search for


Analysis of Massive Transfusion for Trauma Patients and Non-Trauma Patients in a Tertiary Hospital
단일 3차 병원에서 외상환자와 비-외상환자의 대량수혈 분석
Korean J Blood Transfus 2016;27:237−246
Published online December 30, 2016;
© 2016 The Korean Society of Blood Transfusion.

Hyerim Kim1,2, Dong-Won Yoo1,2, Kyung-Hwa Shin1,2, Hyun-Ji Lee2,3, and Hyung Hoi Kim1,2,4
김혜림1,2, 유동원1,2, 신경화1,2, 이현지2,3, 김형회1,2,4

1Department of Laboratory Medicine, Pusan National University Hospital, Busan, Korea,
2Department of Laboratory Medicine, School of Medicine, Pusan National University, Yangsan, Korea,
3Department of Laboratory Medicine, Pusan National University Yangsan Hospital, Yangsan, Korea,
4Medical Research Institute, Pusan National University Hospital, Busan, Korea
1부산대학교병원 진단검사의학과,
2부산대학교 의학전문대학원 진단검사의학교실,
3양산부산대학교병원 진단검사의학과,
4부산대학교병원 의생명연구원
Hyung Hoi Kim Biomedical Research Institute, Pusan National Univestiy Hospital Clinical Trial Center (CTC), Department of Laboratory Medicine and BioMedical Informatics Unit Pusan National University School of Medicine, 1-10 Ami-dong, Seo-gu, Busan 49241, Korea Tel: 82-51-240-7418, Fax: 82-51-247-6560, E-mail:
Received November 23, 2016; Revised December 9, 2016; Accepted December 9, 2016.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.


Management of patients with massive hemorrhage often requires the massive blood transfusions. However, few studies have investigated the effects of massive transfusions on non-traumatic patients. Therefore, this study analyzed mortality and descriptive data for patients receiving massive transfusion, including non-trauma patients and trauma patients.


We reviewed a retrospective audit of massive transfusions to investigate the major causes, patient characteristics, ratio of the blood components, and the mortality of massively transfused patients. The analysis was performed using electronic medical records collected from January 2010 to December 2013. Patients who had received a massive transfusion (≥10 units of RBCs within 24-hours) were categorized into trauma and non-trauma patients. We calculated the ratio of blood components and investigated the relationship between ratio and mortality. Descriptive statistics were used to characterize the patients and the indications.


A total of 532 massive transfusions were performed, including 187 trauma and 345 non-trauma patients. The overall mortality rate was 32.0%, encompassing 36.4% of the trauma patients and 29.6% of the non-trauma patients. The mortality in trauma patients was significantly reduced (P<0.001) within the first 48-hours compared with that in non-trauma patients, which was due to the high FFP: RBC ratio transfusion. The annual FFP: RBC ratio in trauma patients showed an increasing trend. Non-trauma patients showed no relationship between mortality and procedure indication/blood component ratio.


We report clinical data pertaining to massive transfusions. Annual increasing FFP: RBC ratio in trauma patients was associated with a decreasing mortality. Non-trauma patients showed heterogeneous characteristics and a lower FFP: RBC ratio than trauma patients.

Keywords : Massive hemorrhage, Massive transfusion, Trauma, Non-trauma, Mortality

Massive hemorrhage is a significant challenge for all clinicians, and interdisciplinary therapeutic approaches to their treatment must be decided within a limited time frame. Massive transfusion, which has historically been defined as the replacement of blood by the transfusion of 10 units of red blood cells (RBC) in 24 hours, had been conducted in cases of massive and uncontrolled hemorrhage.

In the 1970s, massive transfusion was associated with mortality rates of more than 90%, but advances in care have led to a dramatic decrease in mortality to 30%∼70%.1) Massive hemorrhage is a common complication in some clinical settings. In traumatic injury, hemorrhage is one of the leading causes of morbidity, and it is responsible for almost 50% and up to 80% of deaths occurring within 24 hours of injury and deaths during operation, respectively. Moreover, cardiovascular procedures frequently results in massive bleeding,2) and postpartum hemorrhage events could occur after delivery.3) In these situations, supporting blood components while controlling massive hemorrhage is critical. However, the majority of literature has focused on massive transfusion due to traumatic hemorrhage, while there is little information available regarding massive transfusion in other situations.

There has been recent concern regarding the excessive application of massive transfusion in non-trauma patients who do not ultimately require massive transfusion.4) However, there are virtually lack data available on massively transfused non-trauma patients. To address the paucity of data regarding massive transfusion for non-trauma patients, we reviewed all cases of massive transfusion, regardless of trauma relation. Specifically, we collected massive transfusion data regarding both trauma and non-trauma cases and evaluated the clinical characteristics of patients in both groups. In addition, associations between in-hospital mortality and baseline risk factors, procedural risk factors, and transfused blood component ratio were assessed.

Materials and Methods

1. Study design

This retrospective study was based on data acquired from a tertiary referral hospital that has 1,104 beds and used over 20,000 RBC units per year, from January 2010 to December 2013. Over 500 patients who were given ≥10 red blood cell units within 24 hours, including all massively transfused patients categorized as both trauma and non-trauma. We identified trauma patients as those with a potentially severe physical injury caused by an external source (falling, automobile crash, knife wound, etc.) that required emergency medical intervention to prevent loss of life or limb or substantial, permanent physical impairment. Non-trauma patients were identified as those who suffered moderate or minor injuries, as well as heart attacks, strokes, other internal conditions such as chronic infectious diseases or surgical procedures.

2. TASH scoring system

The Trauma-Associated Severe Hemorrhage (TASH) score is recognized as an easy-to-calculate and valid scoring system to predict an individual’s probability for massive transfusion and thus ongoing life-threatening hemorrhage at a very early stage after severe injury. The TASH scoring system uses seven independent variables to identify patients who will require a massive transfusion.5) These include blood pressure, gender, hemoglobin, focused abdominal sonography for trauma, pulse, base excess, and extremity or pelvic fractures and are weighted. Overall, 16 scores need to be summed to calculate the total score with a possible range of scores from 0 to 23. Recently, a threshold of 18 points as cut off, which corresponded to a 50% probability of massive transfusion in trauma patients.6)

3. Ratio of transfused blood components

Each transfusion requirement for fresh frozen plasma (FFP), platelets (PLT) and RBCs in the massive transfusion were used to determine the FFP: RBC or PLT: RBC ratio. As this was a continuous variable, a specific cut point was required to categorize patients into the comparison groups. Those patients who received at least a ratio of 1:1.0 were categorized as the high ratio group. Patients who received a transfusion ratio of 1:1.0 or less were categorized as the low ratio group.

4. Surgical procedure categorization

The surgical procedure is a reliable predictor of massive transfusion and mortality; thus, it was important to adjust for this variable when assessing the relationship between massive transfusion and outcome. We categorized patient procedures based on 231 clinically meaningful categories by using the clinical classification software (CCS) procedures for ICD-9-CM developed by the Agency for Healthcare Research and Quality.7) We then aggregated the identified CCS categories into one of the following 16 procedure categories: nervous, endocrine, eye, ear, nose, respiratory, cardiovascular, hemic and lymphatic, digestive, urinary, male genital, female genital, obstetrical, musculoskeletal, integumentary, and miscellaneous procedures. Since some of the 16 procedure categories of the patients had low frequencies (n<8), we combined them into a single group in this study to simplify the table.

5. Statistical methods

For each of the outcomes, the proportion of patients experiencing the outcome and its associated 95% binomial exact confidence interval (CI) were estimated. A two-sided P value of <0.05 was considered to be statistically significant. Continuous, normally distributed data were reported as the means (standard deviation [SD]). The univariate association between RBC usage, FFP or PLT to RBC ratio and mortality was performed using the chi-square test, while the Mann-Whitney U or ANOVA was used to identify differences between ranked ordinal data. Statistical analysis was performed using the Statistical Package for Social Sciences version 22.0 (SPSS Inc., Chicago, IL, USA).


From 2010 to 2013, 499 patients received a total of 532 episodes of massive transfusion totaling 23,741 blood components. The mean age was 53.5 years, and the study population consisted of 322 men and 177 women. The overall in-hospital mortality was 32.0% and the 48 hr-mortality was 11.9%. Cases were divided into several subgroups, including trauma, death, and FFP: RBC ratio (Fig. 1). Each graph is divided into death and survivor patients, and each cell shows the number of patients.

Fig. 1.

The number of patients for each trauma or FFP: RBC ratio subgroup. The patients who received at least a ratio of 1:1.0 were categorized as the high ratio group and other patients who received a transfusion ratio of 1:1.0 or less were categorized as a low ratio group. Abbreviations: RBC, red blood cells; FFP, fresh frozen plasma.

1. Demographics of massively transfused patients

Among 532 massive transfusion cases, 187 (35.2%) were due to traumatic injuries and 345 (64.8%) were due to non-traumatic indications. Baseline demographic data comparing trauma with non-trauma in the massive transfusion group are presented in Table 1. Statistical analysis of age, gender, anesthesia level, elective or emergency operation, TASH score, and initial hemoglobin (Hb) level was conducted. There were differences between groups, with the trauma group having a relatively younger age, higher male portion, more frequent emergent operations, and higher TASH score. A total 409 operations were conducted, 298 of which were conducted as emergencies. Based on the American Society of Anesthesiologists (ASA) physical status, 50.7% (72 of 142) of the trauma and 44.0% (118 of 268) in non-trauma for whom general anesthesia was implicated were greater than class four. General anesthesia was performed more frequently in the trauma group (P<0.001), possibly because more patients in the non-trauma group were too sick to undergo surgery.

Characteristics in massively transfused episodes (n=532)

 Trauma (n=187)  Non-trauma (n=345) P value 
Age (years), mean (SD)47.7 (17.7)56.6 (16.9)<0.001*
 <18, N (%)3 (1.6)7 (2.0)
 18∼39, N (%)57 (30.5)56 (16.2)
 40∼64, N (%)96 (51.3)138 (40.0)
 65∼79, N (%)22 (11.8)129 (37.4)
 ≥80, N (%)9 (4.8)15 (4.3)
Male, N (%)132 (70.6)195 (56.5)<0.001
ASA classification, N (%)142 (75.9)268 (77.7)<0.001
Emergency operation, N (%)127 (67.9)171 (49.6)<0.001
TASH score, mean (SD)13.2 (6.4)7.8 (5.3)<0.001
Initial Hb, mean (SD)9.5 (2.6)9.8 (3.0)0.43

*P values based on the Mann-Whitney test;

P values based on one way ANOVA.

Abbreviations: SD, standard deviation; ASA, american society of anesthesiologists; TASH, trauma-associated severe hemorrhage score; Hb, hemoglobin.

2. Mortality and blood usage between trauma and non-trauma groups

There was no difference in overall mortality between the trauma and non-trauma groups; however, a difference in early mortality existed between the two groups. Specifically, the mortality within 12 hours in trauma patients (17.8%) was higher than that in non-trauma patients (9.2%), which was similar to the 2 day mortality of massively transfused patients (P<0.001). The number of RBC transfusions in patients who expired in admission (20.6 [12.9]) was univariably greater than in patients who survived (16.7 [8.5], P<0.001).

Comparison of trauma patients and non-trauma patients revealed a difference in transfused RBC units, with an average of 20.2 units being observed in trauma- and 16.7 units in non-trauma patients (Table 2). Massively transfused trauma patients received FFP: RBC and PLT: RBC at 0.74:1 and 0.61:1, respectively. Although the appropriate product ratio for non-trauma patients is unclear, our transfusion ratio is similar to the protocol goal of 1:1:1 (FFP: RBC: PLT), which was developed based on data pertaining to trauma patients.8) We didn’t faithfully carry out the guidance, even though better ratio was shown in trauma patients.

Mean units and ratios of transfused products and mortality in trauma and non-trauma patients

MortalityRBC (units)FFP/RBC ratioPLT/RBC ratio

TraumaNon-traumaP valueTraumaNon-traumaTraumaNon-traumaTraumaNon-trauma
N (%)N (%)
12 hr mortality34 (17.8)32 (9.2)<0.00123.719.30.750.580.310.32
48 hr mortality47 (24.6)64 (18.5)<0.00126.619.00.740.60.40.44
In-hospital mortality*68 (36.4)102 (29.6)0.11924.418.10.710.630.510.67
 201021/57 (36.8)21/115 (18.3)
 201120/44 (45.5)19/73 (26.0)20.918.80.720.570.770.77
 201215/37 (40.5)30/86 (34.9)22.816.90.830.700.650.94
 201312/39 (24.5)32/71 (45.1)19.915.20.820.640.600.87
P value0.2340.2620.1720.0540.5560.724

*Mortality indicates number of expired patients/number of massively transfused patients (%);

P value determined by a chi-squared test of mortality between trauma and non-trauma patients;

P value based on chi-squared tests of survivor and non-survivor between trauma subgroups using trauma-weight cases.

Abbreviations: RBC, red blood cell; FFP, fresh frozen plasma; PLT, platelet; hr, hour.

3. Mortality with FFP: RBC ratio in trauma patients

The trend of early mortality in trauma patients has decreased since 2012; however, the FFP: RBC ratio has increased with time (Fig. 2). There were no differences in the annual mean TASH score in the trauma and non-trauma groups (9.29 [6.4], P>0.05). According to the blood component ratio, two groups, a low and high ratio, were similar in age, gender, and overall mortality, but trauma sustained, mean TASH score, and average RBC usage was different (data not shown). Although no statistical differences in mortality were observed between those who received a high FFP: RBC ratio versus those who had a low FFP: RBC ratio, the group with the high FFP: RBC ratio had a reduced blood transfusion requirement (P<0.001). Moreover, the high FFP: RBC ratio group had low average RBC usage instead of a high mean TASH score.

Fig. 2.

Annual mortality (A) and FFP: RBC ratio (B) for early (within 2 days) mortality. (A) Mortality of trauma patients was higher than that of non-traumatic patient, except for 2013. (B) Annual FFP:RBC ratio in trauma patients showed an increasing trend. Abbreviations: See Fig. 1 for FFP, RBC and T_12h_M, 12 hours-mortality of trauma; N_12h_M, 12 hoursmortality of non-trauma; T_48h_M, 48 hours-mortality of trauma; N_48h_M, 48 hours-mortality of non-trauma; T_12h_FR, FFP:RBC ratio of trauma patients who died within 12 hours; N_12h_FR, FFP:RBC ratio of non-trauma patients who died within 12 hours; T_48h_FR, FFP:RBC ratio of trauma patients who died within 48 hours; N_48h_FR, FFP:RBC ratio of non-trauma patients who died within 48 hours.

4. Blood transfusion requirement according to surgical procedure categories

According to CCS categories, the individual diagnostic categories were 42.1% of cardiovascular procedures, 17.7% of musculoskeletal operations, 11.8% of gastrointestinal operations, and 5.8% of obstetrical complications. The cardiovascular procedural category included many non-trauma cases (84.8%) and the mean age was older than that of trauma patients. The significant high mortality in several categories disappeared after a partial correlation controlling for age. The categories with a lower proportion (below 50%) of traumatic patients showed higher mortality than traumatic patients, except for the non-procedure category (Table 3). These results could be explained by a large proportion of the categories showing high mortality.

Blood transfusion requirements according to disease categories (n=532)

ProcedureTotal patientsTrauma patientsRBCFFP/RBC ratioPLT/RBC ratio

NMortality (%)NMortality (%)Mean units

*A group of orthopedics, neurosurgery, urology, otolaryngology, plastics, and gynecology.

Abbreviations: RBC, red blood cell; FFP, fresh frozen plasma; PLT, platelet.


As massive hemorrhage is a highly emergent situation that can be life-threatening, many patients who received massive transfusion sustained an overall mortality rate of 32.0%. Mortality rate ranged was lowest (3.2%) for the obstetrical procedure and highest (69.0%) for patients with severe illnesses that could not undergo surgery. Previous studies of massive blood transfusions have mostly been based on trauma patients. For example, there are several clinical scoring systems that can predict the implications of receiving a massive blood transfusion in trauma patients, or known risk factors for prognosis such as the FFP:RBC ratio or sex.1,5,8) However, other large numbers of massively transfused patients are admitted due to massive non-traumatic hemorrhage.3,9) Therefore, there are clinical limitations to applying massive transfusion scoring systems such as the TASH score to non-trauma patients. In this study, the results showed a surprisingly low TASH score in both trauma and non-trauma patients, with a mean (SD) 13.2 (6.4) in trauma patients and 7.8 (5.3) in non-trauma patients. These discrepancies indicate that the predictive scoring system could not be utilized in our clinical situation. According to the TASH score definition, our patients may have received over-activation of massive transfusion, especially non-trauma patients. TASH score is selectively studied in some trauma cases, and the above definition may not reflect the acute resuscitative phase of treatment or non-trauma cases appropriately.5,6) The relationship between blood loss and administered blood volume will be most direct for a large number of RBC units. Use of the blood component during massive transfusion in the expired group was significantly higher than in the survivor group. This might be explained by survivors sustaining relatively mild blood loss relative to expired patients (data not shown, P<0.001).

Although most patients were massively transfused with a low FFP: RBC ratio and low PLT: RBC ratio, there was no significant difference between blood component ratio and mortality, which is contrary to the results of previous studies.10) Unexpectedly, non-trauma patients showed low FFP: RBC ratio and an even higher mean age than trauma patients. Overall usage of blood components during massive transfusion in the high FFP: RBC ratio group was significantly lower than in the low FFP: RBC ratio group. The hospital blood bank and physician hypothesized that a fixed blood component ratio would help reduce utilization of the overall blood component. For our trauma department, education regarding the implications of using a 1:1:1 ratio in massive transfusion started to improve the mortality of massively transfused patients from 2011.8) Thereby, we report a trend of decreasing mortality in trauma patients although non-trauma patients do not have company with inconstant policies on FFP: RBC ratio for non-trauma patients. These findings are supported by those of previous clinical studies that have provided evidence that a higher ratio of FFP: RBC is associated with an improved outcome.11)

The results of this study demonstrate the importance of education regarding a high ratio of FFP: RBC in the case of massive transfusion. Furthermore, efficient massive transfusion protocols must be established to control problems that occur in massive transfusions such as inconstant blood usage or component ratio, regardless of disease categories or trauma. Some known massive transfusion protocols have confirmed that the use of massive transfusion protocol is directly correlated with improved mortality and reduced number of complications in clinically injured patients who receive Rh-O pack RBC and subsequent laboratory tests.12-14) The first goal of the protocol is to improve mortality and reduce blood usage by establishing appropriate implication timing of massive transfusion.

In future studies, we will evaluate the implications of a standardized massive transfusion protocol in our hospital. After several years, we hope to report the appropriations and usefulness of a massive transfusion protocol in a tertiary referral hospital.15)


배경: 대량출혈 환자의 처치시 대량수혈을 필요로 하는 경우가 종종 있다. 하지만 외상환자에 비해 비-외상환자의 대량수혈 연구는 드물게 보고되어 있다. 본 연구에서는 대량수혈을 받은 비-외상환자와 외상환자를 포함하여 사망률과 설명적 데이터를 분석하였다.

방법: 대량수혈을 받은 환자들의 주요 원인, 환자 특성, 혈액제제비 및 사망률 등을 살펴보기 위해 후향적으로 대량수혈 사례를 검토했다. 2010년 1월부터 2013년 12월까지의 전자의무기록을 이용하였다. 대량수혈(24시간 내에 ≥10단위 적혈구제제 수혈) 환자를 외상환자와 비-외상환자로 분류하였다. 혈액제제비를 구하고, 비와 사망률의 상관관계를 확인하였다. 환자 특성과 대량수혈의 적응증에 대해서는 기술적 통계를 사용하였다.

결과: 총 532건의 대량수혈을 수행하였고, 이는 187건의 외상환자와 345건의 비-외상환자로 구분되었다. 전체 사망률은 32.0%였고, 외상환자군은 36.4%, 비-외상환자에서는 29.6%의 사망률을 보였다. 외상환자군에서 높은 신선동결혈장:적혈구 비로 수혈 받은 환자의 48시간 이내 사망률이 유의하게 낮았으며, 연간 신선동결혈장:적혈구 비는 증가하는 경향을 보였다. 비-외상환자군은 혈액제제비나 적응증이 사망률과 상관관계를 보이지 않았다.

결론: 외상환자군에서 연간 경향을 보면 신선동결혈장:적혈구 비가 증가하면서 외상환자의 사망률이 감소하는 경향을 보였다. 비-외상환자군은 이질성을 보이며 외상환자군에 비해 낮은 신선동결혈장:적혈구 비를 보였다.

  1. Kauvar DS, Lefering R, and Wade CE. Impact of hemorrhage on trauma outcome:an overview of epidemiology, clinical presentations, and therapeutic considerations. J Trauma 2006;60:s3-11.
  2. Rose AH, Kotzé A, Doolan D, Norfolk DR, and Bellamy MC. Massive transfusion--evaluation of current clinical practice and outcome in two large teaching hospital trusts in Northern England. Vox Sang 2009;97:247-53.
    Pubmed CrossRef
  3. Burtelow M, Riley E, Druzin M, Fontaine M, Viele M, and Goodnough LT. How we treat:management of life-threatening primary postpartum hemorrhage with a standardized massive transfusion protocol. Transfusion 2007;47:1564-72.
    Pubmed CrossRef
  4. Pepe PE, Potkin RT, Reus DH, Hudson LD, and Carrico CJ. Clinical predictors of the adult respiratory distress syndrome. Am J Surg 1982;144:124-30.
  5. Yücel N, Lefering R, Maegele M, Vorweg M, Tjardes T, and Ruchholtz S et al. Trauma Associated Severe Hemorrhage (TASH)-Score:probability of mass transfusion as surrogate for life threatening hemorrhage after multiple trauma. J Trauma 2006;60:1228-36.
    Pubmed CrossRef
  6. Maegele M, Lefering R, Wafaisade A, Theodorou P, Wutzler S, and Fischer P et al. Revalidation and update of the TASH-Score:a scoring system to predict the probability for massive transfusion as a surrogate for life-threatening haemorrhage after severe injury. Vox Sang 2011;100:231-8.
    Pubmed CrossRef
  7. Hcup ccs homepage on Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project (HCUP). [Online] last visited on 15 November 2016
    Pubmed CrossRef
  8. Borgman MA, Spinella PC, Perkins JG, Grathwohl KW, Repine T, and Beekley AC et al. The ratio of blood products transfused affects mortality in patients receiving massive transfusions at a combat support hospital. J Trauma 2007;63:805-13.
  9. You KY, Kim JY, Choi SH, Yoon YH, Lim CS, and Park SM. Prediction of massive blood transfusion at emergency department in non-trauma patients. Korean J Blood Transfus 2013;24:275-85.
  10. Sperry JL, Ochoa JB, Gunn SR, Alarcon LH, Minei JP, and Cuschieri J et al. An FFP:PRBC transfusion ratio >/=1:1.5 is associated with a lower risk of mortality after massive transfusion. J Trauma 2008;65:986-93.
    Pubmed CrossRef
  11. Choi GR, Choi SJ, Kim S, and Kim HO. An analytical study for cases associated with massive blood transfusion at a tertiary referral hospital. Korean J Blood Transfus 2014;25:123-31.
  12. Young PP, Cotton BA, and Goodnough LT. Massive transfusion protocols for patients with substantial hemorrhage. Transfus Med Rev 2011;25:293-303.
    Pubmed KoreaMed CrossRef
  13. Sinha R, Roxby D, and Bersten A. Experience with a massive transfusion protocol in the management of massive haemorrhage. Transfus Med 2013;23:108-13.
    Pubmed CrossRef
  14. Horst J, Leonard JC, Vogel A, Jacobs R, and Spinella PC. A survey of US and Canadian hospitals'paediatric massive transfusion protocol policies. Transfus Med 2016;26:49-56.
    Pubmed CrossRef
  15. Lee HT, Park PW, Seo YH, Ahn JY, Seo JY, and Jeong JH et al. Massive transfusion protocols for pediatric patients. Lab Med Online 2016;6:60-3.


December 2018, 29 (3)
Full Text(PDF) Free

Social Network Service

Cited By Articles

Author ORCID Information

Funding Information