Review article: blood platelet number and function in chronic
liver disease and cirrhosis
P. WI TTERS* , , K. FRESON, C. VERSLYPE* , , K. PEERLI NCK, , M. HOYLAERTS, F. NEVENS* , , C. VAN GEET, & D. CASSI MAN* , *Laboratory of Hepatology; Depart- ment of Pediatrics, University Hospital Gasthuisberg; Center for Molecular and Vascular Biology; Department of Hepatology, University Hospital Gasthuisberg; Department of Cardiology, University Hospital Gasthuisberg, University of Leuven, Leuven, Belgium Correspondence to: Dr P. Witters, Afd. Hepatologie, O.& N. 1, Herestraat 49-bus 703, Leuven BE-3000, Belgium. E-mail: Peter.witters@gmail.com Publication data Submitted 23 January 2008 First decision 18 February 2008 Resubmitted 28 February 2008 Accepted 29 February 2008 Epub OnlineAccepted 5 March 2008 SUMMARY Background The liver plays a central role in coagulation and brinolysis but is also closely intertwined with the function and number of blood platelets. Aim To describe and integrate all literature concerning blood platelets and liver disease by performing a thorough literature research. Methods A thorough literature research on blood platelets and liver disease was performed. Results Thrombocytopenia is a marked feature of chronic liver disease and cir- rhosis. Traditionally, this thrombocytopenia was attributed to passive platelet sequestration in the spleen. More recent insights suggest an increased platelet breakdown and to a lesser extent decreased platelet production plays a more important role. Besides the reduction in num- ber, other studies suggest functional platelet defects. This platelet dys- function is probably both intrinsic to the platelets and secondary to soluble plasma factors. It reects not only a decrease in aggregability, but also an activation of the intrinsic inhibitory pathways. The net effect, nally, is a decreased platelet function in the various types of chronic liver diseases and cirrhosis. Finally, recent data suggest that platelets are not only affected by but can also contribute to the liver disease process, as for instance, in viral hepatitis and cholestatic liver disease. Conclusion Platelet research in liver disease is a growing area of investigation and could provide new pathophysiological insights. Aliment Pharmacol Ther 27, 10171029 Alimentary Pharmacology & Therapeutics 2008 The Authors 1017 Journal compilation 2008 Blackwell Publishing Ltd doi:10.1111/j.1365-2036.2008.03674.x I NTRODUCTI ON Haemostasis and hepatology are closely related. It is well-known that patients with cirrhosis show a marked decrease in liver synthesis of coagulation factors, which leads to a prolongation of the prothrombin time. 1 Less well studied, but equally important in the physiological process of clot formation, are the defects in the primary haemostasis in patients with liver dis- ease. Blood platelets initiate the haemostatic process by interacting with the damaged vessel wall. 2 These form a haemostatic plug within seconds after injury. The secondary haemostasis (coagulation) starts simul- taneously, but enrolls slower and strengthens this plug by crosslinked-brin within minutes. 2, 3 Cirrhosis is the nal common end-point of many chronic liver diseases and thrombocytopenia is a well- known feature of cirrhosis. However, in addition to the quantitative changes in platelets, there are qualitative defects. These can be assessed in specialized haemo- static laboratories. The most primitive but physiologi- cally meaningful platelet function test is the in vivo bleeding time assessment. 4 More sensitive and repro- ducible tests have been developed, for instance, assays that try to simulate the in vivo platelet function in vitro, under ow conditions, to study shear-induced platelet activation. These tests are technically demanding, but reect aggregation in maximally physiological condi- tions. 5 Less physiological, but more widely used and currently the gold standard for platelet function assessment is the aggregometry. 4, 6 This in vitro assay measures platelet aggregation in response to different sets of agonists [collagen, thrombin, adenosine diphos- phate (ADP), etc.]. Aggregometry is performed with platelet rich plasma (obtained by slight centrifugation of anticoagulated blood) where optical light turbidity is a measure of aggregation. 7 It can also be performed with whole blood by electrical impedance measure- ment. 4 Another sensitive assay is the quantication of membrane molecule expression by ow cytometry [uorescent-activated cell sorting (FACS) analysis]. By assessing the presence of stimulation-dependent anti- gens [e.g. CD62P (P-selectin)] or platelet-leucocyte complexes, the degree of platelet activation can be estimated. 4, 7 Even more specialized, on a molecular level, the concentrations of second messengers [e.g. calcium, cyclic adenosine monophosphate (cAMP)] and the release of platelet granules containing pro- aggregatory molecules [serotonin, adenosine triphos- phate (ATP), platelet factor 4, beta-thromboglobulin, etc.] can be assessed. 4 Finally, because of the technical difculties with platelet function tests, several tests (as are thromboelastography and sonoclot) have been developed for everyday clinical use. These tests require little or no sample preparation and give instant bed- side results to assist in clinical decision making. 7 In thromboelastography, for example, blood clots in a backward and forward rotating sample cup around a suspended pin. Because of haemostasis, the cups motions become limited. These movements are mapped and different variables (depending on coagulation, brinolysis and platelet function) can be calculated. 6, 7 Although spontaneous bleeding caused by low platelet counts in cirrhosis is an infrequent event, low platelet counts become clinically relevant when per- forming liver biopsies, liver transplantation or giving myelosuppressive agents (antiviral treatment like inter- feron or cytostatics). 8 Less well studied is the recent observation that platelets can contribute to the pro- gression of liver disease (e.g. in viral hepatitis and cholestatic liver disease). 9, 10 Finally, platelet derived serotonin has convincingly proven to be involved in the initiation of liver regeneration. 11 The aim of this review was to give a comprehensive overview of the literature covering both quantitative (counts) and qualitative (function) aspects of blood platelets in cirrhosis and chronic liver disease. Most of this review handles cirrhotic liver disease on the whole, as the literature contains few data on the inu- ence of different aetiology or degree of liver brosis on platelet (dys)function. Nevertheless, when available, data are discussed with regard to the different types of cirrhosis separately. We will mainly distinguish between viral liver disease (hepatitis B or C), alcoholic liver disease, biliary cirrhosis [primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC)] and vascular liver disease (e.g. extrahepatic portal vein obstruction, non- cirrhotic portal hypertension, Budd-Chiari syndrome). PLATELET NUMBER I N CI RRHOSI S AND CHRONI C LI VER DI SEASE Theoretically, lowered platelet counts can be the result of decreased platelet production, enhanced splenic sequestration or platelet consumption use. Kinetic studies with radiolabelled platelets in cirrhosis and chronic liver disease indicate that there is a decreased platelet survival. 1215 Depending on the ability of the bone marrow to increase platelet production, platelet counts can be reduced 12 or normal 14 depending on the 1018 P. WI TTERS et al. 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd ability of the bone marrow to increase the platelet pro- duction. These ndings are in accordance with the decreased, 16, 17 normal 18 or increased 19 percentage of reticulated platelets (young platelets) in cirrhosis. The main site of platelet consumption is in the spleen. 12, 13, 15 Here, we review the evidence for the three different mechanisms leading to thrombocytopenia. Platelet production Thrombopoetin (TPO) is the most important growth factor in the regulation of megakaryocyte development and platelet production. However, its serum value is not a reliable indicator of platelet production because upon binding to its receptor on thrombocytes and megakaryocytes, it is internalized and cleared from the circulation. 8, 20 As TPO is mainly produced in the liver and kidney at a constant rate, this clearance from the circulation by platelets is the most important mecha- nism regulating the concentration in the physiological state. 21 So, serum TPO (net effect from production and clearance) should always be interpreted with total platelet count, including immeasurable intrasplenic platelets, 22 to estimate platelet production. Even 13 years after its original description, the function of TPO in liver disease is not clear. As the liver is the major site of TPO production 20, 21 it is reasonable to expect a decreased plasma level and indeed TPO mRNA levels in the liver were slightly decreased in cirrhosis. 23 However, normal, 18, 19, 24, 25 decreased 16, 17, 26, 27 and increased 28 serum TPO levels have also been published. These conicting results can be explained by differences in laboratory techniques: the use of different antibodies (more sensitive poly- clonal vs. monoclonal antibodies) and the lack of an international standard preparation of TPO. 20 TPO levels have been related to Child-Pugh stage and albumin levels in some studies, 26 but not in others, 19, 22 and to portal haemodynamics: a decreased TPO level was found in patients with hepatofugal portal blood ow. 25 Liver transplantation is known to resolve thrombocy- topenia sequentially to an increase in serum TPO and increased platelet production. 18, 22, 23 Other thrombo- poietic cytokines (interleukin 3, interleukin 6 and interleukin 11) do not play a major role, 27 although therapy with recombinant human interleukin 11 increases platelet counts in cirrhotic patients. 29 Beside a decreased TPO production, there is also an increased TPO breakdown in cirrhosis. Partial splenic embolization or splenectomy results in increased TPO and platelet count (presumably through diminished platelet mediated TPO breakdown) and restores the physiological relation between platelet count and TPO. 22, 23, 30 The exact role of impairment of thrombopoiesis in liver disease remains to be determined. Bone marrow megakaryocyte density in cirrhosis was similar to the density in idiopathic thrombocytopenic purpura (ITP) and markedly elevated compared to aplastic anemia, suggesting that a primary production decit is not the only cause of thrombocytopenia. 19 Platelet sequestration (distribution) As splenomegaly is an important feature of portal hypertension, it is not surprising that splenic platelet sequestration was thought to be an important mecha- nism in the aetiology of thrombocytopenia. This the- ory was rst suggested in 1965 by Aster. 31 Although this theory was never proven, it was generally accepted because of lack of other explanations. 32 Kinetic radiolabelled platelet studies do suggest splenic pooling, but they also report shorter platelet survival time, indicative of splenic destruction rather than mere splenic pooling. 1214 The term congestive splenomegaly is a mechanical oversimplication and there is no linear correlation between spleen size and portal pressure. 33 Further- more, decompression of the portal venous system by transjugular portosystemic shunts could not consis- tently demonstrate a benet in terms of platelet count (no benet; 34, 35 some benet 3638 ). In the studies that demonstrated a benet, there was no normalization 36 38 and no correlation with achieved portal pressure or percentage of pressure reduction. 36, 37 Most reports also conclude that portocaval and distal splenorenal shunts do not improve platelet counts in this set- ting. 15, 39, 40 This argues against simple splenic pooling caused by increased portal pressure but does not exclude the spleen from contributing to platelet break- down. As already stated, partial splenic embolization can increase platelet count. 22, 23, 41 Platelet breakdown Kinetic studies indicate that platelet breakdown, illus- trated by decreased platelet survival time, is also an important mechanism. 1215 There is a remarkable homology between ITP and thrombocytopenia in REVI EW: PLATELETS I N CHRONI C LI VER DI SEASE 1019 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd cirrhosis. The reticulated platelet proportion (propor- tion of young platelets) and glycocalicin index (a mar- ker of platelet production) were signicantly higher in both diseases compared with healthy controls. 19, 42 As in ITP, it has been proposed that autoantibody- mediated platelet destruction plays a major role. In chronic liver disease, these platelet-associated IgGs are markedly elevated 24, 4346 as well as the number of B cells producing antibodies against the major autoanti- gen of platelets, GPIIb IIIa. 43 The antibody titre corre- lates inversely with the number of platelets 24, 43, 44 and with spleen volume. 12, 24 Furthermore, splenic emboli- zation increased platelet survival time whereas the platelet-associated IgG decreased. 45 However, it is widely recognized that the specicity of platelet associ- ated IgGs for autoantibody-mediated thrombocytope- nia is quite low, so it is considered an inappropriate test for the diagnosis of ITP. 43 Although not undisputedly proven, it seems an attractive hypothesis. Moreover as in ITP, the antibodies could not only mediate FCc-med- iated clearance in the reticuloendothelial system, but also to some extent suppress megakaryopoiesis. 19 Others found an increased brinogen and plasmino- gen turnover suggesting platelet consumption. 14 Also, increased brinogen degradation products and to a lesser extent, increased D-dimers have been reported, however, without a clear relation to platelet numbers, brinogen levels or prothrombin activity. 47 Finally, soluble P-selectin levels, a marker of in vivo platelet activation, are increased in chronic liver disease 48 (cfr. infra: platelet exhaustion). Overall, it is not yet clear what the exact mecha- nisms leading to thrombocytopenia in cirrhosis are. It is most likely a multifactorial process combining increased splenic platelet breakdown, splenic pooling and the inability of the bone marrow to increase plate- let production adequately. Platelet count in different types of cirrhosis and brotic liver disease Most authors agree on a decrease in platelet count in relation to the severity of cirrhosis. 4951 There are only few reports that relate the platelet number to the dif- ferent aetiologies of cirrhosis. However, in PBC, an immune-mediated disorder, platelet-associated IgGs seem to play an important role in the thrombocytopenia, with recovery after cortico- steroid administration. 52 In one patient with PBC, the antibodies eluted from the platelets bound to GPII- b IIIa and a 70 kDa mitochondrial antigen. Analysis of the peptide sequences showed partial amino acid sequence homology suggesting the possibility of a common antibody binding site. 53 Most likely the thrombocytopenia in viral hepatitis also represents a multifactorial process with altered Figure 1. Platelets in normal (a) and cirrhotic (b) condition. (a) In normal platelets, stimulation by collagen, thrombin, ADP and TXA B2 leads to activation of the PLC (in green, left side of the gure). This results in the formation of DAG and IP3. DAG leads to activation of PLA2 and the production and release of TXA B2. This reinforces the PLC activity and stimulates additional platelets. IP3 results in calcium release from the dense tubular system (stored through the SERCA pump). The increase in cytosolic calcium (also aided by the ADP P2X receptor and the H Ca-Na-antiporter) moderates the platelet acti- vation: the shape change (by actin); the release of stored pro-aggregatory molecules (ATP and 5HT from the dense bodies and BTG, PF4 and P selectin from the alfa granules) recruiting other platelets and the conformational change of the GPII- b IIIa receptor (together with the phosphorylation by the PKC) allowing interactions with brin reinforcing the haemostatic plug. These mechanisms are under tight homeostatic control of the inhibitory pathways (in red, right side of the gure): cAMP (produced by AC: stimulated by adenosine, PGI2 and PGE2 and inhibited by ADP via the P2Y2 receptor) and cGMP (produced by the NO-stimulated GC). (b) In cirrhosis, platelet activation is decreased (in green, left side of the gure). There is decreased PLC and PLA2 activity leading to decreased TXA B2 production and decreased IP3 mediated cytosolic calcium increase. There is also less calcium transport by the H Ca-Na-antiporter because of diminished intraplatelet acidication. The secondarily impaired release of the dense bodies and the alpha granules is less effective because of the storage pool defect: less ATP and 5HT; BTG, PF4 and P-selectin-levels in the granules respectively. The inhibitory pathways are also up- regulated (in red, right side of the gure). This results in more cAMP- and cGMP-mediated inhibition. Finally, there are also negative plasma factors decreasing platelet activation as are the FDP and ApoE (See text for details). 5HT, serotonin; AC, adenylate cyclase; ApoE, Apolipoprotein E; ATP, adenosine-triphosphate; BTG, b-thromboglobulin; Ca, calcium; cAMP, cyclic adenosine monophosphate; cGMP, cyclic guanosine monophosphate; DAG, diacylglycerol; FDP, brin(ogen) degrada- tion products; GPIIb IIIa, glycoprotein IIb IIIa; GTP, Guanosine triphosphate; H, hydrogen; IP3, inositol-triphosphate; Na, sodium; NO, nitric oxide; P2Y1, P2X1, P2TY2, ADP receptors; PC, phosphatidylcholine; PF4, platelet factor 4; PGE2, prosta- glandin E2; PGI2, prostacyclin; PIP2, phosphatidylinositol-bisphosphate; PKC, protein kinase C; PLA2, phospholipase A2; PLC, phospholipase C; SERCA, sarco endoplasmic reticulum Ca-ATPase; TXA2, thromboxane A2; TXB2, thomboxane B2. 1020 P. WI TTERS et al. 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd production of TPO with increasing hepatic brosis 54 and splenomegaly. As in other chronic liver disease, in viral hepatitis, TPO increase has been documented (compared to other causes of chronic liver disease 51 and to controls 55 ). However, there is no (inverse) cor- relation between TPO level and platelet count 55 and RT-PCR and immunohistochemistry showed equal hepatic expression as that of other liver diseases. 55 There is an inverse correlation between platelet count and spleen size, suggesting platelet breakdown or sequestration. 54, 55 However, other studies suggest that platelet autoantibodies are common in individuals with HCV infection and that their detection does not assist in the diagnosis of immune thrombocytopenia. 50 Thrombocytopenia can also arise independent of hypersplenism. 56 So, the exact mechanisms are as yet not clear. PLATELET FUNCTI ON Platelet function in cirrhosis The existence of a functional platelet defect (hypoag- gregability) was described for the rst time by Thomas et al. 57 Bleeding time, the paradigmal platelet function test has been reported to be abnormal in 2.542% of patients with cirrhosis. 5860 This correlates only weakly with the platelet count, suggestive of an additional functional decit. This bleeding time prolongation is closely related to the degree of liver failure. 60 (a) (b) REVI EW: PLATELETS I N CHRONI C LI VER DI SEASE 1021 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd Only few studies on platelets in ow conditions (the most physiologic test in vitro) have been per- formed. 5, 6164 These report conicting results (decreased 5, 64 and normal 6163 platelet function) pos- sibly because of differences in experimental setup as varying shear rates from 600 s, 64 800 s, 5, 63 1600 s 62 to 2600 s 61 and differences in the adhesive surfaces that were perfused. Moreover, because these studies were performed in whole blood or reconstituted blood (with adjusted hematocrit and platelet count), factors other than platelet function could have played an important role. Indeed, plasma factors and haemorhe- ological factors dependent on the hematocrit are important in these experiments. 62 In humans and in animal models of chronic liver dis- ease and cirrhosis, there is a clear hypoaggregability, as demonstrated with in vitro aggregation tests (the gold standard) 5, 61, 6574 and with owcytometric analysis of stimulation dependent antigens (e.g. P-selectin). 70 A decreased response to collagen, 61, 6567, 71, 73, 75 throm- bin, 6669 arachidonic acid, 5, 61, 67, 70, 73 ADP, 5, 61, 65, 70, 72 U46619 (an ADP-mimetic), 5, 61, 75 epineph- rine 5, 61, 70 and ristocetin 5, 61, 76 have been reported. These ndings are also conrmed by thromboelastog- raphy. 77, 78 Crossover experiments (using platelet aggregometry) with patient or control platelets in control or patient plasma hold both an intrinsic plate- let defect and a circulating plasma factor responsible for this hypofunction. 72, 74 The molecular mechanisms underlying this intrin- sic platelet defect leading to hypoaggregability have been studied extensively. The most important defects are depicted in Figure 1: the normal platelet (Figure 1a) as compared to the cirrhotic platelet (Figure 1b). There is evidence supporting a reduced transmem- brane signalling in cirrhotic platelets after stimulation with thrombin or collagen. 66, 67 This leads to a decreased activation of phopholipase C, A2 and cyclo- oxygenase thromboxane synthetase 66, 67 resulting in decreased thromboxane production. 79 There is also evidence for a decreased arachidonic acid availability for prostaglandin and thromboxane production. 71, 79 Moreover, dietary supplementation with arachidonic acid increases the membrane levels of this fatty acid and improves (collagen-induced) platelet aggregation, returning to pre-treatment levels 4 weeks after with- drawal. 71 The IP3 production in response to thrombin stimula- tion is signicantly lower 69 in platelets from patients with cirrhosis. There is a reduced cytosolic calcium increase, also because of a decreased release from intracellular stores. 69 As intracellular calcium is the nal common pathway in platelet aggregation, this leads to platelet hypofunction. There is also a decreased cytosolic alkalinization (Na + H + antiporter activity), which normally facilitates calcium entry from the extracellular environment. 69 Furthermore, there is evidence for a storage pool defect in cirrhosis, decreasing the effect of granule release on the aggregation. There is decreased ATP and serotonin (5HT) in the dense bodies and decreased PF4 (platelet factor 4), b-thromboglobulin (BTG) and P-selectin in the alpha granules. 65, 68, 70, 80 These gran- ules have a normal ultrastructure and normal relative volume. 68 Plasma levels of BTG and PF4 are reported to be elevated in relation to platelet count, 68 although not all authors agree. 80 Finally, there is an upregulation of inhibitory path- ways. Basal cyclic adenosine monophosphate and cyclic guanosine monophosphate, the two main inhibitory messengers, are upregulated in the cirrhotic platelet. 69 There was a normal upregulation upon stimulation indicating an increased stimulation of circulating factors in vivo like PGI2 8183 and nitric oxide (NO) 84 for cAMP and cGMP, respectively. 69 There is one study suggesting a locoregional difference in platelet aggre- gation, i.e. a difference in platelet aggregation between portal and, simultaneously drawn, systemic blood. Splanchnic vasodilatation and enhanced generation of PGI2 shift the collagen (but not the ADP) aggregation curve signicantly to the right in portal blood. There were no differences in coagulation and brinolytic parameters. 82 As possible underlying mechanism platelet exhaus- tion was proposed, as platelets are faced with the por- tal hyperdynamic circulation. Platelets could be damaged during intravascular activation (with loss of granules), which could give rise to their subsequent hypo-function when tested in vitro. 65 This hypothesis is somewhat challenged by the normal microscopy of the platelets, 68 the normal concentration of thrombin- antithrombin complexes, D-dimers and F1+2 [brin (ogen) degradation products] values and the absence of stimulation-dependent antigens on the platelet membrane. 70 The increased urinary excretion of thromboxane A2 (TXA2) metabolites (2,3 dinor-throm- boxane B2 (TXB2) and 11dehydro-TXB2) could be explained by intrasplenic destruction. 70 Currently, defective transmembrane signalling (and secondarily a decrease of the intracellular messengers) 1022 P. WI TTERS et al. 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd is thought to be the most important factor in the platelet hypo-function. 67, 70 Besides this intrinsic platelet defect numerous plasma factors appear to play a role. Numerous nega- tive factors are known. High-density lipoprotein (HDL) apolipoprotein E contents are increased in cirrhosis and this was corre- lated with the inhibition of platelet aggregation. 85 There are also brin(ogen) degradation products, which appear in excess in liver disease, 47 are known to adsorb to the platelet surface and interfere with platelet function. 86 However, their levels may not be well correlated with reduced platelet aggregation. 87 Bile salts can also exert a negative effect. In vitro aggregation studies show an inhibitory effect of bile salts on platelet aggregation and serotonin release induced by ADP or collagen. 88 Only chenodeoxycholic acid is an aggregation-inducing bile salt. 88 In vivo studies in rat models associated with elevated bile acids (bile duct ligation and cholic acid feeding) show a clear inhibitory effect on ADP-epinephrine induced aggregation. 89 Finally, as already stated, primary and secondary haemostasis are closely related. As defects in the coag- ulation are well-studied (for review see 2, 90 ) it is clear that they independently contribute to the haemostatic abnormalities in patients with liver disease. However, their impairment may also affect platelet function. For example, platelets play an important role in thrombin formation and thrombin is a platelet agonist. 1 What effect activated recombinant factor VII (rFVIIa; Novo- seven, Novo Nordisk A S, Copenhagen, Denmark) has on platelets or on thrombin formation remains to be demonstrated. 1 There are also positive factors that promote platelet activation e.g. vWF that is reported to be upregulated in cirrhotic patients. 62 Despite the reduced number of (more active) high molecular weight vWF multi- mers, 42, 62 the reduced collagen binding capacity and relative ristocetin activity (all indicative of loss of function), it is likely that the quantitative effect exceeds the qualitative effect. 62 Glycoprotein Ib (the vWF receptor) has been reported to be increased 47 or decreased 76 in cirrhosis. As already stated, these plasma factors can contribute to the apparent discrep- ancy between the studies under ow conditions and the in vitro aggregation tests. Unconjugated bilirubin is a strong inducer of plate- let aggregation similar to ADP in isolated platelets. This could be completely inhibited by prostaglandin I 2 or E 2 . 91 However, the physiological relevance of this observation has to be questioned as the effect was entirely inhibited by the addition of 0.1% bovine serum albumin, probably because of the high afnity of albumin to unconjugated bilirubin. 91 Tenfold concentrated ascitic uid caused irreversible platelet aggregation and serotonin release of normal platelet-rich plasma similar to collagen and could be abolished by adding collagenase. 92 The coagulopathy after peritoneovenous shunting could be the result of direct and rapid intravenous infusion of procoagulant collagen. 92 Overall, there is a clear platelet dysfunction. As for the intrinsic platelet defect, several steps of platelet activation are shown to be impaired. Underlying mech- anisms remain speculative. Furthermore, the blood platelets from patients with liver disease encounter several agonists and antagonists. This could lead to hyper- and hypoaggregability; it is, however, not clear what the contribution of these isolated ndings is in the complex entity of chronic liver disease and cirrhosis. Platelet function in different types of liver disease As with platelet number, also platelet function seems dependent on the degree of liver brosis. 5, 67, 68 Until now, there has been little attention on the dif- ferent liver diseases in relation to platelet abnormalities. Most reports are on cirrhotic platelets. Here, we review the reports on the different liver disease entities. In cholestatic liver disease, there is some evidence that contrary to other types of cirrhosis, the platelets demonstrate a hyperaggregability. Patients with PBC PSC have a better survival after variceal bleed- ing 93 and demonstrate less blood loss during liver transplantation, 94, 95 although this can also be explained by a better synthetic liver function at the time of transplantation. Thrombosis of portal veins has also been detected in 40% of PBC liver at the time of transplantation. 96 There have been some platelet func- tion studies in these patients (see Table 1) that are indicative of hyperaggregability. This has been docu- mented by the use of thromboelastography, 97, 98 by sonoclot analysis 99 (two newer and less well studied platelet tests that try to evaluate the complete haemo- static system) and PFA-100 analysis. 98, 99 In the latter, there is a lack of prolongation of closure time 98, 99 as is normally seen in cirrhosis. 61, 98 On a molecular level, there is an increased expression of CD42b (an REVI EW: PLATELETS I N CHRONI C LI VER DI SEASE 1023 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd Table 1. Platelet function in different liver diseases References Disease (n) Authors ndings conclusions Functional platelet studies in cholestatic liver disease Ingelberg et al. 65 PBC (10), Alcoholic cirrhosis (10) In PBC and alcoholic cirrhosis, there is a decreased aggregation to collagen and ADP In PBC, there is a smaller release of ATP and total releasable ATP is diminished Platelets are probably damaged during an intravascular activation (loss of granules) Ben-Aril et al. 97 PBC (47), PSC (21), non-cholestatic cirrhosis (40), healthy controls(40) Hypercoagulability on thromboelastography in 28% (PBC), 43% (PSC), 5% (non-cholestatic cirrhosis), 0% (healthy) No correlation with platelet count or brinogen level Not explained by protein C, S or ATIII Pihush et al. 98 PBC PSC (37), Hepatis C alcoholic cirrhosis (53), healthy controls (62) Hypercoagulability on thromboelastography in PBC PSC associated with higher brinogen levels Normal PFA-100 closure time in contrast to prolonged closure time in hepatitis C alcoholic cirrhosis Increased expression of LIBS-1 and CD42b in PBC PSC Biagini et al. 99 PBC (51), healthy controls (102) Sonoclot rate signicantly higher compared with controls Sonoclot rate associated with increased TAT, TF and correlates with hyperhomocysteinemia (presumably due to vitamin deciencies and homozygous TT677 MTHFR genotype) Functional platelet studies in alcoholic liver disease Torres Duart et al. 102 Healthy control (9) Decreased aggregation in whole blood (to collagen) and in platelet rich plasma (to ADP, arachidonate and collagen) by the addition of alcohol Barrison et al. 103 Alcoholic cirrhosis (14), healthy controls Impaired platelet aggregation in 12 14 cirrhotics Higher malondialdehyde production correlated to the higher cholesterol:triglyceride ratio in cirrhotics Hillbom et al. 104 Alcoholics (26, 13 with fatty liver), healthy (13) Decreased platelet aggregation in response to ADP Prolonged bleeding time in alcoholics with fatty liver disease Hilbom et al. 105 Active alcoholics (14, 7 with fatty liver) On admission: prolonged bleeding time, decreased platelet count, aggregability to arachidonate and decreased TXB2 formation 914 days after ethanol withdrawal: normalization of bleeding time, increased TXB2 formation compared to controls after stimulation with ADP Watanabe et al. 106 Alcoholic liver disease Decreased aggregation in response to ADP, normal to collagen possibly because of changes in fatty acid compositions of the platelet membrane Ouwendijk et al. 107 Alcoholic cirrhosis (16) In 12 patients frequently increased serum TXB2 levels Functional platelet studies in viral liver disease Panasiuk et al. 49 Chronic hepatitis (29), viral cirrhosis (27), healthy controls (32) Higher serum concentrations of PF4 (sevenfold) and BTG (twofold) Higher P-selectin expression on resting platelets and normal (viral hepatitis) or higher (viral cirrhosis) after stimulation with thrombin compared with healthy controls Ferroni et al. 108 Chronic hepatitis C (39), healthy controls Higher soluble P-selectin in hepatitis C with higher levels in patients with lower platelet counts Correlation of soluble P-selectin with serum hepatitis C-RNA and lack of correlation with platelet associated immunoglobulins ADP, adenosine diphosphate; ATP, adenosine triphosphate; BTG, betathromboglobulin; LIBS-1, ligand-induced binding site; PBC, primary biliary cirrhosis; PF4, platelet factor 4; PSC, primary sclerosing cholangitis; TAT, thrombin-antithrombin com- plexes; TF, tissue factor; TXB2, thromboxane B2; MTHFR, 5,10-methylenetetrahydrofolate reductase. See text for more details. 1024 P. WI TTERS et al. 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd antigen of the gpIb V IX complex which binds to vWF) and of ligand-induced binding sites both in basal conditions and after stimulation with ADP or epinephrine. 98 This suggests that platelets are preacti- vated in PBC PSC. This is consistent with the higher levels of thrombin-antithrombin complexes and increased levels of homocysteinemia and tissue factor as signs of endothelial activation. 99 In a well-studied animal model of cholestasis (bile-duct ligated rats) there is a rapid rise in the stimulated platelet cytosolic calcium, the nal common pathway in platelet activa- tion, to a greater amplitude compared with control animals. 100 This is because of a greater release from intracellular stores, which store more calcium as a result of increased sacroplasmatic endoplasmatic-retic- ulum calcium-ATPase activity. 100 Possible explana- tions for this hyperaggregability in PBC PSC are the marked systemic inammatory activity and the signi- cantly higher levels of unconjugated bilirubin, which might be pro-aggregable (cfr. supra). 98 Not all reports are unequivocal on this hyperaggrega- bility, as this could not be demonstrated by platelet aggregometry (the gold standard) in both patients 65 and animal models of PBC. 75 Moreover, there is an increased bleeding time with a trend to normalize after inhibition of NO production (a potent platelet antagonist), 75 a smaller amount of releasable ATP possibly caused by loss of platelet granules during intravascular activa- tion, 65 and a lack of increase of CD62P after stimula- tion; 98 all indicative of decreased platelet function. Very recently, the contribution of blood platelets to liver damage in cholestatic liver disease was studied in bile-duct ligated mice. 10 Platelet depletion or blockage of the P-selectin receptor of platelets led to decreased aggregation formation (within the liver), decreased platelet adhesion and less leucocyte accumulation resulting in improved liver transaminase levels. 10 Studies on platelets in alcoholic liver disease (see Table 1) are more difcult to interpret because of the interfering effects of the exogenic toxin (ethanol) on the blood platelets, so active drinking should be taken into account. Ethanol has direct effects on the platelet lipids, the second messenger system (medi- ated by cAMP, inositol trisphosphate and diacylglyc- erol) and the phospholipase A2 system, all resulting in altered platelet aggregability. 101 It is clear that alcohol, at physiologically relevant concentration, has an inhibitory effect on secondary platelet aggre- gation in whole blood as well as in platelet rich plasma. 102 In alcoholic liver disease, most studies indicate a decreased aggregability 103105 although not all agree. 106 Moreover, in alcoholic liver disease, the mal- ondialdehyde production (a by-product of prostaglan- din synthesis) 103 and TXB2 production 107 are elevated. This could indicate an in vivo hyperaggregability, possibly contributing to the thrombocytopenia. In viral liver disease (see Table 1), there is evidence for an in vivo platelet activation demonstrated by the increased concentration of BTG and PF4 in serum. 49 P-selectin expression on resting platelets is also ele- vated in chronic hepatitis and cirrhosis compared to normal controls. 49 Also, plasma soluble P-selectin lev- els where markedly elevated in chronic hepatitis C. 108 This correlated directly with serum hepatitis C virus- RNA and was signicantly higher in patients with low platelet counts and without a relation with platelet associated immunoglobulin G. 108 This suggests that hepatitis C infection might be directly responsible for the in vivo platelet activation in patients with chronic hepatitis C. 108 Currently platelets are increasingly appreciated as inammatory players secreting chemo- tactic factors and mediating cell-to-cell interactions. In mouse models of acute viral hepatitis, it is shown that platelet depletion reduces accumulation of virus specic cytotoxic T lymphocytes and organ damage. Transfusion of platelets in these mice restored accu- mulation of cytotoxic T lymphocytes and severity of disease. 9 In vascular liver diseases as extrahepatic portal vein obstruction and non-cirrhotic portal brosis, parenchymal liver function is usually very well pre- served notwithstanding the presence of portal hyper- tension. Platelet hypoaggregability has been demonstrated in both diseases. 109, 110 Coagulation disorders, suggestive of mild disseminated intravascu- lar coagulation were common. 109 In an experimental rat model of prehepatic portal hypertension (partial portal vein ligation), portal hypertensive rats demon- strated an increased hemorrhagic time and a dimin- ished in vivo platelet activity demonstrated by an in vivo arterial laser lesion model. 111 This could be contradictorily corrected with ultralow dose aspirin administration. 111 In BuddChiari syndrome some patients demon- strate hyperaggregability and increased BTG levels. 110 Also ultrastructural abnormalities like absence of alpha granules and clumping and fusion of the gran- ules in the centre of the platelets have been shown. 112 REVI EW: PLATELETS I N CHRONI C LI VER DI SEASE 1025 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd CONCLUSI ONS In chronic liver disease and cirrhosis, alterations in primary haemostasis (platelet adhesion, activation and aggregation) have received less attention than changes in secondary haemostasis (coagulation). Regarding platelet count, an increased intrasplenic platelet breakdown with variable roles of decreased platelet production and splenic pooling appear to be the most important determinants. Regarding the func- tional change, there is a decreased aggregability attrib- utable to defective (transmembrane and intracellular) signalling, a storage pool defect and an upregulation of the inhibitory pathways. However, there are many con- tradictory reports. This is mainly because of the differ- ent inclusion criteria, different types and degree of liver disease or the small number of patients included. Research in established animal models of liver disease could further help overcome these difculties. Typical for platelet research is the technical dif- culty, the need for fresh blood samples, and the imme- diate execution of the different screening tests. The more specialized the assays (for instance, platelet stud- ies under ow conditions), the more they are prone to artefactual ndings. This is a consequence of the very sensitive nature of platelet tests. The study of platelet numbers and functional changes in relation to (cause and consequence) differ- ent types of chronic liver disease remains an area of interesting research. In view of the importance of blood platelets in disease processes such as tissue repair, inammation, vasculogenesis etc, a consider- able impact on our understanding of the pathophysiol- ogy of liver disease, portal hypertension, liver brosis and therapy thereof can be expected. ACKNOWLEDGEMENTS Declaration of personal interests: P. W. is aspirant for the F. W. O.-Vlaanderen. F. N. and D. C. are funda- mental clinical researchers for the F. W. O.-Vlaander- en. Declaration of funding interests: None. REFERENCES 1 Tripodi A, Mannucci PM. Abnormalities of hemostasis in chronic liver disease: reappraisal of their clinical signicance and need for clinical and laboratory research. J Hepatol 2007; 46: 72733. 2 Amitrano L, Guardascione MA, Bran- caccio V, Balzano A. Coagulation dis- orders in liver disease. Semin Liver Dis 2002; 22: 8396. 3 Caldwell S, Hoffman M, Lisman T, et al. Coagulation disorders and hemo- stasis in liver disease: pathophysiology and critical assessment of current man- agement. Hepatology 2006; 44: 1039 46. 4 Shah U, Ma AD. Tests of platelet func- tion. Curr Opin Hematol 2007; 14: 4327. 5 Ordinas A, Escolar G, Cirera I, et al. Existence of a platelet-adhesion defect in patients with cirrhosis independent of hematocrit: studies under ow con- ditions. Hepatology 1996; 24: 113742. 6 Michelson AD, Frelinger AL III, Furman MI. Current options in platelet function testing. Am J Cardiol 2006; 98: 4N 10N. 7 McKenzie ME, Gurbel PA, Levine DJ, Serebruany VL. Clinical utility of avail- able methods for determining platelet function. Cardiology 1999; 92: 2407. 8 Giannini EG. Review article: thrombo- cytopenia in chronic liver disease and pharmacologic treatment options. Ali- ment Pharmacol Ther 2006; 23: 1055 65. 9 Iannacone M, Sitia G, Isogawa M, et al. Platelets mediate cytotoxic t lympho- cyte-induced liver damage. Nat Med 2005; 11: 11679. 10 Laschke MW, Dold S, Menger MD, Jeppsson B, Thorlacius H. Platelet- dependent accumulation of leukocytes in sinusoids mediates hepatocellular damage in bile duct ligation-induced cholestasis. Br J Pharmacol 2008; 153: 14856. 11 Lesurtel M, Graf R, Aleil B, et al. Plate- let-derived serotonin mediates liver regeneration. Science 2006; 312: 104 7. 12 Aoki Y, Hirai K, Tanikawa K. Mecha- nism of thrombocytopenia in liver cir- rhosis: kinetics of indium-111 tropolone labelled platelets. Eur J Nucl Med 1993; 20: 1239. 13 Schmidt KG, Rasmussen JW, Bekker C, Madsen PE. Kinetics and in vivo distri- bution of 111-in-labelled autologous platelets in chronic hepatic disease: mechanisms of thrombocytopenia. Scand J Haematol 1985; 34: 3946. 14 Stein SF, Harker LA. Kinetic and func- tional studies of platelets, brinogen, and plasminogen in patients with hepa- tic cirrhosis. J Lab Clin Med 1982; 99: 21730. 15 Toghill PJ, Green S. Splenic inuences on the blood in chronic liver disease. Q J Med 1979; 48: 61325. 16 Koike Y, Yoneyama A, Shirai J, et al. Evaluation of thrombopoiesis in throm- bocytopenic disorders by simultaneous measurement of reticulated platelets of whole blood and serum thrombopoietin concentrations. Thromb Haemost 1998; 79: 110610. 17 Panasiuk A, Prokopowicz D, Zak J, Panasiuk B. Reticulated platelets as a marker of megakaryopoiesis in liver cirrhosis; relation to thrombopoietin and hepatocyte growth factor serum concentration. Hepatogastroenterology 2004; 51: 11248. 18 Peck-Radosavljevic M, Wichlas M, Zacherl J, et al. Thrombopoietin induces rapid resolution of thrombocy- topenia after orthotopic liver transplan- tation through increased platelet production. Blood 2000; 95: 795801. 1026 P. WI TTERS et al. 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd 19 Kajihara M, Okazaki Y, Kato S, et al. Evaluation of platelet kinetics in patients with liver cirrhosis: similarity to idiopathic thrombocytopenic pur- pura. J Gastroenterol Hepatol 2007; 22: 1128. 20 Wolber E, Jelkmann W. Thrombopoie- tin: the novel hepatic hormone. News Physiol Sci 2002; 17: 610. 21 Sungaran R, Markovic B, Chong BH. Localization and regulation of throm- bopoietin mrna expression in human kidney, liver, bone marrow, and spleen using in situ hybridization. Blood 1997; 89: 1017. 22 Rios R, Sangro B, Herrero I, Quiroga J, Prieto J. The role of thrombopoietin in the thrombocytopenia of patients with liver cirrhosis. Am J Gastroenterol 2005; 100: 13116. 23 Martin TG III, Somberg KA, Meng YG, et al. Thrombopoietin levels in patients with cirrhosis before and after ortho- topic liver transplantation. Ann Intern Med 1997; 127: 2858. 24 Sanjo A, Satoi J, Ohnishi A, Maruno J, Fukata M, Suzuki N. Role of elevated platelet-associated immunoglobulin G and hypersplenism in thrombocytope- nia of chronic liver diseases. J Gastro- enterol Hepatol 2003; 18: 63844. 25 Sezai S, Kamisaka K, Ikegami F, et al. Regulation of hepatic thrombopoietin production by portal hemodynamics in liver cirrhosis. Am J Gastroenterol 1998; 93: 802. 26 Koruk M, Onuk MD, Akcay F, Savas MC. Serum thrombopoietin levels in patients with chronic hepatitis and liver cirrhosis, and its relationship with circulating thrombocyte counts. Hepatogastroenterology 2002; 49: 16458. 27 Peck-Radosavljevic M, Zacherl J, Wich- las M, et al. Thrombopoietic cytokines and reversal of thrombocytopenia after liver transplantation. Eur J Gastroen- terol Hepatol 1999; 11: 1516. 28 Aref S, Mabed M, Selim T, Goda T, Khafagy N. Thrombopoietin (TPO) lev- els in hepatic patients with thrombocy- topenia. Hematology 2004; 9: 3516. 29 Ghalib R, Levine C, Hassan M, et al. Recombinant human interleukin-11 improves thrombocytopenia in patients with cirrhosis. Hepatology 2003; 37: 116571. 30 Ichikawa N, Kitano K, Shimodaira S, et al. Changes in serum thrombopoietin levels after splenectomy. Acta Haema- tol 1998; 100: 13741. 31 Aster RH. Splenic platelet pooling as a cause of hypersplenic thrombocyto- penia. Trans Assoc Am Physicians 1965; 78: 36273. 32 Peck-Radosavljevic M. Thrombocytope- nia in liver disease. Can J Gastroenterol 2000; 14(Suppl. D): 60D6D. 33 Mannl HFK, Matzander U, Tkocz H, Buchwald W. Portal pressure, spleen size, hypersplenism and the problem of congestive splenomegaly in the surgery of portal hypertension in cirrhotics. Chir Gastroent (Gastroent. Surg.) 1975; 9: 5116. 34 Jabbour N, Zajko A, Orons P, Irish W, Fung JJ, Selby RR. Does transjugular intrahepatic portosystemic shunt (tips) resolve thrombocytopenia associated with cirrhosis? Dig Dis Sci 1998; 43: 245962. 35 Karasu Z, Gurakar A, Kerwin B, et al. Effect of transjugular intrahepatic por- tosystemic shunt on thrombocytopenia associated with cirrhosis. Dig Dis Sci 2000; 45: 19716. 36 Gschwantler M, Vavrik J, Gebauer A, et al. Course of platelet counts in cir- rhotic patients after implantation of a transjugular intrahepatic portosystemic shunt a prospective, controlled study. J Hepatol 1999; 30: 2549. 37 Jalan R, Redhead DN, Allan PL, Hayes PC. Prospective evaluation of haem- atological alterations following the transjugular intrahepatic portosystemic stent-shunt (tipss). Eur J Gastroenterol Hepatol 1996; 8: 3815. 38 Pursnani KG, Sillin LF, Kaplan DS. Effect of transjugular intrahepatic por- tosystemic shunt on secondary hy- persplenism. Am J Surg 1997; 173: 16973. 39 Lawrence SP, Lezotte DC, Durham JD, Kumpe DA, Everson GT, Bilir BM. Course of thrombocytopenia of chronic liver disease after transjugular intrahe- patic portosystemic shunts (tips). a ret- rospective analysis. Dig Dis Sci 1995; 40: 157580. 40 Mutchnick MG, Lerner E, Conn HO. Effect of portacaval anastomosis on hy- persplenism. Dig Dis Sci 1980; 25: 92938. 41 Lee C, Leung T, Wang H, et al. Evalua- tion of the effect of partial splenic embolization on platelet values for liver cirrhosis patients with thrombocy- topenia. World J Gastroenterol 2007; 13: 61922. 42 Beer JH, Clerici N, Baillod P, von Felten A, Schlappritzi E, Bu chi L. Quantitative and qualitative analysis of platelet GPIb and von willebrand factor in liver cirrhosis. Thromb Haemost 1995; 73: 6019. 43 Kajihara M, Kato S, Okazaki Y, et al. A role of autoantibody-mediated platelet destruction in thrombocytopenia in patients with cirrhosis. Hepatology 2003; 37: 126776. 44 Kajiwara E, Akagi K, Azuma K, Onoy- ama K, Fujishima M. Evidence for an immunological pathogenesis of throm- bocytopenia in chronic liver disease. Am J Gastroenterol 1995; 90: 9626. 45 Noguchi H, Hirai K, Aoki Y, Sakata K, Tanikawa K. Changes in platelet kinet- ics after a partial splenic arterial embo- lization in cirrhotic patients with hypersplenism. Hepatology 1995; 22: 16828. 46 Pereira J, Accatino L, Alfaro J, Brahm J, Hidalgo P, Mezzano D. Platelet au- toantibodies in patients with chronic liver disease. Am J Hematol 1995; 50: 1738. 47 Cioni G, Cristani A, Mussini C, et al. Incidence and clinical signicance of elevated brin(ogen) degradation prod- uct and or d-dimer levels in liver cir- rhosis patients. Ital J Gastroenterol 1990; 22: 704. 48 Vardareli E, Saricam T, Demirustu C, Gulbas Z. Soluble p selectin levels in chronic liver disease: relationship to disease severity. Hepatogastroenterology 2007; 54: 4669. 49 Panasiuk A, Prokopowicz D, Zak J, Matowicka-Karna J, Osada J, Wysocka J. Activation of blood platelets in chronic hepatitis and liver cirrhosis p-selectin expression on blood platelets and secretory activity of beta-thrombo- globulin and platelet factor-4. Hepato- gastroenterology 2001; 48: 81822. 50 Panzer S, Seel E, Brunner M, et al. Platelet autoantibodies are common in hepatitis C infection, irrespective of the presence of thrombocytopenia. Eur J Haematol 2006; 77: 5137. 51 Scho ffski P, Tacke F, Trautwein C, et al. Thrombopoietin serum levels are elevated in patients with hepatitis B C infection compared to other causes of chronic liver disease. Liver 2002; 22: 11420. 52 Bassendine MF, Collins JD, Stephenson J, Saunders P, James OF. Platelet associated immunoglobulins in primary biliary cirrhosis: a cause of thrombo- cytopenia? Gut 1985; 26: 10749. REVI EW: PLATELETS I N CHRONI C LI VER DI SEASE 1027 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd 53 Panzer S, Penner E, Nelson PJ, Proc- hazka E, Benda H, Saurugger PN. Iden- tication of the platelet glycoprotein IIb IIIa complex as a target antigen in primary biliary cirrhosis-associated autoimmune thrombocytopenia. evi- dence that platelet-reactive autoanti- bodies can also bind to the mitochondrial antigen m2. J Autoim- mun 1990; 3: 47383. 54 Adinol LE, Giordano MG, Andreana A, et al. Hepatic brosis plays a central role in the pathogenesis of thrombocytope- nia in patients with chronic viral hepati- tis. Br J Haematol 2001; 113: 5905. 55 Okubo M, Shiota G, Kawasaki H. Thrombopoietin levels in serum and liver tissue in patients with chronic viral hepatitis and hepatocellular carci- noma. Clin Sci (Lond) 2000; 99: 207 14. 56 Wang S, Tsai S, Chen G, Lan J, Yen T. In-111-labeled platelet scintigraphies in patients with chronic hepatitis C. Hepatogastroenterology 2002; 49: 10668. 57 Thomas DP, Ream VJ, Stuart PK. Plate- let aggregation in patients with laen- necs cirrhosis of the liver. N Engl J Med 1967; 276: 13448. 58 Blake JC, Sprengers D, Grech P, McCor- mick PA, McIntyre N, Burroughs AK. Bleeding time in patients with hepatic cirrhosis. BMJ 1990; 301: 125. 59 Bonnard P, Vitte RL, Barbare JC, et al. Is bleeding time measurement useful for choosing the liver biopsy route? The results of a pragmatic, prospective multicentric study in 219 patients. J Clin Gastroenterol 1999; 29: 3479. 60 Violi F, Leo R, Vezza E, Basili S, Cor- dova C, Balsano F. Bleeding time in patients with cirrhosis: relation with degree of liver failure and clotting abnormalities. C.A.L.C. group. coagula- tion abnormalities in cirrhosis study group. J Hepatol 1994; 20: 5316. 61 Escolar G, Cases A, Vin as M, et al. Evaluation of acquired platelet dys- functions in uremic and cirrhotic patients using the platelet function analyzer (pfa-100): inuence of hemat- ocrit elevation. Haematologica 1999; 84: 6149. 62 Lisman T, Bongers TN, Adelmeijer J, et al. Elevated levels of von willebrand factor in cirrhosis support platelet adhesion despite reduced functional capacity. Hepatology 2006; 44: 5361. 63 Lisman T, Adelmeijer J, de Groot PG, Janssen HLA, Leebeek FWG. No evi- dence for an intrinsic platelet defect in patients with liver cirrhosis studies under ow conditions. J Thromb Hae- most 2006; 4: 20702. 64 Tonda R, Galan AM, Pino M, et al. Hemostatic effect of activated recombi- nant factor VII (rFVIIa) in liver disease: studies in an in vitro model. J Hepatol 2003; 39: 9549. 65 Ingeberg S, Jacobsen P, Fischer E, Bentsen KD. Platelet aggregation and release of atp in patients with hepatic cirrhosis. Scand J Gastroenterol 1985; 20: 2858. 66 Laf G, Cominelli F, Ruggiero M, Fedi S, Chiarugi V, Gentilini P. Molecular mechanism underlying impaired plate- let responsiveness in liver cirrhosis. FEBS Lett 1987; 220: 2179. 67 Laf G, Cominelli F, Ruggiero M, et al. Altered platelet function in cirrhosis of the liver: impairment of inositol lipid and arachidonic acid metabolism in response to agonists. Hepatology 1988; 8: 16206. 68 Laf G, Marra F, Gresele P, et al. Evi- dence for a storage pool defect in platelets from cirrhotic patients with defective aggregation. Gastroenterology 1992; 103: 6416. 69 Laf G, Marra F, Failli P, et al. Defec- tive signal transduction in platelets from cirrhotics is associated with increased cyclic nucleotides. Gastroen- terology 1993; 105: 14856. 70 Laf G, Cinotti S, Filimberti E, et al. Defective aggregation in cirrhosis is independent of in vivo platelet activa- tion. J Hepatol 1996; 24: 43643. 71 Pantaleo P, Marra F, Vizzutti F, et al. Effects of dietary supplementation with arachidonic acid on platelet and renal function in patients with cirrhosis. Clin Sci (Lond) 2004; 106: 2734. 72 Rubin MH, Weston MJ, Langley PG, White Y, Williams R. Platelet function in chronic liver disease: relationship to disease severity. Dig Dis Sci 1979; 24: 197202. 73 Willis SE, Jackson ML, Meric SM, Rousseaux CG. Whole blood platelet aggregation in dogs with liver disease. Am J Vet Res 1989; 50: 18937. 74 Younger HM, Hadoke PW, Dillon JF, Hayes PC. Platelet function in cirrhosis and the role of humoral factors. Eur J Gastroenterol Hepatol 1997; 9: 98992. 75 Albornoz L, Bandi JC, Otaso JC, Laud- anno O, Mastai R. Prolonged bleeding time in experimental cirrhosis: role of nitric oxide. J Hepatol 1999; 30: 456 60. 76 Sanchez-Roig MJ, Rivera J, Moraleda JM, Garc a VV. Quantitative defect of glycoprotein Ib in severe cirrhotic patients. Am J Hematol 1994; 45: 10 5. 77 Chau TN, Chan YW, Patch D, Tokunaga S, Greenslade L, Burroughs AK. Throm- belastographic changes and early reb- leeding in cirrhotic patients with variceal bleeding. Gut 1998; 43: 267 71. 78 Papatheodoridis GV, Patch D, Webster GJ, Brooker J, Barnes E, Burroughs AK. Infection and hemostasis in decompen- sated cirrhosis: a prospective study using thrombelastography. Hepatology 1999; 29: 108590. 79 Dav ` G, Migneco G, Vigneri S, Tripi S, Scialabba A, Strano A. Platelet throm- boxane production in liver cirrhosis. Prostaglandins Leukot Med 1985; 19: 99104. 80 Luzzatto G, Fabris F, Gerunda GE, Zan- grandi F, Girolami A. Failure of two anti-platelet drugs (indobufen and dipyridamole) to improve thrombocyto- penia in liver cirrhosis. Acta Haematol 1987; 77: 1016. 81 Chamone DA, Van Hoof A, Vermylen J. Increased release of prostacyclin-like activity from rat aorta by plasma from patients with hepatic or renal failure. Arq Gastroenterol 1987; 24: 169. 82 Kunihiro N, Kawai B, Sanjo A, Osaka K, Ohnishi A. Platelet aggregation and coagulation and brinolysis parameters in both portal and systemic circulations in patients with cirrhosis and hepato- cellular carcinoma. Hepatol Res 2001; 19: 529. 83 Laf G, La Villa G, Pinzani M, et al. Altered renal and platelet arachidonic acid metabolism in cirrhosis. Gastroen- terology 1986; 90: 27482. 84 Laf G, Foschi M, Masini E, et al. Increased production of nitric oxide by neutrophils and monocytes from cir- rhotic patients with ascites and hyper- dynamic circulation. Hepatology 1995; 22: 166673. 85 Desai K, Mistry P, Bagget C, Burroughs AK, Bellamy MF, Owen JS. Inhibition of platelet aggregation by abnormal high density lipoprotein particles in plasma from patients with hepatic cir- rhosis. Lancet 1989; 1: 6935. 86 Fuse I. Disorders of platelet function. Crit Rev Oncol Hematol 1996; 22: 125. 1028 P. WI TTERS et al. 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd 87 Solum NO, Rigollot C, Budzyn ski AZ, Marder VJ. A quantitative evaluation of the inhibition of platelet aggregation by low molecular weight degradation products of brinogen. Br J Haematol 1973; 24: 41934. 88 Baele G, Beke R, Barbier F. In vitro inhibition of platelet aggregation by bile salts. Thromb Haemost 1980; 44: 624. 89 Pereira J, Accatino L, Pizarro M, Mezz- ano V, Iban ez A, Mezzano D. In vivo effect of bile salts on platelet aggrega- tion in rats. Thromb Res 1995; 80: 35762. 90 Lisman T, Leebeek FWG, de Groot PG. Haemostatic abnormalities in patients with liver disease. J Hepatol 2002; 37: 2807. 91 Moiny G, Thirion A, Deby C. Bilirubin induces platelet aggregation. Thromb Res 1990; 59: 4136. 92 Baele G, Rasquin K, Barbier F. Coagu- lant, brinolytic, and aggregating activity in ascitic uid. Am J Gastroen- terol 1986; 81: 4403. 93 Gores GJ, Wiesner RH, Dickson ER, Zinsmeister AR, Jorgensen RA, Lang- worthy A. Prospective evaluation of esophageal varices in primary biliary cirrhosis: development, natural history, and inuence on survival. Gastroenter- ology 1989; 96: 15529. 94 Biagini MR, Guadascione M, McCor- mick AP, Raskino C, McIntyre N, Burr- oughs AK. Bleeding varices in PBC and its prognostic signicance. Gut 1990; 31: A1209. 95 Palareti G, Legnani C, Maccaferri M, et al. Coagulation and brinolysis in orthotopic liver transplantation: role of the recipients disease and use of anti- thrombin III concentrates. S. orsola working group on liver transplantation. Haemostasis 1991; 21: 6876. 96 Shannon P, Wanless IR. The site of vascular obstruction in early stage pri- mary biliary cirrhosis with portal vein (pv) thrombosis. Hepatology 1995; 22: 253A. 97 Ben-Ari Z, Panagou M, Patch D, et al. Hypercoagulability in patients with pri- mary biliary cirrhosis and primary scle- rosing cholangitis evaluated by thrombelastography. J Hepatol 1997; 26: 5549. 98 Pihusch R, Rank A, Go hring P, Pihusch M, Hiller E, Beuers U. Platelet function rather than plasmatic coagulation explains hypercoagulable state in cholestatic liver disease. J Hepatol 2002; 37: 54855. 99 Biagini MR, Tozzi A, Marcucci R, et al. Hyperhomocysteinemia and hypercoag- ulability in primary biliary cirrhosis. World J Gastroenterol 2006; 12: 1607 12. 100 Atucha NM, Iyu D, Alcaraz A, et al. Altered calcium signalling in platelets from bile-duct-ligated rats. Clin Sci (Lond) 2007; 112: 16774. 101 Rubin R. Effects of ethanol on platelets. Lab Invest 1999; 63: 72932. 102 Torres Duarte AP, Dong QS, Young J, Abi-Younes S, Myers AK. Inhibition of platelet aggregation in whole blood by alcohol. Thromb Res 1995; 78: 10715. 103 Barrison IG, Viola L, Murray-Lyon IM. Platelet prostaglandin production in alcoholic liver disease. Prostaglandins Leukot Med 1983; 10: 33144. 104 Hillbom M, Muuronen A, Neiman J. Liver disease and platelet function in alcoholics. Br Med J (Clin Res Ed) 1987; 295: 581. 105 Hillbom M, Neiman J. Platelet throm- boxane formation capacity after etha- nol withdrawal in chronic alcoholics. Haemostasis 1988; 18: 1708. 106 Watanabe M, Shiraishi K, Itakura M, Matsuzaki S. Relationship between platelet membrane lipid compositions and platelet aggregability in alcoholic liver disease. Alcohol Clin Exp Res 1998; 22: 97S102S. 107 Ouwendijk RJ, Zijlstra FJ, Wilson JH, Bonta IL, Vincent JE. Raised plasma thromboxane B2 levels in alcoholic liver disease. Prostaglandins Leukot Med 1983; 10: 11522. 108 Ferroni P, Mammarella A, Martini F, et al. Increased soluble p-selectin levels in hepatitis C virus-related chronic hepatitis: correlation with viral load. J Investig Med 2001; 49: 40712. 109 Bajaj JS, Bhattacharjee J, Sarin SK. Coagulation prole and platelet func- tion in patients with extrahepatic portal vein obstruction and non-cirrhotic por- tal brosis. J Gastroenterol Hepatol 2001; 16: 6416. 110 Prasad CV, Kaur U, Marwaha N, Ghosh K, Chawla YK, Dilawari JB. Hemostatic alterations in non-cirrhotic portal brosis, extrahepatic portal venous obstruction and budd-chiari syndrome. Indian J Gastroenterol 1990; 9: 5760. 111 Eizayaga FX, Aguejouf O, Belon P, Doutremepuich C. Platelet aggregation in portal hypertension and its modi- cation by ultra-low doses of aspirin. Pathophysiol Haemost Thromb 2005; 34: 2934. 112 Dayal S, Pati HP, Pande GK, Sharma P, Saraya AK. Platelet ultra-structure study in budd-chiari syndrome. Eur J Haematol 1995; 55: 294301. REVI EW: PLATELETS I N CHRONI C LI VER DI SEASE 1029 2008 The Authors, Aliment Pharmacol Ther 27, 10171029 Journal compilation 2008 Blackwell Publishing Ltd