Fröhlich et al.

Scandinavian Journal of Trauma,

Resuscitation and Emergency Medicine (2016) 24:148
DOI 10.1186/s13049-016-0340-2

ORIGINAL RESEARCH Open Access

Is the shock index based classification of

hypovolemic shock applicable in multiple
injured patients with severe traumatic brain
injury?—an analysis of the TraumaRegister
DGU®
Matthias Fröhlich1,2,6*, Arne Driessen1,6, Andreas Böhmer3,6, Ulrike Nienaber4,6, Alhadi Igressa5,6, Christian Probst1,6,
Bertil Bouillon1,6, Marc Maegele1,6, Manuel Mutschler1,6 and and the TraumaRegister DGU6

Abstract
Background: A new classification of hypovolemic shock based on the shock index (SI) was proposed in 2013. This
classification contains four classes of shock and shows good correlation with acidosis, blood product need and mortality.
Since their applicability was questioned, the aim of this study was to verify the validity of the new classification in
multiple injured patients with traumatic brain injury.
Methods: Between 2002 and 2013, data from 40 888 patients from the TraumaRegister DGU® were analysed. Patients
were classified according to their initial SI at hospital admission (Class I: SI < 0.6, class II: SI ≥0.6 to <1.0, class III SI ≥1.0 to
<1.4, class IV: SI ≥1.4). Patients with an additional severe TBI (AIS ≥ 3) were compared to patients without severe TBI.
Results: 16,760 multiple injured patients with TBI (AIShead ≥3) were compared to 24,128 patients without severe TBI.
With worsening of SI class, mortality rate increased from 20 to 53% in TBI patients. Worsening SI classes were
associated with decreased haemoglobin, platelet counts and Quick’s values. The number of blood units transfused
correlated with worsening of SI. Massive transfusion rates increased from 3% in class I to 46% in class IV. The accuracy
for predicting transfusion requirements did not differ between TBI and Non TBI patients.
Discussion: The use of the SI based classification enables a quick assessment of patients in hypovolemic shock based
on universally available parameters. Although the pathophysiology in TBI and Non TBI patients and early treatment
methods such as the use of vasopressors differ, both groups showed an identical probability of recieving blood
products within the respective SI class.
Conclusion: Regardless of the presence of TBI, the classification of hypovolemic shock based on the SI enables a fast
and reliable assessment of hypovolemic shock in the emergency department. Therefore, the presented study supports
the SI as a feasible tool to assess patients at risk for blood product transfusions, even in the presence of severe TBI.
Keywords: Haemorrhagic shock, Shock index, Traumatic brain injury, Multiple trauma

* Correspondence: froehlichm@kliniken-koeln.de
1
Department of Orthopaedic Surgery, Traumatology and Sports
Traumatology, Cologne-Merheim Medical Centre (CMMC), Witten/Herdecke
University, Ostmerheimer Str. 200, D-51109 Cologne, Germany
2
Institute for Research in Operative Medicine (IFOM), University of Witten/
Herdecke, Cologne Merheim Medical Center (CMMC), Ostmerheimerstr.200,
D-51109 Cologne, Germany
Full list of author information is available at the end of the article

© The Author(s). 2016 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Fröhlich et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2016) 24:148 Page 2 of 9

Background the SI possibly underestimates the extent of haemorrhage

Severe trauma is the leading cause of death among in the presence of acute TBI [14].
younger people. Annual deaths worldwide attributed to Since more than 40% of all severely injured patients
trauma are expected to increase from five million to in Germany sustain an additional TBI, the aim of this
more than eight million by 2020 [1]. Despite all im- study was to determine if the SI based classification of
provements in treatment, uncontrolled post-traumatic hypovolemic shock is applicable in the presence of TBI
bleeding is the leading cause of potentially preventable predicting transfusion requirements reliably in patients
death among these patients [2]. This emphasizes the key with and without TBI.
role of an early recognition and treatment of haemor-
rhage, hypovolemia and disorders of coagulation. There- Methods
fore, the American College of Surgeons has defined four The TraumaRegister DGU®
degrees of hypovolemic shock, which are taught in the The TraumaRegister DGU® of the German Trauma Soci-
Advanced Trauma Life Support (ATLS®) training pro- ety (Deutsche Gesellschaft für Unfallchirurgie, DGU) was
gram. These four classes of hypovolemic shock are based founded in 1993. The aim of this multi-centre database is
upon an estimated blood loss and corresponding vital a pseudonymized and standardized documentation of se-
signs including mental state, blood pressure and pulse verely injured patients. Detailed presentations of the Trau-
rate [3]. Recent analyses from the TraumaRegister DGU® maRegister DGU® have been published previously [15, 16].
and the TARN registry questioned the classification’s Data are collected prospectively starting at the pre-
validity [4, 5]. Only 10% of all trauma patients can be hospital phase, covering the ED and ICU stay until dis-
classified according to the ATLS® classification [4]. In charge of the patient. The documentation includes detailed
order to reflect clinical reality more precisely our group information on demographics, injury pattern, comorbidi-
proposed a new classification of hypovolemic shock ties, pre- and in-hospital management, course on intensive
(Table 1) which is based on the shock index (SI) [6]. As care unit, relevant laboratory findings including data on
the SI is the ratio of heart rate to systolic blood pressure, transfusion and outcome of each individual. The inclusion
this index can be immediately calculated when basic vital criterion is admission to hospital via emergency room with
signs are available. The SI correlates with the extent of subsequent ICU/ICM care or reaching the hospital with
hypovolemia and thus may facilitate the early identification vital signs and death before admission to ICU. Currently,
of severely injured patients threatened by complications approx. 25,000 cases from more than 600 hospitals are
due to blood loss and therefore need urgent treatment, i.e. yearly entered into the database.
blood transfusion [6]. Scientific data analysis is approved according to a peer
Along with haemorrhage, traumatic brain injury (TBI) review procedure established by the German Trauma
deteriorates the outcome and is associated with an Society. The present study is in line with the publication
increased morbidity and mortality, regardless if it occurs guidelines of the TraumaRegister DGU® and registered
with other injuries or as an isolated mono trauma [7]. In as TR-DGU project ID 2011–010.
young people, TBI is the leading cause of death and
disability [8]. Several studies have shown that TBI in Study population
conjunction with haemorrhage might disturb the auto- For the present study, datasets entered between 2002
nomic response to blood loss or the ability to modulate and 2013 into the TraumaRegister DGU® were analysed.
vascular tone [9–12]. Goldstein described the uncoupling In 2002, the online version of the registry was intro-
of the autonomic and cardiovascular system [13]. There- duced, replacing paper form data collection. Inclusion
fore, the use of vital signs such as heart rate or blood pres- criteria were age ≥16 years, primary admission, admis-
sure for the assessment of hypovolemic shock has to be sion to an intensive care unit (ICU) and complete data-
questioned. In line with these results, McMahon et al. sets for systolic blood pressure (SBP), heart rate (HR)
described the effect of acute TBI on the performance of and Glasgow Coma Scale (GCS). Severe TBI was defined
shock index in a small animal model with combined as an AIShead ≥ 3 [17, 18]. The shock index (SI) was
TBI and haemorrhage [14]. The authors concluded that calculated for each individual patient by the ratio of HR
due to disturbance in the central cardiovascular regulation to SBP at emergency department (ED) admission [6].

Table 1 Classification of hypovolemic shock based on the shock index [6]

Class I Class II Class III Class IV
Shock no shock mild shock moderate shock severe shock
SI at admission <0.6 ≥0.6 to <1 ≥1 to <1.4 ≥1.4
Need of blood products Observe Consider use of blood products Prepare transfusion Prepare massive transfusion
Fröhlich et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2016) 24:148 Page 3 of 9

In 2013 our group described and validated the shock In all patients, worsening of SI category was associated
index as a fast guide to transfusion requirements among a with an increased ISS, increased in-hospital mortality.
large cohort of multiple trauma patients [6]. With respect Accordingly a higher rate of multiple organ failure
to previous descriptions of the SI as a predictor for mor- (MOF) and sepsis occurred in higher SI classes. Parame-
tality [19], four classes of SI were defined as follows: Class ters reflecting a complicated treatment such as hospital
I: SI < 0.6–no shock; class II: SI ≥0.6 to <1.0–mild shock; length of stay (LOS) and ICU (intensive care unit) LOS
class III: SI ≥1.0 to <1.4–moderate shock and class IV: SI as well as ventilator-days increased. In all classes, TBI
≥1.4–severe shock (Table 1) [6]. patients showed a higher ISS compared to Non-TBI pa-
Further, demographics, injury pattern and vital signs tients. Within each class, more patients without TBI had
were assessed as present upon ED arrival. Therapeutic in- injuries associated with high blood loss such as severe
terventions such as administration of blood products and abdominal and pelvic injuries. Furthermore, TBI patients
intravenous fluids and vasopressors were analysed. Massive had a significantly increased mortality and showed a
transfusion (MT) was defined by the administration of ≥10 higher rate of multiple organ failure (MOF) and sepsis.
blood products (including packed red blood cells (pRBC),
fresh frozen plasma (FFP) and thrombocyte concentrates Vital signs of TBI patients
(TC)) within 24 h after ED admission. Coagulopathy was As defined a-priori, SBP decreased and HR increased at
defined by a Quick’s value ≤ 70%, which is equivalent to an emergency department admission according to the SI
international normalized ration (INR) of approxi- classes. Differences between TBI and Non-TBI patients
mately ≥ 1.3 [20, 21]. Evaluating the reliability of the SI were not observed as shown in Table 3. The presence of
based classification regardless of the injury pattern, patients TBI was associated with a lower GCS and remarkable
with an AIShead ≥3 were assigned according to their SI higher pre-clinical intubation rate. However, a higher SI
at ED admission and compared to patients without a (≥1.4) was associated with lower Glasgow Coma Scale
significant TBI (AIShead ≤2). and higher intubation rate in all patients. Table 4 pro-
vides the first laboratory findings. Haemoglobin values
Statistical analysis and platelet counts were lower with worsening SI classes.
Data are presented as means ± 95% confidence interval In the presence of TBI, coagulation markers were more
(CI) for continuous variables or percentages for categorical severely impaired compared to Non-TBI patients. In these
variables. Formal statistical testing comparing TBI and patients, coagulopathy, that is characterized by a Quick’s
Non-TBI patients within the respective SI classes was value <70% and prolonged aPTT, appeared in class III and
avoided since due to the large sample size even minor dif- IV, assuming a SI of 1.0 or higher.
ferences would result in highly significant results, which
could mislead to over-interpretation. The clinical relevance
of differences between the observed groups has to be care- Volume management and transfusion requirements in
fully interpreted [22]. For the comparison of the SI based TBI patients
classification in the prediction of transfusion requirements Volume management and transfusion requirements of
in patients with and without TBI, the area under the TBI patients are displayed in Table 5. With worsening SI,
receiving operating characteristics curve (AUROC) was the volume administered and the percentage of patients
calculated with occurrence of transfusion (≥1 blood that received vasopressors increased significantly. Ac-
product) and MT as the state variable. The comparison of cording to the predicted transfusion requirements by the
two areas under the receiving operating characteristics TASH score, the observed transfusion incidence increased
curve was based upon the 95% confidence interval for similarly.
each curve. Data were analysed using SPSS statistical soft-
ware package (Version 21, IBM Inc., Armonk, NY, U.S.A.). Comparison of transfusion requirements according to
injury characteristics
Results The percentage of multiple injured patients receiving at
Demographics and characteristics least one blood product or MT increased stepwise from
During the observed time period 40 888 patients matched class I to class IV regardless the presence of TBI (Fig. 1).
the inclusion criteria. Patients were 46.6 ± 0.2 years old, The percentage of TBI patients who received at least
predominantly male (73%) and severely injured with a one blood product increased from 10% in class I to 70%
mean injury severity score (ISS) of 21.4 ± 0.1. Most in class IV. Accordingly the rate of MT increased from
patients suffered blunt trauma (95%). Severe head injury, 3% in class I to 46% in class IV. In comparison, Non-
displayed by an AIS ≥ 3, occurred in 41% of the cases TBI patients received slightly less blood products in
(n = 16,760). General demographics, injury severities and shock classes I and II, whereas the relation turned in
outcome parameters for the patients are shown in Table 2. class IV with balanced percentages in class III (Fig. 1).
Fröhlich et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2016) 24:148 Page 4 of 9

Table 2 Patients classified by SI calculated at ED admission and the presence of TBI: demographics, injury mechanism and severities
as well as outcome parameters. Continuous variables are presented as mean ± 95% confidence interval; categorical variables are
presented as absolute number and percentage
Class I Class II Class III Class IV
SI <0.6 SI ≥0.6 to <1 SI ≥1 to <1.4 SI ≥ 1.4
Non-TBI TBI Non-TBI TBI Non-TBI TBI Non-TBI TBI
Demographics
n (total, %) 6949 (28.8) 5177 (30.9) 12,780 (53.0) 7861 (46.9) 2015 (8.4) 1741 (10.4) 800 (3.3) 839 (5.0)
Male (n, %) 5324 (77) 3691 (72) 9219 (72) 5678 (72) 1423 (71) 1255 (73) 592 (74) 613 (72)
Age (years; mean ± CI) 48.5 ± 0.5 55 ± 0.6 42 ± 0.3 47 ± 0.5 44 ± 0.8 46 ± 1.0 46 ± 1.3 45 ± 1.4
Blunt trauma (n, %) 6367 (96) 4877 (97) 11,405 (93.1) 7428 (97) 1765 (90) 1605 (95) 702 (91) 768 (94)
Injury Severity
ISS (points; mean ± CI) 13.3 ± 0.2 24.9 ± 0.3 15.8 ± 0.2 28.4 ± 0.3 23.4 ± 0.6 36.2 ± 0.7 30.6 ± 1.0 43.0 ± 1.1
NISS (points; mean ± CI) 16.4 ± 0.2 33.6 ± 0.4 19.4 ± 0.2 35.7 ± 0.3 28.7 ± 0.6 42.9 ± 0.8 36.7 ± 1.1 49.0 ± 1.1
RISC (points; mean ± CI) 4.3 ± 0.2 23.7 ± 0.7 4.5 ± 0.2 26.4 ± 0.6 11.1 ± 0.8 38.3 ± 1.6 23.9 ± 1.9 52.1 ± 2.4
AIS Thorax ≥3 points (n; %) 2669 (38) 1461 (28) 5554 (44) 3182 (40) 1198 (60) 1021 (59) 541 (68) 585 (70)
AIS Abdomen ≥3 points (n; %) 674 (10) 179 (3) 1985 (15) 661 (8) 606 (30) 357 (21) 355 (44) 286 (34)
AIS Pelvis/Extremities ≥3 points (n; %) 1840 (27) 541 (10) 4554 (36) 1569 (20) 1045 (52) 657 (38) 513 (64) 410 (49)
Outcome
Mortality (n; %) 154 (2.2) 1057 (20.4) 373 (2.9) 1523 (19.4) 204 (10.1) 637 (36.6) 196 (24.5) 433 (51.6)
Hospital LOS (days; mean ± CI) 16.6 ± 0.4 18.3 ± 0.5 20.4 ± 0.4 21.0 ± 0.5 30.8 ± 1.3 23.0 ± 1.2 31.7 ± 2.2 22.2 ± 1.9
ICU LOS (days; mean ± CI) 4.2 ± 0.2 9.7 ± 0.3 6.1 ± 0.2 11.8 ± 0.3 12.2 ± 0.7 14.7 ± 0.7 15.6 ± 1.2 14.6 ± 1.3
Ventilatior days (days; mean ± CI) 1.8 ± 0.1 6.4 ± 0.3 3.1 ± 0.1 8.1 ± 0.2 7.5 ± 0.6 14.8 ± 0.6 10.8 ± 1.0 11.5 ± 1.1
MOF (n; %) 220 (4) 676 (15) 670 (6) 1393 (20) 302 (17) 483 (32) 186 (28) 273 (41)
Sepsis (n; %) 176 (3) 345 (8) 515 (5) 680 (10) 226 (13) 214 (14) 143 (21) 119 (17)

ROC curves displaying the predictive value of the SI MT was comparable in both groups (AUROC: TBI 0.756
regarding the occurrence of transfusion and MT are dis- (0.740–0.773) vs. Non TBI 0.764 (0.748–0.779)).
played in Fig. 2. As reflected by an AUROC of 0.706
(0.693–0.719) for TBI patients and 0.718 (0.707–0.730) Discussion
for Non-TBI patients, the accuracy for predicting the The SI is a tried and tested approach recognizing the
transfusion of ≥1 blood product did not differ signifi- presence of haemodynamic shock. Previously, a shock
cantly (Fig. 3). Accordingly, the accuracy for predicting index based classification has been proposed to assess

Table 3 Patients classified by SI calculated at ED admission and the presence of TBI: traditional vital signs as presented on scene
and at ED admission. Continuous variables are presented as mean ± 95% confidence interval; categorical variables are presented as
absolute number and percentage
Class I Class II Class III Class IV
SI <0.6 SI ≥0.6 to <1 SI ≥1 to <1.4 SI ≥ 1.4
Non-TBI TBI Non-TBI TBI Non-TBI TBI Non-TBI TBI
Vital signs
SBP at scene (mmHg; mean ± CI) 138 ± 0.7 141 ± 1.0 125 ± 0.5 123 ± 0.8 107 ± 1.3 105 ± 1.9 96 ± 2.4 95 ± 2.9
SBP at ED (mmHg; mean ± CI) 149 ± 0.6 149 ± 0.7 126 ± 0.3 123 ± 0.5 97 ± 0.7 96 ± 0.8 71 ± 1.1 70 ± 1.1
HR at scene (beats/min; mean ± CI) 84 ± 0.4 80 ± 0.6 95 ± 0.3 92 ± 0.5 106 ± 1.0 102 ± 1.6 112 ± 2.0 108 ± 2.4
HR at ED (beats/min; mean ± CI) 75 ± 0.3 73 ± 0.4 92 ± 0.3 91 ± 0.4 110 ± 0.8 109 ± 0.9 124 ± 1.4 123 ± 1.4
GCS at scene (points; mean ± CI) 13.9 ± 0.1 9.7 ± 0.1 13.6 ± 0.1 8.8 ± 0.1 12.4 ± 0.2 7.0 ± 0.2 11.2 ± 0.3 6.4 ± 0.3
GCS at ED (points; mean ± CI) 12.5 ± 0.1 8.1 ± 0.1 11.4 ± 0.1 6.7 ± 0.1 8.5 ± 0.2 4.6 ± 0.2 6.2 ± 0.4 3.9 ± 0.2
Intubation rate at ED admission (n; %) 1271 (19) 2497 (49) 3396 (27) 4804 (63) 954 (49) 1398 (81) 528 (68) 716 (87)
Fröhlich et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2016) 24:148 Page 5 of 9

Table 4 Patients classified by SI calculated at ED admission and the presence of TBI: laboratory findings at ED admission.
Continuous variables are presented as mean ± 95% confidence interval
Class I Class II Class III Class IV
SI <0.6 SI ≥0.6 to <1 SI ≥1 to <1.4 SI ≥ 1.4
Non-TBI TBI Non-TBI TBI Non-TBI TBI Non-TBI TBI
Laboratory findings
Haemoglobin (g/dl; mean ± CI) 13.3 ± 0.1 12.7 ± 0.1 12.8 ± 0.1 12.3 ± 0.1 11.1 ± 0.1 10.8 ± 0.1 9.6 ± 0.2 9.6 ± 0.2
Platelets (tsd/μl; mean ± CI) 222 ± 1 209 ± 2 223 ± 1 208 ± 2 211 ± 4 192 ± 4 190 ± 7 177 ± 5
Quick (%; mean ± CI) 89 ± 0.5 84 ± 0.6 86 ± 0.4 80 ± 0.5 73 ± 1.0 66 ± 1.2 62 ± 1.8 57 ± 2
pTT (seconds; mean ± CI) 28.5 ± 0.2 30.6 ± 0.4 29.2 ± 0.2 32.4 ± 0.4 33.6 ± 0.8 42.7 ± 1.5 44.2 ± 2.1 53.4 ± 2.8
Lactate (mmol/l; mean ± CI) 2.5 ± 0.1 2.5 ± 0.1 2.9 ± 0.1 3.1 ± 0.1 4.3 ± 0.3 4.6 ± 0.4 5.6 ± 0.5 6.4 ± 0.7

the extent of hypovolemic shock after trauma in order associated with a significant increased risk for MT, the
to realize the need of blood product transfusions [6]. Ac- presented results show the SI’s predictive value at ED
cording to the presented nationwide, population-based admission (Fig. 1) [24, 25]. Although the SI is immediately
prospective database analysis, the SI based classification available at admission, the four classes of hypovolemic
of hypovolemic shock predicts transfusion requirements shock based on SI are equivalent to classifications based
in trauma resuscitation regardless of the presence of on early laboratory findings such as base deficit [6].
severe TBI. Compared to the ATLS® classification, which is a good
During the past few years, several approaches targeting didactic tool to identify critical patients, the SI based
the recognition and evaluation of the extent of hypovol- score enables a better prediction for the need of blood
emic shock after trauma have been proposed In the products. It proves to be a robust indicator of shock
present study, we confirmed the SI as a reliable indicator based on readily available clinical variables. However,
assessing the presence of hypovolemia that is easily one key element of ATLS® is its universal and worldwide
ascertained. There is a growing body of evidence, that application, nearly independent of infrastructure and time
the SI as the ratio of HR and SBP is more sensitive than points of trauma care. The proposed score fulfils these
its underlying vital signs alone. In a systematic review, demands, as no blood tests or point of care diagnostics
Pacagnella assessed the relationship between blood loss (POCT) are required.
and corresponding vital signs [23]. The accuracy in pre- In the present study, regardless of the presence of TBI,
dicting blood loss displayed by the area under receiver an increased SI class was associated with more serious
operating characteristic curves ranged within the injuries depicted by an increased ISS including higher
reviewed studies from 0.56 to 0.74 for HR, from 0.56 to percentages of thoracic, abdominal and pelvic injuries.
0.79 for SBP and from 0.77 to 0.84 for SI [23]. While an This results in a significantly increased mortality according
increased prehospital SI has also been shown to be to the respective SI class. Likewise, data from the British

Table 5 Patients classified by SI and the presence of TBI: blood products and fluid resuscitation. Continuous variables are presented
as mean ± 95% confidence interval; categorical variables are presented as absolute number and percentage
Class I Class II Class III Class IV
SI <0.6 SI ≥0.6 to <1 SI ≥1 to <1.4 SI ≥ 1.4
Non-TBI TBI Non-TBI TBI Non-TBI TBI Non-TBI TBI
Transfusion requirements
All blood products/units (n; mean ± CI) 0.6 ± 0.1 0.9 ± 0.2 1.8 ± 0.2 2.4 ± 0.2 8.1 ± 0.9 8.3 ± 0.9 17.6 ± 2.0 17.3 ± 1.9
pRBC transfusions/units (n; mean ± CI) 0.5 ± 0.1 0.7 ± 0.1 1.3 ± 0.1 1.6 ± 0.1 4.5 ± 0.4 4.6 ± 0.4 9.1 ± 0.9 8.7 ± 0.8
FFP transfusions/units (n; mean ± CI) 0.3 ± 0.1 0.5 ± 0.1 0.9 ± 0.1 1.3 ± 0.2 3.5 ± 0.4 3.7 ± 0.4 6.9 ± 0.8 6.8 ± 0.8
TC transfusion/units (n; mean ± CI) 0.1 ± 0.0 0.1 ± 0.0 0.1 ± 0.0 0.1 ± 0.0 0.5 ± 0.1 0.5 ± 0.1 1.2 ± 0.2 1.1 ± 0.2
TASH Score (points; mean ± CI) 3.1 ± 0.1 2.9 ± 0.1 4.6 ± 0.1 4.8 ± 0.1 10.1 ± 0.2 10.0 ± 0.2 15.3 ± 0.4 14.6 ± 0.3
IV fliuds at scene (ml; mean ± CI) 814 ± 16 811 ± 18 960 ± 13 1027 ± 18 1237 ± 46 1319 ± 46 1487 ± 81 1522 ± 71
IV fliuds at ED (ml; mean ± CI) 1190 ± 34 1258 ± 40 1582 ± 40 1663 ± 40 2578 ± 119 2479 ± 142 3475 ± 221 3279 ± 213
Vasopressors at scene (n; %) 143 (2) 322 (7) 454 (4) 866 (12) 217 (12) 451 (27) 190 (26) 305 (39)
Vasopressors at ED (n; %) 557 (9) 955 (20) 1746 (14) 2225 (30) 781 (41) 954 (57) 543 (71) 629 (79)
Fröhlich et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2016) 24:148 Page 6 of 9

the autonomic and cardiovascular system complicates

the assessment and usability of blood pressure and heart
rate [13]. In an animal model of combined TBI and
haemorrhagic shock, the rise of SI was markedly attenuated
in non surviving animals which suggests a lack of car-
diovascular response to haemorrhage [14]. However, dif-
ferences were not observed until a blood loss of 40% [14].
McMahon and colleagues concluded that the significantly
differing trends in the performance of SI with on-going
haemorrhage might lead to an underestimation of the lost
blood volume in the presence of acute TBI [14]. However,
we could demonstrate that within one shock class the
presence of severe TBI did not influence the transfusion
frequency. Above a SI ≥ 1.4, blood products and MT were
more likely administered to Non-TBI patients, although
according to the ROC analysis, the predictive value did
not differ between both groups. Therefore, an effect of the
observed difference for clinical practice remains ques-
tionable. The main goal of the presented classification
is to increase awareness and to identify patients at risk
for bleeding and to predict reliably the need for blood
products.
During the study period, the practice of trauma resus-
citation and major transfusion changed considerably
Fig. 1 a and b Ratio of multiple injured patients receiving a) any
[15]. A recent analysis from the TraumaRegister DGU®
transfusion or b) mass transfusion regarding the presence of TBI showed that from 2002–2012 the preclinical adminis-
and according to their SI class tered volume decreased dramatically [15]. At the same
time, less severely injured patients received any blood
products or MT [15]. Although small volume resuscita-
Trauma Audit and Research Network (TARN) estab- tion is not recommended in trauma patients with severe
lished the SI among the top markers predicting 48-h TBI, we did not observe any differences in the volume
mortality [26]. But in difference to our study, Bruijns administered. However, the European Guideline on the
and colleagues excluded moderate and severe head ‘Management of bleeding and coagulopathy following
injuries [26]. Doubtless, severe TBI is associated with major trauma” explicitly recommend the same transfu-
increased mortality. But since the incidence of severe sion triggers and Hb-targets for patients with and with-
TBI was 41% in the presented study population, a clinically out TBI [2]. Since this recommendation did not change
useful shock classification should be applicable in all over the study period [28, 29], differences in the early
trauma patients including those patients suffering TBI. treatment should not have influenced the results of the
Unsurprisingly, TBI patients’ outcome was worse com- presented comparison between TBI and non-TBI
pared to Non-TBI patients’ outcome regardless of the shock patients.
class, displayed by ICU LOS, ventilator days and mortality. Within the respective SI classes, vasopressors were
As we focused on patients with AIShead ≥3, the mortality of used more frequently in TBI patients. Regardless of the
20% in shock classes I and II is in line with previous studies presence of TBI, vasopressors should be applied cau-
[8], as well as the stepwise increased mortality in Classes III tiously in addition to volume therapy to maintain the
and IV [6, 27]. However, further outcome parameters, which target arterial pressure [2]. In patients with a severe TBI
depend on the time of death, such as hospital LOS or venti- a mean arterial pressure ≥ 80 mmHg is recommended
lator days, decreased from class III to IV in TBI patients, while in bleeding Non-TBI patients a systolic blood
while these parameters increased across all classes from pressure of 80–90 mmHg should be targeted [2]. This
I to IV in the Non-TBI cohort. The combination of se- could explain the observed difference between both
vere haemorrhagic shock and TBI, which involves a sig- groups. The increased use of vasopressors influences
nificantly increased mortality rate and a supposed time of directly the SI and the consequent SI class by raising
death early after trauma, might explain these differences. the blood pressure. This might result in a false-low SI
The influence of TBI on the reliability of vital signs class and therefore in an underestimation of the extent
and the SI was discussed previously. The uncoupling of of hypovolemic shock and the resulting transfusion
Fröhlich et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2016) 24:148 Page 7 of 9

Fig. 2 ROC curves displaying the predictive value of the SI as continuous variable regarding the occurrence of transfusion (≥1 blood product; a Non-TBI,
b TBI) and massive transfusion (≥10 blood products; c) Non-TBI, d) TBI) as state variable

Fig. 3 Graphic representation of the performance analysis of the SI regarding a any transfusion and b massive transfusion displayed as area under the
receiver operating curve ± 95% confidence interval
Fröhlich et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2016) 24:148 Page 8 of 9

requirements. However, the demonstrated comparison Availability of data and materials

of applied blood products showed no differences within All data that are relevant for the study are included in this published article.
Further datasets analysed during the current study are available from the
the respective SI classes (Fig. 1). Therefore, the theoretically corresponding author on reasonable request.
false-low SI class due to the use of vasopressors does not
seem to influence the applicability of the proposed SI based Authors’ contributions
classification in the rapid assessment of the need of blood MF, MMu and MMa designed the study. UN analysed the data which MF,
MMu, CP, BB and MMa interpreted. MF, AD and MMu did perform the literature
products at ED admission. review and wrote the manuscript. AB, AI, MMa and BB were providing scientific
This study has limitations, as it is a retrospective study support and valuable advice working on the manuscript. AD, AB, UN and CP
of register-data with all the shortcomings associated. We proofread the manuscript and revised it critically. All authors have read and
approved the final manuscript.
have to rely on recorded data and are not able to verify
the validity. The administration of blood products and Competing interests
MT is based on clinicians’ judgments rather than based The authors declare that they have no competing interests.
on an objective measurement of haemorrhage. However,
both parameters are well established and serve as surro- Consent for publication
Not applicable.
gates for critical bleeding. We avoided formal statistical
testing comparing TBI and Non-TBI patients within the Ethics approval and consent to participate
respective SI classes, since due to the large sample size The present study is in line with the publication guidelines of the TraumaRegister
even minor differences would result in highly significant DGU® and registered as TR- DGU project ID 2011–010. As register data are
assessed anonymously, individual informed consent is not required.
results. This might mislead to over-interpretation and a
careful reflection of the differences between the observed Author details
1
groups regarding their clinical relevance is needed. In Department of Orthopaedic Surgery, Traumatology and Sports
Traumatology, Cologne-Merheim Medical Centre (CMMC), Witten/Herdecke
spite of these restrictions, we are confident that the SI University, Ostmerheimer Str. 200, D-51109 Cologne, Germany. 2Institute for
based classification is a feasible tool to assess patients Research in Operative Medicine (IFOM), University of Witten/Herdecke,
in hypovolemic shock and at risk for blood product Cologne Merheim Medical Center (CMMC), Ostmerheimerstr.200, D-51109
Cologne, Germany. 3Department of Anaesthesiology and Intensive Care
transfusions. Medicine, Cologne-Merheim Medical Centre, Witten/Herdecke University,
Ostmerheimer Str. 200, D-51109 Cologne, Germany. 4AUC-Academy for
Trauma Surgery, Straße des 17. Juni 106-108, D-10623 Berlin, Germany.
5
Conclusion Department of Neurosurgery, Cologne-Merheim Medical Centre,
Ostmerheimer Str. 200, D-51109 Cologne, Germany. 6Committee on
Summarizing, the proposed classification of hypovolemic Emergency Medicine, Intensive Care and Trauma Management of the
shock based on the SI proved to perform equally in mul- German Trauma Society (Sektion NIS), Berlin, Germany.
tiple injured trauma patients with and without severe
Received: 27 July 2016 Accepted: 30 November 2016
TBI. Within the four classes of hypovolemic shock, no
clinical relevant differences in transfusion requirements
between TBI and non-TBI patients were observed. References
Therefore, the presented study confirmed that the SI 1. Murray CJ, Lopez AD. Alternative projections of mortality and disability by
cause 1990–2020: global burden of disease study. Lancet. 1997;349(9064):
based classification is an easy and reliable tool to identify 1498–504.
trauma patients at risk for the need of blood products 2. Rossaint R, Bouillon B, Cerny V, Coats TJ, Duranteau J, Fernandez-Mondejar
regardless of the presence of TBI. E, Filipescu D, Hunt BJ, Komadina R, Nardi G, et al. The European guideline
on management of major bleeding and coagulopathy following trauma:
fourth edition. Crit care. 2016;20:100.
Abbreviations 3. Subcommittee A, College of Surgeons’ Committee on T, International Awg.
95% CI: 95% confidence interval; AIS: Abbreviated injury scale; Advanced trauma life support (ATLS(R)): the ninth edition. J trauma acute
ATLS®: Advanced trauma life support; AUROC: Area under the receiving care surg. 2013;74(5):1363–6.
operating characteristics curve; ED: Emergency Department; GCS: Glasgow 4. Mutschler M, Nienaber U, Brockamp T, Wafaisade A, Wyen H, Peiniger S,
coma scale; HR: Heart rate; ICU: Intensive care unit; INR: International Paffrath T, Bouillon B, Maegele M, Traumaregister DGU. A critical reappraisal
normalized ratio; ISS: Injury severity score; LOS: Length of stay; MOF: Multiple of the ATLS classification of hypovolaemic shock: does it really reflect
organ failure; MT: Massive transfusion; NISS: New injury severity score; clinical reality? Resuscitation. 2013;84(3):309–13.
POCT: Point of care diagnostics; PTT: Partial thromboplastin time; 5. Guly HR, Bouamra O, Spiers M, Dark P, Coats T, Lecky FE, Trauma A,
RISC: Revised injury severity classification; SBP: Systolic blood pressure; Research N. Vital signs and estimated blood loss in patients with major
SI: Shock Index; TASH score: Trauma associated severe haemorrhage score; trauma: testing the validity of the ATLS classification of hypovolaemic shock.
TBI: Traumatic brain injury Resuscitation. 2011;82(5):556–9.
6. Mutschler M, Nienaber U, Munzberg M, Wolfl C, Schoechl H, Paffrath T,
Bouillon B, Maegele M, Traumaregister DGU. The shock index revisited-a
Acknowledgements fast guide to transfusion requirement? a retrospective analysis on
Not applicable. 21,853 patients derived from the TraumaRegister DGU. Crit care. 2013;
17(4):R172.
7. Jeremitsky E, Omert L, Dunham CM, Protetch J, Rodriguez A. Harbingers of
Funding poor outcome the day after severe brain injury: hypothermia, hypoxia, and
This research received no specific grant from any funding agency in the public, hypoperfusion. J trauma. 2003;54(2):312–9.
commercial or not-for-profit sectors. 8. Ghajar J. Traumatic brain injury. Lancet. 2000;356(9233):923–9.
Fröhlich et al. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine (2016) 24:148 Page 9 of 9

9. Yuan XQ, Wade CE. Influences of traumatic brain injury on the outcomes of
delayed and repeated hemorrhages. Circ shock. 1991;35(4):231–6.
10. Mcmahon CG, Kenny R, Bennett K, Kirkman E. Modification of acute
cardiovascular homeostatic responses to hemorrhage following mild to
moderate traumatic brain injury. Crit care med. 2008;36(1):216–24.
11. Mcmahon CG, Kenny R, Bennett K, Little R, Kirkman E. Effect of acute
traumatic brain injury on baroreflex function. Shock. 2011;35(1):53–8.
12. Chen B, Mutschler M, Yuan Y, Neugebauer E, Huang Q, Maegele M.
Superimposed traumatic brain injury modulates vasomotor responses in
third-order vessels after hemorrhagic shock. Scand j trauma resusc emerg
med. 2013;21(1):77.
13. Goldstein B, Toweill D, Lai S, Sonnenthal K, Kimberly B. Uncoupling of the
autonomic and cardiovascular systems in acute brain injury. Am j physiol.
1998;275(4 Pt 2):R1287–92.
14. Mcmahon CG, Kenny R, Bennett K, Little R, Kirkman E. The effect of acute
traumatic brain injury on the performance of shock index. J trauma. 2010;
69(5):1169–75.
15. Driessen A, Frohlich M, Schafer N, Mutschler M, Defosse JM, Brockamp T,
Bouillon B, Sturmer EK, Lefering R, Maegele M, et al. Prehospital volume
resuscitation—Did evidence defeat the crystalloid dogma? an analysis of
the TraumaRegister DGU(R) 2002–2012. Scand j trauma resusc emerg med.
2016;24:42.
16. 20 years TraumaRegister DGU((R)): development, aims and structure. Injury
2014, 45 Suppl 3:S6–S13. https://www.ncbi.nlm.nih.gov/pubmed/25284237.
17. Grote S, Bocker W, Mutschler W, Bouillon B, Lefering R. Diagnostic value of
the Glasgow coma scale for traumatic brain injury in 18,002 patients with
severe multiple injuries. J neurotrauma. 2011;28(4):527–34.
18. Mutschler M, Nienaber U, Wafaisade A, Brockamp T, Probst C, Paffrath T,
Bouillon B, Maegele M, Traumaregister DGU. The impact of severe traumatic
brain injury on a novel base deficit- based classification of hypovolemic
shock. Scand j trauma resusc emerg med. 2014;22:28.
19. Zarzaur BL, Croce MA, Fischer PE, Magnotti LJ, Fabian TC. New vitals after
injury: shock index for the young and age x shock index for the old. J surg
res. 2008;147(2):229–36.
20. Wafaisade A, Wutzler S, Lefering R, Tjardes T, Banerjee M, Paffrath T, Bouillon
B, Maegele M, Trauma Registry of DGU. Drivers of acute coagulopathy after
severe trauma: a multivariate analysis of 1987 patients. Emerg med j. 2010;
27(12):934–9.
21. Maegele M, Lefering R, Yucel N, Tjardes T, Rixen D, Paffrath T, Simanski C,
Neugebauer E, Bouillon B, Society AGPotGT. Early coagulopathy in multiple
injury: an analysis from the German trauma registry on 8724 patients. Injury.
2007;38(3):298–304.
22. Lefering R. Strategies for comparative analyses of registry data. Injury. 2014;
45 Suppl 3:S83–8.
23. Pacagnella RC, Souza JP, Durocher J, Perel P, Blum J, Winikoff B, Gulmezoglu
AM. A systematic review of the relationship between blood loss and clinical
signs. Plos one. 2013;8(3):e57594.
24. Vandromme MJ, Griffin RL, Kerby JD, Mcgwin Jr G, Rue 3rd LW, Weinberg
JA. Identifying risk for massive transfusion in the relatively normotensive
patient: utility of the prehospital shock index. J trauma. 2011;70(2):384–8.
discussion 388–390.
25. Olaussen A, Peterson EL, Mitra B, O'Reilly G, Jennings PA, Fitzgerald M.
Massive transfusion prediction with inclusion of the pre-hospital Shock
Index. Injury. 2015;46(5):822–6. Epub 2014 Dec.
26. Bruijns SR, Guly HR, Bouamra O, Lecky F, Lee WA. The value of traditional
vital signs, shock index, and age-based markers in predicting trauma
mortality. J trauma acute care surg. 2013;74(6):1432–7.
27. Cannon CM, Braxton CC, Kling-Smith M, Mahnken JD, Carlton E, Moncure M. Submit your next manuscript to BioMed Central
Utility of the shock index in predicting mortality in traumatically injured
patients. J trauma. 2009;67(6):1426–30. and we will help you at every step:
28. Spahn DR, Bouillon B, Cerny V, Coats TJ, Duranteau J, Fernandez-Mondejar
• We accept pre-submission inquiries
E, Filipescu D, Hunt BJ, Komadina R, Nardi G, et al. Management of bleeding
and coagulopathy following major trauma: an updated European guideline. • Our selector tool helps you to find the most relevant journal
Crit care. 2013;17(2):R76. • We provide round the clock customer support
29. Spahn DR, Cerny V, Coats TJ, Duranteau J, Fernandez-Mondejar E, Gordini G,
• Convenient online submission
Stahel PF, Hunt BJ, Komadina R, Neugebauer E, et al. Management of bleeding
following major trauma: a European guideline. Crit care. 2007;11(1):R17. • Thorough peer review
• Inclusion in PubMed and all major indexing services
• Maximum visibility for your research

Submit your manuscript at

www.biomedcentral.com/submit