The perioperative use of thrombelastography (and, in Europe, specifically, rotational thrombelastometry) is increasing as it provides rapid information on various clot properties.1 The main determinants of the results of activated whole blood thrombelastography are zymogen concentrations,2 fibrinogen concentration,3 and platelet count.4 For discrimination of the platelet-independent coagulation response, thrombelastographic assays are performed before and after platelet inhibition: cytochalasin D prevents activation by inhibiting platelet cytoskeletal reorganization.5 Parallel use of assays with and without cytochalasin D delineates the differential contribution of platelets to overall clot firmness.6 Inhibiting platelets in thrombelastography makes these assays mainly dependent on coagulation factor concentrations,7 specifically fibrinogen.8,9
A retrospective study of unselected surgical patients (n = 1076) found an excellent correlation between fibrinogen results (Clauss method) and platelet-independent activated whole blood thrombelastography.10 For samples with low maximum clot firmness (MCF) (amplitude <10 mm, n = 150), the correlation with fibrinogen levels was only moderate with no signal at all in 27 patients, despite measurable fibrinogen concentrations between 0.8 and 1.67 g/L.10 This variability was confirmed for low-range thrombelastographic results and fibrinogen levels in a prospective study in trauma patients,3 with the amplitude ranging from 4 to 11 mm at 1.5 g/L fibrinogen and fibrinogen ranging from 0.5 to 2 g/L at 5-mm amplitude.
Changes in factor XIII concentration significantly influence thrombelastographic results,11 and platelets are a relevant source of factor XIII.12 Decreased factor XIII activity (i.e., factor XIII deficiency) prolongs clot formation time (CFT) in plasma.13,14 Using an inhibiting anti-factor XIII antibody induces similar changes and will, therefore, further increase the prolongation of the CFT of a sample that is already partially deficient in factor XIII.15 Thus, adding a fixed amount of anti-factor XIII antibody will make a CFT-based assay sensitive for a preexisting factor XIII deficiency.
There is evidence from the literature that platelet function inhibition through monoclonal glycoprotein (GP) IIb/IIIa antibodies and cytochalasin D is mediated through two different pathways,16 and it is therefore recommended that maximum platelet inhibition is best established with a combined approach.7
Therefore, we evaluated (a) whether in vitro inhibition of factor XIII with neutralizing anti-factor XIII antibody or further platelet inhibition with anti-GPIIb/IIIa antibody might influence platelet-inhibited, activated thrombelastography results and (b) whether ex vivo factor XIII inhibition could be used to establish a thrombelastographic assay for detecting factor XIII deficiency in whole blood rather than plasma.
Two sets of experiments were performed: Set A, antibody-mediated inhibition experiments to evaluate the respective influence on “FIBTEM®” (see below) results; and Set B, a pilot evaluation of a thrombelastographic assay using an anti-factor XIII antibody to evaluate its ability to detect factor XIII deficiency. All experiments were performed in accordance with the Declaration of Helsinki.
All samples were collected into Vacutainer tubes containing 0.105 M (3.2%) buffered sodium citrate (BD Diagnostics, Basel, Switzerland). For all thrombelastographic assays, citrated whole blood was used on a ROTEM® analyzer (Pentapharm, Munich, Germany) with the respective commercially available reagents. Factor XIII activity in samples used for whole blood assays was determined as follows: one part of the respective sample was used for performing the whole blood thrombelastographic assays; another part was used to prepare plasma for factor XIII activity determination using the Berichrom® F XIII assay (Siemens, Marburg, Germany). This is a standard chromogenic assay for determination of factor XIII activity, which is used for the clinical and the research setting, the reference range being 70%–140%.17
For Set A, three samples from a healthy volunteer were obtained after informed consent; experiments were performed in triplicate. The sample-size calculation was based on the results by Nielsen et al.,15 who showed a clear reduction in clot strength with the use of an anti-factor XIII antibody in plasma (“total thrombus generation” 165 ± 15 dynes/cm2 with a nonspecific antibody versus 75 ± 12 dynes/cm2 with a specific anti-factor XIII antibody). Using these numbers results in only a small calculated sample size (n = 2) (Type I error 0.05 and Type II error 0.1). Because our planned experiments differed from those performed by Nielsen et al. (we used whole blood instead of plasma and a different device to perform the thrombelastography), we increased the sample size (n = 3). Platelet count, fibrinogen concentration, and factor XIII activity (Multifibren U and Berichrom F XIII on a Behring Coagulation System analyzer, Siemens) were determined using standard analyzers and procedures according to the recommendations of the manufacturers.
For Set B, routine blood samples of unselected surgical intensive care patients were used after IRB approval. The sample-size calculation to detect a significant difference between areas of 0.8 vs 0.5 under receiver operating characteristic (ROC) curves (with a Type I error of 0.05 and a Type II error of 0.2) resulted in a necessary sample size of n = 28. However, only 25 samples could be evaluated with the amount of antibody available. Again, factor XIII activity was determined in these samples using the Berichrom assay.
The anti-factor XIII antibody used in Set A and Set B experiments was a rabbit antihuman polyclonal immunoglobulin (Ig)G antibody against factor XIII subunit A. The inhibiting properties of this antibody were verified in a plasma assay based on the data by Nielsen et al.15 Thirty microliters of various antibody dilutions (no antibody, 1:3, 1:10, 1:30, and 1:100) was incubated in 200 μL of commercially available standard human plasma (Siemens Healthcare Diagnostics, Deerfield, IL) and 50 μL of normal saline at 37°C for 10 min; factor XIII activity was determined (in duplicate) again using the Berichrom assay.
For Set A, three series of in vitro experiments were performed. First, the neutralizing (polyclonal rabbit antihuman IgG) anti-factor XIII antibody was diluted 1:5 with normal saline. Whole blood samples (300 μL) were incubated for 20 min at room temperature with different volumes of the diluted anti-factor XIII antibody, resulting in antibody concentrations of 37, 71, and 104 μg/mL in the samples to be evaluated. A nonspecific anti-human IgA antibody solution diluted to yield identical volumes and final concentrations (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) was used for control.
Second, a monoclonal anti-GPIIb/IIIa antibody (abciximab, ReoPro®, Lilly Deutschland GmbH, Bad Homburg, Germany) was incubated with 300 μL of citrated whole blood for 20 min at room temperature to yield sample antibody concentrations of 1, 2, and 10 μg/mL. The IgA antibody used for control was again diluted to yield identical volumes and final concentrations.
To perform these two lines of experiments, a commercially available, platelet-inhibited, and tissue factor-activated whole blood thrombelastographic assay (“FIBTEM®”) was performed on a ROTEM analyzer according to the recommendations of the manufacturer. The FIBTEM assay contains cytochalasin D for prevention of platelet cytoskeletal reorganization and thus platelet activation.5,7 The coagulation time (CT, i.e., the time [in seconds] from beginning of measurement until a clot formation can be detected), the CFT (i.e., the time [in seconds] between first detection of clot formation until an amplitude of 20 mm is reached), the MCF (i.e., the maximum amplitude of the thrombelastogram [in millimeters]), and the lysis index 60 (LI60, i.e., the amplitude after 60 min in percentage of the MCF, which is a measure of fibrinolytic activity) were analyzed (Fig. 1).
In Set B, the potential of using anti-factor XIII antibody in a thrombelastographic assay to detect factor XIII deficiency was evaluated: the anti-factor XIII antibody (final concentration of 71 μg/mL) was added to whole blood before it was evaluated in a tissue factor-activated thrombelastographic assay (“ExTEM®”) without platelet inhibition and using the CFT as the readout (see also Introduction). Routine blood samples obtained from 25 patients on the surgical intensive care unit were evaluated after IRB approval.
Statistical analysis was performed using SigmaStat Version 3.11 (Jandel, San Rafael, CA). Results are expressed as mean ± sd. Student’s t-test was used to compare results in paired samples. Statistical significance was considered at P ≤ 0.05. ROC curve was calculated and graphed with MedCalc Version 184.108.40.206 (MedCalc Software, Mariakerke, Belgium).
Platelet count (184 G/L), hematocrit (44.7%), international normalized ratio (1.0), activated partial thromboplastin time (28 s), fibrinogen level (2.66 g/L), and factor XIII activity (134%) in the samples used for Set A of experiments were all in the normal range.
Set A of Experiments
Effect of Nonspecific Antibody
Compared with the anti-factor XIII antibody, the nonspecific antibody did not alter CT, CFT, MCF, or LI60 (Table 1).
Effect of Anti-factor XIII Antibody
To ascertain the properties of the factor XIII-inhibiting antibody, its concentration-dependent factor XIII-inhibiting potency was confirmed by incubation in commercially available standard plasma.
Incubation of increasing antibody dilutions with the standard plasma led to almost complete inhibition of factor XIII activity at the highest antibody concentrations used. Residual factor XIII activity as percentage of a control sample was 93%, 67%, 15%, and <7% with increasing antibody concentrations (7, 23, 69, and 228 μg/mL).
In contrast, whole blood incubation with increasing antibody concentrations (0, 37, 71, and 104 μg/mL) resulted in measured factor XIII activity of 133% ± 3%, 67% ± 1%, 42% ± 2%, and 40% ± 2% (mean ± sd).
Thrombelastographic results obtained in whole blood after incubation with anti-factor XIII antibody are shown in Table 1, which shows that the anti-factor XIII antibody, compared with the nonspecific antibody, significantly prolongs CFT and reduces MCF and increases fibrinolytic effects on the clot. It can also be seen that this effect on the CFT is not further increased with an antibody concentration higher than 71 μg/mL, well in line with the results of residual factor XIII activity measured in whole blood after antibody inhibition (see above).
Effect of Anti-GPIIb/IIIa Antibody
Incubation of whole blood samples with increasing concentrations of platelet-inhibiting anti-GPIIb/IIIa antibody18,19 did not lead to results different from those obtained with the nonspecific antibody present in identical concentrations.
Set B of Experiments
Using Anti-factor XIII Antibody in ExTEM for Estimation of Factor XIII Activity
Using the anti-factor XIII antibody in the extrinsically activated whole blood thrombelastographic ExTEM assay (unlike the FIBTEM assay, the ExTEM assay is routinely performed without platelet inhibition), it was possible to distinguish between factor XIII activity below and above the lower cutoff of the normal range (70%)17 with 90% sensitivity and 90% negative predictive value and 60% positive predictive value when using the CFT (cutoff >191 s) as the readout (area under the curve [AUC] for the ROC curve is 0.803, 95% confidence interval 0.597–0.933, P = 0.0015 for difference from AUC 0.5). For determining factor XIII activity below or above 60%, which is believed to be clinically relevant,20–22 the sensitivity and negative predictive value is 100%, and positive predictive value is 42% with an AUC of 0.84 (95% confidence interval 0.639–0.954, Fig. 2).
The main finding of our study is that inhibition of factor XIII with a polyclonal, neutralizing antibody in an activated whole blood thromboelastometry assay with platelet inhibition significantly prolongs CFT, decreases MCF, and increases fibrinolytic effects on the clot. Platelet-inhibited thrombelastographic assays are increasingly used as a surrogate marker for fibrinogen concentration and to trigger and guide procoagulant therapy, especially in Europe. Other authors have shown that factor XIII is an important contributor to clot properties in thrombelastographic experiments: adding factor XIII concentrate to factor XIII-deficient plasma increases the maximum amplitude and the α-angle, a marker of clot propagation velocity, in vitro.11 Using a celite-activated method, antibody-mediated factor XIII inhibition induces a concentration-dependent decrease in the velocity of clot propagation and clot strength, similar to the results observed in factor XIII-deficient plasma.15 Our experiments revealed similar results but were generated in whole blood and with tissue factor activation. This suggests that factor XIII has important influence on routine whole blood thrombelastographic assays (as used in this study), which are, as mentioned, increasingly used for clinical decision making in a point-of-care setting.
The fact that the coagulation time is not prolonged with incubation of anti-factor XIII antibody can be easily explained because it is dependent on the occurrence of fibrin aggregation (which is not factor XIII-dependent) but not dependent on γ chain cross-linking (which is factor XIII-dependent).23 Furthermore, experiments with a synthetic peptide-inhibiting fibrin polymerization indeed show that only clot formation is retarded.24
Confirmatory assays prove that the antibody used here is able to almost completely inhibit factor XIII activity in plasma. However, it was not possible to induce complete inhibition of factor XIII activity in whole blood under the conditions tested. Similar results were observed with the thrombelastographic assay for CFT, MCF, and LI60. This suggests that there are differences in antibody-mediated factor XIII inhibition in whole blood and plasma: obviously, there is residual factor XIII activity that escapes antibody inhibition in whole blood as compared with plasma. One explanation is that incubation of whole blood with the antibody will not reach and thus not inhibit intracellular factor XIII, as shown in a case of an acquired factor XIII inhibitor in which plasmatic factor XIII was inhibited but intracellular factor XIII was unaffected.25 Intracellular factor XIII A subunit is found in significant amounts in macrophages and platelets,26,27 which is immediately active on liberation.28 Also, the antibody used here is polyclonal, and it is well established that such preparations show some heterogeneity,29 which by itself can explain different inhibition results with different conditions, e.g., plasma or whole blood.
Besides the experiments with anti-factor XIII antibodies, our results also demonstrate that the assay used here (FIBTEM) is not sensitive to further platelet blockade with an anti-GPIIb/IIIa antibody. This is important because platelets are a source of factor XIII-associated protein cross-linking activity upon platelet activation.12,30 Of note, however, is that the platelet count was normal in our experiments; it therefore remains to be elucidated whether the same conclusions would hold true for samples with increased platelet counts.
Because thrombelastographic results after platelet inhibition are increasingly used as a surrogate marker for fibrinogen concentration and therefore to trigger and dose fibrinogen therapy during intraoperative bleeding,8,31–33 it seems important to interpret our results in this perspective. Indirect support for the relevance of our results comes from the fact that thrombelastographic assays need to be replenished with activated factor XIII to yield a good linear correlation between the thrombelastographic signal and functional fibrinogen levels.8,34
From a clinical point of view, intervening in patients with inadequate factor XIII concentration seems relevant because a recent study shows that administering factor XIII in patients at high risk for bleeding prevents loss of clot firmness, blood loss, and loss of fibrinogen35; the reduction in loss of fibrinogen is possibly due to factor XIII-mediated protection of fibrinogen from fibrinolytic degradation.36
Because of its short turnaround time, thrombelastography has clearly enhanced the possibility to provide a fast differential diagnosis in perioperative bleeding. Given the potential relevance of a factor XIII deficiency in this setting,35 it seems desirable to develop tests that allow fast detection of a factor XIII deficiency. Our pilot evaluation indicates that such an assay is feasible (Fig. 2).
There are potential limitations to our study. Samples in which a neutralizing antibody is used might behave differently as compared with samples that are deficient in factor XIII; however, factor XIII inhibition by antibodies is an accepted tool to evaluate the influence of an induced deficiency on the clotting process.37 Also, we cannot exclude that other thrombelastographic platforms and assays than the ones evaluated here might show different results after factor XIII inhibition; however, other groups have reported comparable results with a different platform, different reagents, and using plasma instead of whole blood.15 Finally, because the current results are derived from in vitro or ex vivo experiments, they cannot give proof that either the influence of factor XIII on thrombelastographic results or the early detection of factor XIII deficiency by thrombelastography will result in improved patient care; however, the results of a recent randomized placebo controlled trial suggest that this might very well be the case.35
In conclusion, factor XIII is an important contributor to clot firmness in routine thrombelastographic measurements as used in the perioperative setting. The influence of a potential factor XIII deficiency should, therefore, be considered when interpreting such assays. Thrombelastographic assays to detect factor XIII deficiency are feasible. Further clinical studies are needed to determine the potential influence of the detection and treatment of (absolute or relative) factor XIII deficiency on outcome in surgical patients.
1. Ganter MT, Hofer CK. Coagulation monitoring: current techniques and clinical use of viscoelastic point-of-care coagulation devices. Anesth Analg 2008;106:1366–75
2. Nielsen VG, Cohen BM, Cohen E. Effects of coagulation factor deficiency on plasma coagulation kinetics determined via thrombelastography: critical roles of fibrinogen and factors II, VII, X and XII. Acta Anaesthesiol Scand 2005;49:222–31
3. Rugeri L, Levrat A, David JS, Delecroix E, Floccard B, Gros A, Allaouchiche B, Negrier C. Diagnosis of early coagulation abnormalities in trauma patients by rotation thrombelastography. J Thromb Haemost 2007;5:289–95
4. Bowbrick VA, Mikhailidis DP, Stansby G. Influence of platelet count and activity on thromboelastography parameters. Platelets 2003;14:219–24
5. Olorundare OE, Simmons SR, Albrecht RM. Cytochalasin D and E: effects on fibrinogen receptor movement and cytoskeletal reorganization in fully spread, surface-activated platelets: a correlative light and electron microscopic investigation. Blood 1992;79:99–109
6. Nielsen VG, Geary BT, Baird MS. Evaluation of the contribution of platelets to clot strength by thromboelastography in rabbits: the role of tissue factor and cytochalasin D. Anesth Analg 2000;91:35–9
7. Lang T, Toller W, Gutl M, Mahla E, Metzler H, Rehak P, Marz W, Halwachs-Baumann G. Different effects of abciximab and cytochalasin D on clot strength in thrombelastography. J Thromb Haemost 2004;2:147–53
8. Carroll RC, Craft RM, Chavez JJ, Snider CC, Kirby RK, Cohen E. Measurement of functional fibrinogen levels using the Thrombelastograph. J Clin Anesth 2008;20:186–90
9. Kettner SC, Panzer OP, Kozek SA, Seibt FA, Stoiser B, Kofler J, Locker GJ, Zimpfer M. Use of abciximab-modified thrombelastography in patients undergoing cardiac surgery. Anesth Analg 1999;89:580–4
10. Lang T, Johanning K, Mahla E, Nentwig B, Scheinichen D, Nurtsch A, von Depka M, Borscheuer A. Clot firmness cannot be estimated by fibrinogen concentration and platelet count. Haemostaseologie 2007;27:A60
11. Nielsen VG, Gurley WQ Jr, Burch TM. The impact of factor XIII on coagulation kinetics and clot strength determined by thrombelastography. Anesth Analg 2004;99:120–3
12. Francis CW, Marder VJ. Rapid formation of large molecular weight alpha-polymers in cross-linked fibrin induced by high factor XIII concentrations. Role of platelet factor XIII. J Clin Invest 1987;80:1459–65
13. Schroeder V, Chatterjee T, Kohler HP. Influence of blood coagulation factor XIII and FXIII Val34Leu on plasma clot formation measured by thrombelastography. Thromb Res 2001;104:467–74
14. McDonagh J, McDonagh RP Jr, Delage JM, Wagner RH. Factor XIII in human plasma and platelets. J Clin Invest 1969;48:940–6
15. Nielsen VG, Kirklin JK, Hoogendoorn H, Ellis TC, Holman WL. Thrombelastographic method to quantify the contribution of factor XIII to coagulation kinetics. Blood Coagul Fibrinolysis 2007;18:145–50
16. Nomura S, Yanabu M, Miyake T, Miyazaki Y, Kawakatsu T, Kido H, Yamaguchi K, Fukuroi T, Kagawa H, Suzuki M, Kokawa T, Fukuhara S. Effect of cepharanthin and cytochalasin D on platelet internalization of anti-glycoprotein IIb/IIIa antibodies. Autoimmunity 1994;18:23–9
17. Fickenscher K, Aab A, Stuber W. A photometric assay for blood coagulation factor XIII. Thromb Haemost 1991;65:535–40
18. Harder S, Klinkhardt U, Graff J, Westrup D, Kirchmaier CM, Glusa E, Mascelli MA, Marciniak SJ, Just A, Losche W, Breddin HK. In vitro
dose response to different GPIIb/IIIa-antagonists: inter-laboratory comparison of various platelet function tests. Thromb Res 2001;102:39–48
19. Moser M, Bertram U, Peter K, Bode C, Ruef J. Abciximab, eptifibatide, and tirofiban exhibit dose-dependent potencies to dissolve platelet aggregates. J Cardiovasc Pharmacol 2003;41:586–92
20. Gerlach R, Raabe A, Zimmermann M, Siegemund A, Seifert V. Factor XIII deficiency and postoperative hemorrhage after neurosurgical procedures. Surg Neurol 2000;54:260–4
21. Rasche H, Haghou F, Gaus W, Dietrich M, Hoelzer D, Pflieger H, Kurrle E, Pindur G, Seifried E, Heimpel H. [Blood clotting factor XIII substitution in acute leukaemia: result of a randomized and controlled study]. Dtsch Med Wochenschr 1982;107:1882–6
22. Gerlach R, Tolle F, Raabe A, Zimmermann M, Siegemund A, Seifert V. Increased risk for postoperative hemorrhage after intracranial surgery in patients with decreased factor XIII activity: implications of a prospective study. Stroke 2002;33:1618–23
23. Shen L, Lorand L. Contribution of fibrin stabilization to clot strength. Supplementation of factor XIII-deficient plasma with the purified zymogen. J Clin Invest 1983;71:1336–41
24. Chakroun T, Gerotziafas GT, Seghatchian J, Samama MM, Hatmi M, Elalamy I. The influence of fibrin polymerization and platelet-mediated contractile forces on citrated whole blood thromboelastography profile. Thromb Haemost 2006;95:822–8
25. Ajzner E, Schlammadinger A, Kerenyi A, Bereczky Z, Katona E, Haramura G, Boda Z, Muszbek L. Severe bleeding complications caused by an autoantibody against the B subunit of plasma factor XIII: a novel form of acquired factor XIII deficiency. Blood 2009;113:723–5
26. Weisberg LJ, Shiu DT, Conkling PR, Shuman MA. Identification of normal human peripheral blood monocytes and liver as sites of synthesis of coagulation factor XIII a-chain. Blood 1987;70:579–82
27. Adany R. Intracellular factor XIII: cellular distribution of factor XIII subunit a in humans. Semin Thromb Hemost 1996;22:399–408
28. Muszbek L, Polgar J, Boda Z. Platelet factor XIII becomes active without the release of activation peptide during platelet activation. Thromb Haemost 1993;69:282–5
29. Hosoi S, Shinomiya K, Mikawa H. The use of monoclonal antibodies in demonstrating the effect of antibody heterogeneity on immune complex size. Clin Immunol Immunopathol 1984;32:378–86
30. Adany R, Bardos H. Factor XIII subunit A as an intracellular transglutaminase. Cell Mol Life Sci 2003;60:1049–60
31. Haas T, Fries D, Velik-Salchner C, Oswald E, Innerhofer P. Fibrinogen in craniosynostosis surgery. Anesth Analg 2008;106:725–31 , table of contents
32. Gorlinger K. [Coagulation management during liver transplantation]. Hamostaseologie 2006;26:S64–6
33. Kozek-Langenecker S. Management of massive operative blood loss. Minerva Anestesiol 2007;73:401–15
34. Craft RM, Chavez JJ, Bresee SJ, Wortham DC, Cohen E, Carroll RC. A novel modification of the Thrombelastograph assay, isolating platelet function, correlates with optical platelet aggregation. J Lab Clin Med 2004;143:301–9
35. Korte WC, Szadkowski C, Gahler A, Gabi K, Kownacki E, Eder M, Degiacomi P, Zoller N, Devay J, Lange J, Schnider T. Factor XIII substitution in surgical cancer patients at high risk for intraoperative bleeding. Anesthesiology 2009;110:239–45
36. Mosesson MW, Siebenlist KR, Hernandez I, Lee KN, Christiansen VJ, McKee PA. Evidence that alpha2-antiplasmin becomes covalently ligated to plasma fibrinogen in the circulation: a new role for plasma factor XIII in fibrinolysis regulation. J Thromb Haemost 2008;6:1565–70
37. Lukacova D, Matsueda GR, Haber E, Reed GL. Inhibition of factor XIII activation by an anti-peptide monoclonal antibody. Biochemistry 1991;30:10164–70