Fibrinogen levels in trauma patients during the first seven days after fibrinogen concentrate therapy: a retrospective study

The most striking result of the current study is that the evolution of FIB levels from day 3 to day 7 is similar between patients with and without FC as haemostatic therapy. Median FIB levels appeared to reach a plateau at approximately 7 g/l between day 5 and day 7 post-trauma, irrespective of initial fibrinogen supplementation. It is also notable in our study that post-traumatic evolution of FIB levels did not differ between the two groups despite higher ISS, transfusion requirements, allogeneic blood product use, or PCC administration in the treatment group.

It has previously been shown that FIB levels at admission to the ER depend partly on factors such as blood loss/dilution, signs of shock and severity of injury [2]. An analysis of trauma patients, categorized according to whether they received FC alone, FC together with PCC, or FC combined with PCC and FFP, showed that FIB levels 24 h post-trauma (day 1) did not differ between treatment groups [24]. We could recently show in a small number of patients treated with FC and PCC, FC alone, or no coagulation therapy, investigating thrombin generation over the time course of 7 days that FIB levels rose in a similar manner [25]. This finding is in line with data from the Cryostat study which revealed that fibrinogen supplementation with cryoprecipitate had no significant effect on plasma FIB levels in trauma patients from 24 h to 28 days post-treatment [27].

In contrast to the assumption that fibrinogen supplementation may potentially lead to higher FIB levels post-trauma, our current data from 435 patients show for the first time that irrespective of the amount of FC given, FIB levels over the following seven days were not higher in the FC treatment subgroups than in the control group that did not receive FC but were even lower up to day 5 in the group receiving 10 g or more. This finding is in agreement with data from a porcine trauma model which demonstrated that administration of human FC did not down-regulate endogenous fibrinogen synthesis [28].

The rise of FIB levels from day 2 onwards in our study can be attributed to an upregulated fibrinogen synthesis in the liver, occurring in both study groups as part of the acute phase response after tissue injury [29]. Increased FIB levels during the acute phase response has also been shown in patients who received FC as targeted first-line haemostatic therapy in aortic surgery; FC provided a specific, significant, short-lived increase in FIB [23]. This short-lived increase in FIB levels following FC administration has also been shown in patients undergoing cystectomy or cardiopulmonary bypass [7, 43]. In both studies, FIB levels were comparable with the control group 24 h post-operatively. However, the time course of FIB in the subsequent post-operative days was not determined in either study. In the current study FIB levels appeared to reach a plateau between day 5 and day 7 post-trauma (with no significant differences between those days), irrespective of initial fibrinogen supplementation. However, FIB levels after day 7 were not investigated; therefore, we do not know whether (post-operative) FIB levels on day 10, as reported by Solomon et al. [23] in aortic surgery would still be in the plateau or already in the declining phase. It is worth noting that the aforementioned study addresses only two timepoints with an 8-day interval in between. Thus, it is unknown whether FIB levels at day 10 represented a continuing increase, or whether FIB levels might have already peaked between days 2 and 10.

Patients in this study were treated according to a goal-directed algorithm [26] based on quickly available ROTEM parameters, such as FIBTEM clot amplitude at 10 min (A10). On admission to the ER, median FIB levels in the group receiving FC were below 2 g/l. Thus, FIBTEM-based diagnosis of impaired fibrinogen contribution to clotting, as carried out in our institution, appears to be consistent with current internationally recommended plasma level triggers [17, 18] in determining the need for fibrinogen supplementation in trauma patients [5]. The current subgroup analysis shows that admission FIB levels in all patients were significantly related to the amount of FC administered in the first 24 h. Goal-directed therapy with FC has been shown to reduce transfusion requirements in various clinical settings [8, 15, 20, 30‐33]. In addition, FC has been shown to significantly diminish bleeding volume in both the clinical setting and in animal models of trauma [8, 34‐36].

There is evidence to support the effectiveness of FC in a variety of clinical settings. However, only a few studies have reported on the safety profile of FC [23, 30, 32, 33]. The most compelling evidence for the safety of FC has been provided by a 27-year pharmacosurveillance program, which reported a low incidence of thromboembolic events [37]. A preclinical study of trauma has shown that an increase in thrombin generation after FC administration is short-lived (<6 h) [38]. A clinical study did not find an increase in endogenous thrombin potential between 24 h and 7 days post-treatment with FC [25, 38]. Consistent with these findings, the results of our study suggest that for the trauma population, use of FC does not appear to increase the thrombogenic status above that produced by the acute phase response. Furthermore, polymerized fibrin, also known as antithrombin I, can actually inhibit thrombin generation by sequestering thrombin in the forming clot [39], though the antithrombin effect has yet to be quantified with regard to FIB levels.

As was to be expected, haemoglobin was lower at admission in the treatment group at all timepoints and in line with the higher transfusion rate of RBCs in the treatment group. The time course of haemoglobin levels over seven days post-trauma differed slightly between control and treatment groups but mainly remained stable at around 95 and 90 g/l, respectively. This is most likely attributable to a balance of minor blood loss (drains, blood sampling, re-operations) and minor transfusion requirements for RBCs at the ICU, not further investigated for the purpose of this study.

Interestingly, platelet count remained lower in the treatment group at all timepoints and was even more significantly diminished in the subgroups receiving more FC. This may be attributable to the severity of injury and possible higher systemic inflammatory response [40]. This can also be supported by the results of IL-6, CRP and PCT, being higher in the treatment groups in our study. Generally, CRP levels reached maximum levels by day 3 or day 4, which is in agreement with previous studies showing that CRP levels were elevated for several days following trauma and were not correlated with trauma severity [41, 42].

It is of particular interest that in the treatment group PTI was lower throughout and aPTT was prolonged at the majority of timepoints despite the use of PCC in about half of these patients, suggesting these patients have a lower capacity to initiate clotting. It is important to note that a lower PTI or prolonged aPTT may occur despite the higher endogenous thrombin potential that usually exists during haemodilution following trauma and after PCC administration [44, 45]. For example, Dunbar and Chandler showed that when all coagulation factors and inhibitors in plasma are diluted, thrombin generation increased and peaked when the plasma was diluted to 40 % of normal [44].

Interestingly, one patient in the control group received no FC therapy despite receiving massive transfusion. This patient, who was bleeding and on pre-trauma vitamin K antagonist therapy, needed PCC only, and maintained sufficient functional fibrin polymerisation throughout RBC transfusion.

Our study has several limitations. Data were collected retrospectively from a single centre, and the baseline characteristics between the two groups were not comparable. We included only those patients with FIB values available post-trauma up to day 6 or 7. For the purpose of this study we documented FC, PCC, platelet concentrate and RBC use in the first 24 h only. Therefore, the potential effects of similar therapies after 24 h on the study parameters could not be determined–nor could the potential impact of other therapies at any timepoint be excluded (e.g., tranexamic acid). The data in this study were collected over a long time period (2005–2015), during which changes occurred in our clinic’s standard transfusion protocols e.g., initially, PCC was administered more liberally but now this is only administered after FC administration and an ongoing prolonged EXTEM clotting time (CT). Therefore, differences in patient management might have impacted our assessments of patients’ haemostatic outcomes. We cannot exclude the impact of age on the evolution of FIB levels; interestingly, patients in the control group were significantly older than those in the treatment group. However, in the range of observed ages we consider this difference to be of minor importance. It is also noteworthy that some patients in the control group who did not present with bleeding at admission did not receive FC despite FIB levels below the current recommended threshold levels. We did not investigate the prothrombotic effects of FC in this study and were unable to include any information on thromboembolic events as patients are not routinely screened for deep vein thrombosis in our centre. The main problem being that markers of thromboembolic events, such as D-dimers, are not suitable for establishing these events in trauma patients. However, due to the overall equal time course of FIB levels regardless of administration of FC, this should not impact such events per se.

Furthermore, the tendency for lower fibrinogen in the highest FC group may be because of the need for further operations after the initial damage control surgery in those patients with more severe trauma. We cannot exclude that this may have biased the results of this group to some extent. However, it is interesting that the patients who received the highest amount of FC had the most severe injuries and presumably the most secondary “hits” (operations), and still present with a tendency to lower fibrinogen levels.