An automated turbidimetric method for ... - Wiley Online Library

Veterinary Clinical Pathology ISSN 0275-6382

ORIGINAL RESEARCH

An automated turbidimetric method for fibrinogen determination in dogs n1 F. Tecles1, A. Tvarijonaviciute1, M. Caldın2, S. Martınez-Subiela1, J.J. Cero 1

Department of Animal Medicine and Surgery, Veterinary School, Murcia, Spain; and 2San Marco Veterinary Hospital, Padova, Italy

Key Words Acute phase response, ammonium sulfate, canine, fibrinogen depletion, precipitation Correspondence n, Department of Animal Medicine and J.J. Cero Surgery, Veterinary School, University of Murcia, 30100 Espinardo, Murcia, Spain E-mail: [email protected] DOI:10.1111/vcp.12126

Background: Several methods have been used for fibrinogen determination in dogs, but to the authors’ knowledge, methods based on ammonium sulfate precipitation have not yet been reported in this species. Objectives: The aim of this study was to develop and validate an automated method based on ammonium sulfate precipitation for canine fibrinogen determination. Methods: A reagent containing ammonium sulfate, sodium chloride, and K2EDTA was used to precipitate fibrinogen at a final ammonium sulfate concentration of 0.57 M and final turbidity was measured on a Cobas Mira Plus autoanalyzer. Analytic validation included imprecision, accuracy, comparison with reference method (Clauss), limits of detection and quantification, and the evaluation of the influence of different anticoagulants. For diagnostic validation, fibrinogen was determined in a group of Beagle dogs before and after neutering, and in dogs affected by diseases known to produce low fibrinogen plasma concentration, such as liver insufficiency, disseminated intravascular coagulation, and protein-losing enteropathy. Results: Low imprecison (90%), and low bias (0.092 g/L) with respect to Clauss method indicated a high reproducibility and accuracy. Limits of detection and quantification were 0.01 and 0.22 g/ L, respectively. The method was applicable in plasma samples anticoagulated with EDTA, heparin, or sodium citrate. The fibrinogen concentration in Beagle dogs after neutering was increased, and decreased in animals with disseminated intravascular coagulation, liver insufficiency, or gastrointestinal protein loss. Conclusions: The automated method validated in this study represents a rapid, cheap, and easy protocol to quantify canine fibrinogen in routine practice.

Introduction Fibrinogen is a protein synthesized by the liver, which is abundant in plasma of vertebrates. It is composed of 3 polypeptide chains (A-alpha, B-beta, and gamma) bound by disulfide bridges, and it has a molecular weight of 340 kDa.1 In dogs, fibrinogen quantification is used for assessment of acute phase response, a physiologic response to disturbed homeostasis caused by infection, tissue damage, neoplasia, or immunologic disorders. During the acute phase response, some inflammatory mediators such as cytokines are released eliciting variations in plasma concentrations of proteins called acute phase proteins (APPs).2 Fibrinogen is

172

considered a moderate APP based on a slow response and < 10-fold increase during inflammation.3 Fibrinogen is also used to assess coagulation, as it is one of the coagulation factors involved in secondary hemostasis. Some diseases associated with hypofibrinogenemia or afibrinogenemia have been recognized to be related to hypocoagulability in dogs,4 so fibrinogen has been measured to investigate hemorrhagic disorders, such as in cases of disseminated intravascular coagulation (DIC). Finally, fibrinogen may also be measured to assess liver function. In human medicine, the Clauss5 method is considered the method of choice by the National Committee on Clinical Laboratory Standards for fibrinogen mea-

Vet Clin Pathol 43/2 (2014) 172–179 ©2014 American Society for Veterinary Clinical Pathology and European Society for Veterinary Clinical Pathology

Tecles et al

surement.6 In dogs, several assays have been used based on fibrinogen clotting.5,7–9 Other methods are based on heat precipitation,10 which is a cheap and fast method, but accuracy is poor.11 Recently, a novel method based on the precipitation of fibrinogen by vancomycin has been described in dogs with satisfactory results.12 The assay based on ammonium sulfate [(NH4)2SO4] precipitation has been used for human fibrinogen quantification.13–16 The basis of this methodology is the fact that a proper concentration of ammonium sulfate is able to precipitate fibrinogen from plasma samples with minimal effect on other plasma proteins. The amount of precipitated protein can be quantified by turbidimetry and thus independently of its clotting activity. To the authors’ knowledge, this method has not yet been reported in veterinary medicine. The main objective of this work was the development, optimization and validation of an automated turbidimetric assay for fibrinogen determination in canine plasma samples based on ammonium sulfate precipitation.

Turbidimetric fibrinogen determination in dogs

dogs using a 22 Gauge needle and a 5-mL syringe. All dogs were considered healthy based on physical examination, and hematology, and biochemistry profiles. Blood was collected in EDTA tubes (BD Vacutainer; Becton Dickinson, Franklin Lakes, NJ, USA), and a pool of canine plasma was prepared after centrifugation. Fibrinogen was purified from that pool by precipitation with the reagent containing a final concentration of 0.50M ammonium sulfate. After mixing and centrifugation at 2200g for 10 min, the pellet was resuspended in 0.11 M trisodium citrate (Sigma Chemical Co.). The protein concentration of the fibrinogen pool was determined by the biuret method. Sodium dodecyl-sulfate polyacrilamide gel electrophoresis (SDS-PAGE; Bio-Rad, Hemel Hempstead, Hertfordshire, UK) was performed to assess the presence and purity of fibrinogen. Commercially available purified human fibrinogen (Diagnostica Stago, Roche, Meylan, France) was used as control, and the overall evaluation was based on a molecular weight marker (Invitrogen Mark 12 Unstained Standard; Life Technologies, Carlsbad, CA, USA). Optimization of precipitation reagent

Materials and Methods Assay development for fibrinogen determination A previously described method for human fibrinogen measurement in plasma with some modifications was used in this study.16 It was run on an autoanalyzer (Cobas Mira Plus; Abx Diagnostics, Montpellier, France). Essentially, the precipitation reagent consisted of 1.56 M ammonium sulfate (Sigma Chemical Co., St. Louis, MO, USA), 39.55 mM K2EDTA (Sigma Chemical Co.), and 0.37 M NaCl (Sigma Chemical Co.) at pH 4.5. Five microliters of a canine plasma sample were diluted with 125 lL of deionized water. After 36 s, 22 lL of the precipitation reagent were added, resulting in a final concentration of 0.23M ammonium sulfate. After 3 min, another 53 lL of the precipitation reagent were added, resulting in a final ammonium sulfate concentration of 0.57M. After incubation for 9 min, the absorbance was recorded at 340 nm wavelength immediately before addition of the second reagent and during the final incubation period. The difference between both absorbance readings was proportional to fibrinogen concentration in the sample. Definitive sample and reagent volumes were established after the assay was optimized. Preparation of a canine fibrinogen standard Canine plasma samples were obtained by venipuncture of the jugular vein from 6 adult healthy Beagle

As proteins other than fibrinogen could co-precipitate if ammonium sulfate concentration was high enough, optimization of the concentration resulting in maximum fibrinogen precipitation with minimum coprecipitation of other proteins was crucial. For this purpose, a pool of canine EDTA plasma from the 6 Beagle dogs, the previously purified canine fibrinogen, and purified human fibrinogen at a concentration of 4 g/L (Diagnostica Stago) were used. A pool of canine serum from the same animals was used to check the coprecipitation of other proteins. Samples were incubated with serial dilutions/increasing volumes of the precipitation reagent to achieve ammonium sulfate concentrations between 0.34M and 1.41M. The optimal ammonium sulfate concentration was defined as the one that produced the largest proportion of canine fibrinogen precipitate with the least amount of coprecipitation of other serum proteins.

Analytic validation of the method Imprecision The intra-assay coefficient of variation (CV) was calculated by a 10-time repeat precipitation and analysis of 3 pooled EDTA plasmas with different fibrinogen concentrations in a single assay run. The inter-assay CV was determined by precipitation and analysis of the same pooled plasmas in 10 separate runs on different


173


days. All plasma samples were frozen in aliquots, and, in order to avoid possible changes due to repetitive thawing and freezing, only the vials needed for each run were thawed. Accuracy Linearity under dilution was evaluated by serially diluting and analyzing 3 pools of canine EDTA plasma with saline. The results were compared with those expected by linear regression analysis. Spike recovery was determined in different volumes of a pool of canine EDTA plasma with low fibrinogen concentration spiked with purified canine fibrinogen. The rate of recovery was calculated based on the difference between spiked and unspiked samples, relative to the amount of spike in the sample, according to the following formula: (observed resultsunspiked results)/spike amount 9 100. For the comparison with the Clauss standard method, a total of 33 canine citrate plasma samples were analyzed by both the precipitation method and by a Clauss modified method on an automated coagulation analyzer (STA Compact Coagulation Instrument; Diagnostica Stago) with commercially available reagents (Fibrinogen STA; Diagnostica Stago). These samples were obtained from healthy dogs and dogs with different inflammatory and coagulative disorders, providing a wide range of fibrinogen concentrations. The Bland–Altman graph was plotted and the correlation coefficient (r) between methods was calculated. Detection limit The limit of detection of the assay (ie, the lowest concentration of fibrinogen that could be distinguished from a specimen with a value of zero) was calculated on the basis of 20 replicate determinations of the zero standard (deionized water), as the mean value plus 3 standard deviations (SD). Limit of quantification A pool of canine EDTA plasma was serially diluted with saline. Each dilution was measured 10 times in a single assay run and the CV for each dilution was calculated. The quantitative limit was set as the minimum dilution, which provided an acceptable CV (< 15%).

Tecles et al

fugation, serum and plasma samples were analyzed and the results were compared. Effect of urea concentration Aliquots of a pool of canine EDTA plasma with a urea concentration of 6.64 mmol/L were spiked with different amounts of urea (final urea concentrations of 6.64, 12.45, 16.60, 28.22, 49.80, and 94.62 mmol/L; Sigma Chemical Co.). The means of triplicate fibrinogen measurements for each aliquot were then compared.

Clinical validation Samples of EDTA plasma were obtained from 4 male Beagle dogs prior to neutering and at 0, 24, 48, 72, and 120 h after surgery. This experiment was part of a previous research project for the study of the effect of neutering on some biochemistry analytes.17 The health status of the animals was evaluated by physical examination, and hematology and biochemistry profiles. Fibrinogen was analyzed by both the precipitation method and the method of Clauss, and the results were compared. This study protocol was approved by the Ethical Committee of the University of Murcia. The suitability of the assay to detect animals with low fibrinogen concentration was studied in a group of adult dogs of different breeds and ages affected by different diseases known to cause low fibrinogen plasma concentrations, such as liver insufficiency (n = 4), disseminated intravascular coagulation (DIC, n = 2), combined liver insufficiency and DIC (n = 2), and protein-losing enteropathy (n = 1). Diagnoses were made on the basis of clinical history, physical examination, hematology, biochemistry, and urine analyses results; D-dimers were determined for the diagnosis of DIC;. Chronic diarrhea and hypoalbuminemia (after ruling out renal loss) and liver failure were considered indicative for a diagnosis of protein-losing enteropathy that was further confirmed by endoscopy and biopsy. The fibrinogen concentrations determined in these dogs were compared with those obtained from a group of adult healthy dogs of different breeds (n = 15) that were presented to the clinic for routine health checks. They had no evidence for disease at physical examination, and hematology and biochemistry profiles were within the reference intervals of the laboratory.

Effects of anticoagulants Whole blood from the same 6 Beagle dogs used in the previous experiments was obtained as previously described and placed into plain, sodium citrate, lithium heparin, and EDTA blood collection tubes. After centri-

174

Statistical analyses Arithmetic means, standard deviations (SD), intraand inter-assay CVs, and linear regression analyses were performed using routine descriptive statistical


Tecles et al


procedures. All data were assessed for normality by using the Kolmogorov–Smirnov test; logarithmic transformation was applied to nonnormally distributed data to assume Gaussian distribution. One-way ANOVA and Tukey post hoc test were used to assess differences between anticoagulants. Linear regression testing for a slope different from zero was used to study interference with increasing urea concentrations. Pearson correlation coefficients were calculated and Bland–Altman plots were made to compare results obtained between the turbidimetric precipitation method and the reference method by Clauss. One-way ANOVA of repeated measures and Tukey’s multiple comparison tests were used to assess differences in fibrinogen concentration in Beagle dogs before and after neutering. An unpaired t-test was used to assess statistically significant differences between healthy dogs and dogs with fibrinogendepleting diseases. The significance level used in each case was P < .05. All statistical analyses were calculated using Excel 2000 (Microsoft Corporation, Redmond, WA, USA) and Graph Pad Software Inc. (La Jolla, CA, USA).

Results Method optimization Preparation of the standard After ammonium sulfate ((NH4)2SO4) precipitation, the protein content was 6.44 g/L in the precipitate. Polyacrilamide gel electrophoresis of the resuspended precipitate showed 3 bands with molecular weights coinciding with human fibrinogen, indicating that canine fibrinogen was the main constituent of the precipitate (Figure 1). Additional small bands of unidentified proteins were also noted. Optimization of (NH4)2SO4 concentration. Figure 2 shows precipitation curves of purified canine fibrinogen, canine pooled EDTA plasma, canine pooled serum, and human purified fibrinogen with increasing concentrations of the precipitation reagent. Purified fibrinogen and pooled plasma precipitation curves reached maximum absorbance at 0.61M ammonium sulfate. The pooled serum precipitation curve showed negligible absorbance at ammonium sulfate concentrations ≤ 0.57M, but absorbance progressively increased with rising ammonium sulfate concentrations, indicating serum protein precipitation other than fibrinogen. The optimal final ammonium sulfate concentration for the assay was therefore set at 0.57M.

Figure 1. Sodium dodecyl-sulfate polyacrilamide gel electrophoresis (SDS-PAGE) of commercial human fibrinogen (S) and canine fibrinogen (C) purified by the fibrinogen precipitation assay. M: molecular weight marker. There are 3 bands of identical molecular sizes at 50, 60, and 70 kDa in human and canine fibrinogen. There are a few other bands of undisclosed nature at higher molecular weights, probably co-precipitating proteins.

Analytic validation Imprecision was < 4% (Table 1). Linearity under dilution of 2 pooled canine EDTA-plasma samples showed coefficients of linear regression > 0.999 (Figure 3). Recovery of the samples was between 87.03 and 98.37% (Table 2). The Bland–Altman plot indicated a bias of 0.092 g/L between precipitation and the reference method by Clauss, which was within 95% limits of agreement from 1.653 to 1.837 g/L (Figure 4). A sample with very high fibrinogen concentration that was out of this limit of agreement was considered an outlier after Grubbs’ test analysis (P < .05); in spite of this, the value was not excluded from statistical analysis. Pearson correlation coefficient between both methods was r = .864 (P < .001). The detection limit of the assay was fixed at 0.01 g/L, and the limit of quantification was 0.22 g/L. Effects of urea concentration Although linear regression analysis showed no statistically significant differences in fibrinogen


175


Tecles et al

Effects of anticoagulants No significant differences were observed between results obtained with citrated, EDTA, or heparinized plasma from 6 healthy dogs, while serum from the same 6 animals provided residual precipitation that was significantly (P < .001) lower than that obtained with the different plasma samples (Figure 6).

Clinical validation

Figure 2. Precipitation curve of canine fibrinogen (C) purified by the fibrinogen precipitation assay (●), a canine EDTA plasma pool (△), a canine serum pool (♦), and commercial human fibrinogen (□) at increasing ammonium sulfate ((NH4)2SO4) concentrations.

Table 1. Precision of fibrinogen determinations in g/L in 3 canine plasma pools in 10 subsequent replications of the fibrinogen precipitation assay in the same analytic batch (within-run) or on different days (between-run). Within-Run

Pool 1 Pool 2 Pool 3

Discussion

Between-Run

Mean (SD)

CV%

Mean (SD)

CV%

4.4 (0.04) 3.6 (0.13) 1.8 (0.02)

0.8 3.5 1.3

4.2 (0.14) 3.5 (0.08) 1.7 (0.04)

3.2 2.2 2.6

Data are means (standard deviation) and coefficient of variation (CV).

Figure 3. Linearity under dilution and linear regression analyses of fibrinogen determined by the precipitation assay of 3 pooled canine plasma samples serially diluted with distilled water.

concentrations with urea concentrations from 6.64 to 94.62 mmol/L, there was a tendency to significance (P = .0501, Figure 5).

176

Plasma fibrinogen concentration was significantly increased (P < .05, 1.65-fold) at 48 h after neutering compared to presurgical values. The levels then decreased and reached basal levels at 120 h after surgery (Figure 7). There were statistically significant (P < .001) differences between fibrinogen concentrations in healthy dogs (median 3.01 g/L, 25–75 percentiles 2.60–3.40 g/L) and dogs affected by liver insufficiency, DIC, and protein-losing enteropathy (median 1.36 g/L, 25–75 percentiles 1.04–1.73 g/L) (Figure 8).

A method based on a precipitation reagent containing ammonium sulfate, NaCl, and K2EDTA for fibrinogen quantification was optimized and validated in this study. The precipitation of fibrinogen by ammonium sulfate is based on the kosmotropic (the ability of a substance to increase the stability of intermolecular forces in water–water interactions) effect on watery solutions, decreasing the solubility of proteins.18 The optimal precipitation reagent concentration was determined at 0.57M ammonium sulfate. Under these conditions, the method precipitated > 90% of the fibrinogen present in the sample with almost negligible co-precipitation of other serum proteins, similar to previous reports with human plasma at different ammonium sulfate concentrations.16 A canine standard was purified, as there are no commercially available canine standards. Although purified fibrinogen in the standard would have been more convincingly demonstrated by Western blotting, the results of SDS were considered indicative of the presence of fibrinogen based on molecular weight similarity with a human standard. Although the presence of contaminating proteins was low, further analysis based on proteomics would be desirable to identify the nature of these contaminating proteins, and to determine whether their presence could mask fibrinogen changes in inflammatory or coagulation disorders.


Tecles et al


Table 2. Recovery rate in the fibrinogen precipitation assay measured in a canine EDTA plasma pool after spiking with a known amount of purified canine fibrinogen. Added% of the EDTA Plasma Pool (1.46 g/L) 80% 60% 50% 40% 20%

Added% of the Purified Canine Fibrinogen (6.44 g/L)

Unspiked Sample (g/L)

Spike Amount (g/L)

Spiked Sample (g/L)

Recovery (%)

20% 40% 50% 60% 80%

1.2 0.9 0.7 0.6 0.3

1.3 2.6 3.2 3.9 5.2

2.3 3.2 3.9 4.3 5.3

87% 91% 98% 95% 97%

Unspiked sample: known amount of fibrinogen concentration in the EDTA plasma pool; Spike amount: known amount of fibrinogen concentration added to the plasma pool; Spiked sample: final fibrinogen concentration observed after mixing unspiked sample with spike.

Figure 4. Bland–Altman plots of 33 canine samples analyzed with both the fibrinogen precipitation assay and the reference method by Clauss. X-axis indicates the mean of the values obtained with both methods, and y-axis represents the difference between the two methods. Dotted lines show bias and 95% limits of agreement.

Figure 5. Effect of urea concentration on fibrinogen determination in the precipitation assay. The median (●) is shown with the interquartile range (whiskers). The broken line represents linear regression analysis.

The method was easily automated, fast, cheap, and reproducible. Within- and between-run CVs were < 4%, similar to the ones reported in human fibrino-

Figure 6. Effect of anticoagulants on fibrinogen determination in the precipitation assay. Boxes indicate 25–75 percentiles and horizontal lines indicate median value. Whiskers indicate minimum and maximum values. ***: Fibrinogen concentrations were significantly different P < .001.

gen validation studies.15,16 The method was also accurate and highly correlated with the method by Clauss. As a higher bias was observed in one sample with a very high fibrinogen concentration based on both methods, more studies are needed to establish the reason for the discrepancy in samples with fibrinogen concentrations > 6.5–7.0 g/L. In spite of this, both methods detected high fibrinogen values; consequently, the clinical relevance may not be high for this finding. This is in agreement with the high correlation with the Clauss method determined for an immunoturbidimetric method (Dako Fibrinogen, Dako, Denmark).12 Limit of quantification indicated good reproducibility with values > 0.22 g/L. This was sufficient to detect fibrinogen in plasma from dogs with diseases associated with low fibrinogen concentrations. Dogs affected by afibrinogenemia (a rare condition characterized by coagulative disorders due to the lack of fibrinogen) have been reported to have a negligible fibrinogen concentration.4 Whether the diagnosis of afibrinogenemia would be verified in cases with fibrin-


177


Figure 7. Fibrinogen concentration determined with the precipitation assay in Beagle dogs before and after neutering. Boxes indicate 25–75 percentiles and horizontal lines indicate median value. Whiskers indicate minimum and maximum values. *: Fibrinogen concentrations were significantly different P < .05 compared with concentrations of presurgery and 120h postsurgery samples.

Figure 8. Fibrinogen concentration determined with the precipitation assay in a group of healthy dogs (n = 15) and dogs with fibrinogen depleting diseases (FDD, n = 9). Boxes indicate 25–75 percentiles and horizontal lines indicate median value. Whiskers indicate minimum and maximum values. ***: Fibrinogen concentrations were significantly different P < .001.

Tecles et al

One limitation of the fibrinogen precipitation assay could be the presence of other ions affecting fibrinogen solubility. For example, urea due to its chaotropic activity19 could affect the solubility of fibrinogen. As urea concentration could potentially vary widely between samples, its effect was studied, but no significant differences were observed after precipitating a pool of canine EDTA plasma pool with urea concentrations from 6.64 to 94.62 mmol/L, reflecting levels in healthy dogs and dogs with renal insufficiency. Although a tendency for a slight increase was observed, the change was of small magnitude and probably clinically not relevant. The same effect should be also studied for other ions. Median fibrinogen concentration in 6 healthy dogs in this study was 3.01 g/L, similar to previous reports.20 The fibrinogen precipitation method detected increased fibrinogen concentrations in Beagle dogs after neutering, comparable to what has been described in female dogs after ovariohysterectomy.21 The magnitude of the increase (1.65-fold) is considered a moderate response. One limitation of the study was the lack of samples from dogs with fibrinogen functional disorders, such as dysfibrinogenemia. Clearly, the fibrinogen precipitation assay cannot detect functional changes of fibrinogen as a component of the clotting cascade. In conclusion, the automated fibrinogen precipitation assay described here allows a rapid, cheap, and easy fibrinogen determination that can be used in routine canine practice to assess acute phase reactions as well as diseases that are associated with fibrinogen depletion. Disclosure: The authors have indicated that they have no affiliations or financial involvement with any organization or entity with a financial interest in, or in financial competition with, the subject matter or materials discussed in this article.

References

ogen values below this limit of quantification would need to be determined in a further study. The fibrinogen precipitation method can be used in samples collected in different anticoagulants, such as EDTA, heparin, or sodium citrate. This is an advantage with respect to the Clauss method that is based on clotting of fibrinogen in sodium citrate samples only, and with respect to precipitation with vancomycin in which heparin was found unsuitable.12

178

1. Kudryk B, Okada M, Redman CM, Blomb€ ack B. Biosynthesis of dog fibrinogen. Characterization of nascent fibrinogen in the rough endoplasmic reticulum. Eur J Biochem. 1982;125:673–682. 2. Eckersall PD, Bell R. Acute phase proteins: biomarkers of infection and inflammation in veterinary medicine. Vet J. 2010;185:23–27. 3. Bayramli G, Ulutas B. Acute phase protein response in dogs with experimentally induced gastric mucosal injury. Vet Clin Pathol. 2008;37:312–316.


Tecles et al

4. Wilkerson MJ, Johnson GS, Stockham S, Riley L. Afibrinogenemia and a circulating antibody against fibrinogen in a Bichon Frise dog. Vet Clin Pathol. 2005;34:148–155. 5. Clauss VA. Gerinnungsphysiologische schnellmethode zur bestimmung des fibrinogens. Acta Haematol. 1957;17:237–246. 6. Bovill EG, McDonagh J, Triplett DA, et al. Performance characteristics of fibrinogen assays: results of the college of American pathologists proficiency testing program 1988-1991. Arch Pathol Lab Med. 1993;117:58–66. 7. Jacobsson K. Studies in the determination of fibrinogen in human blood plasma. Scand J Clin Lab Invest. 1955;7:1–54. 8. Ratnoff OD, Menzie C. A new method for the determination of fibrinogen in small samples of plasma. J Lab Clin Med. 1951;37:316–320. 9. Mischke R, Menzel D. Measuring fibrinogen concentrations in healthy dogs: standardization, comparison of methods and reference values. Zentralbl Veterin€ armed A. 1994;41:587–598. 10. Thomas JS. Overview of plasma proteins. In: Feldman BF, Zinkl JG, Jain NC, eds. Schalm’s Veterinary Hematology. 5th ed. Philadelphia, PA: Lippincott Williams and Wilkins; 2000:892. 11. Gonzalez FH, Tecles F, Martınez-Subiela S, Tvarijonaviciute A, Soler L, Cer on JJ. Acute phase protein response in goats. J Vet Diagn Invest. 2008;20: 580–584. 12. Ludwig P, Leidinger EF, Hooijberg EH. Evaluation of the fibrinogen antigenic turbidimetric assay as a screen-


ing method for measurement of fibrinogen concentration in dogs. Vet Clin Pathol. 2012;41:243–248. 13. Parfentjev IA, Johnson ML, Cliffton EE. The determination of plasma fibrinogen by turbidity with ammonium sulphate. Arch Biochem. 1953;46:470–480. 14. Loewy AG, Dunathan K, Kriel R, Wolfinger HL. Fibrinase. I. Purification of substrate and enzyme. J Biol Chem. 1961;236:2625–2633. 15. Stakenburg J, Neumann H. A rapid and simple fibrinogen determination. Clin Chim Acta. 1977;80:141–149. 16. Macart M, Koffi A, Henocque G, Mathieu J, Gullbaud J. Optimized microturbidimetric assay for fibrinogen. Clin Chem. 1989;35:211–214. 17. Tvarijonaviciute A, Martınez-Subiela S, CarrilloS anchez JD, Tecles F, Cer on JJ. Effects of orchidectomy in selective biochemical analytes in Beagle dogs. Reprod Domest Anim. 2011;46:957–963. 18. Collins KD. Ions from the Hofmeister series and osmolytes: effects on proteins in solution and in the crystallization process. Methods. 2004;34:300–311. 19. Moelbert S, Normand B, De Los Rios P. Kosmotropes and chaotropes: modelling preferential exclusion, binding and aggregate stability. Biophys Chem. 2004;112:45–57. 20. Piccione G, Casella S, Giannetto C, Giudice E. Effect of storage conditions on prothrombin time, activated partial thromboplastin time and fibrinogen concentration on canine plasma samples. J Vet Sci. 2010;11:121–124. 21. Sobiech P, Targo nski R, Stopyra A, Zarczy nska K. Changes in the blood coagulation profile after ovariohysterectomy in female dogs. Pol J Vet Sci. 2011;14: 289–290.


179