Secondary Logo

Journal Logo

Cardiovascular Anesthesiology: Research Report

Fibrinogen Measurements in Plasma and Whole Blood

A Performance Evaluation Study of the Dry-Hematology System

Ogawa, Satoru MD*; Tanaka, Kenichi A. MD, MSc; Nakajima, Yasufumi MD, PhD*; Nakayama, Yoshinobu MD*; Takeshita, Jun MD*; Arai, Masatoshi; Mizobe, Toshiki MD, PhD*

Author Information
doi: 10.1213/ANE.0000000000000448
  • Free

Fibrinogen plays crucial roles in establishing hemostasis1 and supporting wound healing.2 Patients undergoing cardiopulmonary bypass (CPB) are particularly vulnerable to hypofibrinogenemia due to blood loss and hemodilution.3 Several studies have shown that the risk of excessive bleeding after CPB is inversely correlated with plasma fibrinogen levels.4 It is thus important to accurately determine fibrinogen level in a timely manner to guide fibrinogen replacement.5,6

Plasma fibrinogen is most commonly measured using the modified Clauss method,7 or the prothrombin time (PT)-derived method.8,9 These tests usually take 30 to 60 minutes before results are available.10 Prompt diagnosis of hypofibrinogenemia is necessary because the administration of corrective therapy may be delayed due to the need to thaw plasma or cryoprecipitate.11 Plasma fibrinogen assays are further affected by type of diluents and anticoagulant agents (e.g., heparin) used in surgical settings.12,13 Thromboelastography and thromboelastometry allow a rapid assessment of fibrin polymerization in whole blood,14 but the optimal cutoff value for fibrinogen replacement has not been clearly established.15

Another approach to fibrinogen measurement is the dry reagent method (dry-hematology), which assesses thrombin-induced clot formation in an oscillating magnetic field.16 This technology has been used for whole blood measurements of PT, activated partial thromboplastin time (aPTT), and activated clotting time.16,17 The DRIHEMATO® system (CG02N, A&T Corporation, Kanagawa, Japan) is a cartridge-based fibrinogen assay approved for clinical use by the Ministry of Health in Japan. However, there is a paucity of data on the correlation between fibrinogen assays from plasma compared with whole blood particularly in relation to changes in hematocrit that are common perioperatively.18,19 We hypothesized that plasma or whole blood fibrinogen measurement using the dry-hematology method would be similar to those measured with conventional plasma fibrinogen assays.

METHODS

Part I: Fibrinogen Measurements from Healthy Volunteers

After approval of A&T Corporation IRB (Kanagawa, Japan, IRB#: AANDT-2013-A1), the first part of the study was conducted with blood samples collected from 12 healthy volunteers after informed written consents were obtained. These subjects had not taken medication(s) that affect platelet function or coagulation in the preceding 2 weeks. Whole blood samples were collected into glass tubes including 3.13% sodium citrate (Insepack II-W, Sekisui Medical Co. Ltd., Tokyo, Japan). The hematocrit level was measured by MYTHIC 18-J (A&T Corporation, Kanagawa, Japan). Blood samples were immediately processed for whole blood testing, and the remainder was centrifuged at 2000g for 20 minutes to obtain platelet-poor plasma for plasma assays. Plasma samples were stored at −80°C until batch analysis.

In the initial experiment, acquired hypofibrinogenemia was modeled by serial dilutions using 1 of 3 diluents to achieve 25%, 50%, and 75% volume replacement. The diluent was normal saline, 5% human albumin (CSL Behring, Marburg, Germany), or 6% hydroxyethyl starch (HES) (Voluven, Fresenius Kabi, Uppsala, Sweden). The effect of heparin on plasma fibrinogen assays was assessed by the addition of porcine heparin (Mochida Pharmaceutical, Tokyo, Japan) to obtain 0, 1, 2, 4, and 6 IU/mL (final concentration).

Plasma fibrinogen levels were measured using the dry-hematology method, the Clauss method, the PT-derived method, antigen levels measurement, and by thromboelastometric fibrin formation. The dry-hematology method uses plasma samples added to a disposable card containing bovine thrombin activator (final concentration: 33 IU/mL), and paramagnetic iron oxide particles that move in response to an oscillating magnetic field (prewarmed to 37°C) (Fig. 1A). The principle of this assay is based on the technology formerly used in the Thrombolytic Assessment System (Helena Laboratories, Beaumont, TX).16,17 Twenty-five microliters of well-mixed citrated plasma is manually mixed with the diluent according to the manufacturer’s instruction just before sample loading to the device. The loaded sample is moved via capillary action and is mixed with paramagnetic iron oxide particles and reagents within the testing chamber. As thrombus is formed, the particle movement is decreased. The light (870 nm) illuminates the inner reaction cell, and the signal movements of particles are quantitatively measured as changes in scattered lights. Coagulation time is determined as the decrease of 30% in maximal amplitude (Fig. 1B). Clotting time was measured in seconds, and correlated to fibrinogen concentrations (mg/dL) determined by the reference laboratory method. The test result is available in 1 minute. The precision of this assay was evaluated in quintuple testing for each plasma sample.

Figure 1
Figure 1:
The principle of the dry-hematology system. The blood sample loaded to a disposable cartridge is carried via capillary action and then mixed with the paramagnetic iron oxide particles and thrombin reagent in the testing chamber. By turning the power of the electromagnet on and off, particles move up and down within the reaction cell (A), and the photometric sensor detects the movement of particles as light and shade waveform (B). As the blood sample is clotted, particle movement is decreased. Clotting time is measured in seconds and correlated to fibrinogen concentrations (mg/dL) determined using the reference laboratory method.

The Clauss method was performed on a coagulation analyzer (KC-4 coagulometer, Amelung, Lieme, Germany) using the manufacturer’s directions and thrombin reagents (Thrombocheck Fib, Sysmex, Kobe, Japan). For the PT-derived method, fibrinogen levels were determined using HemosIL RecombiPlastin (Mitsubishi Chemical Medience, Tokyo, Japan) with appropriate calibrations on the STACIA (Mitsubishi Chemical Medience). Antigen levels of fibrinogen were determined by means of nephelometry (7170 Automatic Analyzer, Hitachi High-Technologies, Tokyo, Japan), using human antiserum generated from goats (Nittobo Medical Co., Tokyo, Japan). Fibrin polymerization was also evaluated by thromboelastometry (ROTEM, Tem Innovations GmbH, Munich, Germany) using FIBTEM (Tem Innovations GmbH), which allows fibrinogen-specific maximum clot firmness (mm).20 FIBTEM reagent contains tissue factor, and cytochalasin D, which is an inhibitor of platelet actin polymerization. To evaluate the relationship between plasma and whole blood measurements both determined with the dry-hematology method, 25 μL of citrated whole blood was manually mixed with diluent, and it was subsequently loaded onto the device. The hematocrit-adjusted whole blood fibrinogen values were calculated by including a correction factor as follows:

Adjusted whole blood fibrinogen level = measured whole blood fibrinogen level/{[(100 – Hematocrit)/100] × F} where a coefficient (F) is used to correct for the volume ratio (vol/vol) of assay diluent and whole blood in relation to the vol/vol ratio of diluent and plasma. This is performed because a calibration line for fibrinogen measurements is calculated using plasma as a standard. In this study, the coefficient F was set as ½.

Part II: Fibrinogen Measurements from Cardiac Surgery Patients

The second part of the study was conducted at Kyoto Prefectural University Hospital of Medicine, according to the protocol approved by Kyoto Prefectural University of Medicine IRB (Kyoto, Japan, KPUM-ERB-C-14). After obtaining informed written consent, 20 patients undergoing CPB surgery had whole blood samples collected with glass tubes including 3.13% sodium citrate (Insepack II-W) to evaluate the effect of in vivo hemodilution on each fibrinogen assay. Inclusion criteria were age >20 years, normal preoperative PT (10–12 seconds), aPTT (25–40 seconds), and platelet count (150–400 ×109 per liter). The arterial blood samples were immediately processed for whole blood testing, and the remainder was used for plasma assays after centrifugation at 2000g for 20 minutes. Patients receiving anticoagulation or thrombolytic therapies before surgery, and those with preexisting hepatic dysfunction were excluded. Patients included in the study were given heparin 300 U/kg to achieve activated clotting time >400 seconds (Hemochron Signature Elite, International Technidyne Corporation, Edison, NJ) before instituting CPB. Intraoperative red blood cell salvage was used in all cases. Albumin was the only type of colloid used for intravascular volume replacement. Heparin anticoagulation was neutralized after CPB with 1.0 to 1.2 mg protamine sulphate per 100 IU of the initial bolus. Blood samples were obtained at 3 time points: (1) before heparin administration (baseline), (2) after protamine administration before any hemostatic intervention (post-CPB), and (3) at the end of surgery. If microvascular bleeding persisted, hemostatic interventions were administered at the discretion of the attending anesthesiologist.

Fibrinogen measurements by the dry-hematology method were performed in parallel with the Clauss method, PT-derived method, and fibrinogen antigen measurement. The relationship between plasma and whole blood fibrinogen measurement using the dry-hematology method was evaluated. In addition, fibrin polymerization was assessed by thromboelastometry in the whole blood sample.

Statistical Analysis

Sample size determination for the in vitro dilution part of the study was based on a previously reported comparison between the Clauss and PT-derived fibrinogen level measurements where fibrinogen levels were 283 ± 8 mg/dL 262 ± 10 mg/dL, respectively.12 We estimated that a sample size of 8 per group would detect differences between each fibrinogen test with a power (1-β) of 0.84 between Clauss and PT-derived fibrinogen, with a power of 0.99 between Clauss and antigen levels (α = 0.05). We thus chose to have a sample size of 12 based on our preliminary data (coefficients of variation [CV] on dry-hematology, 8%) to obtain maximum permissible errors ≤5% (confidence coefficient: 95%); n = (22 × CV2)/maximum permissible errors.2 In the second part of the study, 60 measurements were obtained from 20 patients to achieve a correlation coefficient of 0.64 with a 95% confidence interval of (0.47 to 0.76) based on previous studies.20,21 Data are expressed as median (IQR) according to nonnormal distributions by the Kolmogorov-Smirnov test. Pearson correlation coefficient (r2) was used to assess linear association between the different methods for fibrinogen levels. The Lin’s concordance correlation coefficient (Pc) was applied to conduct these comparisons except in evaluating a correlation between fibrinogen levels and thromboelastometry (i.e., mg/dL versus mm).22 The interchangeability of tested method with the reference method was evaluated using the following scale of values: Pc < 0.90 as poor, 0.90 ≤ Pc < 0.95 as moderate, and 0.95 ≤ Pc as substantial.23 The agreement between methods of measuring fibrinogen concentration was analyzed using a Bland-Altman plot for repeated measurements.24,25 The bias is defined as the mean of the difference between 2 measurements, and a corresponding 95% prediction interval (PI) for the bias is presented. The statistical significances of the difference among the groups were assessed by nonparametric Friedman analysis of variance and Mann-Whitney U test, as appropriate. A P value of 0.05 was considered significant, and statistically significant P values were corrected using the Bonferroni correction for repeated measurements. All analyses were performed using Graph-Pad Prism, Version 5.0 (Graph-Pad Software, Inc., San Diego, CA), or MedCalc, Version 13.0.6.0 (MedCalc Software, Mariakerke, Belgium).

RESULTS

Part I: Plasma Fibrinogen Measurements in Healthy Volunteers

The intraindividual coefficient of variation of dry-hematology was 2.5% (IQR: 1.8, 3.2) in plasma samples (n = 12) as determined by quintuple testing. In plasma, there was no difference in fibrinogen concentrations between the dry-hematology and the Clauss method (254 mg/dL [IQR: 222, 284] vs 250 mg/dL [IQR: 224, 280]; P = 0.44). Fibrinogen levels determined with the PT-derived method were higher (288 mg/dL [IQR: 274, 349]) than that measured with the dry-hematology method (P = 0.041). Fibrinogen levels determined with the antigen level method (308 mg/dL [IQR: 260, 348]) were higher than those measured with the dry-hematology method (P = 0.0117). In plasma diluted with normal saline in vitro (n = 48), fibrinogen levels measured with the dry-hematology method and the Clauss method were correlated (Pc = 0.99, Fig. 2A) as were those measured with the PT-derived method (Pc = 0.95, Fig. 2B) and the fibrinogen antigen level method (Pc = 0.91, Fig. 2C). Bland-Altman analysis assessing the agreement among the dry-hematology method and the other methods for measuring fibrinogen demonstrated a bias of −5 mg/dL (95% PI: −29, 19) for the Clauss method, 18 mg/dL (95% PI: −14, 50) for the PT-derived method, and 31 mg/dL (95% PI: −3.2, 64) for the fibrinogen antigen levels method (Fig. 2, A–C). There was a correlation between the dry-hematology and thromboelastometric results (r2 = 0.84; P < 0.0001), but the correlation was weaker than those with other fibrinogen assays (Fig. 3). After in vitro dilution, fibrinogen levels determined with the dry-hematology method were decreased proportionally with 25%, 50%, and 75% dilution, but there were no significant differences in this finding among the type of diluents, or whether fibrinogen was measured with the Clauss or fibrinogen antigen level methods (Table 1). However, fibrinogen levels determined with the PT-derived method were elevated in plasma diluted with albumin and 6% HES compared with the levels after saline dilution (Table 1). Fibrin polymerization assessed with thromboelastometry was extensively reduced after dilutions with albumin and 6% HES compared with saline. The addition of heparin (>4 U/mL) affected the results of the PT-derived method, but not the dry-hematology method (Table 2).

Figure 2
Figure 2:
The relationship between the dry-hematology (DH) method and (A) the Clauss method, (B) the prothrombin time (PT)-derived method, and (C) fibrinogen antigen levels in plasma samples obtained from healthy volunteers. The dry-hematology method yielded acceptable concordance correlation coefficients with the different methods for fibrinogen levels (concordance correlation coefficients [Pc] = 0.99, 0.95, and 0.91, respectively). In the Bland-Altman analysis assessing the agreement among different methods for measuring fibrinogen, the Clauss method showed less bias with the dry-hematology method compared with the PT-derived method or with antigen levels. Gray bold lines in Bland-Altman analyses depict bias and dotted lines depict 95% prediction intervals.
Figure 3
Figure 3:
The correlation between the dry-hematology (DH) method and thromboelastometric fibrin formations using plasma samples obtained from healthy volunteers. There was moderate correlation between the dry-hematology method and thromboelastometric assay (r2 = 0.84; P < 0.0001).
Table 1
Table 1:
The Effects of Diluents on Fibrinogen Assays
Table 2
Table 2:
The Effect of Heparin on Fibrinogen Assays

Comparison Between Whole Blood and Plasma Fibrinogen Measurements in Healthy Volunteers

The intraindividual coefficient of variation from the dry-hematology assays determined by quintuplet measurements was 1.8% (IQR: 0.7, 2.2) when determined from the whole blood samples. The baseline value of hematocrit was 37.5% (IQR: 34.7, 40.0). For the dry-hematology method, the fibrinogen levels in whole blood samples were higher than when measured from plasma samples (315 mg/dL [IQR: 302, 344] vs 254 mg/dL [IQR: 222, 284]; P = 0.0012). There was low correlation between the fibrinogen levels measured from whole blood versus plasma (r2 = 0.77; P = 0.0002 and Pc = 0.39). After adjustment for hematocrit, the correlation of fibrinogen levels determined from whole blood versus plasma was stronger (r2 = 0.85; P < 0.0001 and Pc = 0.92).

Part II: Plasma Fibrinogen Measurements from Cardiac Surgery Patients

In cardiac surgery patients (mean: 75 years [IQR: 71, 79], mean height 160 cm [IQR: 153, 167], mean body weight 58 kg [IQR: 50, 64]), there was no difference in plasma fibrinogen levels determined with the dry-hematology and Clauss methods (185 mg/dL [IQR: 156, 235] vs 180 mg/dL [IQR: 146, 227], respectively; P = 0.58). The fibrinogen levels determined with the PT-derived method (265 mg/dL [IQR: 221, 323]) and the fibrinogen antigen method (225 mg/dL [IQR: 189, 279]), however, were higher compared with the levels determined by the dry-hematology method (P < 0.0001, and P = 0.0015, respectively). There was a correlation between the dry-hematology and Clauss methods (Pc = 0.96). However, correlations were weaker between the dry-hematology and PT-derived methods (Pc = 0.61), and between the dry-hematology and the fibrinogen antigen level methods (Pc = 0.82).

Comparison Between Whole Blood and Plasma Fibrinogen Measurements

The median value of the hematocrit in cardiac patients was 25.9% (IQR: 23.8, 28.0). There was a correlation between the fibrinogen levels measured from whole blood versus plasma (r2 = 0.91; P < 0.0001), but the reproducibility was low (Pc = 0.58) (Fig. 4A). Fibrinogen levels in whole blood samples thus showed significantly higher values than those measured from plasma (260 mg/dL [IQR: 231, 310] vs 185 mg/dL [IQR: 156, 235]; P < 0.0001). Subsequent hematocrit adjustment in the fibrinogen levels obtained from the dry-hematology method decreased the differences between measurements in whole blood versus plasma as assessed by Bland-Altman analysis (bias, 73 mg/dL [95%PI: 40, 106] to −13 mg/dL [95%PI: −35, 8.5]) (Fig. 4, A and B). There was no significant difference between hematocrit-adjusted fibrinogen levels measured in whole blood using the dry-hematology method and those measured in plasma using the Clauss method (183 mg/dL [IQR: 151, 210] vs 180 mg/dL [IQR: 146, 227], respectively; P = 0.60). Hematocrit adjustment further improved the correlations between whole blood and plasma fibrinogen values by the dry-hematology method (Pc = 0.92), as well as the correlation with Clauss fibrinogen levels in plasma (Pc = 0.92) (Fig. 4C). There was a moderate correlation between the dry-hematology measurements from whole blood and the thromboelastometric fibrinogen assay (r2 = 0.72; P < 0.0001).

Figure 4
Figure 4:
The relationship between (A) plasma samples and whole blood (WB) samples in the dry-hematology (DH) method, (B) plasma samples and hematocrit-adjusted whole blood (AWB) samples in the DH method, and (C) AWB samples in the DH method and plasma samples in Clauss method, from cardiac patients. The hematocrit adjustment in fibrinogen levels obtained from the DH method improved concordance correlation coefficient (concordance correlation coefficients [Pc] = 0.58 to 0.92), and decreased the differences in bias obtained from the Bland-Altman analysis between WB and plasma samples. AWB fibrinogen levels in the DH method showed acceptable concordance correlation coefficient (Pc = 0.92) and less bias (5 mg/dL [prediction intervals: −19, 28]) with Clauss method in plasma samples. Gray bold lines in Bland-Altman analyses depict bias and dotted lines depict 95% prediction intervals.

DISCUSSION

In this study, we demonstrated that fibrinogen levels assessed by the dry-hematology method in plasma are comparable to results from established assay methods and that the results are not affected by heparin or colloids. Furthermore, we showed that whole blood fibrinogen levels can be quickly and accurately assessed by the dry-hematology method when corrected for hematocrit level.

There are several approaches for measuring fibrinogen assays including determination of the amount of fibrin generated by thrombus formation, assessing for the amount of fibrinogen antigen present, or the actual measurement of fibrinogen protein in plasma samples. The Clauss method is the most commonly used method for measuring fibrinogen level.7 This method measures fibrinogen levels generated in a diluted plasma sample in the presence of excess thrombin. The dry-hematology method for measuring fibrinogen levels is a modification of the Clauss method.8,9 However, there are several other methods for clinically assessing fibrinogen levels using different approaches.10,12 The PT-derived method is not a direct determination of plasma fibrinogen level per se, but an estimation of its concentration performed during a clotting process triggered by tissue thromboplastin on an automated coagulometer. Although this method is simple and inexpensive, some authors have concluded that the test might be inaccurate and misleading in the presence of hypofibrinogenemia or after thrombolytic therapy.8,9,26 Fibrinogen levels derived from the Clauss method are generally lower than PT-derived fibrinogen levels,9 and this was corroborated in this study (Table 1). Assessing fibrinogen levels using an antigen assay is based on electrophoresis or enzyme-linked immunosorbent assays. However, in the presence of abnormal forms of fibrinogen (i.e., dysfibrinogenemia), the antigen levels show higher fibrinogen levels than the Clauss assay,9,12 a finding that is agreement with our data (Table 1).

The measurement of fibrinogen in the presence of colloids results in erroneously high fibrinogen levels measured by certain coagulation analyzers and methods.12 Indeed, functional photometric assay used for the PT-derived method significantly overestimated levels of fibrinogen after dilution with albumin and HES, and fibrin formations in thromboelastometry were decreased in albumin and HES diluents (Table 1). On the other hand, the dry-hematology method, similar to the Clauss method, showed a limited impact from dilution resulting from saline, HES, or albumin.12,27 Also, high levels of heparin cause an underestimation of fibrinogen levels in some fibrinogen assays.13 This study showed the dry-hematology method was unaffected by a high concentration of heparin (up to 6 U/mL), while the PT-derived method was susceptible to heparin levels (Table 2).

Several whole blood point-of-care devices are currently used to assess coagulation variables (e.g., PT, aPTT, thromboelastometry, and thromboelastography).11,28 The use of transfusion algorithms with predefined variables for hemostatic interventions has been shown to reduce blood loss and transfusion requirements in cardiothoracic surgery.15,29 A simple bedside whole blood fibrinogen measurement is desirable because acquired hypofibrinogenemia below 150 mg/dL is common in perioperative, peripartum, and trauma hemorrhage.5,30,31 In this study, we demonstrated that whole blood fibrinogen levels in patients undergoing cardiac surgery can be quickly assessed by the dry-hematology, and the results after hematocrit adjustment were highly correlated with plasma fibrinogen levels (Fig. 4, B and C).18 These results support the findings of Amukele et al.18 who compared Clauss fibrinogen levels between whole blood and plasma samples using an automated coagulation analyzer (STA and STA fibrinogen reagent, Diagnostica Stago, Parsippany, NJ). In their study, samples obtained from patients (n = 126) showed that the correlation between whole blood and plasma fibrinogen levels improved after hematocrit correction (r2 =0.929 to 0.984); adjusted whole blood fibrinogen = measured whole blood fibrinogen/[1.4294 – (0.0137 × hematocrit)]).

The thromboelastometry FIBTEM test is used perioperatively to assess the extent of fibrin polymerization in whole blood.20,29 Low FIBTEM-clot firmness (<8–10 mm) has been used without hematocrit adjustment as a trigger for fibrinogen replacement.29,30 In this study, the correlation of the Clauss fibrinogen with FIBTEM was moderate (r2 = 0.72), but less than that of the dry-hematology fibrinogen and Clauss methods (r2 = 0.94). This may be in part due to residual factor XIII activity and the lack of hematocrit adjustment in the FIBTEM test.19,32,33 Our findings of high correlations between the Clauss and hematocrit-adjusted dry-hematology values imply that the latter is a practical alternative (Fig. 4C).

This study has several limitations. First, we did not measure plasma fibrin degradation product levels. Increasing fibrin degradation product levels have been found to affect the result of fibrinogen assays, especially the PT-derived method.8 Second, we did not examine the effect of fibrinogen replacement therapy on the dry-hematology method. Indeed, Solomon et al.10 reported that the agreements among fibrinogen assays are diminished after hemostatic interventions, particularly the infusion of fibrinogen concentrate. Finally, the dry-hematology method currently requires sample dilution before testing, and thus proper quality control and technical skills should be assured.

In conclusion, the dry-hematology method can be used to measure fibrinogen levels in a small volume (25 μL) of plasma or whole blood sample. The results of the dry-hematology method in plasma are equivalent to the Clauss method and are hardly affected by heparin or colloids. For whole blood fibrinogen measurements by the dry-hematology method, hematocrit adjustment is necessary to compensate for dynamic changes in hematocrit in perioperative bleeding events. The use of whole blood fibrinogen assay could provide a shorter turnaround time compared with conventional plasma measurements. Further studies including a larger sample size are warranted to examine the clinical efficacy of using whole blood fibrinogen level as a trigger for fibrinogen replacement.

DISCLOSURES

Name: Satoru Ogawa, MD.

Contribution: This author conducted the study, analyzed the data, and wrote the manuscript.

Attestation: Satoru Ogawa is the archival author, and attests to the integrity of the original data and the analysis reported in this manuscript.

Conflicts of Interest: Satoru Ogawa has previously received honoraria for lectures related to DRIHEMATO (A&T, Kanagawa, Japan); the company was involved in the conduction of the present study.

Name: Kenichi A. Tanaka, MD, MSc.

Contribution: This author helped with the concept of the study, and the manuscript preparation.

Attestation: Kenichi A. Tanaka approved the final manuscript.

Conflicts of Interest: Kenichi A. Tanaka has previously received honoraria for lectures related to ROTEM (TEM Innovations, Munich, Germany); the company was not involved in the planning or analysis of the present study.

Name: Yasufumi Nakajima, MD, PhD.

Contribution: This author helped to conduct the study, and the manuscript preparation.

Attestation: Yasufumi Nakajima is the archival author, and approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Yoshinobu Nakayama, MD.

Contribution: This author helped to conduct the study.

Attestation: Yoshinobu Nakayama approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Jun Takeshita, MD.

Contribution: This author helped to conduct the study.

Attestation: Jun Takeshita approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Masatoshi Arai.

Contribution: This author conducted the study, and helped to analyze the data.

Attestation: Masatoshi Arai attests to the integrity of the original data and the analysis reported in this manuscript. Masatoshi Arai is the archival author, and approved the final manuscript.

Conflicts of Interest: Masatoshi Arai is an employee of the A&T Corporation, which manufactures the dry-hematology system.

Name: Toshiki Mizobe, MD, PhD.

Contribution: This author helped to conduct the study.

Attestation: Toshiki Mizobe approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

This manuscript was handled by: Charles W. Hogue, Jr, MD

ACKNOWLEDGMENTS

The authors are grateful for Yoichi Yuki (Department of Clinical Laboratory, Kyoto Prefectural University of Medicine) for his technical assistance with coagulation assays.

REFERENCES

1. Levy JH, Welsby I, Goodnough LT. Fibrinogen as a therapeutic target for bleeding: a review of critical levels and replacement therapy. Transfusion. 2014;54:1389–405
2. Clark RA. Fibrin and wound healing. Ann N Y Acad Sci. 2001;936:355–67
3. Bolliger D, Szlam F, Levy JH, Molinaro RJ, Tanaka KA. Haemodilution-induced profibrinolytic state is mitigated by fresh-frozen plasma: implications for early haemostatic intervention in massive haemorrhage. Br J Anaesth. 2010;104:318–25
4. Blome M, Isgro F, Kiessling AH, Skuras J, Haubelt H, Hellstern P, Saggau W. Relationship between factor XIII activity, fibrinogen, haemostasis screening tests and postoperative bleeding in cardiopulmonary bypass surgery. Thromb Haemost. 2005;93:1101–7
5. Bolliger D, Görlinger K, Tanaka KA. Pathophysiology and treatment of coagulopathy in massive hemorrhage and hemodilution. Anesthesiology. 2010;113:1205–19
6. Levy JH, Szlam F, Tanaka KA, Sniecienski RM. Fibrinogen and hemostasis: a primary hemostatic target for the management of acquired bleeding. Anesth Analg. 2012;114:261–74
7. CLAUSS A. Rapid physiological coagulation method in determination of fibrinogen [in German]. Acta Haematol. 1957;17:237–46
8. Mackie J, Lawrie AS, Kitchen S, Gaffney PJ, Howarth D, Lowe GD, Martin J, Purdy G, Rigsby P, Rumley A. A performance evaluation of commercial fibrinogen reference preparations and assays for Clauss and PT-derived fibrinogen. Thromb Haemost. 2002;87:997–1005
9. Miesbach W, Schenk J, Alesci S, Lindhoff-Last E. Comparison of the fibrinogen Clauss assay and the fibrinogen PT derived method in patients with dysfibrinogenemia. Thromb Res. 2010;126:e428–33
10. Solomon C, Cadamuro J, Ziegler B, Schöchl H, Varvenne M, Sørensen B, Hochleitner G, Rahe-Meyer N. A comparison of fibrinogen measurement methods with fibrin clot elasticity assessed by thromboelastometry, before and after administration of fibrinogen concentrate in cardiac surgery patients. Transfusion. 2011;51:1695–706
11. Tanaka KA, Bader SO, Sturgil EL. Diagnosis of perioperative coagulopathy–plasma versus whole blood testing. J Cardiothorac Vasc Anesth. 2013;27:S9–15
12. Fenger-Eriksen C, Moore GW, Rangarajan S, Ingerslev J, Sørensen B. Fibrinogen estimates are influenced by methods of measurement and hemodilution with colloid plasma expanders. Transfusion. 2010;50:2571–6
13. Gertler R, Wiesner G, Tassani-Prell P, Braun SL, Martin K. Are the point-of-care diagnostics MULTIPLATE and ROTEM valid in the setting of high concentrations of heparin and its reversal with protamine? J Cardiothorac Vasc Anesth. 2011;25:981–6
14. Solomon C, Sørensen B, Hochleitner G, Kashuk J, Ranucci M, Schöchl H. Comparison of whole blood fibrin-based clot tests in thrombelastography and thromboelastometry. Anesth Analg. 2012;114:721–30
15. Bolliger D, Seeberger MD, Tanaka KA. Principles and practice of thromboelastography in clinical coagulation management and transfusion practice. Transfus Med Rev. 2012;26:1–13
16. Oberhardt BJ, Dermott SC, Taylor M, Alkadi ZY, Abruzzini AF, Gresalfi NJ. Dry reagent technology for rapid, convenient measurements of blood coagulation and fibrinolysis. Clin Chem. 1991;37:520–6
17. Gosselin RC, Owings JT, Pollock ME, Larkin E. Evaluation of a point-of-care coagulation device, the thrombolytic assessment system (TAS), for measuring prothrombin times, activated partial thromboplastin times, and high-dose heparin anticoagulation. Lab Hematol. 1998;4:217–24
18. Amukele TK, Ferrell C, Chandler WL. Comparison of plasma with whole blood prothrombin time and fibrinogen on the same instrument. Am J Clin Pathol. 2010;133:550–6
19. Ogawa S, Szlam F, Bolliger D, Nishimura T, Chen EP, Tanaka KA. The impact of hematocrit on fibrin clot formation assessed by rotational thromboelastometry. Anesth Analg. 2012;115:16–21
20. Ogawa S, Szlam F, Chen EP, Nishimura T, Kim H, Roback JD, Levy JH, Tanaka KA. A comparative evaluation of rotation thromboelastometry and standard coagulation tests in hemodilution-induced coagulation changes after cardiac surgery. Transfusion. 2012;52:14–22
21. Roullet S, Pillot J, Freyburger G, Biais M, Quinart A, Rault A, Revel P, Sztark F. Rotation thromboelastometry detects thrombocytopenia and hypofibrinogenaemia during orthotopic liver transplantation. Br J Anaesth. 2010;104:422–8
22. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics. 1989;45:255–68
23. McBride GB. A proposal for strength-of-agreement criteria for Lin’s concordance correlation coefficient. NIWA Client Report. HAM2005-062. 2005:1–10
24. Myles PS, Cui J. Using the Bland-Altman method to measure agreement with repeated measures. Br J Anaesth. 2007;99:309–11
25. Bland JM, Altman DG. Agreed statistics: measurement method comparison. Anesthesiology. 2012;116:182–5
26. Lawrie AS, McDonald SJ, Purdy G, Mackie IJ, Machin SJ. Prothrombin time derived fibrinogen determination on Sysmex CA-6000. J Clin Pathol. 1998;51:462–6
27. Hiippala ST. Dextran and hydroxyethyl starch interfere with fibrinogen assays. Blood Coagul Fibrinolysis. 1995;6:743–6
28. Ganter MT, Hofer CK. Coagulation monitoring: current techniques and clinical use of viscoelastic point-of-care coagulation devices. Anesth Analg. 2008;106:1366–75
29. Weber CF, Görlinger K, Meininger D, Herrmann E, Bingold T, Moritz A, Cohn LH, Zacharowski K. Point-of-care testing: a prospective, randomized clinical trial of efficacy in coagulopathic cardiac surgery patients. Anesthesiology. 2012;117:531–47
30. Schöchl H, Nienaber U, Hofer G, Voelckel W, Jambor C, Scharbert G, Kozek-Langenecker S, Solomon C. Goal-directed coagulation management of major trauma patients using thromboelastometry (ROTEM)-guided administration of fibrinogen concentrate and prothrombin complex concentrate. Crit Care. 2010;14:R55
31. Solomon C, Collis RE, Collins PW. Haemostatic monitoring during postpartum haemorrhage and implications for management. Br J Anaesth. 2012;109:851–63
32. 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
33. Jámbor C, Reul V, Schnider TW, Degiacomi P, Metzner H, Korte WC. In vitro inhibition of factor XIII retards clot formation, reduces clot firmness, and increases fibrinolytic effects in whole blood. Anesth Analg. 2009;109:1023–8
© 2015 International Anesthesia Research Society