TEG 6s system during the study period. Each cartridge consisted of four channels that simultaneously evaluated clot initiation, fibrinogen contribution to clot firmness, platelet contribution to clot firmness, and fibrinolysis.

The TEG 6s results were classified as normal or abnormal based on institutionally predefined thresholds. The TEG 6s system measured four key parameters: CK-R (clot initiation), CFF-MA (fibrinogen contribution to clot firmness), CRT-MA (platelet contribution to clot firmness), and CK-LY30 (fibrinolysis). Reference ranges for each parameter were based on previously published literature [22, 23] and defined as follows: CK-R, 4.6–9.1 min; CFF-MA, 15–32 mm; CRT-MA, 52–70 mm; and CK-LY30, 0–2.6%. CK-R was categorized as shortened, within the reference range, or prolonged based on predefined reference values. A shortened CK-R was defined as a value below 4.6 min, a normal CK-R as 4.6–9.1 min, and a prolonged CK-R as a value greater than 9.1 min. Abnormal CK-R was defined as values outside the reference range.

In our institution, tranexamic acid (TXA) is routinely administered to patients with suspected major hemorrhage immediately after emergency department arrival. Blood sampling for the initial TEG 6s analysis is performed during establishment of vascular access. Following blood sampling, TXA administration and blood product transfusion are initiated, thereby ensuring that the initial TEG 6s results reflect coagulation status prior to pharmacologic antifibrinolysis and transfusion. A TEG 6s–guided hemostatic treatment algorithm is implemented at our institution, whereby abnormal TEG 6s findings inform subsequent component-specific hemostatic interventions according to standardized institutional protocols. Importantly, these targeted interventions are initiated after completion of the initial TEG 6s analysis.

Data analysis

Categorical variables were presented as frequencies and percentages, and comparisons between the initial normal and abnormal TEG 6s groups were performed using the chi-squared test. Continuous variables were expressed as medians with interquartile ranges, and group comparisons were conducted using the Mann–Whitney U test for non-normally distributed data.

Multivariable logistic regression analysis was performed to identify factors independently associated with in-hospital mortality. Variables with a p-value < 0.1 in univariable analysis were included in the multivariable model. In addition, clinically relevant covariates, including age, sex, injury mechanism, systolic blood pressure on arrival, and hemostatic intervention, were included regardless of their univariable significance. Although time to massive transfusion protocol (MTP) activation and transfusion volume showed statistical significance in the univariable analysis (p < 0.1), they were excluded from the multivariable model as potential mediators in the relationship between TEG 6s results and mortality.

To evaluate factors associated with transfusion requirements, additional multivariable linear regression analyses were conducted for RBC transfusion volumes within 4 and 24 h. The selection of independent variables for the linear regression model followed the same criteria used in the logistic regression analysis. Regression coefficients and corresponding 95% confidence intervals were reported to assess statistical significance. In Model 2, instead of using the overall initial TEG 6s classification, abnormal findings in each of the four TEG 6s components were included for additional analysis.

All statistical analyses were performed using SPSS for Windows, version 29.0 (SPSS Inc., Chicago, IL, USA), and a two-sided p-value less than 0.05 was considered statistically significant.

Ethical considerations

This study was approved by the Institutional Review Board of Ajou University Hospital (IRB No. AJOUIRB-DB-2025-270) and was conducted in accordance with the ethical principles of the Declaration of Helsinki. As a retrospective observational study, all patient data were de-identified prior to analysis, and the requirement for informed consent was waived by the IRB. All data were used solely for research purposes, and strict adherence to ethical guidelines for data protection and confidentiality was maintained throughout the study.

Results

Comparison of baseline characteristics according to initial TEG 6s results

During the study period, a total of 385 trauma patients underwent massive transfusion, of whom 194 were included in the final analysis after applying predefined exclusion criteria. The most common reasons for exclusion were inter-hospital transfer and severe head or neck injury (AIS ≥ 3), followed by missing TEG 6s data on arrival.

Table 1 presents a comparison of demographic characteristics, injury mechanisms, laboratory findings, and clinical outcomes between patients with normal and abnormal initial TEG 6s results. The abnormal results group had a significantly lower proportion of blunt injuries (83.8% vs. 94.4%, p = 0.021) and a shorter time to activation of the MTP (6.5 min vs. 10.5 min, p < 0.001). Fibrinogen levels at emergency department admission were also significantly lower in the abnormal group than in the normal group (median 161 mg/dL vs. 209 mg/dL, p < 0.001). Patients in the abnormal group received significantly more red blood cell transfusions within both 4 h (10 vs. 9 units, p = 0.043) and 24 h (12 vs. 10 units, p = 0.029). In-hospital mortality was significantly higher in the abnormal TEG 6s group compared with the normal group (21.9% vs. 6.7%, p = 0.003). Review of causes of death showed that hemorrhage accounted for the majority of in-hospital deaths, with hemorrhagic mortality being higher in the abnormal TEG 6s group (19.0% vs. 6.7%, p = 0.019). Overall, among the 29 in-hospital deaths, hemorrhage accounted for 26 cases (89.7%), indicating that mortality in the analyzed cohort was largely hemorrhage-related. No significant differences were observed between the groups in age, sex, comorbidities, use of antithrombotic agents, initial vital signs, GCS, lactate levels, ISS, timing of surgical or endovascular interventions, or incidence of in-hospital complications.

Table 1.

Characteristics of trauma patients assessed by thromboelastography

Variables	Initial normal TEG 6s (N = 89)	Initial abnormal TEG 6s (N = 105)	P value
Age (y), median [IQR]	54 [40,66]	55 [39–65]	0.690
Sex, male (%)	68 (76.4)	84 (80)	0.545
Injury type, Blunt (%)	84 (94.4)	88 (83.8)	0.021
Underlying disease, n (%)	50 (56.2)	55 (55.0)	0.871
Antithrombotic agent, n (%)	11 (12.8)	12 (12.2)	0.911
Systolic blood pressure, median [IQR]	105 [90.5–134.0]	111 [80.25–126.5]	0.788
Pulse rate, median [IQR]	96 [80.5–115.0]	96 [80.0–120.0]	0.957
Glasgow Coma Scale, median [IQR]	14 [12–15]	14 [11–15]	0.071
Injury Severity Score, median [IQR]	34 [25–38]	29 [19–38]	0.089
Time to MTP (min), median [IQR]	10.5 [6–26.25]	6.5 [4–15]	< 0.001
Time to surgery (min), median [IQR]	60 [50–88.75]	55 [46.75–74.5]	0.067
Time to angiography (min), median [IQR]	161 [126.25–227.5]	154 [120.5–223.5]	0.539
Lactic acid, median [IQR]	4.8 [3.2–6.35]	4.9 [3.57–9.065]	0.121
Fibrinogen, mg/dL [IQR]	209 [168–244]	161 [135–199]	< 0.001
Complication, n (%)	50 (56.2)	49 (46.7)	0.187
RBC < 4 h, median [IQR]	9 [7.5–12.0]	10 [8–16]	0.043
RBC < 24 h, median [IQR]	10 [8–13]	12 [8–19.5]	0.029
In-hospital mortality, n (%)	6 (6.7)	23 (21.9)	0.003
Hemorrhagic mortality, n (%)	6 (6.7)	20 (19.0)	0.019

IQR interquartile range, AIS abbreviated injury scale, MTP massive transfusion protocol, TEG thromboelastography, RBC red blood cell

Hemorrhagic mortality was determined based on review of clinical records, operative findings, and treating physician documentation

Overall cohort and TEG 6s abnormalities

Initial TEG 6s results were abnormal in 54.1% (n = 105) of the patients. Each TEG 6s parameter was categorized according to its physiological characteristics into clot initiation abnormalities, fibrinogen contribution abnormalities, platelet contribution abnormalities, and fibrinolysis-related clotting abnormalities. The frequency and proportion of each abnormality are presented in Table 2.

Table 2.

Distribution of initial abnormal TEG 6s parameters

Initial TEG abnormalities	Clinical implication	N = 105
CK-R abnormality, n	Clot initiation abnormalities	57
CFF-MA abnormality, n	Fibrinogen contribution abnormalities	42
CRT-MA abnormality, n	Platelet contribution abnormalities	45
CK-LY30 abnormality, n	Fibrinolysis-related clotting abnormalities	18

Reference ranges were as follows: CK-R (4.6–9.1 min), CFF-MA (15–32 mm), CRT-MA (52–70 mm), and CK-LY30 (0–2.6%)

CK-R abnormalities were defined as values outside the reference range; among these, 5 patients demonstrated prolonged CK-R, whereas shortened CK-R values were observed but were not analyzed separately due to their unclear clinical relevance in the setting of hemorrhagic trauma

Analysis of factors associated with in-hospital mortality

Table 3 presents the results of univariable and multivariable logistic regression analyses for factors associated with in-hospital mortality. In the multivariable analysis, age (adjusted OR 1.079 per year, 95% CI: 1.031–1.128, p < 0.001), Glasgow Coma Scale score (adjusted OR 0.777, 95% CI: 0.660–0.916, p = 0.003), and Injury Severity Score (adjusted OR 1.054, 95% CI: 1.013–1.098, p = 0.010) were identified as independent predictors of in-hospital mortality.

Table 3.

Logistic regression analysis of factors associated with in-hospital mortality

Variables	Univariable analysis			Multivariable analysis
	OR (95% CI)	P value	Adjusted OR (95% CI)		P value
Age, year	1.020 (0.995–1.045)	0.112	1.079 (1.031–1.128)		< 0.001
Sex, male	1.070 (0.405–2.826)	0.892
Injury type, Blunt	0.537 (0.119–2.432)	0.420
Systolic blood pressure	0.996 (0.983–1.010)	0.579
Glasgow Coma Scale	0.786 (0.699–0.883)	< 0.001	0.777 (0.660–0.916)		0.003
Lactic acid	1.223 (1.119–1.337)	< 0.001
Injury Severity Score	1.026 (1.000-1.053)	0.053	1.054 (1.013–1.098)		0.010
Hemostatic intervention	1.349 (0.359–5.063)	0.657
Initial abnormal TEG 6s	3.880 (1.502–10.022)	0.005	3.505 (1.003–12.249)		0.050

OR odds ratio, CI confidence interval, MTP massive transfusion protocol, RBC red blood cell, TEG thromboelastography

Initial TEG abnormality was defined as the presence of any TEG 6s parameter outside the manufacturer reference range, including CK-R, CFF-MA, CRT-MA, or CK-LY30

All listed covariates were included in the multivariable model

Hosmer–Lemeshow chi-square test: χ² = 7.476, df = 8, p = 0.486

Initial abnormal TEG 6s findings were associated with higher in-hospital mortality in the multivariable analysis (adjusted OR 3.505, 95% CI: 1.003–12.249, p = 0.050). Other variables, including sex, injury mechanism, systolic blood pressure, lactic acid level, and hemostatic intervention, were not significantly associated with in-hospital mortality in the multivariable analysis.

Model calibration demonstrated acceptable fit based on the Hosmer–Lemeshow goodness-of-fit test. These results are visualized in Figs. 2.

Fig. 2.

Multivariable logistic regression for in-hospital mortality

Analysis of factors associated with RBC transfusion volume

Tables 4 and 5 present the results of linear regression analyses for factors associated with RBC transfusion volumes within 4 and 24 h, respectively.

Table 4.

Linear regression analysis of factors associated with RBC transfusion within 4 h (N = 188)†

Variables	Multivariable analysis*
	Model 1 B (95% CI)	Model 1 P value	Model 2 B (95% CI)	Model 2 P value
Age, year	0.019 (-0.026,0.064)	0.408	0.022 (-0.023,0.068)	0.330
Sex, male	-0.078 (-1.832,1.675)	0.930	-0.381 (-2.161,1.400)	0.674
Injury type, Blunt	1.151 (-1.517,3.819)	0.395	1.940 (-0.843,4.723)	0.171
Systolic blood pressure	-0.022 (-0.045,0.002)	0.071	-0.024 (-0.048,0.000)	0.047
Pulse rate	0.000 (-0.034,0.035)	0.983	0.003 (-0.031,0.038)	0.849
Glasgow Coma Scale	-0.037 (-0.308,0.233)	0.786	-0.045 (-0.317,0.228)	0.747
Lactic acid	0.509 (0.259,0.758)	< 0.001	0.467 (0.203,0.732)	< 0.001
Injury Severity Score	0.121 (0.065,0.176)	< 0.001	0.125 (0.068,0.182)	< 0.001
Initial abnormal TEG 6s	1.686 (0.231,3.101)	0.023
CK-R abnormality			0.411 (-1.263,2.085)	0.628
CFF-MA abnormality			2.898 (0.545,5.252)	0.016
CRT-MA abnormality			-0.794 (-3.076,1.489)	0.493
CK-LY30 abnormality			1.612 (-1.694,4.917)	0.337

CI confidence interval, RBC red blood cell, TEG thromboelastography

†Patients who died within 4 h were excluded from the analysis

^* Model 1, R² = 0.261, Adjusted R² = 0.220, P < 0.001; Model 2, R² = 0.272, Adjusted R² = 0.217, P < 0.001

All VIFs < 5, suggesting no collinearity concern

Table 5.

Linear regression analysis of factors associated with RBC transfusion within 24 h (N = 178) †

Variables	Multivariable analysis*
	Model 1 B (95% CI)	Model 1 P value	Model 2 B (95% CI)	Model 2 P value
Age, year	0.019 (-0.060,0.098)	0.635	0.014 (-0.066,0.094)	0.728
Sex, male	0.714 (-2.008,3.436)	0.605	0.403 (-2.362,3.168)	0.774
Injury type, Blunt	1.079 (-3.211,5.370)	0.620	1.889 (-2.532,6.310)	0.400
Systolic blood pressure	-0.012 (-0.049,0.025)	0.513	-0.021 (-0.059,0.017)	0.280
Glasgow Coma Scale	-0.030 (-0.466,0.406)	0.891	0.007 (-0.431,0.446)	0.974
Underlying disease	1.303 (-1.219,3.825)	0.309	1.424 (-1.107,3.954)	0.268
Lactic acid	0.713 (0.304,1.121)	< 0.001	0.627 (0.204,1.049)	0.004
Injury Severity Score	0.159 (0.067,0.252)	< 0.001	0.101 (0.053,0.150)	< 0.001
Time to MTP	0.042 (0.016,0.069)	0.002	0.040 (0.013,0.067)	0.004
Initial abnormal TEG 6s	2.313 (0.085,4.542)	0.042
CK-R abnormality			0.490 (-2.155,3.136)	0.715
CFF-MA abnormality			2.747 (-0.898,6.393)	0.139
CRT-MA abnormality			-0.132 (-3.767,3.503)	0.943
CK-LY30 abnormality			5.853 (0.424,11.281)	0.035

CI confidence interval, RBC red blood cell, AIS abbreviated injury scale, TEG thromboelastography

†Patients who died within 24 h were excluded from the analysis

^* Model 1, R² = 0.232, Adjusted R² = 0.180, P < 0.001; Model 2, R² = 0.251, Adjusted R² = 0.182, P < 0.001

All VIFs < 5, suggesting no collinearity concern

Model 1: Includes overall abnormal TEG 6s findings as a binary variable

Model 2: Incorporates specific TEG 6s abnormalities

Model 1: Includes overall abnormal TEG 6s findings as a binary variable

Model 2: Incorporates specific TEG 6s abnormalities

For 4-hour transfusion volume, significant predictors included lactic acid (p < 0.001), ISS (p < 0.001), and initial abnormal TEG 6s findings (B = 1.686, p = 0.023). Among the specific TEG 6s parameters, CFF-MA abnormality was significantly associated with increased transfusion volume (B = 2.898, 95% CI: 0.545–5.252, p = 0.016).

For 24-hour transfusion volume, significant predictors included lactic acid (p = 0.004), ISS (p < 0.001), and time to MTP activation (p = 0.004). Initial abnormal TEG 6s findings were also significantly associated with increased transfusion volume (B = 2.313, p = 0.042). Among the specific TEG 6s parameters, CK-LY30 abnormality showed a significant association (B = 5.853, 95% CI: 0.424–11.281, p = 0.035).

Multicollinearity was not detected in either model (variance inflation factor < 5), and the explanatory power of the regression models was R² = 0.261 for 4-hour and R² = 0.232 for 24-hour transfusion volumes. These results are visualized in Figs. 3 and 4.

Fig. 3.

Predictors of RBC transfusion volume within 4 h

Fig. 4.

Predictors of RBC transfusion volume within 24 h

Discussion

This study evaluated the clinical utility of initial TEG 6s in a cohort of massively transfused trauma patients without severe traumatic brain injury. Initial TEG 6s abnormalities were associated with in-hospital mortality and early RBC transfusion requirements, and parameter-specific analyses further demonstrated associations with increased transfusion volumes. Although the median RBC transfusion volume differed by only one unit, patients with abnormal initial TEG 6s findings demonstrated a broader interquartile range and a right-skewed distribution of transfusion volumes, indicating that a subset of patients required substantially higher transfusion volumes. Accordingly, the observed statistical significance reflects differences in the overall distribution of transfusion requirements rather than a clinically trivial shift in median values. By excluding patients with severe traumatic brain injury, the analysis minimized the influence of neurologic causes of mortality and focused on outcomes predominantly determined by hemorrhage-related physiology. Accordingly, hemorrhage represented the predominant cause of in-hospital mortality in this cohort, suggesting that the observed associations are largely driven by hemorrhage-related pathophysiological processes.

CCTs, including PT, aPTT, platelet count, and fibrinogen level, have been extensively studied for their prognostic value in trauma patients. Niles et al. proposed a definition of trauma-induced coagulopathy based on the international normalized ratio and demonstrated its independent association with mortality [24]. Similarly, MacLeod et al. reported that early coagulopathy defined by PT and aPTT was independently associated with increased mortality in trauma patients [25]. Among individual coagulation components, fibrinogen is an independent predictor of mortality in trauma patients and the first coagulation factor to reach critically low levels after injury, with low fibrinogen on arrival consistently associated with worse clinical outcomes [26–30]. Recent mechanistic evidence further indicates that hypofibrinogenemia plays a central role in impaired clot initiation. Hofmann et al. demonstrated that prolongation of r-time in viscoelastic assays primarily reflects substrate deficiency rather than reduced clotting factor activity [31]. Consistent with these findings, the present study showed that patients with abnormal initial TEG 6s results had significantly lower fibrinogen levels upon emergency department arrival. This supports the interpretation that early TEG abnormalities, particularly r-time prolongation, reflect underlying hypofibrinogenemia and its association with increased early transfusion requirements.

However, CCT-based markers reflect only isolated aspects of the coagulation cascade and fail to assess the overall coagulation dynamics and fibrinolytic processes occurring in real time in actual patients. Particularly in the setting of acute hemorrhage, the delayed turnaround time and limited physiological representation of CCTs pose significant challenges for timely clinical decision-making [32–34]. As a result, VEAs, such as TEG and ROTEM, have increasingly been utilized for early assessment of coagulation abnormalities and guidance of transfusion strategies [35]. Ongoing studies continue to evaluate the clinical utility of these methods in trauma care.

Prior studies have examined the prognostic relevance of thrombelastography in trauma care; however, their findings are largely derived from selected injury patterns or broad, heterogeneous trauma populations. Kane et al. reported that R-time on TEG was a significant predictor of mortality in patients with pelvic trauma [19]. Nystrup et al. demonstrated that reduced clot strength on admission TEG was associated with higher 30-day mortality in a heterogeneous trauma cohort [36]. Together, these studies support the potential prognostic value of early TEG findings. However, the prognostic interpretation of TEG 6s measurements obtained immediately upon arrival in trauma patients undergoing massive transfusion–specifically in relation to hemorrhage-related outcomes–has not been sufficiently defined. This knowledge gap directly motivated the present analysis.

Recently, Murali et al. conducted a large multicenter retrospective study and proposed a TEG/ROTEM-based TIC score for trauma patients who underwent massive transfusion. Their results demonstrated that higher TIC scores were independently associated with increased in-hospital mortality and greater RBC transfusion requirements [17]. However, that study focused on a composite score and did not evaluate the prognostic implications of individual TEG parameters or sufficiently account for antithrombotic agent use. In addition, it did not adequately account for the use of antithrombotic agents, which may influence coagulation profiles and confound the results. In contrast, the present study specifically targeted trauma patients who received massive transfusion. It quantitatively assessed not only the presence of abnormal TEG findings but also the impact of individual TEG parameters on clinical outcomes and transfusion volume, while adjusting for potential confounders such as antithrombotic agent use.

This gap is particularly relevant for the interpretation of r-time in the early phase of trauma. Shortened r-time has been observed in the very early phase of trauma as a transient hypercoagulable response preceding hypoperfusion-related coagulopathy [37]. However, early thromboelastographic patterns are heterogeneous, and the clinical significance of R-time shortening in relation to hemorrhage-related outcomes remains inconsistent [38, 39]. Accordingly, r-time abnormalities were analyzed as a unified parameter rather than stratified by direction to better capture early coagulation derangement.

The clinical utility of ROTEM has likewise been evaluated in several trauma studies. Smith et al. identified specific ROTEM parameters as significant independent predictors of mortality in trauma patients [40]. In addition, Theusinger et al. reported that hyperfibrinolysis, diagnosed through ROTEM analysis, was independently associated with increased mortality in trauma patients, highlighting the potential of ROTEM not only as a monitoring tool but also as a pathophysiology-based prognostic instrument [41]. Collectively, these ROTEM-based studies support the prognostic relevance of viscoelastic abnormalities in trauma. However, differences in analytic methodology and clinical workflow underscore the need for modality-specific validation, particularly for TEG 6s.

Although multiple observational studies and institutional protocols have suggested that viscoelastic test–guided hemostatic therapy may improve transfusion efficiency and clinical outcomes in trauma patients, evidence from randomized controlled trials remains mixed. In the iTACTIC trial [42], the largest randomized controlled study to date comparing VHA-guided resuscitation with conventional coagulation test–guided management, VHA-guided resuscitation resulted in earlier hemostatic intervention; however, no significant differences were observed in 24-hour or 28-day mortality or other major clinical outcomes, except in a subgroup of patients with severe traumatic brain injury. Accordingly, rather than evaluating therapeutic efficacy, the present study focused on the prognostic performance of early TEG 6s parameters in a carefully defined hemorrhage-related trauma population.

Recently, growing interest has focused on the clinical and outcome-related utility of the next-generation TEG device, TEG 6s, leading to expert consensus efforts aimed at standardizing its use in trauma care [23]. In this context, TEG 6s is a fully automated, cartridge-based device that offers standardized measurements, reducing operator dependency compared to the earlier TEG 5000 system. It enhances reproducibility and reliability, particularly in emergency settings, thereby improving its clinical applicability [43]. Cochrane et al. reported that a transfusion algorithm based on TEG 6s significantly reduced mortality and blood product wastage compared with standard care [44]. Building on these advances, the present study employed TEG 6s on arrival to assess early coagulation abnormalities in trauma patients.

Our study adds to the existing literature by refining the prognostic interpretation of TEG 6s obtained at hospital arrival in trauma patients with massive hemorrhage. Rather than evaluating TEG as a therapeutic guide, we examined its role as an early prognostic tool before substantial resuscitative interventions were initiated. By excluding patients with severe traumatic brain injury and adjudicating causes of death, we were able to interpret TEG abnormalities primarily in the context of hemorrhage-related physiology. Importantly, we moved beyond composite TEG abnormality and demonstrated that individual TEG 6s parameters exhibit distinct, time-dependent prognostic associations. Low fibrinogen-related clot strength (CFF-MA) was associated with increased early transfusion requirements, whereas hyperfibrinolysis (elevated LY30) was linked to greater transfusion burden over 24 h. These findings suggest that arrival-based, parameter-specific TEG 6s abnormalities provide prognostic resolution across different phases of early trauma resuscitation, rather than merely indicating the presence of global coagulopathy.

This study has several limitations. First, it was a retrospective observational study conducted at a single institution, which may limit the generalizability of the findings. To mitigate this limitation, we performed multivariable analyses to adjust for major confounding variables and to derive more objective results. Second, the study included only patients who underwent TEG 6s testing, making it impossible to compare outcomes with trauma patients who experienced massive hemorrhage but did not receive TEG 6s evaluation. This introduces the potential for selection bias. Third, although TEG parameters are dynamic and can change over time, this study analyzed only the initial TEG 6s values obtained upon arrival. As a result, temporal changes in coagulation status were not reflected. Furthermore, the TEG 6s analyses were performed using a kaolin-based cartridge with an integrated heparinase channel as part of the routine institutional workflow. Recent evidence suggests that heparinase-based viscoelastic assays may fail to detect trauma-associated autoheparinization, thereby providing limited incremental diagnostic information in this specific context [45]. However, because the identification or interpretation of heparin-related effects was not a primary objective of this study, this limitation is unlikely to have materially affected the main findings. Finally, cause-specific mortality was not formally adjudicated in this study and was determined based on retrospective chart review. Therefore, the findings should be interpreted as reflecting outcomes in a trauma population predominantly influenced by hemorrhage, but not exclusively attributable to hemorrhage.

Conclusion

This study demonstrates that TEG 6s parameters obtained at hospital arrival are associated with prognostic information regarding mortality and early transfusion requirements in trauma patients undergoing massive transfusion without severe traumatic brain injury. By focusing on a hemorrhage-related trauma population, our findings support the clinical utility of TEG 6s for early risk stratification beyond conventional injury severity markers. In addition, component-specific abnormalities may provide clinically relevant information to support transfusion decision-making during resuscitation.

Author contributions

D.S. and J.K. contributed equally to this work. D.S. and J.K. collected the data and performed the statistical analysis. I.H. prepared the figures and tables. K.J. conceptualized the study and supervised the project. D.S. wrote the original draft, and J.K. revised the manuscript. All authors reviewed and approved the final manuscript.

Funding

This study was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (grant number: HI22C1806).

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to institutional privacy policies but are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.