Evaluation of the Potential for Improvement of Clinical Outcomes in Trauma Patients with Massive Hemorrhage by Maintaining a High Plasma-to-Red Blood Cell Ratio during the First Hour of Hospitalization

Several reports indicate that early plasma transfusion may promote survival and reduce the incidence of traumatic coagulopathy in situations of massive bleeding. Consequently, it is recommended to maintain a plasma and RBC transfusion ratio between 1 : 1 and 1 : 2 at the start of admission. This retrospective study examined the effect of an early high plasma : RBC ratio on mortality rates by adopting a massive transfusion protocol (MTP) that forced an early and rapid issue of plasma products. Patients who received massive transfusions at a single trauma center between January 2014 and May 2020 were included in the study. A new protocol was established in January 2020, wherein a fixed amount of plasma was issued following MTP activation. Patients who underwent massive transfusions before and after the adoption of the new protocol were compared. In total, 1059 patients met the inclusion criteria. Fifty-one patients who underwent MTP were propensity score-matched with the patients who received a nonprotocolized massive transfusion. The MTP group had a higher plasma : RBC ratio at 1 h (0.8 vs. 0.2) and 4 h of hospitalization (1.1 vs. 0.6), with no significant between-group difference in the plasma : RBC ratio at 24 h of hospitalization. The MTP group had a lower 24 h mortality rate than the control group. There was no significant difference in the 30-day mortality. Using MTP to achieve a high plasma : RBC ratio in the early period of hospitalization appeared to affect 24-hour mortality; however, 30-day mortality did not change.


Introduction
Massive transfusion is defned as follows: (1) the replacement of the total blood volume in 24 h, (2) the replacement of 50% of the blood volume in 3 h, or (3) an emergency transfusion for a clear case of critical bleeding. Critical bleeding refers to when at least 4 units of red blood cells (RBCs) are transfused within 4 h or blood is transfused at 150 mL/min in an adult [1]. Several reports suggest that early transfusion of plasma and platelets can beneft survival and reduce the incidence of traumatic coagulopathy in massive bleeding situations [2][3][4][5]. A Pragmatic, Randomized Optimal Platelet, and Plasma Ratio (PROPPR) study achieved rapid hemostasis and fewer deaths attributed to exsanguination [6]. Based on the results of this study, the Trauma Quality Improvement Program (TQIP) of the American College of Surgeons (ACS) recommends maintaining a plasma and RBC transfusion ratio between 1 : 1 and 1 : 2 at the beginning of admission [7]. However, since plasma is usually stored in a frozen state, its prompt administration similar to that of RBC is difcult; hence, administering enough plasma at the beginning of hospitalization during traumatic coagulopathy is considerably challenging.
Tus, this study aimed to achieve a high plasma-to-RBC ratio in the early phase of massive transfusion using a newly updated massive transfusion protocol (MTP) with a quick issue of plasma products at a level-I trauma center. Furthermore, we aimed to compare the survival beneft of patients who received transfusions using the new MTP with that of patients who received transfusions according to the previous protocol.
Tis study was approved by the Institutional Review Board of Ajou University (AJIRB-MED-MDB-21-373), and the requirement for patient consent was waived owing to the retrospective study design.

Materials and Methods
Adult patients aged ≥15 years hospitalized at the Ajou University Trauma Center between January 2014 and May 2020 who received a massive transfusion were included in the study. Te Ajou University Trauma Center is a level-I trauma center designated by a governmental health organization that covers approximately 9 million people across the South Gyeonggi Province, South Korea. Te center admits approximately 2,000 patients every year, and 45% of all the patients admitted to the center have an injury severity score (ISS) of ≥16 points.
Massive transfusion was defned as a transfusion of ≥10 units of RBC in the frst 24 h of hospitalization or 4 units of RBC in the frst 4 h of hospitalization [1,7]. Te protocol used previously was merely a "recommendation" for transfusing plasma as promptly as possible for the maximum number of RBCs used (Supplement 1). However, it was not efcient since the composition and quantity of each MTP batch were randomly determined each time. Tis was the main cause of the logistic problem, resulting in the plasma delivery delay. Te protocol updated in January 2020 forced a fxed ratio of plasma to be drawn from a blood bank according to the RBC prescription amount in the MTP situation. Moreover, the composition of the batch was decided for an amount that could be promptly thawed and provided in consensus with the blood bank (Supplement 2). Demographic and clinical data were obtained from electronic medical records, which were collected by the Korean Trauma Data Registry and the hospital. All trauma deaths were reviewed by a mortality review committee operating for quality improvement, and the "types of death" were determined according to the results of the meeting.

Statistical Analysis.
All the statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) v24. Independent t-tests were performed to measure the diferences between continuous variables, and a chisquare test was used to measure the diferences between categorical variables. Kaplan-Meier curves were used for the survival analysis, and p < 0.05 was considered statistically signifcant. Te control group was matched with the MTP group based on propensity scores in a 1 : 1 ratio using age, sex, initial systolic blood pressure, and ISS (Injury Severity Score). Te matching tolerance was set to 0.01 to minimize the diference between the matched control group and the MTP group. Additional analyses, including mortality analysis, were performed using the control group thus created.

Results
A total of 16,305 patients were admitted to the level-I trauma center and enrolled in the Korean Trauma Data Bank from January 2014 to May 2020. Of these, 1,093 patients who received massive transfusions were enrolled in this study. We excluded 11 patients who were considered "dead on arrival" with a heart rate of 0, systolic blood pressure of 0, and Glasgow coma scale score motor component of 1 [8]. In addition, 22 patients aged ≤14 years and one patient whose transfusion amount was not precisely recorded was excluded. Finally, the remaining 1,059 patients were deemed eligible and included in this study ( Figure 1). Table 1 presents the general characteristics of the 51 patients who received MTP and the 1,008 patients who received nonprotocolized massive transfusion (NPMT). Te MTP group was younger and had a lower initial blood pressure than the NPMT group. Te MTP group also had a higher percentage of patients with the assessment of blood consumption scores ≥2 points, a higher ISS, and a higher mean score for extremity on the abbreviated injury scale (AIS) than did the NPMT group (Table 1).
Te NPMT group was matched with the MTP group based on propensity scores in a 1 : 1 ratio using age, sex, initial systolic blood pressure, and ISS. After matching, we found no signifcant diferences between the two groups in most variables, including hemoglobin level, for which differences were observed before matching. Diferences in the mean scores for extremities on the AIS remained after matching ( Table 2).
In a comparison of the initial blood transfusion time and the transfusion amount and ratio over time, the mean initial fresh frozen plasma (FFP) transfusion time was 57.5 minutes shorter in the MTP group than in the control group; moreover, the MTP group demonstrated signifcantly higher RBC and FFP transfusion amounts at 1 h of hospitalization (Table 3).
Te MTP group had a higher FFP-to-RBC ratio at 1 h of hospitalization than the control group (0.8 vs. 0.2). Te group also had a higher FFP-to-RBC ratio in the frst 4 h of hospitalization than the control group (1.1 vs. 0.6). No diference was found in the FFP-to-RBC ratio in the frst 24 h of hospitalization between the two groups ( Figure 2).
During the frst 1 h and at 4 h, the plasma-to-RBC ratio was low, and the cumulative ratio at 24 h reached 1 : 1 in the NPMT control group, whereas in the MTP group, the plasma-to-RBC ratio was maintained at 1 : 1 from the frst 1 h, at 4 h, and until 24 h.
In the comparison of the outcomes, the MTP group had a signifcantly lower 24 h mortality rate than the control group. No diferences were found based on the 30-day mortality, other outcomes, and procedures for hemostasis, including surgery and angioembolization, between the two groups (Table 4, Figure 3).
Te 30-day mortality was not signifcantly diferent; however, the 24 h mortality was signifcantly lower in the MTP group than in the NPMT group.

Discussion
Several studies dealing with transfusion strategies for patients with massive bleeding trauma report the advantages of the fxed-ratio transfusion strategy. Tus, many trauma centers have adopted an MTP using a 1 : 1 transfusion ratio. However, observational studies examining the efect of a high plasma-to-RBC ratio in trauma patients are subject to survivor bias since plasma administration generally starts after RBC administration [9]. Survivor bias or survivorship bias is a type of selection bias based on the logical error of concentrating on the people or things that made it past a selection process and overlooking those that did not, typically owing to their lack of visibility. Based on the survivor bias, the results can be interpreted as having an association between high plasma administration and a high survival rate merely because the survivors had the opportunity to receive more plasma. In a study of 806 patients analyzed using logistic regression, Scalea et al. [10] demonstrated that a high plasma-to-RBC ratio had no benefts on survival except in cases of early death. However, the duration of intensive care unit hospitalization was not investigated, and the mortality rate in the massive transfusion group was too low (6%). A review of studies on massive transfusion performed on trauma patients from 1966 to 2011 reported that 11 of the 26 included articles were prone to survival bias [11]. Many studies on trauma resuscitation reported that most of the early deaths were due to bleeding within 2-3 h of hospitalization [2,5,[12][13][14]. Terefore, to examine the efect of early plasma administration via a fxedratio transfusion without survival bias, the plasma-to-RBC transfusion ratio should be examined at 2-3 h rather than at 24 h of hospitalization. Terefore, to circumvent the risk of survivorship bias, the MTP group (which received sufcient plasma during the frst hour of the patient's admission) was set as the study group.
In this study, the MTP using a fxed-ratio transfusion strategy maintained a stable plasma-to-RBC ratio in the frst 1 and 4 h of hospitalization. Before the adoption of the fxedratio transfusion strategy, most plasma transfusions were administered after the frst 4 h of hospitalization, thus, failing to reach the 1 : 1 ratio in the frst 1-4 h of hospitalization. Although both groups reached a plasma-to-RBC ratio of 1 : 1 after 24 h of hospitalization, the fxed-ratio transfusion group had lower 24 h mortality, while several studies report the beneft of a high plasma-to-RBC ratio in 12-and 24-hour survival rates following trauma [1,9], others do not report a 30-day survival beneft [8]. No statistical diference was observed in the 30-day mortality in this study. Although the "PROPPR study," a large-scale, multicentric randomized clinical trial, did not report the overall mortality beneft of the strategy of using a high plasma-to-RBC ratio, the reduction in mortality due to exsanguination was also noted in its results [10].

Emergency Medicine International
Tere is a lack of conclusive studies on optimal plasmato-RBC ratios and specifc patient subgroups for which early plasma administration may provide clear benefts. However, early plasma transfusion has been reported to reduce trauma-induced coagulopathy and systemic endothelial damage caused by traumatic hemorrhage [10,11]. Tis strategy is efective against hyperfbrinolysis, one of the most lethal phenotypes of trauma-induced coagulopathy [12]. Large-scale studies on the MTP for patients with severe traumas, such as "PROMMT [5]" and "PROPPR [6]," clearly demonstrated the importance of early plasma administration. However, there are obstacles to early plasma administration, such as logistical problems related to universal donor AB plasma, the short expiry date of thawed plasma, and a potential increase in the blood disposal rate [15]. Tese obstacles hinder early plasma administration in many trauma centers that use their own protocols adapted from an MTP.
In the trauma center where this study was conducted, frequent delays in FFP delivery were discovered in the quality improvement activities, which was the reason for initiating the work for updating the previous transfusion protocol. Tis study was conducted to implement such a quality improvement program. Te trauma center in which this study was conducted computerized the blood application process after updating the MTP and determined the   Emergency Medicine International optimal amount of FFP to avoid thawing delays and deliver promptly upon request. Furthermore, the blood bank, nursing department, blood transportation workforce, and the quality improvement team of the trauma center worked together to improve the blood-issuing processes. Te initial time of administration of the type-matched FFP was signifcantly shortened. Tis study achieved a fxed-ratio transfusion in the early period of hospitalization merely by improving the logistics of the type-matched plasma-issuing process without the use of thawed universal plasma and signifcantly improved the mortality rates.
Our study has some limitations. First, we found no improvement in platelet administration, which is another major blood product used in the 1 : 1 : 1 balanced transfusion. Batch administration was not possible based on the number of transfused RBCs, due to insufcient storage of the platelet concentrates. Second, all types of deaths were included in the mortality analysis, and deaths attributable to causes other than hemorrhage, such as severe brain trauma, could not be excluded. As excessive bleeding leads to sudden death from a haemorrhagic shock or death from multiorgan failure resulting in an irreversible ischemic insult on multiple vital organs; hence, bleeding cannot be completely excluded as    the cause of death. Tird, due to the retrospective nature of the study, we could not control the efects of other variables, excluding the implementation of the improved MTP using a fxed-ratio transfusion, on the study and control groups. To overcome this limitation, propensity score-matching using early vital signs and other variables related to trauma intensity was used. However, the small diference in AIS scores at extremity was not statistically fully adjusted. Finally, the control and experimental groups in this study were not collected at the same time. In the meantime, there have been several medical advances, so the improvement in outcomes may not all be due to the efect of the revised MTP.

Conclusions
In conclusion, we updated the MTP to maintain plasma levels similar to RBCs in the frst and fourth hours, which did not lower the 30-day mortality rate in trauma patients requiring massive transfusions but did appear to afect 24hour mortality.

Data Availability
Te datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.