See “Thrombin Generation and Microparticles in Inflammatory Bowel Diseases” by Stadnicki on page 343.
The risk of venous thromboembolism (VTE) is increased in patients with inflammatory bowel disease (IBD) compared with non-IBD subjects representing a relevant cause of morbidity and mortality among patients with IBD (1,2). The absolute risk of VTE increases with age. A recent report by Kappelman et al, stratifying patients with IBD by age, showed that younger patients (age 0–20 years) have a higher relative risk for VTE compared with controls than individuals 60 years of age or older (3).
There are several qualitative and quantitative abnormalities in hemostatic parameters observable in patients with IBD. Fibrinogen as an acute-phase reactant is elevated as well as platelets counts, factor V, VII, and VIII (4–6). Greater levels of surrogate markers of thrombin generation such as prothrombin fragment 1 + 2 (F1 + 2) and thrombin antithrombin complex (TAT) have been described (7); however, a decrease in natural anticoagulant factors, such as tissue factor pathway inhibitor and antithrombin, has also been described (8). Additionally, hypofibrinolysis prevails (9–11). In adult patients with IBD, Saibeni et al (12) described an increased endogenous thrombin potential (ETP) for the first time using calibrated automated thrombography (CAT).
Recently, we have shown an increased ETP and higher levels of F1 + 2 and TAT also in pediatric patients with Crohn disease (CD) compared with healthy controls (13). In contrast to F1 + 2 and TAT, which are indicators of thrombin already generated in the circulation, the ETP quantifies the enzyme thrombin activity that can be triggered in plasma in vitro, and allows the detection of the influence of procoagulants and anticoagulants on the formation of thrombin. It may become a useful global test for the estimation of the risk of thrombosis because coagulation tests such as prothrombin time and activated partial thromboplastin time do not represent in vivo balance of coagulation.
Additional mechanisms may contribute to the prothrombotic condition in IBD. The role of cell components called “microparticles” (MPs) in the thrombus development has been recently discussed. MPs, initially called “platelet dust,” are microvesicles that are released from the plasma membrane of many eukaryotic cells (leukocytes, endothelial cells, red cells, and platelets). MPs are generated upon cell activation or apoptosis and bear surface antigens reflecting their cell origins (14–17). In resting cells, phosphatidylserine (PS) is located on the inner membrane layer (18). Upon cell activation, the lipid asymmetry is lost, resulting in exposure of PS on the outer cell membrane (19). The concurrent shedding of MPs from the plasma membrane results in a high content of PS on their surface. Therefore, MPs optimally support the formation of catalytic clotting factor complexes such as the tissue factor/VIIa complex. MPs have been implicated to play a direct and indirect role in the complex coagulation process.
Elevated numbers of MPs have been described in several diseases associated with thromboembolic complications (20–26). Investigations on MPs in IBD are scarce and controversial results are reported. A first report in 2005 showed elevated circulating MP levels in patients with CD or infectious colitis compared with healthy controls and patients with ulcerative colitis (UC) (27). In the same year, Andoh et al (28) described elevated MPs in active UC and CD compared with patients in remission. A correlation between activity index and levels of MPs was observed. Platelet-derived MPs in patients with UC compared with controls showed significantly higher levels in the control group than in the inactive group; no difference was observed between controls and the active UC group (29).
In the present study, first we were looking for a confirmation of our previous results on thrombin generation in pediatric patients with CD and, second, we measured thrombin generation also in pediatric patients with UC using CAT. We investigated procoagulant function of MPs and their influence on thrombin generation during active and quiescent disease states compared with controls. We wanted to clarify whether elevated MPs are a major cause for the changes of thrombin generation observed in pediatric patients with IBD.
In the present study, we compared 53 pediatric patients with IBD (33 with CD and 20 with UC) with 59 healthy controls (31 girls, mean age ± SD: 15 ± 0.3 years). In the group of the 33 patients with CD, 13 patients, already published (13), were measured again for thrombin generation at a different time point. The remaining 20 patients with CD were included for the first time. Our controls were recruited from the general pediatric outpatient clinics and from the nursing school. The children were healthy and had routine coagulation screening before minor elective surgery such as tonsillectomy and adenoidectomy. The patients with IBD were recruited at the Department of Pediatrics, Medical University of Graz, which is a tertiary center, during episodes of active and quiescent disease states. The patients were consecutively enrolled from our outpatient's clinic. UC extension and CD location were classified according to the Paris Classification (30). The disease activity was estimated for CD using the Pediatric Crohn Disease Activity Index (PCDAI) (31). Quiescent disease state was defined by a score <10. For patients with UC, the Pediatric Ulcerative Colitis Activity Index (PUCAI) was used, whereby a PUCAI score <10 differentiated between patients with active versus quiescent disease (32). The clinical scoring was done by an experienced physician. CD and UC were confirmed by clinical, endoscopic, and histopathological findings. No medication that may have interfered with test results (eg, nonsteroidal anti-inflammatory drugs) was administered. None of the patients had any clinical signs of thrombosis. The present study was approved by the local ethics committee.
Blood Collection and Preparation of Platelet-free Plasma
After informed consent, blood was drawn with a 21-gauge needle from the antecubital vein, without applying venostasis, into percitrated S-Monovette premarked tubes (3 mL) from Sarstedt, containing 0.30 mL of 0.106 mol/L trisodium citrate solution. Blood counts were measured before and after preparing platelet-free plasma (PFP), by using a Sysmex KX 21 cell counter (Sysmex Corp, Kobe, Japan). The second blood cell count measurement was performed to exclude platelet contamination of the samples. To prepare PFP, blood was centrifuged at room temperature for 20 minutes at 1550g without brake. The supernatant plasma was carefully removed, aliquotted (1 mL), snap-frozen, and stored at −80°C for further analysis.
Erythrocyte sedimentation rate, hemoglobin level included in the PCDAI, and C-reactive protein (CRP) were measured by standard laboratory methods.
The ZYMUPHEN MP-Activity kit was purchased from HYPHEN BioMed (ZAC Neuville Universite, France). Measurements of TG were performed by use of a thrombin calibrator, platelet-poor plasma (PPP) reagent with a content of 5 pmol/L TF and 4 μmol/L phospholipids in the final reaction mixture, as well as TF-free MP reagent with a phospholipid content of 4 μmol/L phospholipids in the final reaction mixture, which were purchased from Thrombinoscope BV (Maastricht, the Netherlands). The fluorogenic substrate Z-Gly-Gly-Arg-amino-methyl-coumarin, purchased from Bachem (Bubendorf, Switzerland), was solubilized in pure dimethyl sulfoxide purchased from Sigma-Aldrich (St Louis, MO). Fluobuffer consisted of 60 mg/mL bovine serum albumin (BSA) and 20 mmol/L HEPES, which were both obtained from Sigma-Aldrich. Calcium chloride was purchased from Merck (Darmstadt, Germany).
Microparticle Detection by ELISA
The ZYMUPHEN MP-Activity kit is a functional assay to detect the procoagulant activity of MPs in human plasma resulting from their catalytic phospholipid surface. This method is based on the conversion of prothrombin into thrombin in the presence of a factor Xa-Va mixture, calcium, and thrombin inhibitors. There is a direct relation between the phospholipid concentration and the amount of thrombin that is generated during the measurement. The diluted samples are placed into the microplate wells coated with Streptavidin and biotinylated Annexin V. If MPs are present in the sample, then they will be bound by Annexin V. The phospholipid surface of the MPs allows factor Xa-Va to cleave prothrombin into thrombin. The phospholipid concentration is then the limiting factor. PFP was thawed at 37°C and centrifuged for 2 minutes at 1,3000g at room temperature to obtain MPs in the supernatant. The samples were further processed as described by the manufacturer. The MP concentration was expressed as PS equivalents in nanomoles per liter (nmol/L PS Eq).
Determination of Thrombin Generation
For automated measurement of thrombin generation, we used the method that was developed and described by Hemker et al (33). This method is based on the conversion of the fluorogenic thrombin–specific substrate Z-Gly-Gly-Arg-amino-methyl-coumarin.
Procoagulant MPs can have an influence on thrombin generation, which can be assessed using CAT and specific reagents tailored at detecting the effect of MPs (34,35). Measurements of thrombin generation were performed according to manufacturer‘s instructions as previously described (36). We performed 2 thrombin generation measurements for each study subject using 2 different activation reagents. The PPP reagent resulted in a final reaction mixture of 5 pmol/L TF and 4 μmol/L phospholipids, whereas the MP reagent resulted in a final reaction mixture of 4 μmol/L phospholipids and no TF. The absence of TF in the MP reagent makes the assay particularly sensitive to the influence of MPs because they are the only source of TF in the reaction mixture.
After the measurement, the program calculates all parameters of the thrombogram and displays the concentration of thrombin developed in time. Thrombin generation is monitored automatically, on line, in clotting PPP. The resulting thrombogram visualizes the entire process of the overlapping steps of initiation, amplification, propagation, and termination of coagulation. After a period in which no observable thrombin is formed (lag time), the concentration steeply increases (time to peak), rises to a peak (peak height), and then decreases again. The area under the curve of generated thrombin represents the ETP and has been shown to correlate with plasma-based hypercoagulable states and the individual's risk of being affected by thrombosis (see thrombogram in Fig. 1B).
All statistical analyses were performed using PASW Statistics 18.0 software (SPSS Inc, Chicago, IL) and P values <0.05 were considered significant. Data were evaluated for normality of distribution using the Kolmogorov-Smirnov test and for equality of variances using the Levene test. If necessary, data were log transformed to fulfill requirements for parametric testing. Means of all groups (CD active/quiescent, UC active/quiescent, controls) were compared using 1-way analysis of variance, followed by a Bonferroni post-hoc test or Dunnett 2-tailed t test if groups were compared with the control group.
Correlations between disease activity and thrombin generation values and different variables were calculated using Pearson or Spearman rank-order correlation.
The baseline clinical and demographic characteristics of our patients with IBD and controls are seen in Table 1.
Measurement of Microparticles Procoagulant Activity
A significantly increased MP activity in plasma of patients with active CD (median [interquartile range or IQR] 14.40 nmol/L PS Eq [6.70–24.25]; P < 0.05) and inactive CD (median [IQR] 17.65 nmol/L PS Eq [12.00–24.10]; P < 0.001) compared with control subjects (median [IQR] 7.60 nmol/L PS Eq [4.40–19.20]) was shown (Fig. 2). There was no significant difference between active and inactive CD.
In UC, we found a significantly increased MP activity in patients with active disease (median [IQR] 29.80 nmol/L PS Eq [21.70–44.30]; P < 0.001) compared with controls, and in patients with active compared with inactive disease (median [IQR] 13.40 nmol/L PS Eq [8.10–21.10]; P < 0.05) (Fig. 2). There was no significant difference between inactive disease state and controls.
In the IBD patient group, a correlation between CRP levels and MP activity was found (P < 0.05; r = 0.191). There was also a correlation between PUCAI and MP activity (P < 0.001; r = 0.452). No correlation was found between PCDAI and MP activity (P = 0.56; r = 0.69).
ETP was significantly higher in patients with active CD (median [IQR] 1666 nM·min [1499–1732]; P < 0.001) and quiescent CD (median [IQR] 1510 nM·min [1303–1828]; P < 0.05) compared with controls (median [IQR] 1339 nM·min [1175–1570]). There was no significant difference between active and quiescent disease status in CD. In UC there was a significant difference between active disease (median [IQR] 2039 nM·min [1897–2360]) and quiescent disease status (median [IQR] 1512 nM·min [1475–1676]; P < 0.05), and also between active disease as well as quiescent disease and controls (median [IQR] 1339 nM·min [1175–1570]; P < 0.001; P < 0.05) (Fig. 1A and B). There was a significant positive correlation among PCDAI (P < 0.001; r = 0.401), PUCAI (P < 0.05; r = 0.521), and ETP. Also, CRP levels of patients with IBD correlated significantly with ETP (P < 0.001; r = 0.586).
Each sample was tested with PPP reagent, containing a defined concentration of TF, and MP reagent, containing no TF. To determine MPs’ effect on thrombin generation, we calculated ratios between values with MP and PPP reagent for each parameter (ETP, lag time, time to peak, peak height, and start tail). Lower values for the ratios of lag time, time to peak, and start tail point to a higher effect of MPs, whereas lower values of ETP ratios and peak height ratios would suggest lower effect of MPs on thrombin generation.
There was no significant difference in the ratios for ETP, lag time, time to peak, peak height, and start tail in patients with CD in active disease state versus inactive disease state or controls, which supposes no influence of MPs on thrombin generation in patients with CD.
In UC, there was a significant difference in the lag time ratio between patients in active (median [IQR] 2.71 minutes [2.43–2.99]) and inactive disease state (median [IQR] 3.15 minutes [2.86–3.43]; P < 0.05), whereas ETP ratios were not significantly different.
Patients with IBD have an increased risk of thromboembolic complications. The role of MPs in prothrombotic conditions has been discussed (20–22).
In the present study, we found an increased procoagulant function of MPs in pediatric active and quiescent CD and pediatric active UC by using the enzyme-linked immunosorbent assay method. Apparently, in CD, independent from disease status, MP activity was compared with controls. In agreement with the already published literature on adults using also the enzyme-linked immunosorbent assay method for investigating MP activity, we found a correlation with the disease activity, but only in UC (27). In general, in the pediatric IBD patient group, a correlation between CRP levels and MP activity was found. Recently, we have shown that thrombin generation is elevated in pediatric patients with active CD compared with controls (13). These results were reproduced for patients with CD. Even in quiescent disease status in CD, we found an elevated thrombin generation in the present study. In addition, elevated thrombin generation was also shown in pediatric patients with active and quiescent UC compared with controls. A positive correlation was found between PCDAI, PUCAI, CRP, and ETP. Generally, these findings support the clinical observation that the risk of VTE is greater at the time of a flare but still there is a higher risk in remission compared with controls (37). Saibeni et al, who described an increased ETP in adult patients with IBD for the first time, performed the measurements with and without the addition of thrombomodulin to investigate the influence of protein C-activity. Interestingly, increased ETP in patients with IBD compared with controls was only observed after thrombomodulin addition. Therefore, a resistance to thrombomodulin was proposed as a reason for hypercoagulability (12). In our study, we showed a significantly higher ETP in patients with IBD compared with controls without the addition of thrombomodulin. This may argue for additional factors besides thrombomodulin resistance causing the prothrombotic phenotype in IBD.
We were interested in the influence of MPs in increased thrombin generation as a possible cause for hypercoagulability in pediatric IBD. Therefore, we used 2 different types of CAT experiments for each sample. The classic CAT experiment, which involves addition of phospholipids and TF, allows for the assessment of procoagulants and anticoagulants in each individual sample. The second experiment involves just the addition of phospholipids and, therefore, relies on MP derived TF as a limiting MP-sensitive factor; however, the second experiment is also dependent on all the other factors of the first experiment, which differ from sample to sample. Only a comparison of the 2 experiments for each sample blanks out the intraindividual differences in these other factors and leads to a valid evaluation of MP's effect on thrombin generation. This is reflected in the ratio we calculated for each parameter of the thrombogram. An influence of MPs on thrombin generation would be reflected by a higher rate of thrombin generation represented by a shorter lag time, shorter time to peak, and shorter start-tail time; however, only in patients with UC, lag time ratios were shown to be shorter in active versus inactive disease state, which was not observable in the CD patient group. Whether this reflects the difference in the pathophysiology of these 2 diseases, or depends on the procoagulant function of MPs, which was most increased in active UC, remains an open question. In general, MPs did not show a strong influence on thrombin generation. ETP depends on various pro- and anticoagulants, mainly on prothrombin, antithrombin, and tissue factor pathway inhibitor (38–41). To increase ETP, MPs, besides being a strong trigger, would have to add procoagulatory factors to the thrombin generation process, which we could not observe in the present study. These findings suggest that mostly other factors than MPs enhance the overall amount of generated thrombin in IBD.
There is in vitro and epidemiologic evidence that corticosteroids increase the risk of hypercoagulability (42). There are no data available about F1 + 2, TAT, or ETP during corticosteroid treatment in healthy controls. We observed in our patients with CD that, by lowering disease activity index, under therapy with corticosteroids, ETP decreased. There are no data about procoagulant effects of azathioprine. It is extremely difficult to assess whether there is an association between antitumor necrosis factor-α therapy and thrombosis because of multiple confounders in the patient groups (43).
Inflammation probably plays an important role because it is crucially linked with coagulation in patients with IBD. Inflammation and coagulation cross-talk and influence each other (9). The role of inflammation-induced coagulation is underlined by the correlation between elevated ETP, inflammation markers, and disease activity index as we have recently shown in a previous study (13). Again, there was a positive correlation among PCDAI, PUCAI, CRP levels, and elevated ETP in the present study. The identity of the mediators inducing this hypercoagulable phenotype in IBD is unknown. Cytokines are able to enhance the expression of TF on endothelial cells (44). Recently, it has been shown that the enhanced extraintestinal thrombus formation associated with dextran sodium sulfate–induced colitis in wild-type mice can be completely attenuated by immunoblockage of the 2 cytokines tumor necrosis factor-α and interleukin-1β. This observation underlines the effect of these 2 cytokines on the coagulation system (45,46).
Elevated MPs may play a role in inflammation (47). They contain several mediators of inflammation or affect their formation (48). It was shown that platelet-derived MPs may modulate inflammation and adaptive immunity at sites distant from the location of platelet activation (49). Therefore, even if we have not shown a strong effect of MPs on thrombin generation, the role of elevated MPs and their pleiotropic effects on the immune system in the inflammation process in IBD may be an important field for further investigation. Our cohort is not large enough to investigate the influence of disease location and different therapeutic measures that will have to be done in larger studies.
In conclusion, we found an increased procoagulant function of MPs in pediatric patients with IBD as compared with controls, but demonstrated that they are not a major cause for the higher thrombin generation, which is a feature in patients with IBD.
1. Bernstein CN, Blanchard JF, Houston DS, et al. The incidence of deep venous thrombosis and pulmonary embolism among patients with inflammatory bowel disease: a population-based cohort study. Thromb Haemost
2. Miehsler W, Reinisch W, Valic E, et al. Is inflammatory bowel disease an independent and disease specific risk factor for thromboembolism? Gut
3. Kappelman MD, Horvath-Puho E, Sandler RS, et al. Thromboembolic risk among Danish children and adults with inflammatory bowel diseases: a population-based nationwide study. Gut
4. Hudson M, Chitolie A, Hutton RA, et al. Thrombotic vascular risk factors in inflammatory bowel disease. Gut
5. Novacek G, Vogelsang H, Genser D, et al. Changes in blood rheology caused by Crohn's disease. Eur J Gastroenterol Hepatol
6. van Wersch JW, Houben P, Rijken J. Platelet count, platelet function, coagulation activity and fibrinolysis in the acute phase of inflammatory bowel disease. J Clin Chem Clin Biochem
7. Chamouard P, Grunebaum L, Wiesel ML, et al. Prothrombin fragment 1+2 and thrombin-antithrombin III complex as markers of activation of blood coagulation in inflammatory bowel diseases. Eur J Gastroenterol Hepatol
8. Souto JC, Martinez E, Roca M, et al. Prothrombotic state and signs of endothelial lesion in plasma of patients with inflammatory bowel disease. Dig Dis Sci
9. Danese S, Papa A, Saibeni S, et al. Inflammation and coagulation in inflammatory bowel disease: the clot thickens. Am J Gastroenterol
10. Gris JC, Schved JF, Raffanel C, et al. Impaired fibrinolytic capacity in patients with inflammatory bowel disease. Thromb Haemost
11. Saibeni S, Bottasso B, Spina L, et al. Assessment of thrombin-activatable fibrinolysis inhibitor (TAFI) plasma levels in inflammatory bowel diseases. Am J Gastroenterol
12. Saibeni S, Saladino V, Chantarangkul V, et al. Increased thrombin generation in inflammatory bowel diseases. Thromb Res
13. Bernhard H, Deutschmann A, Leschnik B, et al. Thrombin generation in pediatric patients with Crohn's disease. Inflamm Bowel Dis
14. Piccin A, Murphy WG, Smith OP. Circulating microparticles
: pathophysiology and clinical implications. Blood Rev
15. Shet AS. Characterizing blood microparticles
: technical aspects and challenges. Vasc Health Risk Manag
16. Freyssinet JM, Toti F. Formation of procoagulant microparticles
and properties. Thromb Res
2010; 125 (suppl 1):S46–S48.
17. VanWijk MJ, VanBavel E, Sturk A, et al. Microparticles
in cardiovascular diseases. Cardiovasc Res
18. Zwaal RF, Schroit AJ. Pathophysiologic implications of membrane phospholipid asymmetry in blood cells. Blood
19. Bevers EM, Comfurius P, Zwaal RF. Changes in membrane phospholipid distribution during platelet activation. Biochim Biophys Acta
20. Morel O, Hugel B, Jesel L, et al. Sustained elevated amounts of circulating procoagulant membrane microparticles
and soluble GPV after acute myocardial infarction in diabetes mellitus. Thromb Haemost
21. Mallat Z, Benamer H, Hugel B, et al. Elevated levels of shed membrane microparticles
with procoagulant potential in the peripheral circulating blood of patients with acute coronary syndromes. Circulation
22. Nieuwland R, Berckmans RJ, McGregor S, et al. Cellular origin and procoagulant properties of microparticles
in meningococcal sepsis. Blood
23. Meziani F, Tesse A, David E, et al. Shed membrane particles from preeclamptic women generate vascular wall inflammation and blunt vascular contractility. Am J Pathol
24. Sellam J, Proulle V, Jungel A, et al. Increased levels of circulating microparticles
in primary Sjogren's syndrome, systemic lupus erythematosus and rheumatoid arthritis and relation with disease activity. Arthritis Res Ther
25. Van Aalderen MC, Trappenburg MC, Van Schilfgaarde M, et al. Procoagulant myeloblast-derived microparticles
in AML patients: changes in numbers and thrombin generation potential during chemotherapy. J Thromb Haemost
26. Pattanapanyasat K, Gonwong S, Chaichompoo P, et al. Activated platelet-derived microparticles
in thalassaemia. Br J Haematol
27. Chamouard P, Desprez D, Hugel B, et al. Circulating cell-derived microparticles
in Crohn's disease. Dig Dis Sci
28. Andoh A, Tsujikawa T, Hata K, et al. Elevated circulating platelet-derived microparticles
in patients with active inflammatory bowel disease. Am J Gastroenterol
29. Pamuk GE, Vural O, Turgut B, et al. Increased circulating platelet-neutrophil, platelet-monocyte complexes, and platelet activation in patients with ulcerative colitis
: a comparative study. Am J Hematol
30. Levine A, Griffiths A, Markowitz J, et al. Pediatric modification of the Montreal classification for inflammatory bowel disease: the Paris classification. Inflamm Bowel Dis
31. Hyams JS, Ferry GD, Mandel FS, et al. Development and validation of a pediatric Crohn's disease activity index. J Pediatr Gastroenterol Nutr
32. Turner D, Otley AR, Mack D, et al. Development, validation, and evaluation of a pediatric ulcerative colitis
activity index: a prospective multicenter study. Gastroenterology
33. Hemker HC, Giesen P, AlDieri R, et al. The calibrated automated thrombogram (CAT): a universal routine test for hyper- and hypocoagulability. Pathophysiol Haemost Thromb
34. Ollivier V, Wang J, Manly D, et al. Detection of endogenous tissue factor levels in plasma using the calibrated automated thrombogram assay. Thromb Res
35. Sturk-Maquelin KN, Nieuwland R, Romijn FP, et al. Pro- and non-coagulant forms of non-cell-bound tissue factor in vivo. J Thromb Haemost
36. Schweintzger S, Schlagenhauf A, Leschnik B, et al. Microparticles
in newborn cord blood: slight elevation after normal delivery. Thromb Res
37. Grainge MJ, West J, Card TR. Venous thromboembolism
during active disease and remission in inflammatory bowel disease: a cohort study. Lancet
38. Butenas S, van’t Veer C, Mann KG. “Normal” thrombin generation. Blood
39. Cvirn G, Gallistl S, Leschnik B, et al. Low tissue factor pathway inhibitor (TFPI) together with low antithrombin allows sufficient thrombin generation in neonates. J Thromb Haemost
40. Cvirn G, Gallistl S, Rehak T, et al. Elevated thrombin-forming capacity of tissue factor-activated cord compared with adult plasma. J Thromb Haemost
41. Fritsch P, Cvirn G, Cimenti C, et al. Thrombin generation in factor VIII-depleted neonatal plasma: nearly normal because of physiologically low antithrombin and tissue factor pathway inhibitor. J Thromb Haemost
42. Brotman DJ, Girod JP, Posch A, et al. Effects of short-term glucocorticoids on hemostatic factors in healthy volunteers. Thromb Res
43. Bessissow T, Renard M, Hoffman I, et al. Review article: non-malignant haematological complications of anti-tumour necrosis factor alpha therapy. Aliment Pharmacol Ther
44. Szotowski B, Antoniak S, Poller W, et al. Procoagulant soluble tissue factor is released from endothelial cells in response to inflammatory cytokines. Circ Res
45. Yoshida H, Yilmaz CE, Granger DN. Role of tumor necrosis factor-alpha in the extraintestinal thrombosis associated with colonic inflammation. Inflamm Bowel Dis
46. Yoshida H, Russell J, Senchenkova EY, et al. Interleukin-1beta mediates the extra-intestinal thrombosis associated with experimental colitis. Am J Pathol
47. Meziani F, Tesse A, Andriantsitohaina R. Microparticles
are vectors of paradoxical information in vascular cells including the endothelium: role in health and diseases. Pharmacol Rep
48. Ratajczak J, Wysoczynski M, Hayek F, et al. Membrane-derived microvesicles: important and underappreciated mediators of cell-to-cell communication. Leukemia
49. Sprague DL, Elzey BD, Crist SA, et al. Platelet-mediated modulation of adaptive immunity: unique delivery of CD154 signal by platelet-derived membrane vesicles. Blood