Article Text
PDF
Abstract
Background Fibrinogen concentrate (FC) can be administered during massive transfusions to manage trauma-induced coagulopathy. However, its effectiveness in survival remains inconclusive due to scarce high-level evidence. This study aimed to investigate the hemostatic effects of FC regarding mortality in massive hemorrhage caused by trauma.
Methods This retrospective study analyzed 839 patients who received massive transfusions (red blood cells (RBCs) ≥5 units in 4 hours or ≥10 units in 24 hours) at a level I trauma center between 2015 and 2022. Patients who were transferred to other hospitals or were deceased upon arrival, suffered or died from severe brain injury, and were aged 15 years or less were excluded (n=334). 1:2 propensity score matching was performed to compare the ‘FC (+)’ group who had received FC in 24 hours (n=68) with those who had not (‘FC (−)’, n=437). The primary outcome was mortality, and the secondary outcomes included transfusion volume.
Results The variables for matching included vital signs, injury characteristics, prehospital time, implementation of resuscitative endovascular balloon occlusion of the aorta, and blood gas analysis results. The administration of FC did not significantly reduce or predict mortality (in-hospital, 24 hours, 48 hours, or 7 days). The FC (−) group received more units of RBC (25.69 units vs. 16.71 units, p<0.001, standardized mean difference [SMD] 0.595), fresh frozen plasma (16.79 units vs. 12.91 units, p=0.023, SMD 0.321), and platelets (8.76 units vs. 5.46 units, p=0.002, SMD 0.446) than the FC (+) group.
Conclusion The use of FC did not show survival benefits but reduced transfusion requirements in traumatic massive hemorrhages, highlighting a need for future investigations. In the future, individualized goal-directed transfusion with FC may play a significant role in treating massive bleeding.
Level of evidence IV, retrospective study having more than one negative criterion.
- fibrinogen
- blood transfusion
- hemorrhagic shock
- blood coagulation disorders
Data availability statement
All data relevant to the study are included in the article.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Hypofibrinogenemia increases mortality in trauma patients. Fibrinogen concentrate (FC) is used to manage this, but its effect on survival is inconclusive.
WHAT THIS STUDY ADDS
This study found that although FC reduces the need for blood transfusions, it does not significantly impact mortality rates in massively hemorrhaging trauma patients.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
These findings underscore the need for further studies and suggest a potential usage of FC in trauma care to achieve individualized goal-directed transfusion. These findings are worth considering for institutions that are not yet using FC in traumatic bleeding or are under-resourced.
Introduction
Initial hypofibrinogenemia is an independent predictor of in-hospital mortality in trauma.1 Early consumption and lysis of fibrinogen are linked to worsened trauma-induced coagulopathy (TIC) and adverse outcomes, including increased transfusion requirements and higher mortality rates.2 3 A recent open-label randomized controlled trial demonstrated that early fibrinogen supplementation is crucial in severe multiple trauma to prevent prolonged bleeding and multiple organ failure.4 Hence, there is consensus that fibrinogen concentrate (FC) needs to be administered to hemorrhaging patients once hypofibrinogenemia is documented by standard laboratory test (SLT) or viscoelastic hemostatic assay (VHA), although the triggering serum level varies by guideline.5 Therefore, administering FC can also be a rational therapeutic option for managing TIC.6
Several guidelines have advocated fibrinogen replacement with either FC or cryoprecipitate during massive transfusion protocol (MTP) implementation in trauma.7–13 However, others have yet to since clinical evidence regarding the survival gain of FC supplements remains inconclusive.14 15 Informed by prior observations and literature, this study hypothesized that using FC with massive transfusion may correlate with rapid control of hemorrhage or coagulopathy. This hypothesis was based on the potential of FC to enhance blood product utilization efficiency, thereby minimizing the risks associated with high-volume transfusions. Thus, this study aimed to evaluate the impact of FC administration on survival in massively bleeding trauma patients, with a propensity score matching (PSM) analysis to minimize the effect of patient group heterogeneity.
Methods
This study was a retrospective review of patients who received a massive transfusion after trauma. The level I trauma center involved in this study is a 1060-bed tertiary teaching hospital covering an area with approximately 2 million people. More than 2500 trauma patients are treated annually, and about 30% are severely injured (Injury Severity Score (ISS) >15). It established MTP according to the standards required for trauma center verification and has adhered to the protocol since 2020.16 We applied the protocol to the trauma patients with suspected or proven massive hemorrhage showing a low systolic blood pressure (SBP) lower than 90 mm Hg (figure 1).
A flow chart (A) and the details (B) of the massive transfusion protocol (MTP) at the level I trauma center involved in this study. The trauma team leader initiates each step sequentially, as required: emergency transfusion (ET) 1, then ET2, followed by massive transfusion (MT) 1 and MT2. If further transfusions are necessary, either MT2 or MT3 is repeated. This MTP reflects the reality of the platelet shortage, with platelets being supplied from the MT2. ABC, assessment of blood consumption; ABGA, arterial blood gas analysis; aPTT, activated partial thromboplastin; CBC, complete blood count; D/W, dextrose in water; e-FAST, extended focused assessment with sonography for trauma; FFP, fresh frozen plasma; Hb, hemoglobin; IO, intraosseous; IV, intravenous; N/S, normal saline; PC, platelet concentrate; PT, prothrombin time; RBCs, red blood cells.
Patient selection
We analyzed the electronic medical records of 18 405 patients registered to the trauma database of the institute from January 2015 to August 2022. We identified patients who received a massive transfusion, defined as receiving 5 or more units of red blood cells (RBCs) within 4 hours or 10 or more units within 24 hours.17 Then, we screened patients who received a massive transfusion subsequently. The following patients were excluded: (1) patients who were transferred from the trauma center to other hospitals or deceased upon arrival; (2) those who suffered or died from severe brain injury (Abbreviated Injury Scale (AIS) head score of 5); and (3) those who were 15 years old or younger.
Intervention
FC was administered to the massive transfusion recipients, which involves the transfusion of two or more units of RBCs, independent of initial serum fibrinogen level. Three grams of FC was dissolved in 150 mL of injection water (20 mg/mL) to be infused intravenously at a ≤5 mL/min rate over 30 minutes. The goal was to reach a serum fibrinogen level of ≥100 mg/dL. Additional administration of the FC was decided at the discretion of the trauma team leader. A more detailed protocol of the adjunct therapy, including other therapeutic agents, can be found in figure 1B.
Outcome measures
The primary outcome measure was mortality (in-hospital, 24 hours, 48 hours, and 7 days). Secondary outcomes included transfusion volume (the number of units of RBC, fresh frozen plasma (FFP), and platelets transfused). Other outcomes measured included hemostatic interventions, reason for death, length of hospital and intensive care unit stay, and thromboembolic events. For patients who died, their length of stay would be the duration from their admission to the hospital until their time of death. The prehospital time was defined as the time between injury and presentation to the trauma center. The outcomes were only analyzed during the patients’ hospitalization, and no post-discharge follow-up period was necessary for the purpose of the study. None of the variables we analyzed had missing data.
Statistical analysis
To minimize the impact of confounding factors, such as different injury panoramas and the progression of physiologic derangement between the groups, a 1:2 PSM analysis was used. Considering the total number of patients included in the study, propensity scores were estimated using multivariable logistic regression to avoid overfitting in the matching. The covariates used in the model were initial parameters regarding vital signs and consciousness (SBP, mean arterial pressure, Revised Trauma Score (RTS) and Glasgow Coma Scale (GCS) score); injury characteristics (ISS and AIS chest/abdomen); prehospital time; implementation of resuscitative endovascular balloon occlusion of the aorta; and results of initial blood gas analysis (pH, arterial oxygen pressure, base excess, and lactate). These variables were chosen as they were identified as potentially influential in determining whether FC was administered and affecting study outcomes. Covariate balance was evaluated by standardized mean differences (SMDs).
We compared baseline characteristics and clinical outcomes between the two groups before and after PSM using Χ2 tests for categorical variables and independent t-tests for continuous variables. To describe continuous variables, a mean with an SD and a median with an IQR were used for normal and non-normal distribution, respectively. We performed additional logistic regression analysis on the post-match data for the primary outcomes. A p value of less than 0.05 was considered statistically significant. All statistical analyses were performed using the R package (V.5.1-1; The R Project for Statistical Computing).
Results
Patient enrollment
A flow chart of the patient enrollment process is detailed below (figure 2). Among the 18 405 patients registered in the database, 505 were eligible for analysis and finally enrolled. These patients were divided into two groups based on whether or not they had received FC within the first 24 hours of care: the FC (+) group (n=68) and the FC (−) group (n=437).
A flow chart of the patient enrollment process. FC fibrinogen concentrate; RBC red blood cell.
Pre-match patient characteristics
The pre-match patient characteristics for the two groups exhibited specific differences (table 1A). Both groups had a similar average age, gender distribution, and prevalence of underlying diseases such as hypertension, diabetes mellitus, cardiac disease, and cerebrovascular disease. However, the FC (+) group showed a higher incidence of blunt and more severe trauma, as indicated by higher ISS, lower RTS, and increased scores on the AIS for the chest, abdomen, and extremities. This group also exhibited longer prehospital times and a higher incidence of cardiac arrest in the emergency department. Initial vital signs and laboratory tests showed more significant physiological derangement in the FC (+) group, including lower blood pressure, GCS score, higher lactate levels, and acidosis.
(A) The pre-match patient characteristics and (B) the pre-match clinical outcomes
Pre-match clinical outcomes
In examining the pre-match clinical outcomes between the FC (+) group and the FC (−) group, several key differences were found (table 1B). The FC (+) group demonstrated a lower requirement for platelet transfusion within the initial 24 hours (5.46 units vs. 7.1 units; p=0.049, SMD=0.236). They also underwent significantly more temporary abdominal closures and preperitoneal pelvic packing procedures, indicative of more complex injuries. Moreover, higher in-hospital and 24-hour mortality rates were noted in the FC (+) group (p=0.046, SMD=0.271; p=0.031, SMD=0.282, respectively), suggesting that the severity of trauma in this group might be more intense, despite their lower transfusion needs. Hemorrhagic shock was also a more common cause of death in the FC (+) group (p=0.018, SMD=0.309).
Post-match patients’ characteristics
For the post-match patient characteristics, no significant differences were found between the FC (+) and FC (−) groups in all variables (table 2A). Overall, both groups were well matched and comparable across the evaluated parameters.
(A) The post-match patient characteristics and (B) the post-match clinical outcomes
Post-match logistic regression predicting four mortality rates
Post-match clinical outcomes
There were significant differences in transfusion volume within the first 24 hours of post-match clinical outcomes (table 2B). The FC (−) group received more units of RBC (25.69 units vs. 16.71 units, p<0.001, SMD 0.595), FFP (16.79 units vs. 12.91 units, p=0.023, SMD 0.321), and platelets (8.76 units vs. 5.46 units, p=0.002, SMD 0.446) compared with the FC (+) group. Mortality rates, whether in-hospital, 24 hours, 48 hours, or 7 days, also did not differ significantly between the groups. Although transfusion requirements were significantly higher in the FC (−) group, other clinical outcomes, including emergency operations, mortality rates, causes of death, hospital stay duration, and thromboembolic events, did not show significant differences between the two groups.
Figure 3 outlines the pre-match and post-match SMD for the patient characteristics and clinical outcomes. Most SMD values were low post-match, addressing the efficacy of the PSM conducted in this study.
The pre-match and post-match standardized mean differences (SMDs) for the patient characteristics and clinical outcomes. AIS, Abbreviated Injury Scale; ED, emergency department; MODS, multiorgan dysfunction syndrome; PaO2, arterial oxygen pressure; PT-INR, prothrombin time-international normalized ratio; REBOA, resuscitative endovascular balloon occlusion of the aorta; RTS, Revised Trauma Score.
Post-match logistic regression analysis predicting mortality
Table 3 elaborates on the post-match logistic regression analysis outcomes, predicting four types of mortality rates: in-hospital, 24-hour, 48-hour, and 7-day mortality. Each mortality rate was assessed for its relationship with four variables: FC (+) versus FC (−), RBC, FFP, and platelets. The use of RBC showed a statistically significant association with in-hospital (adjusted OR (adj. OR) 1.07, 95% CI 1.02 to 1.12; adj. p=0.008), 24-hour (adj. OR 1.07, 95% CI 1.02 to 1.12; adj. p=0.007) and 7-day mortality (adj. OR 1.08, 95% CI 1.03 to 1.14; adj. p=0.001). Similarly, platelets exhibited a significant association in reducing in-hospital (adj. OR 0.94, 95% CI 0.89 to 0.98; adj. p=0.01), 24-hour (adj. OR 0.84, 95% CI 0.78 to 0.91; adj. p<0.001) and 7-day mortality (adj. OR 0.92, 95% CI 0.87 to 0.97; adj. p=0.001). However, FC (+) versus FC (−) did not show any statistically significant impact on any of the mortality rates in the adjusted models.
Discussion
MTP has been globally used in trauma practice since the American College of Surgeons provided high-quality evidence for the benefits of MTP.7 However, a lack of consistent practices and regular updates on MTP can hinder the best treatment outcome of trauma-related bleeding. The question of when and how to transition from a traditional fixed-ratio strategy to goal-directed resuscitation introduces another layer of complexity to this issue. The wide variability in hemostatic defects in different individuals highlights a need for individualized resuscitation guided by VHA.6 18
VHA, such as thromboelastography and rotational thromboelastometry, provides a real-time assessment of hypofibrinogenemia and allows targeted FC administration.19 20 VHA-guided blood transfusions were associated with fewer blood product transfusions and reduced mortality after trauma.21 However, limited access to the VHA is a barrier to goal-directed resuscitation at some institutions, and ours is one of them. Therefore, organizations with limited equipment and resources may find our findings instructive. Moreover, caution is needed when using VHA as it has limitations. It has potential bias from the properties of each device, and there is no robust evidence that the VHA is superior to SLT in the treatment of coagulopathy.20
Our study showed an association between FC administration and the reduced transfusion volume of RBC, FFP, and platelets. This finding aligns with several previous studies that proposed the role of FC in decreasing the need for transfusion of blood products.4,19 22–25 In the post-match logistic regression analysis of the factors predicting mortality, the use of RBC appeared to have a significant positive correlation with all four assessed mortality types in the adjusted models. Conversely, platelets showed an association with reducing in-hospital and 24-hour mortality. Although this might be due to the protective effect of platelet transfusions, it is more likely that the patients who died prematurely did not receive platelet transfusions due to inadequate platelet supply. Although higher transfusion volumes were observed in the FC (−) group, this might be attributed to several factors, such as more severe injuries or implementing different hemostatic interventions. Furthermore, the nature of our study design—a retrospective analysis—inherently limits causal inference. Therefore, this result should be interpreted as indicative of a correlation for future research rather than definitive evidence of causation.
In contrast, our study did not demonstrate a significant impact on mortality rates with FC usage despite lower transfusion needs. A previous multicenter study with a similar design has reported the same results.26 27 This can be attributed to the complex nature of trauma and its management regarding TIC, or the varied nature of the study cohorts, such as injury patterns, severity, comorbidities, and the timeliness and effectiveness of trauma care. Another aspect of this finding is that although FC helps control bleeding and reduces the need for transfusions, its effect may not correlate directly with mortality benefits due to the multifaceted nature of mortality in trauma patients. This assumption underscores the necessity for large-scale, randomized controlled trials to definitively determine the impact of FC on trauma-related mortality. Another study using PSM analysis reported increased survival in the group receiving FC within 1 hour of arrival but was limited by the small number of patients and matching variables and should be interpreted with caution.28 Our study is significant in that it is the only one to exclude the severe brain injury group, which may affect the progression of coagulopathy, among studies using the PSM assay on the same subject.
The single-center retrospective design is the most critical limitation of the present study. Other limitations include the FC group’s relatively small sample size and the potential selection bias from individual surgeon discretion. These may have also introduced survival bias, underscoring the need for further studies with larger, more diverse cohorts. Other major limitations include survival and selection bias from transfusion strategies based on the clinician’s judgment. Not all control patients were treated during the same period as the FC group because FC was introduced to our institution in the fifth year of the 8-year study period. The evolution of medical care and institutional protocols during this period could have influenced the utilization of FC and clinical outcomes. Moreover, we could not control for potential confounders, such as variations in initial resuscitation strategies or the timing and dose of FC administration. When a patient received more than two units of RBC, the FC was administered based on a subjective decision by the trauma team leader, not based on results obtained from VHA or an existing scoring system. Unfortunately, data on general complications during trauma care, serum fibrinogen levels, and the effect of other hemostatic supplements, such as cryoprecipitate or tranexamic acids, were absent and not studied. In addition, the inclusion of an early mortality (eg, within 24 or 48 hours of admission) group may have reduced the average length of hospital stay despite adverse outcomes. Separation of early deaths from the overall analysis or subgroup analyses would minimize the limitations in the future study. Lastly, the impact of COVID-19 was not controlled in this study during the pandemic.
In the field of trauma care, ongoing research continually improves our understanding of the most effective strategies for managing hemorrhagic patients. A recent study investigating the efficacy and safety of four-factor prothrombin complex concentrate (4F-PCC), which includes human coagulation factors II, VII, IX, X, and proteins S and C, shows how our knowledge about this area is growing.29 Although our study mainly focused on the role of FC, it is also essential to understand how various hemostatic agents like 4F-PCC work in different clinical situations. As trauma care continues to advance, in-depth studies are key to figuring out the best ways to resuscitate patients, thereby guiding the development of more refined and targeted therapeutic interventions.
Conclusion
Although our study suggests an association between FC and reduced transfusion requirements, the definitive impact of FC on mortality still needs to be defined. Future investigations are warranted to further establish this relationship. Simultaneously, integrating novel point-of-care coagulation tests into trauma care protocols would refine our understanding of FC usage and facilitate more effective, patient-specific resuscitation strategies. Pairing these tools with novel biomarkers of fibrinogen function could significantly improve our knowledge of TIC and refine therapeutic interventions, including FC administration.
Data availability statement
All data relevant to the study are included in the article.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and the institutional review board (IRB) and the ethics committee of Dankook University Hospital approved the study (no. DKUH 2023-03-029). The IRB waived the written informed consent for publication from the patient, because we conducted a retrospective review of the chart.
Footnotes
Contributors DHK developed the study concept and design as guarantor, revised the article, and had responsibility for the overall content. YH performed statistical analyses and wrote the article. SWC contributed to developing the study design and interpretation of data. SWL and DSM contributed to data acquisition.
Funding This study was supported by Research Program of Korean Association for Research, Procedures and Education on Trauma for 2021 (KARPET-21-00).
Competing interests DHK has received honoraria for lecturing and expenses for travel and hotel accommodations from the GC Pharma Corporation (Yongin, Korea). SWC, DSM and SWL have received expenses for travel and hotel accommodations from the GC Pharma Corporation (Yongin, Korea). YH has no conflict of interest. Study design, conduction, data analysis, and preparation of the article were independently done by the authors with no outside influence.
Provenance and peer review Not commissioned; internally peer reviewed.