Fibrinogen concentrate as first-line therapy in aortic surgery reduces transfusion requirements in patients with platelet counts over or under 100×10⁹/L

Abstract

Background

Administration of fibrinogen concentrate, targeting improved maximum clot firmness (MCF) of the thromboelastometric fibrin-based clot quality test (FIBTEM) is effective as first-line haemostatic therapy in aortic surgery. We performed a post-hoc analysis of data from a randomised, placebo-controlled trial of fibrinogen concentrate, to investigate whether fibrinogen concentrate reduced transfusion requirements for patients with platelet counts over or under 100×10⁹/L.

Material and methods

Aortic surgery patients with coagulopathic bleeding after cardiopulmonary bypass were randomised to receive either fibrinogen concentrate (n=29) or placebo (n=32). Platelet count was measured upon removal of the aortic clamp, and coagulation and haematology parameters were measured peri-operatively. Transfusion of allogeneic blood components was recorded and compared between groups.

Results

After cardiopulmonary bypass, haemostatic and coagulation parameters worsened in all groups; plasma fibrinogen level (determined by the Clauss method) decreased by 43–58%, platelet count by 53–64%, FIBTEM maximum clot firmness (MCF) by 38–49%, FIBTEM maximum clot elasticity (MCE) by 43–54%, extrinsically activated test (EXTEM) MCF by 11–22%, EXTEM MCE by 25–41% and the platelet component of the clot by 23–39%. Treatment with fibrinogen concentrate (mean dose 7–9 g in the 4 groups) significantly reduced post-operative allogeneic blood component transfusion requirements when compared to placebo both for patients with a platelet count ≥100×10⁹/L and for patients with a platelet count <100×10⁹/L.

Discussion

FIBTEM-guided administration of fibrinogen concentrate reduced transfusion requirements when used as a first-line haemostatic therapy during aortic surgery in patients with platelet counts over or under 100×10⁹/L.

Introduction

Bleeding during complex surgery can increase the need for allogeneic blood component transfusions, which are associated with increased morbidity and mortality^1,2. It is, therefore, important to consider haemostatic treatments which can reduce the requirement for transfusion following surgery. Fibrinogen, a word coined in 1847 by Rudolf Virchow³, is a glycoprotein which is essential for coagulation and is the first coagulation factor to reach critically low levels during major surgical bleeding⁴. In complex cardiac surgery, both plasma fibrinogen levels^5,6 and fibrin-based clot quality⁷ have been shown to decrease during surgery involving cardiopulmonary bypass (CPB). Low fibrinogen levels are associated with increased postoperative blood loss in cardiac surgery^5,8–10. Hence, maintaining adequate plasma fibrinogen levels may be an effective approach to reducing coagulopathic bleeding following CPB, and the 2013 European Society of Anaesthesiology guidelines recommend that fibrinogen concentrate infusion guided by point-of-care viscoelastic coagulation monitoring should be used to reduce peri-operative blood loss in complex cardiovascular surgery¹⁰.

Fibrinogen concentrate provides a useful therapeutic option for maintaining a high-normal plasma fibrinogen level^6,11, comparable to the patients’ pre-operative levels. We previously performed a prospective, randomised, double-blind, placebo-controlled trial demonstrating that fibrinogen concentrate is effective as a first-line haemostatic therapy during major aortic surgery¹². We administered fibrinogen concentrate using a goal-directed approach in which dosing was guided by thromboelastometry (ROTEM^® [Tem International GmbH, Munich, Germany])-based measurement of fibrin-based clot quality (FIBTEM test). Efficacy analyses showed that, compared with placebo, fibrinogen concentrate infusion reduced intra-operative bleeding and overall transfusion requirements^12,13. These findings were attributed to increases in plasma fibrinogen levels and fibrin-based clot firmness following fibrinogen concentrate administration, after both parameters had initially decreased during CPB¹⁴. Our data showed that other haemostatic factors also varied peri-operatively¹⁴. For example, platelet count and function decreased during CPB, although it is not clear what effect these changes may have had upon peri-operative bleeding or patients’ responses to haemostatic therapy.

Here we present a post-hoc analysis of transfusion requirements in all patients, together with haematological and coagulation parameters. The main aim was to identify any differences in transfusion requirements for patients with high or low platelet counts following fibrinogen concentrate administration.

Materials and methods

Patients and their subgroups

The original trial from which the data for this study were obtained was a prospective, randomised, double-blind, placebo-controlled, single-centre study, performed at the Hannover Medical School. The study was approved by the local Ethics Committee and the German Regulatory Authorities, and was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice (assigned Local Ethics Committee reference code 4891M-mono, EudraCT trial number 2007-004612-31 and clinicaltrials.gov identifier number NCT00701142).

Full details of enrolment, the sample-size calculation and randomisation procedures have been published previously¹². Briefly, patients ≥18 years old and scheduled to undergo elective aortic replacement surgery with CPB were screened for eligibility between June 2008 and April 2010. Of the 80 patients enrolled, 29 and 32 were blindly randomised to fibrinogen concentrate or placebo treatment, respectively. Signed consent was obtained from all patients.

Procedures and study design

Surgical methods and the full treatment algorithm have been described previously¹². Briefly, anaesthesia was induced and aortic replacement surgery was performed with CPB. After removal from CPB, administration of protamine, and surgical control of focal bleeding, but before administration of study medication, the rate of bleeding was measured using 5-minute bleeding mass as described previously¹³. Patients with bleeding between 60 g/5 min and 250 g/5 min were judged to have coagulopathic bleeding and were administered FIBTEM-guided fibrinogen concentrate or placebo. Anaesthesiologists and surgical staff were blinded to the patients’ study medication, which was delivered in opaque syringes.

If coagulopathic bleeding continued after administration of study medication, patients were transfused according to platelet count, recorded upon removal of the aortic clamp, with either four units of fresh-frozen plasma (FFP) for platelet counts ≥100×10⁹/L or two units of apheresis platelet concentrate for platelet counts <100×10⁹/L. If a second transfusion cycle was required, patients received four units of FFP if they had previously received platelets, or two units of platelets if they had initially received FFP. Patients who continued to bleed received further cycles of transfusion with two units of FFP plus one unit of platelets, with bleeding measured after each cycle of transfusion, until it fell below 60 g/5 min.

Blood samples were drawn at different peri-operative time points and used to measure haematological and coagulation parameters as described below. The samples included in the study were as follows: the first sample, drawn before induction of anaesthesia; the second sample, drawn after removal from CPB, administration of protamine, and surgical control of focal bleeding, but before administration of study medication; the final sample, drawn at last suture, after haemostasis had been achieved and chest closure was complete.

Haematology and coagulation monitoring assays

The devices, assays and reagents used for haematological and coagulation assessments have been previously described¹⁴. Briefly, the following tests were performed: (i) a laboratory-based coagulation test: plasma fibrinogen concentration (by the Clauss method); (ii) a laboratory-based haematology test: platelet count and (iii) thromboelastometry analysis: point-of-care viscoelastic tests were performed using a ROTEM^® device (Tem International GmbH). The fibrin-based clot test (FIBTEM) and the extrinsically activated test (EXTEM) were used to record coagulation parameters, including maximum clot firmness (MCF) and clotting time (CT; time taken to reach a clot amplitude of 2 mm). Each assay was conducted using 0.3 mL citrated whole blood. The “platelet component” of clot strength was calculated as the difference in maximum clot elasticity (MCE; MCE=[100×MCF]/[100−MCF]) between EXTEM and FIBTEM, as previously reported¹⁵:

$$\text{Platelet\hspace{0.17em}component}={\text{MCE}}_{\text{EXTEM}}-{\text{MCE}}_{\text{FIBTEM}}$$

The platelet index (PI) was then calculated to estimate the extent of interactions between fibrinogen and platelets, as previously reported¹⁵:

$$\text{PI}=\frac{(\text{platelet\hspace{0.17em}component}\times {10}^3)}{\text{platelet\hspace{0.17em}count\hspace{0.17em}in\hspace{0.17em}}{\text{mm}}^3}$$

Statistical analysis

Primary and secondary end-points of the trial upon which this study is based have previously been analysed and published¹². In this study, patients were analysed according to their platelet count after CPB (≥100×10⁹/L or <100×10⁹/L) and treatment received (fibrinogen concentrate or placebo).

For the majority of test parameters, the recorded values were normally distributed. These data are therefore presented as mean (±standard deviation). EXTEM CT values were not normally distributed, so this parameter is presented as a median (with inter-quartile range).

Statistical comparisons were performed for the transfusion requirements between the four groups, using pre-determined pairs of groups as follows: between treatment groups for patients with either high or low platelet counts, and between the patients with high or low platelet counts within the same treatment groups. The distribution of the values was positively skewed, therefore the Mann-Whitney test was used for comparisons between groups.

Results

Patients’ disposition and haematology and coagulation parameters

The patients’ demographic and surgical characteristics are presented in Table I; patients in all groups had the typical characteristics of patients undergoing aortic surgery. Most of the patients were male (62–89%), the average age ranged from 58.4 years to 62.9 years, and the mean dose of fibrinogen concentrate or placebo administered ranged from 7 g to 9 g between the four groups. Haematology, coagulation, and thromboelastometry measurements taken before surgery, after removal from CPB and administration of protamine (before administration of the study medication), and after last suture are presented in Table II. Measurements were generally comparable between groups. When comparing measurements between time points, haematology and coagulation measurements tended to be reduced following CPB; fibrinogen levels decreased between 43–58%, platelet count decreased 53–64%, FIBTEM MCF decreased 38–49%, FIBTEM MCE decreased 43–54%, EXTEM MCF decreased 11–22%, EXTEM MCE decreased 25–41% and platelet component decreased 23–39%. In contrast, EXTEM CT increased by 23–39% and platelet index increased by 51–84%. Despite the prolongation of EXTEM CT after CPB (mean value of each group was >70 seconds), at the end of the operation the median EXTEM CT values were comparable with the peri-operative baseline ones for both the fibrinogen concentrate and the placebo group.

Table I.

Patients’ demographic and surgical characteristics.

Characteristic	Fibrinogen, platelet count ≥100×109/L	Placebo, platelet count ≥100×109/L	Fibrinogen, platelet count <100×109/L	Placebo, platelet count <100×109/L
	N=18	N=19	N=11	N=13
Males, n (%)	12 (67%)	17 (89%)	7 (64%)	8 (62%)
Age, mean±SD, years	58.4±14.6	59.8±11.8	60.4±14.8	62.9±11.8
Weight, mean±SD, kg	83.8±17.2	92.5±20.8	94.5±17.6	79.4±17.3
Body mass index, mean±SD, kg/m2	27.8±4.3	29.0±4.8	29.3±5.0	25.5±4.2
Obesity (body mass index >30 kg/m2), n (%)	4 (22%)	6 (32%)	3 (27%)	2 (15%)
Smoking, n (%)	6 (33%)	3 (16%)	2 (18%)	5 (38%)
Operation type: TAAA, n (%)	5 (28%)	4 (21%)	3 (27%)	6 (46%)
Operation type: ascending with arch, n (%)	6 (33%)	5 (26%)	6 (55%)	5 (38%)
Operation type: ascending without arch, n (%)	7 (39%)	10 (53%)	2 (18%)	2 (15%)
Time on CPB, mean±SD, min	118.9±35.0	157.6±40.7	147.5±29.8	144.2±60.5
Dose of fibrinogen concentrate or placebo administered, mean±SD, g	7.1±2.4	7.7±2.4	9.1±1.9	7.7±2.1
Platelet count upon removal of the aortic clamp, mean (range), ×109/L	143.1 (104–185)	134.4 (101–199)	78.2 (35–99)	80.9 (22–92)

Data presented as mean±standard deviation, or absolute number and percentage frequency. Ascending with arch, aortic valve with root/ascending aorta and aortic arch replacement; ascending without arch, aortic valve with root/ascending aorta replacement.

CPB: cardiopulmonary bypass; N: total number of patients in group; n: number of patients in given category; SD: standard deviation; TAAA: thoraco-abdominal aortic aneurysm.

Table II.

Peri-operative haematology and coagulation measurements.

Parameter recorded	Pre-operative baseline (before induction of anaesthesia)				Before haemostatic therapy (after CPB and protamine administration)				After haemostatic therapy (last suture)
	Fibrinogen concentrate, platelet count ≥100×109/L	Placebo, platelet count ≥100×109/L	Fibrinogen concentrate, platelet count <100×109/L	Placebo, platelet count <100×109/L	Fibrinogen concentrate, platelet count ≥100×109/L	Placebo, platelet count ≥100×109/L	Fibrinogen concentrate, platelet count <100×109/L	Placebo, platelet count <100×109/L	Fibrinogen concentrate, platelet count ≥100×109/L	Placebo, platelet count ≥100×109/L	Fibrinogen concentrate, platelet count <100×109/L	Placebo, platelet count <100×109/L
	N=18	N=19	N=11	N=13	N=18	N=19	N=11	N=13	N=18	N=19	N=11	N=13
Laboratory coagulation tests
Clauss fibrinogen, mg/dL*	308±122	279±57	298±46	303±70	157±41	159±19	156±42	152±49	261±44	190±30	257±56	188±41
Haematology tests
Platelet count, ×109/L*	238±65	203±42	175±44	177±34	109±31	96±20	63±16	67±19	116±24	158±43	89±27	132±26
ROTEM®-based tests
FIBTEM MCF, mm*	18.2±7.0	16.2±4.0	17.4±4.2	16.7±3.5	10.2±3.2	10.0±2.2	8.9±3.7	8.9±2.8	16.4±2.1	12.3±2.9	16.0±3.8	10.3±3.5
FIBTEM MCE*	23.2±10.9	19.6±5.8	21.3±6.4	20.2±5.1	11.4±4.0	11.2±2.7	9.9±4.6	9.9±3.4	19.7±3.0	14.2±3.7	19.3±5.3	11.7±4.5
EXTEM MCF, mm*	62.6±6.7	60.2±6.1	59.8±4.6	60.8±4.4	52.9±4.9	53.6±3.2	46.5±7.4	49.0±7.1	58.1±2.8	56.7±6.1	53.8±6.2	53.5±8.0
EXTEM MCE*	176±54	157±41	152±31	158±26	115±23	117±14	90±27	100±28	140±16	135±30	120±27	120±33
EXTEM CT, sec†	61 (54, 66)	59 (56, 66)	58 (55, 68)	59 (55, 66)	75 (69, 86)	76 (69, 86)	74 (71, 101)	82 (72, 115)	62 (59, 69)	61 (58, 71)	63 (61, 64)	66 (59, 78)
Platelet component*	153±44	137±36	131±26	138±22	103±20	105±12	80±23	90±25	120±15	121±27	101±24	108±29
Platelet index*	0.65±0.11	0.68±0.14	0.77±0.14	0.80±0.18	0.98±0.19	1.13±0.25	1.28±0.29	1.47±0.67	1.06±0.16	0.80±0.18	1.18±0.23	0.83±0.20

^*Mean±standard deviation reported.

^†Median (inter-quartile range) reported.

CPB: cardiopulmonary bypass; CT: clotting time; MCF: maximum clot firmness; MCE: maximum clot elasticity.

Comparison of transfusion requirements between groups

The amount of blood products used in each group was quantified and the results are summarised in Table III. Statistical analyses showed significantly higher FFP, platelet concentrate and total allogeneic blood product usage in the placebo group than in the group treated with fibrinogen concentrate, for patients with high or low platelet counts. Red blood cell transfusion requirements were also found to be significantly greater in the placebo group for patients with a high platelet count while no significant difference was observed in patients with a low platelet count.

Table III.

Transfusion requirements.

Product administered in 24 hours (units)	Platelet count	Fibrinogen concentrate group	Placebo group	p value
Red blood cells	≥100×109/L	0 (0, 2)	2 (2, 4)	0.01
	<100×109/L	2 (0, 4)	2 (2, 5)	0.28
Fresh frozen plasma	≥100×109/L	0 (0, 4)	6 (4, 10)	<0.001
	<100×109/L	0 (0, 4)	8 (4, 8)	<0.001
Platelet concentrate	≥100×109/L	0 (0, 2)	4 (2, 5)	<0.001
	<100×109/L	2 (0, 3)	4 (2, 4)	0.03
Total allogeneic blood products	≥100×109/L	0 (0, 7)	13 (8, 22)	<0.001
	<100×109/L	3 (2, 9)	14 (8, 20)	0.008

Data presented as median (inter-quartile range). The statistical comparison was performed using a Mann-Whitney test; p values are for the comparison of the fibrinogen group to the placebo group.

Transfusion requirements were also compared between patients with high or low platelets counts within each treatment group (fibrinogen concentrate or placebo). Some variation was observed, for example increased usage of platelet concentrate and red blood cells in the low platelet count patients treated with fibrinogen concentrate, and increased FFP use in the low platelet count patients treated with placebo. However, the study was not powered to determine whether these differences were statistically significant.

Discussion

In this post-hoc analysis of data from a randomised, controlled trial we have shown that fibrinogen concentrate, administered as first-line haemostatic therapy targeting improved fibrin-clot firmness to a pre-specified target of 22 mm FIBTEM MCF, reduced the transfusion requirement in patients with platelet counts over or under 100×10⁹/L. This supports both the previous results of this study¹², and earlier trials which demonstrated a reduction in transfusion of allogeneic blood components in patients treated with fibrinogen concentrate^16–18.

During CPB, a number of haemostatic and coagulation parameters worsened in all groups. This is not surprising given that CPB disrupts haemostasis via the anticoagulant effects of heparin and the consumption of coagulation factors and platelets induced by the extracorporeal circulation¹⁹. We previously demonstrated that patients in our study generally showed a widespread decline in haemostatic activity during CPB¹⁴. In this analysis, fibrinogen levels, platelet count, FIBTEM MCF, and FIBTEM MCE appeared to be the parameters most affected, with EXTEM MCF, EXTEM MCE and platelet component showing smaller percentage changes. This is in line with previous studies demonstrating that fibrinogen and FIBTEM clot quality were the most affected by CPB, followed by platelet contribution to clot, while thrombin generation parameters such as endogenous thrombin potential were largely unchanged⁷.

The rationale for improving fibrin clot quality in platelet-deficient patients is based on the observation that although diminished platelet count can result in a reduction in clot firmness²⁰, laboratory and clinical studies have shown that the consequences (both reduced clot firmness and increased bleeding) may be partially compensated by supplementation of fibrinogen^15,21. This effect has also been observed post-CPB in cardiac surgery patients²², and FIBTEM-guided administration of fibrinogen concentrate has been shown to decrease transfusion requirements in patients with decreased platelet function after CPB²³. In the current study, some variation in allogeneic blood products usage was observed when comparing patients with high or low platelet counts receiving the same treatment; however, this study was not powered to detect whether these differences were statistically significant.

When comparing transfusion requirements between the fibrinogen concentrate and placebo groups, it is interesting to note that the difference tended to be slightly greater for red blood cells and platelets in patients with high platelet counts compared with patients with low platelet counts, but slightly lower for FFP. The difference in total transfusion requirements between the fibrinogen concentrate and placebo groups was also greater for the patients with high platelet counts compared with patients with low platelet counts. However, further analysis would be required to confirm this trend. Nevertheless, the main finding of the study was that fibrinogen concentrate administration resulted in decreased transfusion in patients with platelet counts over or under 100×10⁹/L.

It appears noteworthy that although EXTEM CT was prolonged following CPB, its values returned to those comparable with the pre-operative baseline following haemostatic therapy in all subgroups. The finding that fibrinogen concentrate administration results in a shortening of EXTEM CT in vivo confirms previous observations in the experimental setting, including those in animal models^24,25.

The main limitation of this study is that it is an exploratory post-hoc sub-group analysis and the analyses were not pre-defined, which introduces the potential for bias in the results. The study was designed to confer statistical power on the primary efficacy analysis¹² and in the present study the fibrinogen concentrate and placebo group was further subdivided to allow comparison of patients with different platelet counts, resulting in relatively small subgroups of variable size. The study was not designed to address whether the effect of fibrinogen substitution is modified at very low platelet counts. The study was performed at a single centre, possibly restricting the general applicability of the results, although this does have the advantage of ensuring standardisation of clinical practice throughout the study.

Conclusion

In summary, this analysis indicates that FIBTEM-guided administration of fibrinogen concentrate is effective as first-line haemostatic therapy in reducing transfusion requirements following aortic surgery in patients with platelet counts over or under 100×10⁹/L at removal of the aortic clamp.