Detection of the imbalance of procoagulant versus anticoagulant factors in cirrhosis by a simple laboratory method

Authors

  • Armando Tripodi,

    Corresponding author
    1. Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Department of Internal Medicine and Medical Specialties, Università degli Studi di Milano, Milan, Italy
    • Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, University and IRCCS Maggiore Hospital, Via Pace 9, 20122-Milano, Italy
    Search for more papers by this author
    • fax: +39 02 503 20723

  • Massimo Primignani,

    1. Third Division of Gastroenterology, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Cà Granda Ospedale Maggiore Policlinico Foundation and Università degli Studi di Milano, Milan, Italy
    Search for more papers by this author
  • Laura Lemma,

    1. Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Department of Internal Medicine and Medical Specialties, Università degli Studi di Milano, Milan, Italy
    Search for more papers by this author
  • Veena Chantarangkul,

    1. Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Department of Internal Medicine and Medical Specialties, Università degli Studi di Milano, Milan, Italy
    Search for more papers by this author
  • Alessandra Dell'Era,

    1. Third Division of Gastroenterology, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Cà Granda Ospedale Maggiore Policlinico Foundation and Università degli Studi di Milano, Milan, Italy
    Search for more papers by this author
  • Francesca Iannuzzi,

    1. Third Division of Gastroenterology, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Cà Granda Ospedale Maggiore Policlinico Foundation and Università degli Studi di Milano, Milan, Italy
    Search for more papers by this author
  • Alessio Aghemo,

    1. First Division of Gastroenterology, Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Cà Granda Ospedale Maggiore Policlinico Foundation and Università degli Studi di Milano, Milan, Italy
    Search for more papers by this author
  • Pier Mannuccio Mannucci

    1. Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Department of Internal Medicine and Medical Specialties, Università degli Studi di Milano, Milan, Italy
    Search for more papers by this author

  • Potential conflict of interest: A.T., V.C., and P.M.M. received speaker fees on the occasion of scientific educational meetings organized by Instrumentation Laboratory (Orangeburg, NY).

Abstract

Patients with cirrhosis possess an imbalance in procoagulant versus anticoagulant activity due to increased factor VIII and decreased protein C. This imbalance can be detected by thrombin-generation assays performed in the presence/absence of thrombomodulin (predicate assay) that are not readily available in clinical laboratories. We sought to assess this hypercoagulability with a simpler thrombin-generation assay performed in the presence/absence of Protac, a snake venom that activates protein C in a manner similar to thrombomodulin (new assay). We analyzed blood from 105 patients with cirrhosis and 105 healthy subjects (controls). Results for the predicate-assay or the new-assay were expressed as ratio (with:without thrombomodulin) or as Protac-induced coagulation inhibition (PICI%). By definition, high ratios or low PICI% translate into hypercoagulability. The median(range) PICI% was lower in patients (74% [31%-97%]) than controls (93% [72%-99%]; P < 0.001), indicating that patients with cirrhosis are resistant to the action of Protac. This resistance resulted in greater plasma hypercoagulability in patients who were Child class C than those who were A or B. The hypercoagulability of Child C cirrhosis (63% [31%-92%]) was similar to that observed for patients with factor V Leiden (69% [15%-80%]; P = 0.59). The PICI% values were correlated with the levels of protein C (rho = 0.728, P < 0.001) or factor VIII (rho = −0.517, P < 0.001). Finally, the PICI% values were correlated with the predicate assay (rho = −0.580, P < 0.001). Conclusion: The hypercoagulability of plasma from patients with cirrhosis can be detected with the new assay, which compares favorably with the other markers of hypercoagulability (i.e., high factor VIII and low protein C) and with the predicate-assay based on thrombin-generation with/without thrombomodulin. Advantages of the new assay over the predicate assay are easy performance and standardized results. Prospective trials are needed to ascertain whether it is useful to predict thrombosis in patients with cirrhosis. HEPATOLOGY 2010

Chronic liver disease is characterized by impaired synthesis of most coagulation factors and prolonged conventional coagulation tests such as the prothrombin and activated partial thromboplastin time.1 Recently, the long and widely used belief that there is a causal relationship between abnormal coagulation tests and the risk of bleeding has been challenged by showing that under appropriate experimental conditions, liver disease patients generate as much thrombin as healthy subjects provided that platelets numbers are sufficient (>60 × 109/L) to support the normal thrombin generation elicited by plasma.2-4 More recently, it has been shown that patients with cirrhosis display a procoagulant imbalance that may be detected by measuring thrombin generation performed with and without thrombomodulin.5 These observations are in keeping with an earlier observation that patients with chronic liver disease, despite their substantial prolongation of the conventional coagulation times, are not protected from venous thromboembolism (VTE)6 and with those of a recent population-based case-control study showing that patients with chronic liver disease (both cirrhotic and noncirrhotic) have a relative risk of VTE nearly two-fold higher than that of the general population.7 The detection of the procoagulant versus anticoagulant imbalance might have important practical implications in assessing the risk of VTE, especially in patients with cirrhosis awaiting liver transplantation. Cirrhosis is the main cause of portal vein thrombosis (PVT),8 with a prevalence of 1%9 in compensated cirrhosis, but much higher in advanced cirrhosis or in patients awaiting transplantation (from 8%-25%).10 PVT is a multifactorial process, in which local inflammatory foci and systemic prothrombotic factors concur. Its pathogenetic factors are those recognized for a long time as leading to deep vein thrombosis of the lower limbs: damaged vessel wall, slowing of blood flow, and procoagulant versus anticoagulant imbalance. Thus far, the laboratory method available to detect the procoagulant imbalance in cirrhosis is the thrombin generation test5 which requires expertise and equipment that are not readily available in clinical laboratories. This article reports results on a large series of patients with chronic liver disease investigated for their procoagulant imbalance by means of a standardized, easy-to-run, and commercially available method.

Abbreviations:

ETP, endogenous thrombin potential; PVT, portal vein thrombosis; OD, optical density; PICI, Protac-induced-coagulation inhibition; VTE, venous thromboembolism.

Patients and Methods

Patients

A total of 105 adult patients with cirrhosis (81 males, median age [range] 62 years, [39-80 years] and 24 females, median age 63 years [42-83 years]) were enrolled in this cross-sectional study after approval of our Institutional Review Board and informed consent of the patients. Cirrhosis was diagnosed on the basis of clinical, laboratory, and ultrasound evidence. Criteria for exclusion were the use of drugs known to interfere with blood coagulation, ongoing bacterial infections, hepatocellular carcinoma, and extrahepatic malignancy. The severity of the disease was estimated according to Child-Turcotte-Pugh score.11

Control Subjects

One-hundred-five healthy individuals matched with the patient population for age and sex, were enrolled in this study as controls. In addition to the above controls, a population of 37 individuals (4 with and 33 without a previous history of thrombosis) known to be carriers of the factor V Leiden mutation was included for comparative purposes.

Blood Collection and Plasma Preparation

Blood for laboratory analyses was drawn by clean venipuncture and collected in vacuum tubes (Becton Dickinson, Meylan, France) containing 109 mM trisodium citrate as anticoagulant in the proportion of 1/9 parts of anticoagulant/blood. Blood was centrifuged within 30 minutes from collection at 2880g for 15 minutes at room temperature. Platelet-free plasma was then harvested, quick frozen in liquid nitrogen and stored at −70°C until tested.

Measurements

Thrombopath.

This is a commercially available chromogenic assay (HemosIL Thrombopath; Instrumentation Laboratory, Orangeburg, NY) designed to globally evaluate the functionality of the protein C anticoagulant system.12 It is based on the ability of endogenous activated protein C generated after activation of protein C by a snake venom extract (Protac) to reduce the tissue factor–induced thrombin generation. The amount of thrombin is evaluated by recording changes in optical density (OD) at 405 nm in the presence (A) or absence (B) of Protac after adding a thrombin-specific chromogenic substrate. The assay kit contains all the reagents that are needed to run the test. All reagents, except for the diluent are lyophilized and were reconstituted with distilled water before use according to the manufacturer's specification. Briefly, 10 μL of the plasma sample and 40 μL of the Thrombopath diluent were incubated with either the Thrombopath activator A or the Thrombopath activator B (45 μL) for 120 seconds, before the Thrombopath thromboplastin reagent (50 μL) was added. After 90 seconds of incubation, 50 μL of the Thrombopath substrate was added, and change in OD at 405 nm was recorded for 45 seconds. Results are expressed as the Protac-induced coagulation inhibition percentage (PICI%) calculated by the following equation: PICI% = ([B − A]/B) × 100, where B and A are the OD for plasma tested in the absence (B) or presence (A) of Protac. The smaller the PICI% value, the greater the procoagulant imbalance. Testing with Thrombopath was performed by a fully automated coagulation analyzer (ACL 10000; Instrumentation Laboratory, Bedford, MA) according to the manufacturer's specification. To minimize methodological variability, equal numbers of plasmas from patients and controls were included on each test occasion.

Thrombin Generation.

This was assessed as endogenous thrombin potential (ETP) according to Hemker and coworkers13 as described.14 The test is based on the activation of coagulation in platelet-free plasma after addition of human relipidated recombinant tissue factor (Recombiplastin; Instrumentation Laboratory, Orangeburg, NY), which triggers coagulation in the presence of synthetic phospholipids 1,2-dioleoyl-sn-glycero-3-phosphoserine; 1,2-dioleoyl-sn-glycero-3-phosphoetanolamine; and 1,2-dioleoyl-sn-glycero-3-phosphocholine (Avanti Polar Lipids Inc., Alabaster, AL) in the proportion of 20/20/60 (M/M). The concentrations of tissue factor and phospholipids in the test system were 1 pM and 1.0 μM, respectively.

Tests were repeated in a second aliquot of plasma by adding to the test system soluble thrombomodulin (ICN Biomedicals, Aurora, OH) at a final concentration of 4 nM. Continuous registration of the generated thrombin was obtained by means of a fluorogenic synthetic substrate (Z-Gly-Gly-Arg-AMC HCl; Bachem, Switzerland) added to the test system at a final concentration of 617 μM. The procedure was carried out by means of an automated fluorometer (Fluoroskan Ascent; ThermoLabsystem, Helsinki, Finland). Readings from the fluorometer were automatically recorded and calculated by means of dedicated software (Thrombinoscope; Thrombinoscope BV, Maastricht, The Netherlands), which displays thrombin generation curves (time versus generated thrombin) and calculates the area under the curve, defined as ETP and expressed as nanomolar concentration of thrombin times minutes (nM × minute). Thrombin generation is measured as function of an internal calibrator for thrombin (Thrombin Calibrator; Thrombinoscope BV). To minimize methodological variability, equal numbers of plasmas from patients and controls were included on each test occasion. ETP values were used to calculate the ratios between the values obtained with and without thrombomodulin. These ratios reflect the efficiency of thrombomodulin in the activation of protein C and were taken as indexes of hypercoagulability (the greater the ratios, the higher the hypercoagulability).

Other Measurements.

Other parameters to assess procoagulant (factors II and VIII) and anticoagulant factors (antithrombin and protein C) were measured as previously reported with results expressed as percentage of a normal pooled plasma arbitrarily set at 100% of normal.2 The ratio of factor VIII activity to protein C was taken as an index of hypercoagulability. Prothrombin time was measured with recombinant thromboplastin (Recombiplastin; Instrumentation Laboratory) and results were expressed as ratio of patient-to-normal coagulation time.

Data Analyses

Continuous variables were expressed as medians and ranges and tested for statistical significance with the nonparametric Mann-Whitney U and Wilcoxon tests. Correlation between values was assessed by means of the Spearman rho correlation test. P values of 0.05 or less were considered as statistically significant. Statistical analyses were performed with the SPSS software package, version 17.0 (SPSS Inc., Chicago, IL).

Results

The main characteristics of the patient population are reported in Table 1. The distribution of severity of liver disease according to the Child-Pugh classification identified three groups each of 35 patients for Child class A, B, and C. Information on the patient population concerning plasma levels of procoagulant and anticoagulant factors is given in Table 2 and Fig. 1. When compared to controls, patients had significantly reduced levels of antithrombin, protein C, and factor II and increased levels of factor VIII (Table 2). Factor II, antithrombin, and protein C decreased progressively from patients with Child class A to C cirrhosis (Fig. 1). Factor VIII increased progressively from Child class A to C, reaching a median value close to 200% in the latter (Fig. 1).

Figure 1.

Box plots of the distribution of values (median, lower, and upper quartile, and outliers identified as dots) for individual procoagulant and anticoagulant factors for healthy subjects and patients with cirrhosis stratified according to Child class.

Table 1. Demographic Characteristics of the Study Population Separated According to Child-Pugh Classification
CharacteristicsAll (n =105)Child-Pugh A (n = 35)Child-Pugh B (n = 35)Child-Pugh C (n = 35)
  1. Values are expressed as median (range). Abbreviations: HBV, hepatitis B virus; HCV, hepatitis C virus; PT, prothrombin time, (patient-to-normal coagulation time) ratio.

Age (years)62 (39-83)61 (42-77)62 (42-83)59 (39-78)
Etiology    
 HCV (n)63241821
 HBV (n)13823
 Alcoholic (n)2221010
 Other (n)7151
Bilirubin (mg/dL)1.5 (0.4-42.5)0.9 (0.4-2.9)1.5 (0.5-4.2)3 (1-42.5)
PT (ratio)1.31 (0.94-4.29)1.19 (0.97-1.6)1.39 (0.94-3.80)1.49 (1.14-4.29)
Albumin (g/L)3.5 (2-4.7)3.9 (2.6-4.7)3.4 (2.3-4.7)2.8 (2-4.3)
Creatinine (mg/dL)0.8 (0.5-8.3)0.8 (0.5-8.3)0.8 (0.5-2.7)0.9 (0.5-2.8)
Hemoglobin (g/dL)12.3 (7.6-17.6)12.8 (8-16.5)12.9 (9.7-17.6)10.9 (7.6-17.1)
Leukocytes (109/L)4.0 (1.2-9.1)3.8 (1.2-7.2)3.6 (2-9)5.4 (1.9-9.1)
Platelets (109/L)76 (15-251)79 (15-238)70 (27-251)83 (48-141)
Table 2. Hemostatic Parameters for Patients with Cirrhosis and Healthy Subjects
ParametersPatientsControlsP Value
NMedian (range)NMedian (range)
  • *

    Percent activity relative to the reference normal plasma set at 100%. N, numbers of observations.

Factor VIII (%)*89161 (84-378)105110 (50-193)<0.001
Protein C (%)*8945 (12-110)105112 (50-177)<0.001
Antithrombin (%)*8953 (23-107)10595 (60-124)<0.001
Factor II (%)*8945 (18-105)9397 (70-124)<0.001

PICI% Thrombopath

Figure 2 shows PICI% Thrombopath values for patients and controls. The median (range) value for the patient population (74% [31%-97%]) was significantly lower (P < 0.001) than that for controls (93% [72%-99%]) and similar to that for a population of patients with the gain-of-function factor V Leiden mutation, i.e., 69% (15%-80%, P = 0.10) (Fig. 2). Figure 3 shows PICI% Thrombopath for the patient population subdivided according to the Child-Pugh score. Median values decreased progressively from Child A (79% [35%-97%]) to C, with Child C (63% [31%-92%]) displaying slightly lower median value than that for patients with factor V Leiden mutation (69% [15%-80%]), P = 0.59 (Fig. 3).

Figure 2.

Box plots of the distribution of values (median, lower, and upper quartile, and outliers identified as dots) for PICI% Thrombopath for healthy subjects, patients with cirrhosis, and carriers of the factor V Leiden mutation. FVL, factor V Leiden mutation; PICI, Protac-induced coagulation inhibition.

Figure 3.

Box plots of the distribution of values (median, lower, and upper quartile, and outliers identified as dots) for PICI% Thrombopath for healthy subjects, patients with cirrhosis (stratified according to Child class), and carriers of the factor V Leiden mutation. FVL, factor V Leiden mutation; PICI, Protac-induced coagulation inhibition.

Correlation of PICI% Thrombopath Versus Procoagulant or Anticoagulant Factors

The PICI% values were significantly and directly correlated with the levels of protein C (rho = 0.728, P < 0.001) and inversely correlated with the levels of factor VIII (rho = −0.517, P < 0.001). PICI% levels were significantly and inversely correlated with the ratio of factor VIII-to-protein C activity (rho = −0.739, P < 0.001), the latter being taken as an index of the procoagulant activity (Table 3). Finally, the levels of PICI% were significantly and inversely correlated (rho = −0.580, P < 0.001), with thrombin generation assessed as ETP ratio measured with/without thrombomodulin (Table 3). The ETP ratio has been taken as an index of the procoagulant versus the anticoagulant imbalance.

Table 3. Correlation of PICI Percent Thrombopath Versus Pro- and Anticoagulant Factors or Versus the ETP Ratio (With/Without Thrombomodulin)
ParameterNumbers*Rho ValueP Value
  • *

    Numbers of observations refer to patients plus controls. ETP, endogenous thrombin potential; PICI, Protac-induced coagulation inhibition.

Protein C1940.728<0.001
Factor VIII194−0.517<0.001
Ratio of factor VIII to protein C194−0.739<0.001
ETP ratio (with/without thrombomodulin)194−0.580<0.001

Discussion

The balance of coagulation in normal conditions is ensured by the tight control of thrombin generation. This control results from two opposing drivers: the procoagulant and the anticoagulant. Among the procoagulant drivers, factor VIII plays a key role, being responsible together with factor IX and the negatively-charged phospholipids of activated platelets to boost thrombin generation.15 On the other side, protein C, upon activation by thrombin in complex with its endothelial receptor thrombomodulin, acts as a powerful thrombin-quenching protease by inhibiting the activated forms of factor V and VIII.16 Antithrombin, another naturally occurring anticoagulant, inhibits activated factor X and thrombin, thus contributing to the anticoagulant drive. Interestingly, cirrhosis is characterized by increased levels of factor VIII accompanied by decreased levels of protein C and antithrombin.5 Although protein C and antithrombin are reduced because of the impaired synthetic liver capacity, the increased levels of factor VIII are not explained by increased synthesis, but by decreased clearance from the circulation (reviewed in Hollestelle et al.17). High factor VIII levels and reduced levels of protein C and antithrombin are thought to be responsible for a procoagulant versus anticoagulant imbalance,5 thus explaining the increased risk of VTE in patients with liver disease, as recently shown in retrospective6 as well as population-based case-control studies.7

Investigating such an imbalance by laboratory methods would be the necessary step toward undertaking clinical trials aimed at evaluating the thrombotic risk in such patients, especially in those awaiting liver transplantation. Although PVT is no longer considered as an absolute contraindication, its occurrence may preclude liver transplantation. Survival after transplantation is reduced in patients with PVT as compared to those without.18-20 Recently, in a cross-sectional study, we showed that the ETP, a parameter of the thrombin-generation assay, can be useful in this respect if assessed as the ratio of the values measured in plasma with-to-without thrombomodulin.5 Heightened ETP ratios were, in fact, associated with increased levels of factor VIII and decreased levels of protein C and antithrombin.5 However, the thrombin-generation method with/without thrombomodulin requires expertise and equipment that are not readily available in the general clinical laboratory. Recently, an assay meant to explore the anticoagulant protein C pathway was made available.12 It is based on the ability of endogenous activated protein C, generated after activation of protein C by Protac, to reduce the tissue factor–induced thrombin generation. Results for this test are conveniently expressed as PICI% (i.e., Protac-induced coagulation inhibition), and low PICI% values can be taken as an index of hypercoagulability. The new test resembles a thrombin-generation test performed with/without thrombomodulin,5 but it is much simpler because it employs Protac, which is an extract from snake venom commonly used to activate protein C in vitro, instead of thrombomodulin.21 In a multicenter evaluation, the PICI% Thrombopath proved effective in detecting hypercoagulability secondary to congenital as well as acquired defects of the protein C anticoagulant pathway such as protein C, protein S, and the gain-of-function factor V Leiden mutation.12 Interestingly, the method was also found to be sensitive to increased plasma levels of factor VIII.12 We hypothesized that the above characteristics could make it suitable to investigate the procoagulant versus anticoagulant imbalance that occurs in cirrhosis because of the partial deficiency of protein C combined with the relative increase of factor VIII.

Results of this study show that the Thrombopath method is suitable for the following reasons. First, the PICI% levels for patients with cirrhosis are significantly lower than that for controls (Fig. 2). Second, patients classified as Child C had lower PICI% than both controls and patients of the Child A-B class (Fig. 3). According to clinical observations, patients classified as Child C are those who are more susceptible to develop thrombosis.8-10, 18-20 Third, PICI% levels observed for patients with cirrhosis were equivalent to those observed for patients with the factor V Leiden mutation (Fig. 3), a condition associated with an impaired protein C pathway,22 reduced PICI%,12 and an increased risk of VTE.23 Fourth, PICI% were significantly (negatively) correlated with the levels of factor VIII, significantly (positively) correlated with the levels of protein C, and significantly (negatively) correlated with the ratio of factor VIII-to-protein C, which can be taken as an index of the procoagulant versus anticoagulant imbalance (Table 3). Finally, PICI% were significantly (negatively) correlated with the levels of the ETP ratio measured with/without thrombomodulin (Table 3) that is an index of hypercoagulability5 and was taken in this study as the reference procedure to detect the procoagulant versus anticoagulant imbalance. A further advantage of PICI% Thrombopath over the ratio of thrombin generation (with/without thrombomodulin) is the fact that it is standardized in kit form, it is commercially available, and can be easily implemented on a regular coagulometer.

All these features make this method a suitable candidate to be employed in clinical trials to see whether the procoagulant versus anticoagulant imbalance as detected by lower than normal PICI% is a good predictor of peripheral VTE and/or PVT in patients with advanced cirrhosis who are awaiting liver transplantation. However, it should be acknowledged that few patients enrolled in this cross-sectional study had a history of thrombosis (9 of 105) because most of those who experienced recent episodes were being treated with anticoagulant drugs and were, therefore, not eligible for the study. In addition, information on thrombosis in these patients is retrospective, and the events were not objectively documented. Therefore, conclusive evidence on the association between the hypercoagulability as detected by the new assay and the risk of thrombosis requires further studies. These studies should have a prospective design and clinical endpoints. Patients should be recruited, have their PICI% value measured, and then be followed up to ascertain whether they develop objectively documented peripheral VTE and/or PVT. Because of the relatively low event rates and limited follow-up extensions (if the patients are on the waiting list for transplantation), multicenter clinical studies are warranted.

Acknowledgements

Thrombopath reagents were kindly provided by the manufacturer (Instrumentation Laboratory, Orangeburg, NY). A.T., V.C., and P.M.M. received speaker fees on the occasion of scientific educational meetings organized by Instrumentation Laboratory.

Note Added in Proof

While processing this paper the following article addressing the same issue has been published: Lisman T, Bakhtiari K, Pereboom ITA, Hendriks HGD, Meijers JCM, Porte RJ. Normal to increased thrombin generation in patients undergoing liver transplantation despite prolonged conventional coagulation tests. J Hepatol 2010;52:355–61.

Ancillary