a b s t r a c t

Background: The correction of coagulopathy with Fresh frozen plasma is one of the main issues in the treat- ment of multiple-injured patients. Infectious and septic complications contribute to an adverse outcome in multiple-injured patients. Here, we investigated the role of FFP in the development of inflammatory complica- tions given within the first 48 hours.

Methods: A total of 2033 patients with multiple injuries and an Injury Severity Score greater than 16 points and aged 16 years or older were included. The population was subdivided into 2 groups: those who received FFP and those who did not. The data were analyzed using SPSS version 22.0. Associations between the data were tested using Pearson correlation. Independent predictivity was analyzed by binary logistic regression and multivariate regression. Data were considered as significant if P b .05.

Results: The prothrombin time at admission was significantly lower (68.5% +- 23.3% vs 81.8% +- 21.0% normal;

P b .001) in the group receiving FFP. The application of FFP led to a more severe Systemic Inflammatory Response Syndrome grade (3.0 +- 1.2 vs 2.2 +- 1.4; P b .001), to a higher Infection rate (48% vs 28%; P b .001), and to a higher sepsis rate (29% vs 13%; P b .001) in the patients receiving FFP. The correlations between SIRS and the in- cidence of infections and sepsis increased with the amount of FFP applied (P b .001).

Conclusions: Treatment with FFP of bleeding patients with multiple injuries enhances the risk of SIRS, infection, and sepsis; however, a multifactorial genesis has to be postulated.

(C) 2016

1. Introduction

Trauma is the major cause of death among the young adult popula- tion. Blood loss and coagulopathy of trauma shock are the key issues in multiple-injured patients [1,2]. Necrotic tissues and open wounds not only are at a high risk for infectious complications but also lead to the development of Systemic Inflammatory Response Syndrome [3]. The transfusion of at least 4 erythrocyte concentrates increases the inci- dence of SIRS and transfusion-associated lung injuries [4]. Correction of trauma-associated coagulopathy is pivotal for the survival of a patient with multiple injuries. In recent years, replacement of recombinant blood coagulation factors or factor mixtures has been suggested instead of fresh frozen plasma for the treatment of trauma-associated co- agulopathy. The application of coagulation factors in a concentrated form using high volumes overcomes the effects of coagulopathy very

* Corresponding author at: Division of Trauma Surgery, University Hospital of Zurich, Ramistrasse 100, 8091 Zurich, Switzerland. Tel.: +41 44 255 41 98.

E-mail address: [email protected] (L. Mica).

quickly [5-9]. Factors such as recombinant factor VII are very effective compared with FFP, but they are very expensive. This is limiting their widespread use [9-11]. It has been shown that the trauma-associated reduction in prothrombin time is a risk factor for SIRS and sepsis in pa- tients with multiple injuries and is associated with an increased mortal- ity [12]. The prothrombin time analyzed in this study might reflect not only the trauma-associated coagulopathy but also indirectly the loss of cytokines, parts of the complement system and carrier proteins. The ap- plication of FFP may lead to the application of anaphylatoxins produced by the apheresis technique during FFP production [13]. Therefore, cor- rection of trauma-associated coagulopathy with FFP might affect the immunological state and outcomes of patients with multiple injuries. Suggestions were made to transfuse FFP, platelets and red blood cells in a ratio of 1:1:1 or 1:1:2 in multiple-injured patients with good out- comes; however, the Injury Severity Score (ISS) was much lower than in the presented study [14]. Therefore, the thesis was elaborated to an- alyze the effect of FFP administration within the first 48 hours on the de- velopment of SIRS, infectious complications, sepsis, and mortality in patients with multiple injuries.

http://dx.doi.org/10.1016/j.ajem.2016.04.041

0735-6757/(C) 2016

1. Diagnostic protocol“>Patients and methods
   1. Patient sample

A total of 2033 patients with multiple injuries admitted to the emer- gency department of the University Hospital of Zurich (Switzerland) in the period 1996 to 2011 were included in this retrospective cohort study. The inclusion criteria were an ISS greater than 16 points, age 16 years or older, and admission within at least 24 hours of incurring the multiple injuries. The patient sample was subdivided into 2 groups (Table 1) according to administration of FFP. All patient data were col- lected retrospectively. The patient data were retrieved from patient re- cords with the approval of the local institutional review board according to the University of Zurich Institutional Review Board guidelines and the World Medical Association Declaration of Helsinki, and the study was conducted according to our institutional guidelines for good clinical practice (Ethics Committee of the University Hospital of Zurich Permis- sion: “Retrospektive Analysen in der Chirurgischen Intensivmedizin” Nr. St.V. 01-2008).

Diagnostic protocol

Unstable patients underwent resuscitative procedures according to the advanced trauma life support (ATLS) standards of the American Col- lege of Surgeons [15]. Hemodynamically stable patients received diag- noses according to the clinical findings or a whole-body computed tomographic scan in uncertain situations. hemodynamically unstable patients received focus-oriented diagnostics with immediate problem solving according to the ATLS guidelines.

Primary care

The treatment of all patients admitted was according to the ATLS guidelines and to the previously assessed trauma management proto- col, after appropriate indications had been identified [16,17].

Scoring systems

The Acute physiology and chronic health evaluation II score was used to evaluate the overall physiological impairment of the patient at admission [18]. The ISS and the New Injury Severity Scale

(NISS) were used to define the severity of trauma [19,20]. The Abbrevi- ated Injury Scale (AIS; 2005 version) was used to describe injuries in specific anatomical regions.

Laboratory parameters

Blood lactate levels, pH, and hematocrits were measured at intervals with a Blood gas analyzer (ABL 800 Flex; Radiometer GmbH, Thalwil, Switzerland). The prothrombin time was measured using a standard- ized method [21].

Treatment of trauma-associated coagulopathy

The treatment of trauma-associated coagulopathy in association with or without acquired or idiopathic coagulopathy followed interna- tional diagnostics and guidelines [22-33].

Transfusion resuscitation of Multiple trauma patients

The infusion and Transfusion therapy of multiple trauma patients was applied according to Damage control resuscitation criteria [34] and according to the guidelines of the University Hospital of Zurich [35].

Assessment of SIRS and sepsis

The worst parameters of leukocyte count, respiratory rate, heart rate, and temperature were taken to determine the SIRS score each day [36]. Systemic inflammatory response syndrome was measured during the first 31 days after admission or as long as the patients were hospitalized. Sepsis was defined as a SIRS score greater than or equal to 2 with an infectious focus.

Hypothetical sources of bias

All patients were selected retrospectively. The documentation of all parameters followed our Good Clinical Practice guidelines. Many differ- ent persons collected the data under the guidance of the personnel selecting the patients. The 15-year time span could have led to a bias in time-related changes in the treatment of trauma-associated coagu- lopathy; however, the definition of SIRS and sepsis remained the

Table 1

The characteristics of the patient groups at admission for those receiving FFP vs those not

Characteristics	FFP received	FFP not received	Total	P
n	609	1424	2033	b.001a
Age (y)	39.9 +- 17.6	45.0 +- 20.0	43.8 +- 19.5	b.001b
Sex (male/female)	463/146	1040/384	1503/530	b.001a
AIS head	2.8 +- 2.0	3.3 +- 1.9	3.1 +- 1.9	b.001b
AIS face	0.63 +- 1.1	0.66 +- 1.1	0.65 +- 1.1	.361c
AIS thorax	2.3 +- 1.7	1.8 +- 1.7	1.9 +- 1.7	b.001b
AIS abdomen	1.9 +- 2.1	1.0 +- 1.7	1.3 +- 1.9	b.001b
AIS pelvis	2.0 +- 1.6	1.3 +- 1.4	1.5 +- 1.5	b.001b
AIS extremities	1.0 +- 1.5	0.5 +- 1.1	0.6 +- 1.2	b.001b
AIS skin	0.6 +- 1.0	0.5 +- 0.8	0.5 +- 0.8	.004c
ISS	36.8 +- 12.7	31.0 +- 12.0	32.7 +- 12.5	b.001b
NISS	46.2 +- 14.1	41.6 +- 15.3	43.0 +- 15.1	b.001b
pH	7.27 +- 0.14	7.32 +- 0.14	7.30 +- 0.14	b.001b
Base excess (mEq/L)	-3.5 +- 5.4	-5.9 +- 5.8	-4.3 +- 5.7	b.001b
Lactate (mmol/L)	2.9 +- 2.7	3.8 +- 2.7	3.2 +- 2.7	b.001b
Hemoglobin (g/L)	9.6 +- 3.1	11.8 +- 4.6	11.0 +- 4.3	b.001b
Prothrombin time (%0	68.5 +- 23.3	81.8 +- 21.0	77.6 +- 22.6	b.001b
APACHE II score	18.3 +- 8.2	15.1 +- 9.0	16.1 +- 8.9	b.001b
Predicted mortality from APACHE II (%)	34 +- 22	27 +- 22	29 +- 23	b.001b
Erythrocytes 24 h (U)	13.7 +- 14.2	0.6 +- 2.3	4.4 +- 9.9	b.001b
Erythrocytes 48 h (U)	15.0 +- 15.0	0.8 +- 2.5	4.9 +- 10.6	b.001b
a ?2 Test.

^b Analysis of variance.

^c Mann-Whitney U test (Kolmogorov-Smirnov, P N .05).

Table 2

The development of SIRS, infection, and sepsis and associated complications among patients receiving FFP vs those not

	FFP received	FFP not received	Total	P
SIRS at admission	2.6 +- 1.0	2.2 +- 1.1	2.4 +- 1.1	b.001a
SIRS maximal value	3.0 +- 1.2	2.2 +- 1.4	2.4 +- 1.4	b.001a
SIRS maximal value (d)	4.6 +- 5.0	2.9 +- 4.2	3.5 +- 4.6	b.001a
SIRS duration (d)	2.8 +- 2.8	2.1 +- 2.1	2.2 +- 2.2	.054b
Infection rate (% of the group)	48	28	34	b.001c
Sepsis rate (% of the group)	29	13	17	b.001c
Sepsis duration (d)	8.3 +- 5.2	7.3 +- 5.9	7.8 +- 5.6	.094a
Septic shock (% of the group)	8	3	5	b.001c
a Analysis of variance. (Kolmogorov-Smirnov, PN.05).

^b Mann-Whitney U test.

^c ?² Test.

same. The scores and values were calculated from a single Excel sheet (Microsoft Office 2010, Redmond, WA).

Statistical analysis

Data are presented as the mean +- SD for continuous variables and as percentages for categorical variables. Cases with an incomplete data set were analyzed by missing completely at random test. Two-tailed Kolmogorov-Smirnov tests were used for testing normality, and if P N .05, the data were considered as normally distributed. The data for the FFP groups were compared using the ?² test, and the Kruskal- Wallis test was used for ordinal data; and the 1-way analysis of variance (ANOVA), for continuous data. If the Kolmogorov-Smirnov test showed P N .05, the Mann-Whitney nonparametric U test was used for continu- ous data. Results were considered statistically significant at P b .05. Pear- son correlation was calculated and is given as Pearson r with the corresponding P value (2 tailed). The Predictive ability of FFP was re- ported as the area under the receiver operating characteristic (ROC) curve (AUC) +- SE with a corresponding confidence interval (CI) of 95%. Odds ratios (ORs) were calculated for categorical data. The inde- pendent predictivity was analyzed by binary logistic regression and multinominal logistic regression. The multinomial binary logistic re- gression was used to adjust the FFP effect for the possible cofounders. The data were considered as independent predictive variables if P b .05. The data were analyzed using SPSS for IBM software (version 22.0; IBM Corp, Armonk, NY).

1. Predictive power of the application”>Results
   1. Patient sample

The patient group not receiving FFP within the first 48 hours was sig- nificantly larger (1424 vs 609; P b .001). There were significantly more male (1503) than female patients (530) in both groups (male vs female; P b .001). The patients who received FFP were significantly younger than the patients who did not (39.9 +- 17.6 vs 45.0 +- 20.0 years; P b .001; Table 1). The patients receiving FFP were significantly more se- verely injured (ISS, 36.8 +- 12.7 vs 31.0 +- 12.0, P b .001; NISS, 46.2 +-

14.1 vs 41.6 +- 15.3, P b .001; Table 1). Although the same distribution was observed in the AIS values for each anatomical region, the FFP- treated patients had a significantly higher AIS (Table 1), except for the

face. The physiological state at admission was generally worse in the pa- tients who received FFP. These patients were more acidotic (7.27 +- 0.14 vs 7.32 +- 0.14 pH; P b .001) and had a greater degree of base excess (-3.5 +- 5.4 vs -5.9 +- 5.8 mEq/L; P b .001), and their mean hemoglo- bin level was significantly lower (9.6 +- 3.1 vs 11.8 +- 4.6 g/L; P b .001). Interestingly, the lactate value at admission was significantly lower in the FFP group (2.9 +- 2.7 vs 3.8 +- 2.7 mmol/L; P b .001). The calculated predicted mortality based on the APACHE II score was increased signif- icantly in the FFP group (34% +- 22% vs 27% +- 22%; P b .001; Table 1).

Analysis of SIRS, infection, and sepsis

The SIRS score at admission was significantly elevated in the group receiving FFP (2.6 +- 1.0 vs 2.2 +- 1.1; P b .001). An increase in the SIRS score was observed over time in the same group (3.0 +- 1.2 vs 2.2 +- 1.4; P b .001); however, the maximum was reached more slowly in these patients (4.6 +- 5.0 vs 2.9 +- 4.2 days; P b .001; Table 2). The

rates of infection (48% vs 28%; P b .001) and sepsis (29% vs 13%;

P b .001) were significantly elevated in the FFP group (Table 2).

Correlations between SIRS, the rates of infection and sepsis, and the application of FFP

Analyses of the associations between the application of FFP with SIRS and the development of infection and sepsis over 48 hours revealed a strong correlation between FFP use and the duration of SIRS, the maxi- mum SIRS grade, and the incidence of sepsis (P b .001; Table 3).

Predictive power of the application of FFP for SIRS, infection, and sepsis

Receiver operating characteristic analysis showed the highest pre- dictive power for SIRS (AUC, 0.664 +- 0.013; P b .001; 95% CI, 0.639-

0.69; OR, 5.04), followed by infection (AUC, 0.6 +- 0.014; P b .001; 95%

CI, 0.572-0.628; OR, 2.36) and sepsis (AUC, 0.504 +- 0.015; P b .001;

95% CI, 0.553-0.61; OR, 2.81) (Figure). The multistep binary logistic re- gression revealed FFP (P b .001) as an independent predictor for SIRS, in- fection, and sepsis in patients with multiple injuries (Table 4). However, FFP was not the only independent predictive variable for SIRS, infection, and sepsis (Table 4).

Table 3

Pearson correlations between the given amount of FFP units (n) within the first 48 hours and the development of SIRS, infection, and sepsis shown as Pearson correlation coefficient r; significant if P b .05

Pearson r/P value	FFP 1 h	FFP 2 h	FFP 3 h	FFP 4 h	FFP 6 h	FFP 8 h	FFP 12 h	FFP 24 h	FFP 48 h	FFP total
Total FFP units (n)	0.1 +- 0.7	0.6 +- 2.3	1.2 +- 3.7	1.6 +- 4.7	2.1 +- 5.8	2.4 +- 6.4	3.0 +- 7.6	3.5 +- 8.9	3.9 +- 10.0	19.4 +- 48.7
SIRS maximum	0.021/.348	0.061/.008	0.109/b.001	0.134/b.001	0.167/b.001	0.181/b.001	0.188/b.001	0.205/b.001	0.224/b.001	0.182/b.001
Day of SIRS maximum	-0.039/.100	0.005/.831	0.048/.042	0.065/.006	0.075/b.001	0.094/b.001	0.105/b.001	0.111/b.001	0.127/b.001	0.098/b.001
SIRS duration (d)	-0.039/.592	0.238/b.001	0.173/.006	0.172/.006	0.155/.013	0.162/.009	0.207/b.001	0.234/b.001	0.234/b.001	0.215/b.001
Infection rate (%)	-0.046/.038	0.037/.098	0.082/b.001	0.100/b.001	0.117/b.001	0.133/b.001	0.154/b.001	0.173/b.001	0.192/b.001	0.148/b.001
Sepsis rate (%)	-0.031/.172	0.047/.034	0.098/b.001	0.115/b.001	0.132/b.001	0.148/b.001	0.182/b.001	0.209/b.001	0.234/b.001	0.177/b.001

Table 5

SIRS: Odds: 5.04 ; AUC: 0.664; P < .001; CI 95% 0.639, 0.690

Infection: Odds: 2.36; AUC: 0.600; P < .001; CI95% 0.572, 0.628

SEPSIS: Odds: 2.81; AUC: 0.504; P < .001; CI 95% 0.553, 0.610

Distribution of the infectious side

Infection site	FFP received	FFP not received	P
Pneumonia	32.8%	16.3%	b.001a
Urinary tract infection	8.4%	5.1%	.006a
Wound infection	15.4%	5.3%	b.001a
Abdominal infection	4.1%	1.0%	b.001a
Bacteremia	15.3%	7.8%	b.001a
Central nervous system	5.0%	2.3%	.003a
Catheter infection	11.6%	5.2%	b.001a
Osteomyelitis	3.9%	3.5%	.670a
Other	3.6%	3.9%	.706a
a Kruskal-Wallis test.

Analysis of the infection site

The analysis of the infection site revealed significant more infection in FFP-treated patients than in not treated patients (Table 5). Interestingly, no differences were found in the rate of osteomyelitis (P = .670; Table 5).

Outcome of the patient sample

Figure. The predictive power of the FFP dose given for the occurrence of SIRS and the rates of development of infection and sepsis based on ROC curves.

Table 4

Identification of possible co-founders for SIRS, infection and Sepsis.

SIRS Odds^a P^a Odds^b P^b

FFP	1.030	.006	0.997	.124
ISS	1.016	.052	0.983	.037
NISS	1.004	.548	0.996	.551
pH	1.637	.525	0.613	.528
Base excess (mEq/L)	0.974	.277	1.029	.234
Lactate (mmol/L)	0.926	.048	1.081	.044
Hematocrit (%)	0.986	.604	1.016	.562
Hemoglobin (g/L)	0.995	.950	1.007	.933
Prothrombin time (%)	1.008	.034	0.992	.043
Erythrocytes 24 h	0.889	.048	1.124	.048
Erythrocytes 48 h	1.169	.005	0.855	.005
Infection FFP	1.047	b.001	0.993	b.001
ISS	1.011	.105	0.988	.077
NISS	1.011	.057	0.990	.059
pH	0.865	.851	1.146	.859
Base excess (mEq/L)	0.992	.718	1.009	.689
Lactate (mmol/L)	0.923	.039	1.087	.032
Hematocrit (%)	0.969	.255	1.032	.252
Hemoglobin (g/L)	1.053	.538	0.951	.549
Prothrombin time (%)	1.016	b.001	0.984	b.001
Erythrocytes 24 h	0.958	.254	1.044	.254
Erythrocytes 48 h	1.083	.024	0.924	.024
Sepsis FFP	1.036	b.001	1.005	=.001
ISS	1.021	.009	1.022	.006
NISS	1.011	.109	1.011	.108
pH	0.240	.100	0.234	.093
Base excess (mEq/L)	1.007	.794	1.005	.843
Lactate (mmol/L)	0.921	.090	0.920	.080
Hematocrit (%)	0.904	.011	0.903	.011
Hemoglobin (g/L)	1.313	.024	1.311	.024
Prothrombin time (%)	1.012	.006	1.012	.007
Erythrocytes 24 h	0.976	.523	1.025	.523
Erythrocytes 48 h	1.071	.058	0.934	.058

Depicted is the influence of FFP given within the first 48 hours on the development of SIRS, infection, and sepsis. As shown, the genesis of SIRS in multiple injured patients has a mul- tifactorial origin. Possible cofounders for SIRS and sepsis are highlighted in boldface. If P b .05, the factor was considered as an independent predictor for SIRS, infection, and sepsis.

^a Binary logistic regression.

^b Multinominal logistic regression.

Interestingly, the overall mortality was lower in the FFP-treated group (30% vs 33%; P b .001) (Table 6) and was lower than the mortality predict- ed by the APACHE II score (Table 1). The hospitalization time (24.8 +- 23.2 vs 15.8 +- 19.0 days; P b .001), intensive care unit stay (14.1 +- 13.0 days; P b .001), and the ventilator-assisted duration (9.8 +- 10.0 vs 4.9 +- 8.0 days; P b .001) were significantly increased in the group of FFP-treated patients (Table 4). However, those patients who died had survived longer in the FFP-treated group (6.1 +- 12.2 vs 2.7 +- 5.9 days; P b .001; Table 6). The effect of FFP was adjusted for cofounders and revealed only signifi- cance as a contributor for sepsis (Table 7).

1. Discussion

Trauma is one of the major causes of early death among young adults. The prothrombin time in patients with multiple injuries serves not only as an independent predictor of death but also as a risk factor for sepsis if it is less than 70% of normal [3,12]. There were significant correlations be- tween SIRS and the rates of infection and sepsis after treatment with FFP as well as strong predictive values in ROC analysis with ORs greater than 2 (Figure). Thus, it appears that FFP treatment worsened SIRS and the rates of infection and sepsis. This adverse side effect might be based in the composition of FFP, which does not just consist of proteins from the coagulation cascade. The major components are water and albumin, followed by ?₁-, ?₂-, ?-, and ?-globulins. The ?-globulins from the donor might cross-react with host epitopes and activate the complement system in a classic way. Complement peptides C3a and C5a act as anaphylatoxins and activate the immune system leading to unspecific hy- peractivity [37]. The alternative pathway for activation of the complement system is the spontaneous decay of peptides C3 to C3a catalyzed by C3 convertase, which is possible in FFP units stored for a long time. In healthy persons, C3 convertase is neutralized by factor H and factor I [37]. In pa- tients with multiple injuries, it might remain intact for longer because of the loss of plasma upon exsanguination and act as an anaphylatoxin

Table 6

The outcome of the patient groups between those receiving FFP vs those not Outcome FFP received FFP not received Total P

Hospitalization (d)	24.8 +- 23.2	15.8 +- 19.0	18.5 +- 20.8 b.001a
ICU (d)	14.1 +- 13.0	7.8 +- 10.5	9.7 +- 11.6 b.001a
Ventilator (d)	9.8 +- 10.0	4.9 +- 8.0	6.4 +- 8.9 b.001a
Dead (d)	6.1 +- 12.2	2.7 +- 5.9	3.6 +- 8.3 b.001a
Dead (% of the group)	30	33	32 b.001b

^a Analysis of variance. (Kolmogorov-Smirnov, Pb.05).

^b ?² Test.

Table 7

Multinominal binary logistic regression of the investigated factors for SIRS, sepsis, and death

	SIRS			Sepsis			Death
	Odds	P		Odds	P		Odds	P
FFP	0.997	.425	0.991		.037	0.990		.538
ISS	1.000	.997	1.047		.005	0.926		.303
NISS	1.022	.228	0.994		.645	0.994		.094
pH	9.271	.342	2.349		.686	0.046		.392
Base excess (mEq/L)	0.935	.290	1.002		.979	1.034		.730
Lactate (mmol/L)	0.988	.890	1.019		.823	1.118		.428
Hematocrit (%)	0.976	.622	0.980		.691	0.941		.469
Hemoglobin (g/L)	1.182	.255	1.041		.798	1.100		.698
Prothrombin time (%)	1.003	.742	1.005		.597	0.974		.051
Hospitalization (d)	1.036	.027	0.982		.026	0.652		b.001
ICU (d)	0.973	.576	1.062		.052	0.792		.074
Ventilator (d)	1.141	.030	1.037		.292	1.782		b.001
Age (y)	1.022	.084	1.007		.425	1.022		.084
Infection	0.848	.714	0.048		b.001	1.725		.357
Sex	0.652	.245	1.370		.398	2.769		.107
Erythrocytes 24 h	0.899	.200	0.979		.705	1.079		.489
Erythrocytes 24 h	1.117	.175	1.066		.253	0.922		.437
Constant	7,763E10	.349	0.000		.558	7.763E10		.349

The influence of FFP for SIRS, sepsis, and death is corrected for cofounders.

enhancing the inflammatory reaction. However, the FFP-induced activa- tion of the complement system might not be sufficient. The patient with multiple injuries has much necrotic tissue, and many hidden antigens such as collagen and DNA can be exposed to the circulation. These 2 mol- ecules lead to a further activation of the complement system in a classic way, leading to further production of the anaphylatoxins C3a and C5a [37]. Furthermore, it is a well-known fact that FFP has certain concentra- tion of anaphylatoxins leading to inflammation [13,38]. A multifactorial origin for the increased infection and sepsis rates in the FFP-treated group has to be postulated, such as a higher ISS and APACHE II score at ad- mission (Table 1), but also an adverse effect of FFP as reflected by the bi- nary regression analysis. The multifactorial origin of the SIRS, infection, and sepsis is supported by a low AUC of the ROC. Consumption of the complement system and the compensatory anti-inflammatory response syndrome [39] causing infections seems to be a good explanation. The hy- peractivation of the immune system might also be reflected in the out- come for this patient sample. Thus, the prolonged ICU stay, longer ventilator-assisted treatment, and hospitalization might have been asso- ciated with these more severe inflammatory reactions. Interestingly, the mortality rate was significantly reduced in the group treated with FFP even when those patients were more severely injured. Whether this was because they were younger or had a more favorable medical history could not be decided by this analysis.

Treatment with FFP of bleeding patients with multiple injuries enhances

the risk of SIRS, infection, and sepsis. These effects in this patient sample might have had a multifactorial origin aside from their multiple injuries.

1. Limitations

The main limitations of this study were seen in the multifactorial genesis of SIRS and sepsis and the retrospective character.

Competing interests

The authors Ladislav Mica, Hanspeter Simmen, Clement ML Werner, Michael Plecko, Catharina Keller, Stefan H. Wirth and Kai Sprengel declare no conflict of interest, and no conflict of interest will arise upon publication of this manuscript.

References

1. Frith D, Brohi K. The acute coagulopathy of trauma shock: clinical relevance. Surgeon

2010;8:159-63.

1. Brohi K, Cohen MJ, Ganter MT, Matthay MA, Mackersie RC, Pittet JF. Acute traumatic coagulopathy initiated by hypoperfusion: modulated through the protein C path- way? Ann Surg 2007;245:812-8.
2. Mica L, Furrer E, Keel M, Trentz O. Predictive ability of ISS, NISS, and APACHE II score for SIRS and sepsis in polytrauma patients. Eur J Trauma Emerg Surg 2012;38: 665-71.
3. Beale E, Zhu J, Chan L, Shulman I, Harwood R, Demetriades D. Blood transfusion in critically injured patients: a prospective study. Injury 2006;37:455-65.
4. Joseph B, Amini A, Friese RS. Factor IX complex for the correction of traumatic coag- ulopathy. J Trauma Acute Care Surg 2012;72:828-34.
5. Joseph B, Hadjizacharia P, Aziz H, Kulvatunyou N, Tang A, Pandit V, et al. Prothrom- bin complex concentrate: an effective therapy in reversing the coagulopathy of trau- matic brain injury. J Trauma Acute Care Surg 2013;74:248-53.
6. Dickneite G, Pragst I. prothrombin complex concentrate vs fresh frozen plasma for reversal of dilutional coagulopathy in a porcine trauma model. Br J Anaesth 2009; 102:345-54.
7. Dickneite G, Dorr B, Kaspereit F, Tanaka K. Prothrombin complex concentrate versus recombinant factor VIIa for reversal of hemodilutional coagulopathy in a porcine trauma model. J Trauma 2010;68:1151-7.
8. Stein DM, Dutton RP, Kramer ME, Handley C, Scalea TM. Recombinant factor VIIa: decreasing time to neurosurgical intervention in patients with severe traumatic brain injury. J Trauma 2008;64:620-8.
9. Brown CV, Foulkrod KH, Lopez D. Recombinant factor VIIa for the correction of coag- ulopathy before emergent craniotomy in blunt trauma patients. J Trauma 2010;68: 348-52.
10. Stein DM, Dutton RP, Kramer ME, Scalea TM. Reversal of coagulopathy in critically ill patients with traumatic brain injury: recombinant factor VIIa is more cost-effective than plasma. J Trauma 2009;66:63-72.
11. Mica L, Rufibach K, Keel M, Trentz O. The risk of early mortality of polytrauma pa- tients associated to ISS, NISS, APACHE II values and prothrombin time, a retrospec- tive cohort study. J Trauma Manag Outcomes 2013;7:6.
12. Sonntag J, Stiller B, Walka MM, Maier RF. Anaphylatoxins in fresh-frozen plasma. Transfusion 1997;37:798-803.
13. Holcomb JB, Tilley BC, Baraniuk S, Fox EE, Wade CE, Podbielski JM, et al. Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma: the PROPPR randomized clinical trial. JAMA 2015; 313:471-82.
14. ATLS Subcommittee, American College of Surgeons’ Committee on Trauma, Interna- tional ATLS working group. Advanced trauma life support (ATLS(R)): the ninth edi- tion. J Trauma Acute Care Surg 2013;74:1363-6.
15. Wang K, Liu B, Ma J. Research progress in traumatic brain penumbra. Chin Med J (Engl) 2014;127:1964-8.
16. Ertel W, Trentz O. Causes of shock in the severely traumatized patient: emergency treatment. In: Goris RJA, Trentz O, editors. The integrated approach to trauma care: the first 24 hours. Berlin: Springer; 1995. p. 78-87.
17. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease

classification system. Crit Care Med 1985;13:818-29.

1. Baker S, O’Neill B, Haddon Jr W, Long WB. The injury severity score: a method for de- scribing patients with multiple injuries and evaluating emergency care. J Trauma 1974;14:187-96.
2. Champion HR, Copes WS, Sacco WJ, Lawnick MM, Bain LW, Gann DS, et al. A new characterization of injury severity. J Trauma 1990;30:539-45.
3. Jackson CM, White GC. Scientific and standardization committee communication: a ref- erence system approach to future standardization of laboratory tests for hemostasis. https://c.ymcdn.com/sites/www.isth.org/resource/resmgr/ssc/positionpaper.pdf.
4. Society of thoracic surgeons Blood Conservation Guideline Task Force, Ferraris VA, Brown JR, Despotis GJ, Hammon JW, Reece TB, et al. 2011 update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists blood conser- vation clinical practice guidelines. Ann Thorac Surg 2011;91:944-82.
5. Gorlinger K, Dirkmann D, Hanke AA, Kamler M, Kottenberg E, Thielmann M, et al. First-line therapy with coagulation factor concentrates combined with point-of- care coagulation testing is associated with decreased allogeneic blood transfusion in cardiovascular surgery: a retrospective, single-center cohort study. Anesthesiolo- gy 2011;115:1179-91.
6. Rossaint R, Bouillon B, Cerny V, Coats TJ, Duranteau J, Fernandez-Mondejar E, et al. Task force for advanced bleeding care in trauma management of bleeding following major trauma: an updated European guideline. Crit Care 2010;14:R52.
7. Schmid P. Low dose recombinant factor VIIa for Massive bleeding: a single center ob- servational cohort study with 73 patients. Swiss Med Wkly 2011;141:w13213.
8. Chassot PG, Delabays A, Spahn DR. Perioperative Antiplatelet therapy: the case for con- tinuing therapy in patients at risk of myocardial infarction. Br J Anaesth 2007;99:316-28.
9. Vincent JL, Rossaint R, Riou B, Ozier Y, Zideman D, Spahn DR. Recommendations on the use of recombinant activated factor VII as an adjunctive treatment for massive bleeding-a European perspective. Crit Care 2006;10:R120.
10. Despotis G, Renna M, Eby C. Risk associated with bleeding and transfusion: rationale for the optimal management of bleeding after cardiac surgery. Eur J Anaesthesiol 2007;24(Suppl. 40):15-36.
11. Szefner J. Preventing and managing postoperative bleeding in cardiac surgery by controlling disseminated intravascular coagulation. Eur J Anaesthesiol 2007; 24(Suppl. 40):59-66.
12. Innerhofer P. Perioperative management of coagulation. Hamostaseologie 2006;26: 3-14.
13. Fries D. Coagulation management in massive transfusion. Hamostaseologie 2006;26:

15-20.

1. Lang T, von Depka M. Possibilities and limitations of thrombelastometry/ thrombelastography. Hamostaseologie 2006;26:21-9.
2. Tsai HM. Shear stress and von Willebrand factor in health and disease. Semin Thromb Hemost 2003;29:479-88.
3. Jansen JO, Thomas R, Loudon MA, Brooks A. Damage control resuscitation for pa- tients with major trauma. BMJ 2009;338:b1778.
4. Theusinger OM, Madjdpour C, Spahn DR. Resuscitation and transfusion management in trauma patients: emerging concepts. Curr Opin Crit Care 2012;18:661-70.
5. [No authors listed] American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and

guidelines for the use of innovative therapies in sepsis Crit Care Med 1992;20: 864-74.

1. Walport MJ. Complement. N Engl J Med 2001;344:1058-66.
2. Sonntag J, Emeis M, Vornwald A, Strauss E, Maier RF. Complement activation during plasma production depends on the apheresis technique. Transfus Med 1998;8: 205-8.
3. Shubin NJ, Monaghan SF, Ayala A. Anti-inflammatory mechanisms of sepsis. Contrib Microbiol 2011;17:108-24.