Article, Hematology

Modified traumatic bleeding severity score: early determination of the need for massive transfusion

a b s t r a c t

Background: Determination of the need for massive transfusion (MT) is essential for early activation of a MT pro- tocol. The traumatic bleeding Severity Score (TBSS) predicts the need for MT accurately, but takes time to deter- mine because systolic blood pressure after a 1000 mL of crystalloid infusion is used. The aim of this study is to determine the how well the Modified TBSS (age, sonography, pelvic fracture, serum lactate and systolic blood pressure on arrival) predicts the need for MT (accuracy).

Methods: This is a single-center retrospective study of trauma patients (Injury Severity Score ?16) admitted be- tween 2010 and 2014. The TBSS, the Trauma Associated Severe Hemorrhage (TASH) Score, and the Modified TBSS were calculated. MT is defined as ? 10 U packed red blood cell transfusion within 24 hours of injury, and

the predictive value of the need for MT was compared by Area Under the Receiver Operating Characteristic Curve analysis among three scores.

Results: Three hundred patients were enrolled, and MT given to 25% of patients. Although the AUC of the TBSS was higher than that of the TASH score (0.956 vs 0.912, P = .006) and the Modified TBSS (0.956 vs 0.915, P =

.001), there was no difference between the AUC of the Modified TBSS and the TASH score. The Modified TBSS has high accuracy, within an AUC N 0.9.

Conclusion: The predictive value of the Modified TBSS of the need for MT is still high and is equivalent to the TASH score. The Modified TBSS is calculated earlier in resuscitation than the original TBSS.

(C) 2016

Background

massive hemorrhage is a major cause of death due to traumatic inju- ry [1-3], and early administration of blood products with higher plasma and platelet ratios in trauma resuscitation are associated with decreased mortality [4]. Many major trauma centers in the United States, have in- stituted massive transfusion (MT) protocols to initiate the transfusion of blood products in the appropriate ratio as early as possible [5]. Conve- nient, early and accurate determination of the need for MT are essential

? Source of support: This work received no financial support.

?? Conflicts of interest: All authors report no conflicts of interest.

? Previous presentations: No

* Corresponding author at: 371-0014 Asahi-cho 3-21-36, Maebashi, Gunma, Japan. Tel.: +81 27 224 4585; fax: +81 27 243 3380.

E-mail addresses: [email protected] (T. Ogura), [email protected] (A.K. Lefor), [email protected] (M. Masuda),

[email protected] (S. Kushimoto).

for early activation of a MT protocol, and several appropriate triggers for MT protocol initiation have been developed, such as scoring systems [6] and rotational thromboelastgraphy [7].

Scoring systems to predict the need for MT are generally easy to use and not expensive. A high accurate scoring system is ideal for this pur- pose. There are a number of such scoring systems in use today. Brockamp and colleagues conducted a detailed review of six such scor- ing systems including the Trauma-Associated Severe Hemorrhage (TASH) score, the Prince of Wales score, the Vandromme score, the As- sessment of Blood Consumption score, the Schreiber score and the Larsen score [6]. They reported that the TASH score has the highest overall accuracy.

The Traumatic Bleeding Severity Score (TBSS) is a simple and highly accurate scoring systems to predict the need for MT [8] (Fig. 1), which use only five variables, including patient age, systolic blood pressure after rapid infusion of 1000 mL of crystalloid, the result of the focused assessment with sonography for trauma scan, severity of pelvic frac- ture, and lactate concentrate on arrival. The TBSS is considered as a very useful scoring system for use in the Japanese trauma population

http://dx.doi.org/10.1016/j.ajem.2016.02.072

0735-6757/(C) 2016

Fig. 1. The original traumatic bleeding severity score SBP: systolic blood pressure (after rapid infusion of 1000 mL crystalloid), FAST, Focused Assessment with Sonography for Trauma; AO, Arbeitsgemeinschaft fur Osteosynthesefragen/Orthopedic Trauma Association classification.

[8]. The TBSS is determined using a convenient Smartphone application which is easy to use.

However, the TBSS takes time to determine because it is based on the systolic blood pressure after rapid infusion of 1000 mL of crystalloid. That is a problem for the early determination of the score and identifica- tion of the need for MT. If the systolic blood pressure on arrival can be used, instead of the systolic blood pressure after rapid infusion of 1000 mL of crystalloid, the TBSS is determined earlier, allowing the MT protocol to be instituted earlier. The aim of this study is to test the accuracy of the Modified TBSS, which use the systolic blood pressure on arrival instead of the systolic blood pressure after rapid infusion of 1000 mL crystalloid.

Patients and methods

This study was conducted at a single institution (Japan Red Cross Maebashi Hospital, Japan) that admits approximately 150 to 200 se- verely injured patients per year (Injury Severity Score N/= 16). The In- stitutional Review Board approved this review of patient data prior to starting the study. This is a retrospective study of severely injured pa-

tients (Injury Severity Score ? 16) admitted between January 2010 and March 2014. Due to the epidemiology of the patient population, nearly

all patients in this study suffered blunt traumatic injuries. Patients who presented initially with cardiopulmonary arrest or isolated head trauma were excluded from this study, in accordance with the original TBSS study [8].

This study was conducted in the same way as the original TBSS val- idation study [8]. The TBSS, the TASH score [9], and the Modified TBSS at the time of arrival were retrospectively calculated. The Modified TBSS was defined as a score using the same algorithm as used to calculate the TBSS, but using the systolic blood pressure on arrival instead of the systolic blood pressure after rapid infusion of 1000 mL of crystalloid. MT is defined as a transfusion of 10 or more units of Packed red blood cells, within 24 hours of the time of injury. (In Japan, one unit of packed red blood cells is approximately 120 mL.) This study was performed to

determine how well the Modified TBSS can predict the need for MT, as defined. Transfusion and crystalloid infusions were administered with- out a specific protocol, and determined according to the clinical judg- ment of the physician in charge.

The predictive value of the need for MT was evaluated by area under the receiver operating characteristic curve analysis. High accura- cy in predicting the need for MT is defined by an area under the receiver operating characteristic curve of more than 0.9, moderate accuracy is defined as AUC less than 0.9, but greater than 0.7, and low accuracy is defined as an AUC less than 0.7 [10]. Receiver operating characteristic curve comparison was performed to compare the accuracy of predicting the need for MT comparing the TBSS, TASH score, and Modified TBSS. The method of Bonferroni correction was used to adjust the P values in multiple testing for each AUC comparison among the three groups, and a significant difference was defined as P b .0167 in this study. All statistical calculation in this study was performed using MedCalc(R) (Ostend, Belgium).

Result

A total of 300 patients were enrolled in this study and their clinical characteristics are summarized in Table 1. Nearly all patients (97%) suf- fered blunt traumatic injuries, and just 3% (10/300) suffered Penetrating injuries. The median age was 60 years old (range 37-70), and 77% (230/ 300) were male. The median Injury Severity Score was 27 (20-36), MT (10 U or more of PRBC transfusion within 24 hours of injury) was given to 28% (84/300) of patients. The overall mortality was 14% (42/ 300). Of the 42 patients who died, 30 were in the MT group (30/84) and 12 were in the Non-MT group (12/216) (Table 2). The mortality in the MT group was 36% (30/84) and that in Non-MT group was 6% (12/216) (P b .001.)

The AUC of the Modified TBSS for predicting the need for MT was 0.915 (95% CI, 0.887-0.944; cutoff 14, sensitivity 80.0%, specificity 91.1%). The AUC of the TBSS was 0.956 (95% CI, 0.926-0.976; cutoff 14,

sensitivity 93.3%, specificity 92.4%), and of the TASH score was 0.912

Table 1

Patient characteristics (total N = 300)

Age (y) 60 (37-70)

Discussion

major traumatic injuries are a significant cause of morbidity and

Sex (male) Mechanism (blunt)

230 (76.7%)

290 (96.7%)

mortality throughout the world. A Survival benefit has been shown

with trauma resuscitation that includes the early transfusion of blood

Systolic blood pressure on Arrival (mmHg)

121 (98-141)

products, and a survival benefit has also been shown for trauma resus-

Systolic blood pressure after infusion of

121.0 (98.0-140.5)

citation with the early administration of adequate amounts of plasma

1000 mL of crystalloid (mmHg)

Heart rate (/minute) 86 (73-109)

Glasgow Coma Scale 14 (10-15)

Hemoglobin (g/dL) 12.7 (11.0-14.3)

Platelet (x105/mL) 22.1 (7.4-26.9)

Prothrombin Time International normalized ratio 1.12 (1.04-1.27) Lactate Concentration (mmol/L) 2.2 (1.5-3.3)

or other blood products such as platelet concentrates [4], because acute traumatic coagulopathy is a result of tissue trauma and systemic hypoperfusion [11-13] which presents in the hyper-acute phase of traumatic injury. MT prediction scores are used to facilitate the early de- cision to activate a MT protocol with an appropriate blood product infu-

sion ratio (such as the ratio of packed red blood cells/fresh frozen

Positive Focused assessment with sonography in trauma scan

55 (18.3%)

plasma/platelet concentrates = 1/1/1, or packed red blood cells/fresh

Anticoagulation 21 (7.0)

Injury Severity Score 27 (20-36)

Abbreviated Injury Score

Head 3 (0-4)

Face 0 (0-0)

Chest 3 (0-4)

Abdomen 0 (0-3)

Limbs 0 (0-3)

Surface 0 (0-0)

Massive transfusion 84 (28.0%)

Red blood cells (U) 0 (0-8)

Fresh frozen plasma (U) 0 (0-10)

Platelet concentrate (U) 0 (0-0)

Mortality 42 (14.0%)

Data are expressed as number (%) or median (interquartile range).

(95% CI, 0.874-0.941; cutoff 7, sensitivity 86.7%, specificity 83.6%) (Fig. 2). According to the AUC comparison study, the AUC of the TBSS was significantly higher than that of the TASH score and the Modified TBSS (difference of areas between the TBSS and the Modified TBSS; 0.042, 95% CI, 0.016-0.067, Bonferroni adjusted P = .006, difference in area between the TBSS and the TASH score; 0.044, 95% CI, 0.012-0.077, Bonferroni adjusted P = .001). There was no significant difference between the AUC of the Modified TBSS and that of the TASH score (Table 3).

frozen plasma >= 1/1.5 [14,15]), although the optimal ratio of transfused

blood products is still undefined.

Several MT prediction scores have been developed in recent years [6,8,9,16-18], and some can accurately predict the need for MT in trau- ma patients [8,9]. Convenience and rapid calculation are essential for a scoring system to be useful in predicting the need for MT, and accuracy is also important, since complications of transfusion can be severe [19,20]. In this study, the originally described TBSS was modified and systolic blood pressure on arrival was used to calculate the Modified TBSS, instead of the blood pressure after rapid infusion of 1000 mL of crystalloid. This change allows earlier calculation of the Modified TBSS. The result of this study show that the accuracy of the Modified TBSS is still high (Fig. 2, Table 2) [10] and is equivalent to that of the TASH score (Table 2), although it is significantly decreased compared to the TBSS. The Modified TBSS is determined earlier in resuscitation than the original TBSS and still remains highly accurate in predicting the need for MT [10].

There are some acknowledged limitations to this study. Similar to the original TBSS study, this is a validation study of three MT prediction scores based on retrospective data from a single center, with nearly all patients suffering blunt traumatic injuries [8]. This study population did not include the same patient data used for creation of the original TBSS, and serves as an internal validation study of the TBSS and Modi- fied TBSS, and also an external validation study for the TASH Score. A

Table 2

Comparison between MT group and non-MT group

MT (n = 84) Non-MT (n = 216) P

?Variables included in TBSS and Modified TBSS?

Age+

68.5 (43,7-80.0)

56.0 (34.0-71.0)

b.001?

Systolic blood pressure on Arrival (mmHg)+

84.5 (72.0-102.0)

130.0 (115.0-150.0)

b.001?

Systolic blood pressure after infusion of 1000 mL of crystalloid (mmHg)+

92.0 (75.5-114.0)

129.0 (112.0-150.0)

b.001?

Focused assessment with sonography in trauma (number of lesion)+

0 (0-11.0)

0 (0-0)

b.001?

Pelvic fracture

Stable fracture (n,%)?

11 (13.1%)

16 (7.4)

.208

Partially unstable fracture (n,%)?

25 (29.8%)

23 (10.6%)

b.001?

Completely unstable fracture (n,%)?

6 (7.1%)

1 (0.5%)

.002?

Lactate concentration (mmol/l)+

3.1 (2.4-4.9)

1.9 (1.2-2.7)

b.001?

?Variables included in TASH Score?

Sex (male, %)?

62 (73.8%)

168 (77.7%)

.649

Hemoglobin (g/dL)+

10.7 (9.2-12.5)

13.3 (11.9-14.6)

b.001?

Base Excess (mmol/L)+

-3.7 (-5.6 – -2.0)

-1.1 (-2.5-0.0)

b.001?

Systolic blood pressure on arrival (mmHg)+

84.5 (72.0-102.0)

130.0 (115.0-150.0)

b.001?

Positive focused assessment with sonography in trauma (n, %)?

28 (33.3%)

27 (12.5%)

b.001?

Clinically unstable pelvic fracture (n, %)?

31 (36.9%)

24 (11.1%)

b.001?

Open or dislocated femur fracture (n, %)?

14 (16.7%)

25 (11.6%)

.362

TBSS+

20.0 (17.0-28.0)

6.0 (3.0-10.0)

b.001?

Modified TBSS+

19.5 (14.0-27.0)

6.0 (3.0-6.0)

b.001?

TASH score+

12.0 (8.0-16.0)

3.0 (1.0-6.0)

b.001?

Mortality?

30 (35.7%)

12 (5.6%)

b.001?

Data are expressed as number (%) or median (interquartile range).

+ Mann-Whitney’s U test.

? ?2 test.

* P b .05.

Fig. 2. The receiver operating characteristic curve of the TBSS, the Modified TBSS, and TASH scores for the need for the massive transfusion. *Bonferroni adjusted P b .0167

large prospective multicenter study is necessary to compare the predic- tive value of these scoring systems.

Conclusion

The Modified TBSS is determined on patient arrival, much earlier than the original TBSS, because there is no need to complete a crystal- loid infusion for determination. The MT protocol can then be instituted earlier in the resuscitation process. The predictive value of the Modified TBSS to determine the need for MT is still high and is equivalent to the TASH score. Large prospective multicenter studies are necessary to fur- ther validate the ability of the Modified TBSS to predict the need for MT.

Author Contributions

T. Ogura, M. Masuda collected the data and analyzed. T. Ogura, M. Masuda and A. T. Lefor contributed to the interpretation of result. T. Ogura, A.T. Lefor, and S. Kushimoto drafted the manuscript, which all authors read and approved.

Table 3

Each AUC and pairwise comparison of ROC

TBSS Modified TBSS TASH Score

Disclosures

This work received no financial support and all authors report 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: Chicago, July 11-13, 2008. J Trauma 2010;68:225-30.
  2. Sauaia A, Moore FA, Moore EE, Moser KS, Brennan R, Read RA, et al. Epidemiology of trauma death: a reassessment. J Trauma 1995;38:185-93.
  3. Callcutt RA, Cotton BA, Muskat P, Fox EE, Wade CE, Holcomb JB, et al. Defining when to initiate MT: a validation study of individual MT triggers in PROMMTT patients. J Trauma Acute Care Surg 2013;74:59-68.
  4. Holcomb JB, del Junco DJ, Fox EE, Wade CE, Cohen MJ, Schreiber MA, 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;148:127-36.
  5. Schuster KM, Davis KA, Lui FY, Maerz LL, Kaplan LJ. The status of MT protocols in United States trauma centers: MT or massive confusion? Transfusion 2010;50: 1545-51.
  6. Brockamp T, Nienaber U, Mutschler M, Wafaisade A, Peiniger S, Lefering R, et al, TraumaRegister DGU. Predicting on-going hemorrhage and transfusion requirement after severe trauma: a validation of six scoring systems and algorithms on the TraumaRegister DGU(R). Crit Care 2012;16:R129.
  7. Hagemo JS, Christiaans SC, Stanworth SJ, Brohi K, Johansson PI, Goslings JC, et al. De- tection of acute traumatic coagulopathy and MT requirements by means of rotation- al thromboelastometry: an international prospective validation study. Crit Care 2015;19:97.
  8. Ogura T, Nakamura Y, Nakano M, Izawa Y, Nakamura M, Fujizuka K, et al. Predicting the need for MT in trauma patients: the traumatic bleeding severity score. J Trauma Acute Care Surg 2014;76:1243-50.
  9. Yucel N, Lefering R, Maegele M, Vorweg M, Tjardes T, Ruchholtz S, et al, Polytrauma

    AUC

    0.956

    (cutoff, 14)

    0.915

    (cutoff, 14)

    0.912

    (cutoff, 7)

    Study Group of the German Trauma Society. Trauma associated severe hemorrhage (TASH)-score: probability of mass transfusion as surrogate for life threatening hem-

    Sensitivity

    93.3%

    80.0%

    86.7%

    orrhage after multiple trauma. J Trauma 2006;60:1228-37.

    Specificity

    92.4%

    91.1%

    83.6%

    [10] Akobeng AK. Understanding diagnostic tests 3: receiver operating characteristic

    TBSS-Modified TBSS

    TBSS-TASH

    score

    Modifed TBSS-TASH score

    curves. Acta Paediatr 2007;96:644-7.

    Hess JR, Brohi K, Dutton RP, Hauser CJ, Holcomb JB, Kluger Y, et al. The coagulopathy of trauma: a review of mechanisms. J Trauma 2008;65:748-54.

  10. Chesebro BB, Rahn P, Carles M, Esmon CT, Xu J, Brohi K, et al. Increase in activated

    Difference between AUC

    0.042 0.044 0.003

    protein C mediates acute traumatic coagulopathy in mice. Shock 2009;32:659-65.

    Frith D, Goslings JC, Gaarder C, Maegele M, Cohen MJ, Allard S, et al. Definition and

    95% CI 0.016-0.067 0.012-0.077 -0.033 to 0.039

    Significance level .001? .007? .875

    ROC, receiver operating characteristic curve.

    * Bonferroni-adjusted P b .0167.

    drivers of acute traumatic coagulopathy: clinical and experimental investigations. J

    Thromb Haemost 2010;8:1919-25.

    Holcomb JB, Wade CE, Michalek JE, Chisholm GB, Zarzabal LA, Schreiber MA, et al. In- creased plasma and platelet to red blood cell ratios improves outcome in 466 mas- sively transfused civilian trauma patients. Ann Surg 2008;248:447-58.

  11. Sperry JL, Ochoa JB, Gunn SR, Alarcon LH, Minei JP, Cuschieri J, et al, Inflammation the Host Response to Injury Investigators. An FFP:PRBC transfusion ratio N =1:1.5 is associated with a lower risk of mortality after MT. J Trauma 2008;65:986-93.
  12. Nunez TC, Voskresensky IV, Dossett L, Shinall R, Dutton WD, Cotton B. Early predic- tion of MT in trauma: simple as ABC (assessment of blood consumption)? J Trauma 2009;66:346-52.
  13. Vandromme MJ, Griffin RL, McGwin G, Weinberg J, Rue LW, Kerby JD. Prospective identification of patients at risk for MT: an imprecise endeavor. Am Surg 2011;77: 155-61.
  14. Larson CR, White CE, Spinella PC, Jones JA, Holcomb JB, Blackbourne LH, et al. Asso- ciation of shock, coagulopathy, and initial vital signs with MT in combat casualties. J Trauma 2010;69:S26-32.
  15. Carson JL, Grossman BJ, Kleinman S, Tinmouth AT, Marques MB, Fung MK, et al, Clinical Transfusion Medicine Committee of the AABB. Red blood cell transfusion: a clinical practice guideline from the AABB. Ann Intern Med 2012;157:49-58.
  16. Kaufman RM, Djulbegovic B, Gernsheimer T, Kleinman S, Tinmouth AT, Capocelli KE, et al, AABB. Platelet transfusion: a clinical practice guideline from the AABB. Ann In- tern Med 2015;162:205-13.