The development of acquired coagulopathy as a consequence of severe traumatic or intraoperative bleeding is an independent risk factor for increased mortality (1,2). Besides consumption and loss of both clotting factors and platelets, hemodilution contributes to aggravation of the coagulopathy. During acute bleeding, application of crystalloid and colloid solutions to maintain normovolemia as well as transfusion of erythrocyte concentrates to assure oxygen supply leads to dilution of clotting factors and platelets. Analysis of clotting factors during progressive hemodilution revealed that fibrinogen was the first factor to become critically low (3,4). Moreover, all synthetic colloid solutions are believed to impair fibrin polymerization, which can additionally reduce blood clot stability (5–7). Practice guidelines emphasize treatment of acquired coagulopathy after severe blood loss and volume replacement by infusion of fresh-frozen plasma and platelet concentrates but do not mention selective fibrinogen replacement as a first choice of treatment (8,9). Highly concentrated fibrinogen is available as a purified factor concentrate in Europe and as cryoprecipitate in the United States.
The first aim of the present study was to analyze the effect of progressive hemodilution on the coagulation system, comparing a crystalloid (NaCl 0.9%) and a colloid solution. As colloid we used hydroxyethyl starch (HES) 6% 130/0.4 because this low molecular weight HES has been shown to only minimally affect factor VIII and activated partial thromboplastin time (aPTT) (10). The second aim was to examine the therapeutic effect of administration of fibrinogen concentrate and/or platelets on correcting severe dilutional coagulopathy. Thrombelastography (TEG®) was used for coagulation testing, as this method records the interaction of fibrinogen and platelets and is additionally sensitive to disturbances of fibrin polymerization.
With approval of the local Ethics Committee, 10 mL of citrated blood (1:10) was drawn from 8 healthy volunteers after obtaining written informed consent. The blood donors had not taken acetylsalicylic acid or other nonsteroidal antiinflammatory drugs within the previous week, and their medical history was without pathological findings concerning renal and hepatic function. For all 8 volunteers, baseline measurements for fibrinogen (160–400 mg/dL), platelet count (150–400 G/L) and hematocrit (36%–46%), as well as baseline values of the TEG® (clotting time [CT], 100–240 s; clot formation time [CFT], 35–110 s; maximum clot firmness [MCF], 53–72 mm for the INTEM® test) were in the respective physiological ranges.
The test samples were analyzed by means of a modified TEG® coagulation analyzer (ROTEM®; Pentapharm CO, Munich, Germany) based on the TEG® system described by Hartert (11,12). In the ROTEM® system the sensor pin is guided by a ball-bearing system that makes it less susceptible to movement and vibration. Activation of the test samples accelerates the measurement process and enhances the reproducibility in comparison with native TEG® measurement (13). The thrombelastogram is characterized by specific variables (Fig. 1). The CT corresponds to the reaction time in a conventional TEG® and describes the initiation of clot formation. The CFT is equivalent to the coagulation time and reflects the dynamics of the coagulation process. It is defined as the time from the onset of clotting until a clot firmness of 20 mm is reached. Finally, the MCF refers to the maximum amplitude and represents the quality of the clot.
In the first experimental series, we diluted samples of citrated whole blood with either NaCl 0.9% (Fresenius, Bad Homburg, Germany) or HES 6% 130/0.4 (Voluven®, Fresenius). Both solutions were supplemented with sodium citrate in a 1:10 proportion before use to attain the same citrate concentration in all test samples. For each experiment, 300 μL of whole blood, blood-NaCl 0.9%, or blood-HES 6% dilution was analyzed. Dilutions of 0% (control), 20%, 40%, 60% and 80% were prepared (volume proportion of NaCl 0.9% or HES 6% in the tested dilutions).
In the second experimental series, we used 60% dilution of whole blood to analyze the influence of application of fibrinogen concentrate, platelet concentrate, or both together on ROTEM® variables. We diluted 1.2 mL of citrated whole blood with either 1.8 mL HES 6% or with 1.8 mL NaCl 0.9% and added either 200 μL of fibrinogen concentrate (Hemocomplettan® HS, ZLB Behring, Bern, Switzerland) with a stock concentration of 20 mg/mL, 100 μL of a commercial platelet apheresis concentrate, or both components. The chosen dosages of platelets and fibrinogen corresponded to an application of 6 g fibrinogen or one platelet apheresis concentrate (150 mL) to a total blood volume of 5000 mL. One sample of citrated whole blood remained undiluted, serving as a baseline measurement (control). Two others were diluted with either HES 6% or NaCl 0.9% without addition of fibrinogen or platelets to serve as dilution controls.
According to the manufacturer’s instructions, citrated whole blood was recalcified with buffered CaCl2 (0.2 M, 20 μL) and the intrinsic pathway activator ellagic acid (20 μL) was added (INTEM®, Pentapharm, Munich, Germany). In a separate measurement, the effect of platelets on the clot formation was blocked by the addition of cytochalasin D (FIBTEM®, Pentapharm). The resulting clot firmness was based only on the polymerized and cross-linked fibrin. All ROTEM® measurements were performed at a test temperature of 37°C.
Hematocrit and platelet count were measured by an automated analyzer (Sysmex SE 9000B; Sysmex Europe, Norderstedt, Germany). Fibrinogen concentration was determined with the Clauss method.
To display the effects of dilution and restitution of fibrinogen on blood clot formation, scanning electron microscopy was performed on one set of blood clots. After measurement by ROTEM®, blood clots were extracted from the test cup and immediately placed in 3% glutaraldehyde. After 2 hours of fixation, glutaraldehyde was diluted with distilled water to a concentration of 1% for further storage overnight. The next day, the fixed clots were repeatedly washed in phosphate buffered NaCl 0.9% and then postfixed for another 2 h in 2% osmium tetroxide. After several washing procedures and dehydration in a graded series of ethanol (30%–100%), the samples were dried, spattered with gold, and examined with a JEOL scanning electron microscope (JEOL Ltd., Tokyo, Japan).
For statistical analysis of the changes induced by hemodilution, and comparison of the two intervention groups, one-way repeated measures analysis of variance was performed with post hoc procedures according to the Student-Newman-Keuls test for all pairwise testing and Bonferroni/Dunn’s test for testing versus a control group. All variables are expressed as mean ± sd.
Increasing dilution led to dose-dependent impairment of clot formation of whole blood. The effects were more pronounced with HES 6% than with NaCl 0.9%. Figure 2 shows the effects on TEG® variables. The normal range for the CT is ≤240 s. An extension to ≥400 s is a sign of severe impairment of initiation of the coagulation process. The CFT is normally ≤110 s. A CFT ≥250 s indicates a major disorder of clot formation. Healthy individuals show an INTEM® MCF between 53 mm and 72 mm. A reduction of the MCF to ≤40 mm is a sign of strongly impaired hemostasis. For the FIBTEM® a MCF of ≤8 mm indicates a lack of fibrinogen or a disturbance of the fibrin polymerization. In our dilution series, only 80% dilution with HES 6% resulted in a severe impairment of the CT. Thus, the initiation of clot formation seems to be quite resistant to dilution effects. In contrast, CFT and MCF in the INTEM® test were already strongly impaired after 40% dilution with HES 6%, whereas 60% dilution with NaCl 0.9% was necessary to induce the same effect. Even more pronounced was the situation concerning the MCF in the FIBTEM® test, which relies only on fibrin clot formation and fibrin polymerization. Twenty percent dilution with HES 6% led to severe impairment, and 40% dilution even resulted in a complete inhibition of clot formation. In the case of NaCl 0.9%, 60% dilution scarcely disturbed clot formation. However, 80% dilution inhibited it severely. With respect to CFT and MCF in each of the 20%–60% dilutions, HES 6% impaired coagulation significantly stronger than an equivalent dilution with NaCl 0.9%.
A dilution grade of 60% was chosen for the intervention series, as this led to a significant impairment of coagulation for both diluents. The results of these experiments are depicted in Figure 3. Supplementation with fibrinogen, platelets, or both reduced CT significantly in the case of HES 6%, whereas only the combined application resulted in a significant shortening of CT after dilution with NaCl 0.9%. However, this statistically significant improvement would seem to be of minor clinical relevance, as it only represents a return to normal from a mildly extended CT. In contrast, CFT was massively prolonged after dilution with HES 6% but not critically by NaCl 0.9%. In the case of HES 6%, only the application of platelets, but not of fibrinogen, was able to reduce the CFT significantly. The combination of both components was not superior to the addition of platelets alone. CFT impairment after dilution with NaCl 0.9% improved significantly after the application of fibrinogen or platelets, but no additional effect was detectable when both components were added together. A 60% dilution of whole blood deteriorated MCF in the INTEM® test to a statistically significant and clinically relevant extent. The changes were significantly more pronounced for HES 6% than for NaCl 0.9%. In the presence of NaCl 0.9% only the application of fibrinogen was able to improve the INTEM® MCF to a clinically relevant extent. The combination with platelets was not superior to the sole addition of fibrinogen. In contrast, after dilution with HES 6% no intervention improved the MCF in the INTEM® test to clinically desirable values. Qualitatively identical results were obtained for the measurement by the FIBTEM® test. The addition of fibrinogen reconstituted the fibrin clot firmness fully in the presence of NaCl 0.9% but had only a minor effect after equivalent dilution with HES 6%. Again, platelets were without detectable benefit in either case.
The mean hematocrit decreased significantly from 37% ± 2% to 14% ± 1% after 60% dilution with either HES 6% or NaCl 0.9%. Platelet count was 158 ± 67 G/L in the control samples and decreased to 68 ± 19 G/L after dilution with HES 6% and to 61 ± 12 G/L after dilution with NaCl 0.9% (no significant difference). The addition of a platelet bolus increased the platelet concentration by approximately 25 G/L in both cases. Fibrinogen concentrations determined in the control samples after dilution with either HES 6% or NaCl 0.9% and after fibrinogen substitution are shown in Figure 4. Fibrinogen concentrations of the undiluted whole blood samples were all well within the physiological range. A 60% dilution induced a significant decrease to concentrations less than 100 mg/dL, which is a trigger limit for substitution (3). The addition of fibrinogen concentrate returned fibrinogen concentrations to physiological values, although the pre-dilutional values were not completely reattained. It is noteworthy that after dilution of whole blood with HES 6%, the measured fibrinogen concentrations were in all cases significantly larger than in the corresponding NaCl 0.9% samples. This phenomenon has been previously described (14): fibrinogen concentration is over-estimated in laboratory assays when HES or dextran molecules are present.
After dilution of whole blood with either NaCl 0.9% or HES 6%, the formed clot showed rarefication of the fibrin network (Fig. 5). The individual fibers also seemed to thin out. Addition of fibrinogen resulted in a close fibrin network in both settings. It appears that, after dilution with NaCl 0.9%, the fibrin fibers formed a more dense and planar network than in the presence of HES 6%, where a more porous mesh developed. Nevertheless, the picture revealed by scanning electron microscopy does not explain the considerable difference of the clot firmness detected between both diluents after fibrinogen application.
Our results on progressive hemodilution show that both NaCl 0.9% and HES 6% impaired whole blood clot formation as measured by TEG®. The variable CT, which reflects the initiation of clot formation, was widely resistant to dilution. In contrast, CFT and MCF, markers for the dynamics of clot formation and the quality of the formed clot, displayed progressive dilutional coagulopathy. There is, however, a remarkable difference between the two diluents with respect to their extent of impairment of coagulation for a given degree of hemodilution. In the INTEM® test we found severely impaired coagulation (MCF <40 mm) when 40% dilution was performed using HES 6%, whereas 60% dilution with NaCl 0.9% was necessary to alter the clot formation to the same extent. Our finding that in vitro dilution with HES 6% induced a more pronounced coagulopathy than NaCl 0.9% is in accordance with previous studies. Fries et al. (5) compared the impact of progressive hemodilution on coagulation in vitro using different solutions and found that HES 6% 130/0.4 induced a more pronounced coagulopathy than lactated Ringer’s solution. Ekseth et al. (15) reported development of coagulopathy after in vitro dilution of whole blood with ≥40% of HES 10%, whereas ≥50% of crystalloids were required to induce an equivalent impairment.
In addition, we found a much more pronounced effect of HES 6% on clot formation in the FIBTEM® assay compared with the standard INTEM® measure. For both tests identical activation was applied, but in the FIBTEM® assay platelet function was completely inhibited by cytochalasin D. An abnormal fibrin clot formation expressed as clearly reduced MCF (<8 mm) was observed with HES 6% using a dilution as small as 20%. At a dilution of 40% no fibrin clot formation at all was detectable. When NaCl 0.9% was used, the MCF did not decline to values <8 mm until 60% dilution was reached. Such a drastic effect of HES 6% on fibrin clot formation has not been reported. This finding may have the following clinical consequences:
When platelets link into the fibrin clot they compensate partly for the inhibition of polymerization induced by colloids such as HES 6%. This would explain the more moderate effects of HES 6% on clot firmness seen in INTEM® compared with FIBTEM® testing (Fig. 3) and should account for the high degree of safety of HES 6% when given to patients with no other coagulopathy. However, based on our results, caution may be advisable in applying even moderate amounts of HES 6% to patients with compromised platelet function or with thrombocytopenia.
Several studies have denoted a hypercoagulable state induced by hemodilution up to 20% (16–18), whereas others have denied this phenomenon (5,19). In our experimental setting we did not observe a hypercoagulable state at a 20% dilution. Entholzner et al. (19) speculated that hypercoagulability might be an artifact related to the use of the native thrombelastogram. There is no such complication when using activated TEG®. This argument is supported by our own results.
Sixty percent dilution of whole blood reduced the fibrinogen concentration to values of approximately 70 mg/dL (Fig. 4). This concentration is below the critical threshold of 100 mg/dL for sufficient clot formation (3). In parallel, platelet count decreased to mean values of approximately 60 G/L after dilution. Contrary to the post-dilutional values for fibrinogen, this concentration is above the stipulated critical level of 50 G/L (9). This observation is in accordance with results of Hiippala et al. (3), who found that fibrinogen was the first coagulation factor to reach a critical level (<100 mg/dL) after a calculated blood loss of 142%, whereas thrombocytopenia was of late occurrence, ≤50 G/L being reached after an estimated blood loss of 240%.
In the case of dilution with NaCl 0.9%, the application of fibrinogen led to a statistically significant and clinically relevant improvement of clot formation and stability. Administration of platelets alone had only minor effects on CFT and MCF compared with the use of fibrinogen. Simultaneous application of both components had no additional effect on clot formation in comparison to fibrinogen alone. Failure to find a significant additive effect of platelets indicates that a certain lower level, still present even at 60% dilution, suffices for a full action. As mentioned above, only the fibrinogen concentration, but not the platelet count, was critical in clinical terms. After 60% dilution with HES 6%, the application of fibrinogen did not reduce CFT significantly, whereas platelets did. Increase of the MCF caused by addition of platelets and/or fibrinogen was in each case insufficient to increase the MCF to values >40 mm. Although physiological concentrations of fibrinogen were determined after substitution, only minimal fibrin clot formation was detectable in the FIBTEM® test. A possible reason for this finding might be a disturbance by HES molecules of the polymerization of fibrin. The inhibition of polymerization by HES molecules has been repeatedly reported in the literature (6,19), but the mechanism is still unknown. Mardel et al. (20) proposed that a reduced linking of fibronectin into the fibrin network decreases the size and density of the fibrin fibers, but this is controversial (6). Scanning electron microscopy in the present study revealed a rarefication of the fibrin network after dilution with both solutions. Astonishingly, addition of fibrinogen resulted, in both settings, in a close fibrin network. It appears that the fibrin fibers formed a more planar and denser network in the presence of NaCl 0.9%, whereas a more porous fibrin mesh formed after hemodilution with HES 6%. However, the microscopic analysis of the clots did not show sufficient morphologic differences to explain the striking differences in clot firmness as determined by TEG®.
In a recently published study by Fenger-Eriksen et al. (21), the effects of hemodilution of 55% with colloid and crystalloid solutions on coagulation were evaluated using TEG®. Furthermore, the reversal of dilutional coagulopathy by treatment with fibrinogen concentrate or platelets was studied. The authors stated that fibrinogen, but not platelets, was capable of improving coagulopathy induced by colloids. However, precise evaluation of their results reveals that fibrinogen concentrate only had a minor clinical effect on coagulation in the presence of HES 6% in comparison with NaCl 0.9%. Thus, the results of Fenger-Eriksen et al. are in accordance with our observations. In contrast to our study, only clot formation in whole blood was evaluated but not the isolated fibrin clot formation (FIBTEM®). Thus, the inhibition of fibrin polymerization by HES 6% was missing in their study. Moreover, the combined application of platelets and fibrinogen was not examined.
In the present work, the chosen dose of fibrinogen applied to the diluted blood corresponded to an administration of 6 g fibrinogen to a total blood volume of 5000 mL. Assuming a corresponding body weight of 70 kg, the added 6 g fibrinogen would correspond to a dose of 85 mg/kg fibrinogen. There is little information in the literature about the application and dosage of fibrinogen for treatment of dilutional coagulopathy. In a study by Sawirez et al. (22), an in vitro dose of 105 mg/kg fibrinogen reversed the decreased blood clot firmness after a 66% dilution with a combination of lactated Ringer’s and gelatin solutions. Another study by Fries et al. (23) used a porcine model in which 60% of the estimated blood volume was withdrawn and replaced with gelatin solution. Impaired clot strength was normalized after infusion of 250 mg/kg fibrinogen concentrate. With respect to the work of Fenger-Eriksen et al. (21), their dosage of fibrinogen appears comparable to ours. In our clinical experience, an initial dose of 80–100 mg/kg fibrinogen is sufficient to effectively treat severe dilutional coagulopathy after massive transfusion (24).
Our in vitro results must not be extrapolated to clinical practice without due caution. Limitations of our experimental set-up have to be acknowledged. It does not account for physiological responses to hemodilution (e.g., recruitment of platelets, coagulation factors) and certainly misses the effects of pharmacokinetics (e.g., degradation and excretion of HES). Furthermore, a 60% hemodilution solely performed using a colloid solution is not routine clinical practice.
In conclusion, acquired coagulopathy induced in vitro by progressive hemodilution has an earlier onset and is more severe when induced with HES 6% than with NaCl 0.9%. A 60% dilution reduces blood fibrinogen concentrations to levels below 100 mg/dL, which are insufficient for clot formation. Administration of large-dose fibrinogen (85 mg/kg) is able, in vitro, to quickly correct clot firmness impaired by hemodilution with crystalloid solutions. After 60% hemodilution with HES 6%, restitution of clot formation using this dose of fibrinogen appears incomplete. Combined application of fibrinogen concentrate and platelets was not superior to the exclusive use of fibrinogen concentrate during crystalloid hemodilution. A moderate additive effect of platelet substitution was found during hemodilution with HES 6%; however, such large intravascular HES concentrations are rarely encountered clinically. Our in vitro findings support the application of fibrinogen concentrate for clinical treatment of dilutional coagulopathy.
1. Brohi K, Singh J, Heron M, Coats T. Acute traumatic coagulopathy. J Trauma 2003;54:1127–30.
2. MacLeod JB, Lynn M, McKenney MG, et al. Early coagulopathy predicts mortality in trauma. J Trauma 2003;55:39–44.
3. Hiippala ST, Myllyla GJ, Vahtera EM. Hemostatic factors and replacement of major blood loss with plasma-poor red cell concentrates. Anesth Analg 1995;81:360–5.
4. Singbartl K, Innerhofer P, Radvan J, et al. Hemostasis and hemodilution: a quantitative mathematical guide for clinical practice. Anesth Analg 2003;96:929–35.
5. Fries D, Innerhofer P, Klingler A, et al. The effect of combined administration of colloids and lactated Ringer’s solution on the coagulation system: an in vitro
study using Thrombelastograph® Coagulation Analysis (ROTEM®). Anesth Analg 2002;94:1280–7.
6. Innerhofer P, Fries D, Margreiter J, et al. The effects of perioperatively administered colloids and crystalloids on primary platelet-mediated hemostasis and clot formation. Anesth Analg 2002;95:858–65.
7. Egli GA, Zollinger A, Seifert B, et al. Effect of progressive haemodilution with hydroxyethyl starch, gelatine and albumin on blood coagulation. Br J Anaesth 1997;78:684–9.
8. Crosby ET. Perioperative haemotherapy: I. Indications for blood component transfusion. Can J Anaesth 1992;39:695–707.
9. Practice guidelines for blood component therapy: a report by the American Society of Anesthesiologists Task Force on Blood Component Therapy. Anesthesiology 1996;84:732–47.
10. Langeron O, Doelberg M, Ang ET, et al. Voluven, a lower substituted novel hydroxyethyl starch (HES 130/0.4), causes fewer effects on coagulation in major orthopedic surgery than HES 200/0.5. Anesth Analg 2001;92:855–62.
11. Hartert H. Blutgerinnungsstudien mit der Thrombelastographie, einem neuen Untersuchungsverfahren. Klin Wochenschrift 1948;26:577–83.
12. Mallett SV, Cox DJ. Thrombelastography. Br J Anaesth 1992;69:307–13.
13. Calatzis A, Haas S, Gödje O, et al. Thrombelastographic coagulation monitoring during cardiovascular surgery with the ROTEG Coagulation analyzer. In: Pifarré R, ed. Management of bleeding in cardiovascular surgery. Philadelphia: Hanley & Belfus, 2000;215–26.
14. Hiippala ST. Dextran and hydroxyethyl starch interfere with fibrinogen assays. Blood Coagul Fibrinolysis 1995;6:743–6.
15. Ekseth K, Abildgaard L, Vegfors M, et al. The in vitro
effects of crystalloid and colloids on coagulation. Anaesthesia 2002;57:1102–8.
16. Gibbs NM, Crawford GP, Michalopoulos N. Thrombelastographic patterns following abdominal aortic surgery. Anaesth Intensive Care 1994;22:534–8.
17. Ng KF, Lam CC, Chan LC. In vivo effect of haemodilution with NaCl 0.9% on coagulation: a randomized controlled trial. Br J Anaesth 2002;88:475–80.
18. Ruttmann TG, James MF, Viljoen JF. Haemodilution induces a hypercoagulable state. Br J Anaesth 1996;76:412–14.
19. Entholzner EK, Mielke LL, Calatzis AN, et al. Coagulation effects of a recently developed hydroxyethyl starch (HES 130/0,4) compared to hydroxyethyl starches with higher molecular weight. Acta Anaesthesiol Scand 2000;44:1116–21.
20. Mardel SN, Saunders FM, Allen H, et al. Reduced quality of clot formation with gelatin-based plasma substitutes. Br J Anaesth 1998;80:204–7.
21. Fenger-Eriksen C, Anker-Moller E, Heslop J, et al. Thrombelastographic whole blood clot formation after ex vivo
addition of plasma substitutes: improvements of the induced coagulopathy with fibrinogen concentrate. Br J Anaesth 2005;94:324–9.
22. Sawirez M, Fries D, Klingler A, et al. The effect of the fibrinogen and factor XIII substitution on dilutional coagulopathy: an in vitro
model using thrombelastography (ROTEG). Ann Hematol 2002;81(suppl. 1):P76.
23. Fries D, Krismer A, Klingler A, et al. The effect of fibrinogen on reversal of dilutional coagulopathy: a porcine model. Br J Anaesth 2005;95:172–7.
24. Heindl B, De Lorenzo C, Spannagl M. Hochdosierte Fibrinogengabe zur Akuttherapie von Gerinnungsstörungen bei perioperativer Massivtransfusion. Anaesthesist 2005;54:787–90.