Following local ethics committee approval (415-EP/73/197-2013) we performed a retrospective analysis of major trauma patients admitted to the AUVA Trauma Centre Salzburg between 2010 and 2012. The inclusion criterion was administration of fibrinogen concentrate as first-line coagulation therapy, which also included patients who additionally received PCC and FFP. Exclusion criteria were burns, pregnancy and participation in other studies. Patients receiving anticoagulation medication pre-trauma were not excluded from the study as this information is not always known when a patient arrives at the ER. Patients are usually treated with a ROTEM guided algorithm, receiving fibrinogen concentrate and PCC according to their actual needs.
Demographic data, laboratory data, trauma scores such as injury severity score (ISS), new injury severity score (NISS), Glasgow coma scale (GCS) and outcomes data were obtained from the electronic trauma database.
Blood samples were drawn as soon as possible following ER admission, either from arterial lines or, in less severe cases, from venous lines. For coagulation monitoring, citrated blood was collected in two standard coagulation tubes. Viscoelastic testing was run within minutes on a ROTEM analyser in the ER. The following tests were performed: an extrinsically activated test using tissue factor (EXTEM) and a fibrin polymerization test (FIBTEM) which inhibits platelets’ contribution to clot elasticity. The following parameters were investigated: EXTEM clotting time (CT, normal range 38–79 seconds), EXTEM clot formation time (CFT, normal range 34–159 seconds), EXTEM alpha angle (normal range 63–83°), EXTEM maximum clot firmness (MCF, normal range 50–72 mm) and FIBTEM CA10 (normal range 7–23 mm).
A second citrated blood sample was drawn for standard coagulation tests (SCTs) in the central laboratory. The following parameters were assessed after centrifugation of the sample: fibrinogen concentration (Clauss Method, normal range 200–450 mg/dL), prothrombin time (PT, normal range 11.0–16.1 seconds) and activated partial thromboplastin time (aPTT, normal range 23.7–34.9 seconds).
Blood cell count and blood gas analyses were performed upon ER admission and ICU admission; haemoglobin (Hb, normal range 13.0–17.7 g/dL) and platelet count (normal range 150–350,000/μL) were also assessed. ROTEM analyses or SCTs were performed upon ICU admission for patients with suspected coagulopathy. We used the timepoint “ICU” to describe assessments made at the end of surgery or during the first few hours after ICU admission.
SCTs were also run every morning at 7 am and, depending on the patient’s condition, at 4 pm. For the study, we applied the timepoint “24 hours” to describe analyses performed at either 7 am or 4 pm, depending which was closer to 24 hours after hospital admission.
Patients were treated according to our institutional algorithm which was published recently . Briefly, the indication for fibrinogen concentrate is to increase FIBTEM CA10 to 10–12 mm for patients with low FIBTEM CA10 (<7 mm). Values below 7 mm indicate reduced fibrin polymerization, often resulting from a low plasma fibrinogen level. Fibrinogen concentrate (Haemocomplettan® P; CSL Behring, Marburg, Germany) is then administered, at a dose of 2–6 g (2–4 g if initial FIBTEM CA10 4–6 mm; 6 g if initial FIBTEM CA10 0–3 mm). Platelet concentrate is transfused in patients whose EXTEM CA10 remains low (<40 mm) after increasing FIBTEM CA10 to 10–12 mm. The algorithm also recommends considering PCC if EXTEM CT is >80 seconds after raising FIBTEM CA10 to 10–12 mm, or if EXTEM CA10 <30 mm . Prothromplex® (Baxter, Vienna, Austria) was used in most cases. According to our algorithm, tranexamic acid (TXA) should be given if the patient is in shock or if ISS is greater than 15 . However, TXA was not included in the algorithm until 2012, meaning that most of the patients included in this study did not receive TXA and that the use of this product increased during the observation period. Coagulation therapy data were obtained from the anaesthesia records and ICU medical charts. FFP was used mainly on individual discretion of the attending anaesthetist, particularly in cases of ongoing major bleeding, based on the consideration that factors not present in PCC e.g., factor V are missing in the late stages of a major trauma.
The target haemoglobin concentration during the initial operative procedure was 10 g/dL. In the postoperative phase, lower levels of 7–8 g/dL were accepted, depending on the hemodynamic condition of the patient. All transfused allogeneic blood products were recorded in a database (DataLab, Bartels, Graz, Austria), in accordance with standard practice at our centre. Additionally, the amount of cell-saver blood administered was obtained from anaesthesia charts. Each 250 mL of cell-saver blood transfused was considered equivalent to 1 unit of red blood cells (RBCs).
Mortality was defined as death within 30 days of ER admission.
The primary endpoint in this study was plasma fibrinogen concentration and fibrin clot parameters on ICU admission. For all parameters, normality of the data distribution was tested using the Kolmogorov-Smirnov test. Normally distributed parameters were reported as mean ± standard deviation, and those with non-normal distribution were expressed as median and interquartile range (IQR; 25th percentile – 75th percentile). For categorical variables, p-values were derived from Fisher´s exact or the chi-square test. For continuous variables including those with time dependency, between-group differences were analysed using analysis of variance (ANOVA) and the Newman-Keuls test (comparisons of three groups) or the Mann–Whitney U test (comparisons of two groups). For within-group comparison of timepoints, the t-test or Mann–Whitney U test was used.
Data were analysed for all patients with a value at the timepoint of interest (e.g. patients who died within 24 hours were not included in 24-hour outcomes). Statistical calculations were performed using GraphPad Prism 5.03 (GraphPad Software, La Jolla, CA, USA). The level of significance was set at p < 0.05.