ORIGINAL ARTICLE

Predicting the need for massive transfusion in trauma patients: The Traumatic Bleeding Severity Score Takayuki Ogura, MD, Yoshihiko Nakamura, MD, Minoru Nakano, MD, PhD, Yoshimitsu Izawa, MD, Mitsunobu Nakamura, MD, PhD, Kenji Fujizuka, MD, Masayuki Suzukawa, MD, PhD, and Alan T. Lefor, MD, MPH, Maebashi, Japan

The ability to easily predict the need for massive transfusion may improve the process of care, allowing early mobilization of resources. There are currently no clear criteria to activate massive transfusion in severely injured trauma patients. The aims of this study were to create a scoring system to predict the need for massive transfusion and then to validate this scoring system. METHODS: We reviewed the records of 119 severely injured trauma patients and identified massive transfusion predictors using statistical methods. Each predictor was converted into a simple score based on the odds ratio in a multivariate logistic regression analysis. The Traumatic Bleeding Severity Score (TBSS) was defined as the sum of the component scores. The predictive value of the TBSS for massive transfusion was then validated, using data from 113 severely injured trauma patients. Receiver operating characteristic curve analysis was performed to compare the results of TBSS with the Trauma-Associated Severe Hemorrhage score and the Assessment of Blood Consumption score. RESULTS: In the development phase, five predictors of massive transfusion were identified, including age, systolic blood pressure, the Focused Assessment with Sonography for Trauma scan, severity of pelvic fracture, and lactate level. The maximum TBSS is 57 points. In the validation study, the average TBSS in patients who received massive transfusion was significantly greater (24.2 [6.7]) than the score of patients who did not (6.2 [4.7]) ( p G 0.01). The area under the receiver operating characteristic curve, sensitivity, and specificity for a TBSS greater than 15 points was 0.985 (significantly higher than the other scoring systems evaluated at 0.892 and 0.813, respectively), 97.4%, and 96.2%, respectively. CONCLUSION: The TBSS is simple to calculate using an available iOS application and is accurate in predicting the need for massive transfusion. Additional multicenter studies are needed to further validate this scoring system and further assess its utility. (J Trauma Acute Care Surg. 2014;76: 1243Y1250. Copyright * 2014 by Lippincott Williams & Wilkins) LEVEL OF EVIDENCE: Prognostic study, level III. KEY WORDS: Trauma; hemorrhage; massive transfusion. BACKGROUND:

M

assive hemorrhage is a major cause of acute coagulopathy of trauma shock and early trauma death.1Y3 To correct the acute coagulopathy of trauma shock or dilution coagulopathy4 balanced resuscitation with red blood cells, fresh frozen plasma, platelet concentrates, and clotting factors has been recommended.3,5 Approximately 85% of major trauma centers have instituted massive transfusion protocols to transfuse blood products in the appropriate ratio in the United States.6 However, there are no specific criteria to activate a massive transfusion protocol. The activation of massive transfusion protocols is provider dependent, and great variability exists even in highvolume centers. Any possible survival benefit of receiving fresh frozen plasma or platelet concentrates is related to early

Submitted: November 13, 2013, Revised: January 17, 2014, Accepted: January 22, 2014. From the Advanced Medical Emergency Department and Critical Care Center (T.O., Y.N., M. Nakano, M. Nakamura, K.F.), Japan Red Cross Maebashi Hospital, Maebashi; and Departments of Emergency Medicine (T.O., Y.I., M.S.), and Surgery (A.T.L.), Jichi Medical University, Tochigi, Japan. This study was presented at the American Heart Association Resuscitation Science Symposium 2012 in Los Angeles, California. Address for reprints: Takayuki Ogura, MD, 371-0014 Asahi-cho 3-21-36, Maebashi, Gunma, Japan; email: [email protected]. DOI: 10.1097/TA.0000000000000200

administration of these products.3,7 A rapid and simple scoring system to guide the activation of massive transfusion protocols may help providers know when it is likely that the patient will require massive transfusion.8 There are a number of models to predict the need for massive transfusion. Although studies showed that the TraumaAssociated Severe Hemorrhage (TASH) score and Assessment of Blood Consumption (ABC) score are accurate in predicting the necessity of massive transfusion for trauma patients, a Japanese study reported that these scoring systems are not reliable in elderly patients.9 Since the society is rapidly aging and the number of elderly trauma patient is increasing in Japan, a new scoring system to predict the necessity of aggressive transfusion for trauma patients in any age group may improve outcomes. The aims of this study were to create a new scoring system and to compare it with the TASH and ABC scores. This new scoring system should be simple to calculate and should be assessed quickly in the trauma bay with easily available data to rapidly identify patients who will benefit from massive transfusion.

PATIENTS AND METHODS This study was conducted at a single institution, which cares for approximately 250 severely injured patients per year

J Trauma Acute Care Surg Volume 76, Number 5

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

1243

J Trauma Acute Care Surg Volume 76, Number 5

Ogura et al.

(Injury Severity Score [ISS] 9 16). There are 12 mixed intensive care unit beds, and approximately 140 trauma patients are admitted to this intensive care unit per year. The institutional review board approved the review of patient data before starting this study. Because of the nature of trauma epidemiology in Japan, all patients in this study experienced blunt traumatic injuries. Patients with out-of-hospital cardiac arrest or isolated head trauma were excluded from this study. In the first analysis, the development phase, the scoring system was created. We reviewed data from severely injured trauma patients admitted between January 2008 and December 2009. The following parameters were evaluated: age, sex, systolic blood pressure (after rapid infusion of 1,000-mL crystalloid), heart rate, Glasgow Coma Scale (GCS) score, hypothermia (temperature G 35-C), serum lactate, hemoglobin level, platelet count, prothrombin time, results of the Focused Assessment with Sonography for Trauma (FAST) scan, pelvic fracture, antithrombotic agents, ISS, and mortality. Massive transfusion was defined as the transfusion of 10 or more units of packed red blood cells, within 24 hours of the time of injury. (In Japan, 1 U of packed red blood cells is approximately 120 mL.) Massive transfusion predictors were identified by comparing the parameters listed earlier in patients who underwent massive transfusion (the MT group) with those who did not (the non-MT group). Variables composing the new scoring system were selected from massive transfusion predictors based on logistic regression analysis. Each variable was converted into a simple score based on the odds ratio. Multivariate logistic regression analysis was performed on data classified by severity with reference to previous reports.10Y18 The Traumatic Bleeding Severity Score (TBSS) was defined as the total of the five component scores, including age (G60, 0 points; 960, 6 points), systolic blood pressure (9110, 0 points; 100Y110, 4 points; 90Y100, 8 points; G90, 12 points), number of regions positive on FAST scan (0Y6 areas including pericardium, left thoracic, right thoracic, perihepatic, perisplenic and pelvis, 3 points each, 0Y18 points), pelvic fracture (Type A, 3 points; Type B, 6 points; and Type C, 9 points), and serum lactate concentration (G2.5, 0 points; 2.5Y5.0, 4 points; 5.0Y7.5, 8 points; and 97.5, 12 points), for a total score ranging from 0 to 57 (Fig. 1). In the second analysis (validation phase), data from severely injured trauma patients admitted between January 2010 and March 2012 were reviewed. Severe trauma was defined as an ISS of 16 or greater. All data were available for all patients reviewed in this study. To allow rapid calculation of the TBSS, an application was developed for iOS devices and is available in the Apple App Store (Apple Corp., Cupertino, CA). This application provides a rapid intuitive interface to calculate the TBSS.

Statistical Analysis

The Mann-Whitney U-test and W2 test were used to compare parameters, and the level of significance was set at p G 0.05. The predictive ability of the TBSS, TASH score, and ABC score for massive transfusion was evaluated by calculating the area under the receiver operating characteristic curve, sensitivity, and specificity. Receiver operating characteristic curve comparison was performed to compare the accuracy of

1244

the TBSS with the TASH or ABC score. High accuracy was defined as area under the receiver operating characteristic curve more than 0.9, moderate accuracy was defined as area under the receiver operating characteristic curve less than 0.9 but more than 0.7, and low accuracy was defined as area under the receiver operating characteristic curve less than 0.7.19 Goodness of fit of the models was tested using the Cox-Snell R2 and the Nagelkerke R2 tests. All statistical analyses in this study were performed using MedCalc (Ostend, Belgium).

RESULTS Data from 119 patients were evaluated in the development phase to develop the TBSS. Age, the number of regions positive on FAST scan, arterial blood serum lactate concentration on admission, the presence of a pelvic fracture, and mortality in the MT group were all significantly greater than in the non-MT group. The systolic blood pressure (after rapid infusion of 1,000 mL of crystalloid), hemoglobin concentration, platelet count, and prothrombin time on admission were significantly less in the MT group than in the non-MT group (Table 1). Age, systolic blood pressure after rapid infusion of 1,000 mL of crystalloid, the number of regions with positive FAST scan result, presence of pelvic fracture, and arterial blood serum lactate concentration on admission were selected as the variables to compose the TBSS. The score for each variable was then based on the odds ratio in the results of the multivariate logistic regression analysis (Table 2) and then added to calculate the TBSS. The scores for each component variable constituting the TBSS are shown in Figure 1. The data from 113 patients were then evaluated to validate the TBSS. The mean (SD) age was 56.5 (24) years, and 81% were male. The mean (SD) ISS was 31.8 (13.3), and all patients in the study experienced blunt traumatic injuries (Table 3). There were no significant differences in the characteristics of patients in the development phase and the validation phase (Table 3). The average TBSS of patients in the MT group was significantly greater (24.2 [6.7]) than in the non-MT group (6.2 [4.7]) ( p G 0.01). The area under the receiver operating characteristic curve, sensitivity, and specificity of the TBSS for massive transfusion was 0.985 (cutoff, 15 points), 97.4%, and 96.2% respectively; for the TASH score, values were 0.892 (cutoff, 8 points), 81.6%, and 78.2%, respectively; and for the ABC score, values were 0.813 (cutoff, 1 point), 79.0%, and 78.2%, respectively (Table 5). When comparing the receiver operating characteristic curve, the area under the receiver operating characteristic curve of the TBSS was significantly higher than both the TASH and ABC scores (both Bonferroni adjusted p G 0.05) (Table 4). In addition to the area under the receiver operating characteristic curve, we calculated the goodness of fit of the models using the Cox-Snell R2 and the Nagelkerke R2 tests. The results of the Cox-Snell R2 test show a p value of 0.637, 0.400, and 0.308 for the TBSS, TASH, and ABC scoring systems, respectively. These data suggest that all three models are a good fit for the data. The results of the Nagelkerke R2 test show p values of 0.887, 0.557, and 0.429 for the TBSS, * 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

J Trauma Acute Care Surg Volume 76, Number 5

Ogura et al.

Figure 1. Calculating the TBSS. SBP, systolic blood pressure (after rapid infusion of 1,000 mL crystalloid), AO, Arbeitsgemeinschaft fu¨r Osteosynthesefragen/Orthopedic Trauma Association classification.

TABLE 1. Demographic and Clinical Characteristics of Patients by MT Status in the Development Phase Age, mean (SD), y* Age, Q60 y† Sex (male)† Systolic blood pressure, mean (SD), mm Hg* Heart rate, mean (SD), beats/min* GCS score e 8† Temperature G 35-C† Serum lactate, mean (SD), mmol/L* Hemoglobin, mean (SD), mg/dL* Platelet count, mean (SD) 105/KL* Prothrombin time, mean (SD), %* No. regions with FAST scan, mean (SD), positive* Pelvic fracture† Antithrombotic agents† ISS, mean (SD)* Mortality†

Non-MT (n = 57)

MT (n = 62)

p

50.6 (26.6) 19 (33.3%) 47 (82.5%) 137.7 (28.2) 97.6 (24.2) 11 (19.3%) 3 (5.7%) 2.0 (1.3) 15.1 (15.5) 23.7 (8.8) 81.9 (13.3) 1.6 (0.8) 16 (28.1%) 3 (5.3%) 26.1 (76.5) 0 (0%)

64.0 (20.7) 43 (69.4%) 49 (79.0%) 100.34 (33.3) 99.5 (23.7) 15 (25%) 4 (9.1%) 5.0 (4.1) 10.4 (2.6) 18.4 (7.2) 60.8 (21.3) 2.2 (1.2) 37 (59.7%) 8 (12.9%) 30.1 (14.4) 19 (30.6%)

0.003 0.003 0.636 G0.001 0.660 0.603 0.798 G0.001 0.031 G0.001 G0.001 0.001 0.003 0.262 0.068 G0.001

*Mann-Whitney U-test. †W2 test. Systolic blood pressure after 1-L volume resuscitation.

* 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

1245

J Trauma Acute Care Surg Volume 76, Number 5

Ogura et al.

TABLE 2. The Result of Multivariate Logistic Regression Analysis and Simplified Scoring According to Odds Ratio Age (Q60 y) Stage 1 Stage 2 Systolic blood pressure Stage 1 Stage 2

Stage 3

Stage 4 No. regions with FAST (positive) Severity of pelvic fracture Stage 1 Stage 2 Stage 3 Serum lactate Stage 1 Stage 2 Stage 3 Stage 4

Coefficient

Odds Ratio

Simplified Score

95% Confidence Interval

1.8926

6.6367

Á 6 points

1.5010Y29.3443

Age Q 60 y Age G 60 y

Score 6 0

1.3294 Systolic blood pressure 9 110 mm Hg 100 mm Hg G systolic blood pressure e 110 mm Hg 90 mm Hg G systolic blood pressure e 100 mm Hg Systolic blood pressure e 90 mm Hg 1.1502

1.0687

Á 4 points

3.7786

1.1657Y12.2485 0 4

8

12 Á 3 points

3.1566

1.2202Y8.1773

[No. region with positive FAST scan result]  3 points = Á 3 points 1.1846Y7.1568

2.9116

Stable fracture patterns Incomplete unstable fracture patterns Complete unstable fracture patterns 1.4553 4.2853 Lactate G 2.5 mmol/L 2.5 mmol/L e Lactate G 5.0 mmol/L 5.0 mmol/L e Lactate G 7.5 mmol/L Lactate Q 7.5 mmol/L

3 6 9 Á 4 points

1.4009Y13.1129 0 4 8 12

Systolic blood pressure after 1-L volume resuscitation. Lactate, arterial blood serum lactate concentration at emergency department.

TASH, and ABC scoring systems, respectively, which also support the fact that the models are a good fit for the data.

DISCUSSION Despite improvements in the resuscitation of trauma patients with the appropriate use of blood products, one factor that may further improve patient outcomes is to identify

the patients most likely to require massive transfusion.7 The activation of massive transfusion protocols is provider dependent and shows great variability in their use even in highvolume centers.8 In this study, transfusions were similarly based on a provider-dependent decision. The mortality of patients in the validation phase was 12.4% in this study but may be improved if the decision to start massive transfusion is made earlier.

TABLE 3. Characteristics of Patients in Data Sets in the TBSS Development and Validation Phases Age (mean)* Sex, male** Blunt trauma** Antithrombotic agents** ISS, mean (SD)* Unit of packed red blood cells, mean (SD)* Mortality**

Development (n = 119)

Validation (n = 113)

p

57.6 (24.5) 96 (80.5%) 118 (99.2%) 11 (9.2%) 28.2 (12.1) 12.7 (13.8) 19 (16.0%)

56.5 (24.5) 92 (81.4%) 113 (100%) 9 (7.9%) 31.8 (13.3) 8.1 (13.8) 14 (12.4%)

0.157 0.885 0.329 0.885 0.242 0.224 0.701

*Mann-Whitney U-test. **W2 test.

1246

* 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

J Trauma Acute Care Surg Volume 76, Number 5

Ogura et al.

TABLE 4. Each AUC and Pairwise Comparison of ROC Curves AUC Sensitivity Specificity Difference in AUC 95% CI Significance

TBSS

TASH Score

ABC Score

0.985 (cutoff, 15) 97.4% 96.2% TBSS vs. TASH 0.093 0.041Y0.146 p G 0.001*

0.892 (cutoff, 8) 81.6% 78.2% TBSS vs. ABC 0.172 0.097Y0.248 p G 0.001*

0.813 (cutoff, 1) 79.0% 78.2% TASH vs. ABC 0.0791 0.009Y0.149 p = 0.027*

*Bonferroni adjusted p G 0.05. AUC, area under the ROC curve; ROC, receiver operating characteristic.

Development of the TBSS The aging of society is progressing rapidly, and the aging of the trauma patient has also progressed over time.20 Japan has the highest longevity in the world and is thus faced with a super-aged society with approximately 22% of the population over the age of 65 years.21 It is estimated that the number of elderly trauma patients will continue to increase in Japan. One characteristic of the TBSS is that it accounts for the patient’s age. Normal levels of blood pressure are not reassuring22 in geriatric trauma patients, and advanced age is a predictor of bleeding in patients with stable pelvic fractures.23 Age-related physiologic or anatomic loss of organ function,20,24 muscle atrophy, osteoporosis, and reduction in the average amount of subcutaneous tissue25 may lead to more serious effects of a traumatic injury. Many elderly people have comorbidities, which is one of the factors leading to a worse prognosis in these patients.26 Therefore, hemorrhage in elderly trauma patients can be serious even with a minor injury, and

it is difficult to predict the severity of hemorrhage. Consideration of the patient’s age is important to predict the necessity of massive transfusion and, we believe, contributes to the ability of the TBSS to predict the need for massive transfusion in these patients. The TASH score and the ABC score were developed with younger trauma patients (the mean [SD] age for the TASH score is 39.2 [19.06],27 and the mean [SD] age for the ABC score is 40 [1.8]8). The TBSS was created in the development phase of this study, with generally older patients (the mean [SD] age is 57.6 [24.5] years). Patients enrolled during the validation phase were also older (Table 1), which may partially contribute to the accuracy of the TBSS. Since the aging of society is progressing and the number of elderly trauma patient is increasing, the TBSS may be a more appropriate scoring system to predict the necessity of massive transfusion in trauma patients with an aging society, such as Japan.

TABLE 5. References Used to Classify the Severity of Potential Predictive Variables of Massive Transfusion Variable Systolic blood pressure

Reference Eastridge et al.10

Edelman et al.11

Lalezarzadeh et al.12

Pelvic fracture

Hamill et al.13

Eastridge et al.14 Magnussen et al.15

Lactate concentration

Jansen et al.16

Vandromme et al.17

Guyette et al.18

Comments Systolic blood pressure e110 mm Hg is a more clinically relevant definition of hypotension and hypoperfusion than is 90 mm Hg. Patients with a systolic blood pressure of 90Y109 mm Hg following trauma should be considered as a special group requiring aggressive resuscitation and surgery. Prehospital and emergency department hypotension is a strong predictor of in-hospital mortality and need for emergent surgical intervention in trauma patients. Embolization was required in 44.4% of the patients with fracture patterns indicative of major pelvic ligament disruption, whereas 7 of 18.4% of the patients without these fracture patterns required embolization. Transfusion of 6 U packed red blood cells in first 23 h of unstable fracture patterns group was greater than that of stable fracture pattern group. Higher-energy pelvic ring fractures classified as APC II or III, LC III, vertical shear, or combined mechanism require more frequent transfusion than other pelvic fractures. Nonsurvivors had significantly higher lactate levels than survivors, and mortality was significantly higher in patients with lactate levels of Q3.5 mmol/L compared with those with lactate levels of G3.5 mmol/L. Emergency department blood lactate level is a better predictor than is systolic blood pressure in identifying patients requiring significant transfusion. Lactate may improve the identification of patients who require monitoring, resources, and resuscitation.

* 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

1247

J Trauma Acute Care Surg Volume 76, Number 5

Ogura et al.

Clinical Parameters Constituting the TBSS The TBSS was created focused on the ability to quickly calculate the score in the trauma bay with rapidly available information. There were significant differences in age, systolic blood pressure after rapid infusion of 1,000 mL of crystalloid, the number of regions positive on FAST scan, presence of pelvic fracture, serum lactate concentration, hemoglobin concentration, platelet count, and prothrombin time between the MT and non- MT groups (Table 1). Therefore, age, systolic blood pressure after rapid infusion of 1,000 mL of crystalloid, the number of regions with positive FAST scan result, presence of pelvic fracture, and serum lactate concentration were selected as variables to constitute the TBSS. We set a limit on the number of variables composing the TBSS to allow a rapid and easy calculation in the trauma bay with a minimum number of laboratory values. Hemoglobin concentration was not selected as one of the clinical variables for the TBSS, although it can be determined rapidly on blood gas analysis. Moore et al.28 reported that hemoglobin concentration is not useful to predict the severity of the hemorrhage in trauma patients. Rainer et al.29 mentioned that decreasing hemoglobin concentration is a strong predictor for the necessity of massive transfusion. Thus, opinions regarding the usefulness of hemoglobin concentration as a predictor of massive transfusion are not unified. Similarly, platelet count was not selected as a one of the variables composing the TBSS. Brown et al.30 reported that admission platelet count was inversely correlated with 24-hour transfusion of packed red blood cells, but a normal platelet count may be insufficient after severe trauma. Platelet count may not be a useful predictor of massive transfusion. Prothrombin time or prothrombin time minus international normalized ratio (INR) is sometimes used as a parameter for predicting massive transfusion in trauma patients3,31 and is included in several algorithms. However, standard coagulation assays were available after a mean of 66 minutes (29Y174 minutes), according to a study of admission prothrombin time in trauma patients reported by David et al.32 Prothrombin time or prothrombin time minus INR is not quickly available as a predictor of massive transfusion. To further evaluate these three parameters that were not included in the final TBSS model, we constructed another set of models that include hemoglobin, platelet count, and prothrombin time, to generate a revised TBSS (data not shown). Using these parameters in TBSS did not improve the accuracy or predictive value of the model and, in some cases, resulted in a lower percentage of patients correctly classified. This further supports the validity of the five parameters used to calculate the TBSS, while excluding hemoglobin, platelet count, and prothrombin time. The TBSS accounts for the source and amount of bleeding by FAST scan and x-ray imaging and describes its severity by the lactate concentration and the systolic blood pressure. The clinical variables constituting the TBSS were classified based on their severity with reference to previously published reports (Table 5). Kimbrell et al.33 reported that patients 60 years and older have a high likelihood of active retroperitoneal bleeding in blunt trauma patients with significant pelvic fractures. Henry 1248

et al.34 similarly reported that 55 years and older trauma patients with pelvic fracture were more likely to undergo transfusion, and those undergoing transfusion required more blood even after adjusting for ISS. By considering these reports, 60 years or older was taken as the age limit for patients more likely to undergo massive transfusion in this study. Additional studies will be needed to more precisely define the appropriate age based on the magnitude of the risk for massive transfusion. The systolic blood pressure in trauma patients reflects the severity of the hemorrhage. Eastridge et al.10 proposed that the traumatic hypotension does not begin at 90 mm Hg but rather at 110 mm Hg. Therefore, minor, moderate, and severe hypotension was defined as a systolic blood pressure between 101 mm Hg and 110 mm Hg, 91 mm Hg and 110 mm Hg, and less than 90 mm Hg, respectively. The number of positive regions on FAST scan indicates the macroscopic distribution of intra-abdominal and/or intrathoracic hemorrhage and may be proportional to the severity of hemorrhage. The pelvic fracture pattern13Y15 and serum lactate concentration16Y18 also reflect the severity of hemorrhage. The risk of massive hemorrhage in a patient with an unstable pelvic fracture is higher than that in patients with a stable pelvic fracture. Similarly, an elevated serum lactate level can objectively indicate a mismatch in oxygen supply and demand in the peripheral circulation. The pelvic fracture pattern and the serum lactate concentration were classified based on the severity of each factor.

Comparison of the TBSS With Other Scoring Systems Compared with the TASH score or the ABC score, calculation of the TBSS uses different parameters. Early treatment of massive hemorrhage improved the prognosis in trauma patients.35 Convenience and rapid calculation are essential for a scoring system to be useful in predicting the need for massive transfusion. High accuracy is also important, since complications of transfusion (e.g., infectious disease,36 acute lung injury or adult respiratory distress syndrome,37,38 multiple-organ failure) may be severe.40 The ABC score is the simplest scoring system, but its accuracy is lower than that of the TBSS as shown in this study. ABC score is composed only of nonlaboratory data and is thus easily calculated in the trauma bay.8 It is composed of the mechanism of injury, positive FAST scan result, arrival systolic blood pressure, and arrival heart rate. The TASH score uses seven factors to calculate the total score, including the systolic blood pressure, hemoglobin level, presence/absence of intra-abdominal fluid, presence/absence of long bone/complex fractures, heart rate, base excess, and sex. This is notably different from the ABC score in that calculation of the TASH score, like the TBSS, requires laboratory data. The accuracy of the TBSS was higher than that of TASH score in this study (Table 4). In the PROMMTT study, investigators identified eight clinical parameters as triggers for the initiation of massive transfusion protocols, with one point assigned to each trigger.3 These parameters included INR, systolic blood pressure, hemoglobin, base excess, positive FAST scan result, pulse, temperature, and mechanism of injury. Similar to the TBSS and the TASH score, this system includes both laboratory and * 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

J Trauma Acute Care Surg Volume 76, Number 5

Ogura et al.

nonlaboratory values. Notably, the TASH score, the ABS score, or the PROMMTT triggers did not account for the age of the patient. FAST scan, the rapid infusion of crystalloid, and x-ray imaging of the pelvis are usually performed rapidly, and serum lactate concentration can be checked on blood gas analysis during this initial evaluation. Thus, the TBSS is easily calculated using easily available data, and the necessity of massive transfusion is determined without significant delay. To further simplify the process, we created a TBSS application for iOS devices. This application (TBSS) can be downloaded without charge from the Apple App Store and further simplifies the use of the TBSS. Of the scoring systems that have been described, a convenient system for calculation is available only for the TBSS. There are some limitations to the TBSS described in this study. Development and validation of the TBSS are based on retrospective data from a single center, with all patients experiencing blunt traumatic injuries. It is necessary to perform additional multicenter studies to validate the accuracy of TBSS and compare this scoring system with others. In particular, it will be important to determine if the use of the TBSS reduces the time necessary to implement massive transfusion protocols and determine what, if any, effect it has on patient outcomes. Further studies will also be needed to evaluate the utility of the TBSS in patients with injuries from a variety of mechanisms. Finally, the TBSS was developed based on the survival of patients in Japan, where a unit of packed red blood cells is approximately one third the volume of a unit in the United States (assuming an equal hematocrit). This difference will not affect the calculation of the TBSS at all but may require that the transfusion threshold is recalibrated in other countries and supports the need for further studies.

CONCLUSION The TBSS was developed to be rapidly calculated and to predict the need for massive transfusion in patients after traumatic injuries. It is rapidly determined from age, systolic blood pressure, results of the FAST scan, presence of a pelvic fracture, and serum lactate concentration, which are entered into an iOS application that quickly provides the TBSS. The TBSS has 97% sensitivity and 96% specificity for predicting the need for massive transfusion in patients after blunt trauma for a score of greater than 15 points. More clinical studies are needed to further validate the TBSS for predicting the need for massive transfusions and avoiding the problem of undertriage.

AUTHORSHIP T.O., Y.N., and K.F. collected the data, which T.O., M.Nakamura, and Y.I. analyzed. M.Nakano and M.S. contributed to the interpretation of result. T.O., Y.N., M.Nakamura, and A.T.L. drafted the manuscript, which all authors read and approved.

DISCLOSURE The authors declare no conflicts of interest.

REFERENCES 1. Bouillon B, Brohi K, Hess JR, Holcomb JB, Parr MJ, Hoyt DB. Educational initiative on critical bleeding in trauma. J Trauma. 2010;68: 225Y230. 2. Sauaia A, Moore FA, Moore EE, Moser KS, Brennan R, Read RA, Pons PT. Epidemiology of trauma death: a reassessment. J Trauma. 1995;38: 185Y193. 3. Callcutt RA, Cotton BA, Muskat P, Fox EE, Wade CE, Holcomb JB, Schreiber MA, Rahbar MH, Cohen MJ, Knudson M, et al. Defining when to initiate massive transfusion: a validation study of individual massive transfusion triggers in PROMMTT patients. J. Trauma Acute Care Surg. 2013;74:59Y68. 4. Ho AM, Karmakar MK, Dion PW. Are we giving enough coagulation factors during major trauma resuscitation? Am J Surg. 2005;190:479Y484. 5. Holcomb JB, del Junco DJ, Fox EE, Wade CE, Cohen MJ, Schreiber MA, Alarcon LH, Bai Y, Brasel KJ, Bulger EM, et al. PROMMTT Study Group. The prospective, observational, multicenter, major trauma transfusion (PROMMTT) study: comparative effectiveness of a time-varying treatment with competing risks. JAMA Surg. 2013;148:127Y136. 6. Schuster KM, Davis KA, Lui FY, Maerz LL, Kaplan LJ. The status of massive transfusion protocols in United States trauma centers: massive transfusion or massive confusion? Transfusion. 2010;50:1545Y1551. 7. Callcut RA, Johannigman JA, Kadon KS, Hanseman DJ, Robinson BR. All massive transfusion criteria are not created equal: defining the predictive value of individual transfusion triggers to better determine who benefits from blood. J Trauma. 2011;70:794Y801. 8. Nunez TC, Voskresensky IV, Dossett LA, Shinall R, Dutton WD, Cotton BA. Early prediction of massive transfusion in trauma: simple as ABC (assessment of blood consumption). J Trauma. 2009;66:346Y352. 9. Ohmori T, Murata A, Kanamaru A, Saisaka Y, Tanaka K, Ishihara J, Sugimoto K. Analysis of patients with multiple trauma receiving massive blood transfusionVaccuracy of prediction scoring system in massive blood transfusion. Nippon Gaishou Gakkai Zasshi. 2011;25:348Y352. 10. Eastridge BJ, Salinas J, McManus JG, Blackburn L, Bugler EM, Cooke WH, Convertino VA, Wade CE, Holcomb JB. Hypotension begins at 110 mm Hg: redefining ‘‘hypotension’’ with data. J Trauma. 2007;63: 291Y297; discussion 297Y299. 11. Edelman DA, White MT, Tyburski JG, Wilson RF. Post-traumatic hypotension: should systolic blood pressure of 90Y109 mmHg be included? Shock. 2007;27:134Y138. 12. Lalezarzadeh F, Wisniewski P, Huynh K, Loza M, Gnanadev D. Evaluation of prehospital and emergency department systolic blood pressure as a predictor of in-hospital mortality. Am Surg. 2009;75:1009Y1014. 13. Hamill J, Holden A, Paice R, Civil I. Pelvic fracture pattern predicts pelvic arterial haemorrhage. Aust NZ J Surg. 2000;70:338Y343. 14. Eastridge BJ, Starr A, Minei JP, O’Keefe GE, Scalea TM. The importance of fracture pattern on guiding therapeutic decision-making in patients with hemorrhagic shock and pelvic ring disruptions. J Trauma. 2002;53: 446Y451. 15. Magnussen RA, Tressler MA, Obremskey WT, Kregor PJ. Predicting blood loss in isolated pelvic and acetabular high-energy trauma. J Orthop Trauma. 2007;21:603Y607. 16. Jansen TC, van Bommel J, Mulder PG, Rommes JH, Schieveld SJ, Bakker J. The prognostic value of blood lactate levels relative to that of vital signs in the pre-hospital setting: a pilot study. Crit Care. 2008;12:R160. 17. Vandromme MJ, Griffin RL, Weinberg JA, Rue LW 3rd, Kerby JD. Lactate is a better predictor than systolic blood pressure for determining blood requirement and mortality: could prehospital measures improve trauma triage? J Am Coll Surg. 2010;210:861Y867, 867Y869. 18. Guyette F, Suffoletto B, Castillo JL, Quintero J, Callaway C, Puyana JC. Prehospital serum lactate as a predictor of outcomes in trauma patients: a retrospective observational study. J Trauma. 2011;70:728Y786. 19. Akobeng AK. Understanding diagnostic tests 3: receiver operating characteristic curves. Acta Paediatr. 2007;96:644Y647. 20. Victorino GP, Chong TJ, Pal JD. Trauma in the elderly patient. Arch Surg. 2003;138:1093Y1098. 21. Statistical Research and Training Institute, Ministry of Internal Affairs and Communications. Japan Statistical Yearbook 2008. Statistics Bureau, Ministry of Internal Affairs and Communications. This was published on

* 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

1249

J Trauma Acute Care Surg Volume 76, Number 5

Ogura et al.

22.

23.

24. 25. 26. 27.

28.

29.

30.

line at: http://www.stat.go.jp/english/data/jinsui/2008np/. Access date is 15 Feb 2014. Oyetunji TA, Chang DC, Crompton JG, Greene WR, Efron DT, Haut ER, Cornwell EE 3rd, Haider AH. Redefining hypotension in the elderly: normotension is not reassuring. Arch Surg. 2011;146:865Y869. Bramos A, Velmahos GC, Butt UM, Fikry K, Smith RM, Chang Y. Predictors of bleeding from stable pelvic fractures. Arch Surg. 2011;146: 407Y411. Aschkenasy MT, Rothenhaus TC. Trauma and falls in the elderly. Emerg Med Clin North Am. 2006;24:413Y432, vii. Pudelek B. Geriatric trauma: special needs for a special population. AACN Clin Issues. 2002;13:61Y72. Jacoby SF, Ackerson TH, Richmond TS. Outcome from serious injury in older adults. J Nurs Scholarsh. 2006;38:133Y140. Yu¨cel N, Lefering R, Maegele M, Vorweg M, Tjardes T, Ruchholtz S, Neugebauer EA, Wappler F, Bouillon B, Rixen D, Polytrauma Study Group of the German Trauma Society. Trauma Associated Severe Hemorrhage (TASH)-Score: probability of mass transfusion as surrogate for life threatening hemorrhage after multiple trauma. J Trauma. 2006;60:1228Y1237. Moore FA, McKinley BA, Moore EE, Nathens AB, West M, Shapiro MB, Bankey P, Freeman B, Harbrecht BG, Johnson JL, Minei JP, Maier RV; Inflammation and the Host Response to Injury Collaborative Research Program. Inflammation and the Host Response to Injury, a large-scale collaborative project: patient-oriented research coreVstandard operating procedures for clinical care. III. Guidelines for shock resuscitation. J Trauma. 2006;61:82Y89. Rainer TH, Ho AM, Yeung JH, Cheung NK, Wong RS, Tang N, Ng SK, Wong GK, Lai PB, Graham CA. Early risk stratification of patients with major trauma requiring massive blood transfusion. Resuscitation. 2011;82: 724Y729. Brown LM, Call MS, Margaret Knudson M, Cohen MJ, Trauma Outcomes Group, Holcomb JB, Wade CE, Brasel KJ, Vercruysse G, MacLeod J,

1250

31.

32.

33.

34. 35. 36.

37. 38. 39. 40.

Dutton RP, et al. A normal platelet count may not be enough: the impact of admission platelet count on mortality and transfusion in severely injured trauma patients. J Trauma. 2011;71:S337YS342. Schreiber MA, Perkins J, Kiraly L, Underwood S, Wade C, Holcomb JB. Early predictors of massive transfusion in combat casualties. J Am Coll Surg. 2007;205:541Y545. S57YS63. David JS, Levrat A, Inaba K, Macabeo C, Rugeri L, Fontaine O, Cheron A, Piriou V. Utility of a point-of-care device for rapid determination of prothrombin time in trauma patients: a preliminary study. J Trauma. 2012;72: 703Y707. Kimbrell BJ, Velmahos GC, Chan LS, Demetriades D. Angiographic embolization for pelvic fractures in older patients. Arch Surg. 2004;139: 728Y732; discussion 732Y733. Henry SM, Pollak AN, Jones AL, Boswell S, Scalea TM. Pelvic fracture in geriatric patients: a distinct clinical entity. J Trauma. 2002;53:15Y20. Maegel M. Frequency, risk stratification and therapeutic management of acute post-traumatic coaglopathy. Vox Sang. 2009;97:39Y49. Dunne JR, Riddle MS, Danko J, Hayden R, Petersen K. Blood transfusion is associated with infection and increased resource utilization in combat casualties. Am Surg. 2006;72:619Y625. Nathens AB. Massive transfusion as a risk factor for acute lung injury: association or causation? Crit Care Med. 2006;34:S144YS150. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet. 1967;12:319Y323. Moore FA, Moore EE, Sauaia A. Blood transfusion: an independent risk factor for postinjury multiple organ failure. Arch Surg. 1997;132:620Y625. Dunne JR, Malone DL, Tracy JK, Napolitano LM. Allogenic blood transfusion in the first 24 hours after trauma is associated with increased systemic inflammatory response syndrome (SIRS) and death. Surg Infect. 2004;5:395Y404.

* 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Predicting the need for massive transfusion in trauma patients: the Traumatic Bleeding Severity Score.

The ability to easily predict the need for massive transfusion may improve the process of care, allowing early mobilization of resources. There are cu...
649KB Sizes 0 Downloads 3 Views