Platelet production and destruction in liver cirrhosis .pdf
Nom original: Platelet production and destruction in liver cirrhosis.pdf
Titre: Platelet production and destruction in liver cirrhosis
Auteur: Paola Pradella
Ce document au format PDF 1.7 a été généré par Elsevier / Acrobat Distiller 8.1.0 (Windows), et a été envoyé sur fichier-pdf.fr le 05/08/2011 à 21:50, depuis l'adresse IP 41.200.x.x.
La présente page de téléchargement du fichier a été vue 1558 fois.
Taille du document: 635 Ko (7 pages).
Confidentialité: fichier public
Aperçu du document
Platelet production and destruction in liver cirrhosis
Paola Pradella1, Stefania Bonetto2, Stefano Turchetto1, Laura Uxa1, Consuelo Comar2,
Francesca Zorat2, Vincenzo De Angelis1, Gabriele Pozzato2,⇑
Blood Bank Service, Azienda Ospedaliero-Universitaria Ospedali Riuniti di Trieste, Trieste, Italy; 2Struttura Complessa Seconda
Medicina/Ematologia, Azienda Ospedaliero-Universitaria Ospedali Riuniti di Trieste, Trieste, Italy
Background & Aims: Thrombocytopenia is common in liver
cirrhosis (LC) but the mechanisms are not fully understood. The
purpose of our work was to evaluate platelet kinetics in LC with
different etiologies by examining platelet production and
Methods: Ninety-one consecutive LC patients (36 HCV, 49 alcoholics, 15 HBV) were enrolled. As controls, 25 subjects with idiopathic thrombocytopenic purpura, 10 subjects with aplastic
anemia, and 40 healthy blood donors were studied. Plasma
thrombopoietin (TPO) was measured by ELISA. Reticulated platelets (RP) were determined using the Thiazole Orange method.
Plasma glycocalicin (GC) was measured using monoclonal antibodies. Platelet associated and serum antiplatelet antibodies
were detected by ﬂow cytometry. B-cell monoclonality in PBMC
was assessed by immunoglobulin ﬁngerprinting.
Results: Serum TPO was signiﬁcantly lower in LC (29.9 ± 18.1
pg/ml) compared to controls (82.3 ± 47.6 pg/ml). The GC levels
were higher in LC (any etiology) than in healthy cases. Conversely, the absolute levels of RP were lower in LC (any etiology)
than in healthy controls. The platelet-associated and serum
anti-platelet antibodies were higher in HCV+ LC compared to
healthy subjects (p <0.0064), alcoholic LC (p <0.018), and HBV+
LC (p <0.0001). B-cell monoclonality was found in 27% of the
HCV + LC, while it was not found in HBV+ or alcoholic LC.
Conclusions: Patients with LC present decreased plasma TPO,
accelerated platelet turnover, and reduced platelet production.
This indicates that LC thrombocytopenia is a multifactorial condition involving both increased platelet clearance and impaired
Ó 2010 European Association for the Study of the Liver. Published
by Elsevier B.V. All rights reserved.
Keywords: Liver cirrhosis; Glycocalicin; Thrombopoietin; HCV; HBV; Aplastic
anemia; Idiopathic thrombocytopenic purpura.
Received 12 December 2009; received in revised form 2 August 2010; accepted 9
August 2010; available online 22 December 2010
⇑ Corresponding author. Address: Struttura Complessa Seconda Medicina, Ospedale ‘‘Maggiore’’, Piazza Ospedale 1, 34100 Trieste, Italy. Tel.: +39 040 3992002;
fax: +39 040 3993560.
E-mail address: email@example.com (G. Pozzato).
Abbreviations: LC, liver cirrhosis; HCV, hepatitis C virus; HBV, hepatitis B virus;
TPO, thrombopoietin; RP, reticulated platelets; GC, glycocalicin; ITP, idiopathic
thrombocytopenic purpura; AA, aplastic anemia; PBMC, peripheral blood mononuclear cells; TBS, Tris-buffered saline; PRP, platelet-rich plasma; BSA, bovin
serum albumin; ACD, acid Na-citrate dextrose.
Thrombocytopenia is a frequent feature in patients with liver
cirrhosis (LC). It has traditionally been attributed to the sequestration of platelets by the spleen [1–3], a situation known as
‘‘hypersplenism’’ secondary to portal hypertension. Nevertheless,
no clear correlation between the portal pressure and the platelet
count has ever been observed. Furthermore, in some cirrhotic
patients thrombocytopenia may persist even after splenectomy
or after portal decompression [4–9]. Since the return to a normal
platelet count has been observed following liver transplantation
, it is likely that other mechanisms apart from hypersplenism,
such as reduced thrombopoietin (the cytokine which regulates
megakaryocyte maturation and platelet production) release by
the liver [11–14] or bone marrow suppression [15,16], are
involved in the thrombocytopenia of cirrhotic patients.
To increase the complexity of this problem, the etiology of the
liver disease should be taken into account since HBV, HCV, and
ethanol (the most common causes of liver cirrhosis) all induce
liver damage via different mechanisms; therefore, even if the ﬁnal
outcome is the same (cirrhosis), the biochemical and immunological abnormalities are quite different, especially in HCV-infected
patients. In addition, the possibility for studying platelet turnover
‘‘in vivo’’ is limited. In this study, we assessed reticulated platelet
counts and glycocalicin levels as measures of platelet production
and destruction, respectively. We also quantiﬁed levels of serum
thrombopoietin and anti-platelet antibodies in groups of patients
affected by cirrhosis of different etiologies (HBV-related, HCVrelated or alcohol induced-LC) and in two control groups of
patients affected by hematological diseases known to affect the
platelet count, such as idiopathic thrombocytopenic purpura
(ITP) and aplastic anemia (AA), as well as in healthy control subjects. In a fraction of the patients, we assessed for the presence
of an expanded monoclonal B-cell population in the peripheral
blood mononuclear cells (PBMC) and for immunoglobulin gene
usage via isotype-speciﬁc immunoglobulin ﬁngerprinting.
Patients and methods
Ninety-one consecutive patients with liver cirrhosis and platelet counts less than
100 109/L were enrolled in the study. Of these patients, 36 had a HCV-related
liver disease, 40 were alcoholics, and 15 had a HBV-related liver disease. None
of the patients had previously undergone interferon therapy or were taking drugs
Journal of Hepatology 2011 vol. 54 j 894–900
JOURNAL OF HEPATOLOGY
known to interfere with bone marrow function. The diagnosis of cirrhosis was
based on the grounds of either liver biopsy (58 cases, 64%) or on ﬁndings resulting
from a physical examination of the patient, results of laboratory tests, imaging
studies, and the presence of portal hypertension (splenomegaly, ascites, and
esophageal varices). The liver biopsies were placed in buffered formalin and
stained with haematoxylin and eosin or with Gomori methenamine silver for
reticulum staining. In HCV and HBV patients, the disease activity and ﬁbrosis
were assessed according to the METAVIR scoring system . Patients carrying
anti-nuclear, anti-mitochondrial, or anti-smooth muscle antibodies were discarded from the study.
The study protocol was approved by the ethical committee and conforms to
the ethical guidelines of the 1975 Declaration of Helsinki. All patients gave written informed consent for participation in this medical research.
Twenty-ﬁve cases of idiopathic thrombocytopenic purpura (ITP) were enrolled in
the study. Each subject presented the active phase of the disease but had not yet
initiated immunosuppressive treatment. This disease is characterized by a high
platelet turnover and the presence of anti-platelet antibodies. The diagnosis of
ITP was based on commonly adopted criteria  involving: the patient’s medical
history, physical examination report, complete blood cell count, and cytomorphologic examination of the peripheral-blood smear in which no alterations of erythrocytic and leukocytic series should be present. To conﬁrm the diagnosis, bone
marrow aspiration was performed on all patients and only those with the presence of a normal or increased number of megakaryocytes without pathologic
alterations of the erythroblastic, granuloblastic, or lymphocytic series were
included in the study. Moreover, we had the opportunity to study 10 patients
affected by severe aplastic anemia, a disease exhibiting a very low platelet count
due to a reduced platelet production. The deﬁnition of ‘‘severe’’ aplastic anemia
was determined according to currently used criteria : the disease was considered severe if at least two of the following were noted: a neutrophil count less
than 0.5 109/L, a platelet count less than 20 109/L, and a reticulocyte count
less than 60 109/L with hypocellular (<20%) bone marrow.
Forty volunteer blood donors, serologically negative for HIV, HBV, and HCV, with
normal platelet counts were used as normal healthy control subjects. Informed
consent was obtained prior to all blood donations.
In this observational study, we had the opportunity to apply the same up-to-date
methods for studying platelet turnover to different groups of patients. The antiHCV antibodies were detected by a third-generation enzyme immunoassay (Ortho
HCV SAVe 3.0, Raritan, NJ). Positivity for anti-HCV antibodies was conﬁrmed by
strip immunoblot assay (RIBA HCV 3.0, Chiron Corp., Emeryville, CA). In antiHCV-positive patients, the serum HCV-RNA was assessed by means of nested
reverse transcription polymerase chain reaction. Commercially available
enzyme-linked immunoassays were used to measure HBeAg, anti-HBe, HbsAg,
and anti-HBs (Abbott Laboratories, USA). The Serum HBV-DNA was measured using
the TaqMan polymerase chain reaction assay (COBAS TaqMan, Roche Molecular
System, lowest limit of detection: 70 copies/ml).
The maximum longitudinal diameter of the spleen was estimated by means of
an ultrasound scan by an experienced operator. Ultrasonography was carried out
using an AU 570 Asynchronous (Esaote Biomedica, Genoa, Italy).
healthy subject control samples. The same threshold was used for the analysis of
all samples. The percentage of reticulated platelets was obtained by subtracting
the percentage of reticulated platelets of unstained preparations from that of
stained samples. A normal reference range (5.9 ± 2.25) was obtained using blood
samples from 60 healthy normal subjects.
Blood was collected in EDTA at a 50 mM ﬁnal concentration. The samples were processed within 2 h following blood collection since preliminary tests showed that no
change in glycocalicin values could be observed within this time period. The plasma
was centrifuged at 12,000g for 5 min and the supernatant was stored at –70 °C. The
monoclonal antibodies LJ-P3 and LJ-P19 were kindly given by Z.M. Ruggeri (Scripps
Clinics, La Jolla, CA, USA). LJ-P3 and LJ-P19 monoclonal antibodies interact with epitopes only present in the native glycoproteins located within residues 1-290. The
epitopes for the two monoclonal antibodies are distinct since no cross-reactivity
was observed. To determine the speciﬁcity of each monoclonal antibody, we measured glycocalicin (GC) concentration in pooled control plasma from normal subjects by co-incubating the plasma with various concentrations of mouse IgG and
LJ-P19. There was no decrease in the concentration of GC following co-incubation
of the plasma at different dilutions (1:500–1:25,000) of mouse IgG at room temperature for 1 h. By contrast, the GC concentration was decreased in plasma co-incubated in concentrations of nonbiotinylated LJ-P19 ranging between 3.3 and
0.066 lg/ml. GC was puriﬁed as previously described by Vicente et al. . GC concentration was determined by the Micro BCA-method (Pierce, Rockford, USA) using
albumin as a standard. The GC value could be underestimated by approximately
30–40% in the protein assay because of the high glycosylation of GC. Five to eight
fresh random platelet concentrates, to which a 15% acid Na-citrate dextrose
(ACD) solution was added, were centrifuged at 800g for 20 min at room temperature. The platelet pellet was then washed twice with TBS, pH 7.4, containing
2 mM EDTA and 15% ACD, resuspended in TBS containing 2 mM CaCl2 and
0.1 mM phenyl-methyl-sulphonyl-ﬂuoride and sonicated on ice. The suspension
was then stirred (2 h at RT and 18 h at 4 °C) and centrifuged at 23,000g for
30 min. The supernatant was applied to an afﬁnity column of wheat germ agglutinin-Sepharose 6 MB (Pharmacia, Uppala, Sweden), and the bound proteins were
eluted with TBS containing 1 mM EDTA, 0.1 mM PMSF, 3 mM NaN3, and 100 mM
N-acetyl-glucosamine. The eluate was then immunopuriﬁed with an LJ-P3 Sepharose column (Pharmacia, Uppala, Sweden). GC was eluted with 50 mM diethylamine, 1 mM EDTA, and 0.1 mM PMSF. Glycine 0.55 M was added and the eluted
protein was dialysed for 18 h against TBS and concentrated with Centiprep 100.
Puriﬁed GC was stored at –80 °C. GC was assayed in plasma by ELISA following
the methods of Kunishima et al.  and Beer et al. , with some modiﬁcations.
Brieﬂy, a ﬂat-bottomed microtitre plate (Costar, Cambridge, USA) was coated with
the monoclonal antibody LJ-P3 (1.0 lg/ml) overnight at 4 °C, washed 3 times with
PBS containing 0.05% Tween (PBS-T), blocked with PBS containing 2% BSA and
washed again with PBS-T. The standard curve was prepared using immunopuriﬁed
GC ranging from 0.05 to 1 lg/ml in PBS-T. The plasma samples were then diluted 4fold and 8-fold in PBS-T and 100 ll of each added to individual wells. After incubation for 60 min at RT, 100 ll of the biotin-labelled LJ-P19 (1.5 lg/ml in PBS-T) were
added and incubated for a further 60 min. After three washes, each well was ﬁlled
with 50 ll of the streptavidin-HRPO (Caltag, South San Francisco, CA, USA), diluted
1000-fold in PBS-T, and incubated for 60 min at RT. After three washes, 100 ll of the
o-phenylenediamine dihydrochloride reaction was stopped by adding 50 ll of
2NH2SO4 and the absorbance was measured at 492 nm with a microtitre plate
reader. The results were plotted and expressed as GC concentration in lg/ml
and as GC indices. The GC index represents the standardization of the GC value
for a platelet count of 250 109/L and is calculated as follows: GC index = GC
lg/ml (250 109/L/patient’s platelets).
Peripheral blood was collected using ACD-A (at the ratio 8.5/1.5 v/v) and centrifuged at 100g for 10 min at 22 °C to obtain platelet-rich plasma (PRP). 10 ll PRP
from normal subjects and 20–50 ll from all other patients were added to 1 mL of
Thiazole Orange diluted 1:4 in Tris-buffered saline, pH 7.4, 10 mM EDTA, 0.1%
bovine serum albumin (TBS/EDTA/BSA), and incubated for 60 min at 22 °C in
the dark. PRP incubated with TBS/EDTA/BSA without Thiazole Orange was used
as a blank for each sample. Fluorescence was analysed using an Epics-Elite ﬂow
cytometer (Coulter, USA). Logarithmic ampliﬁcation was used for both light scatter and ﬂuorescence signals. For each sample, ﬂuorescence signals were collected
from 10,000 cells and a frequency histogram created. Three normal controls were
run in each assay; a threshold was positioned giving 1% of ﬂuorescent platelets in
Serum thrombopoietin (TPO) levels (pg/ml) were measured by means of ELISA,
(Quantikine. R&D Systems Europe Ltd., Oxon, UK) according to the manufacturer’s
instructions. The lower limit of detection of the kit is 15 pg/ml. Sera were collected
in a serum-separator tube and allowed to clot for approximately 45–60 min; they
were then centrifuged, chilled rapidly, and stored at 20 °C until TPO determination. In our laboratory the intra-assay coefﬁcient of variation was 8.7%.
Detection of antiplatelet antibodies by ﬂow cytometry
The direct labeling procedures for platelet-associated Ig, and the indirect labeling
procedures for serum circulating Ig, were performed according standard
Journal of Hepatology 2011 vol. 54 j 894–900
Using an EPICS-Elite ﬂow cytometer (Coulter) equipped with a 488 nm argonlaser, data were acquired as 5000 ungated events and a ﬂow rate of 150–300
events was maintained. Platelets were gated using forward- and side-angle light
scatter and log ampliﬁcation with the gains adjusted to include the whole platelet
population. Fluorescent signals were then obtained with platelets sensitized to
surface IgG (direct method), patients, and control samples (indirect method).
Total cellular RNA was isolated from PBMC using the procedure described by
Chomczynski and Sacchi . PBMC was obtained by fractionation of whole
blood (or bone marrow) on the Ficoll/Hypaque gradient. One lg of mRNA was
reverse transcribed (RT) using random examers. The entire RT product was next
subjected to PCR with 50 pmol of a degenerate VH primer placed in FW1
(50 -AGGTGCAGCTGGA(T)GG(C)AGT C(G)T(G)GG-30 ) and a speciﬁc primer for the
constant region of the gene located eight nucleotides downstream from the
beginning of the CH1 exon (50 -GAAAGGGTTGGGGC GGA T-30 ). The ampliﬁed
PCR fragments were puriﬁed, ligated into a Sma site of a pUC18 plasmid, and used
to transform the Escherichia coli strain DH5a. After expansion, the clones were
randomly picked and double-stranded DNA templates were sequenced using
the T7 sequencing kit. The procedure was performed in 30 HCV and 10 HBV cirrhosis patients, and 10 patients with alcoholic cirrhosis.
Statistical analysis was carried out using the Chi-square-test to compare proportions and the Mann–Whitney U-test to compare means. Spearman’s rank correlation tests were used to look for correlations between the various parameters. Data
are shown as the mean ± SD. A p value of 0.05 or less was considered statistically
The main characteristics of the 80 patients with liver cirrhosis are
given in Table 1. As shown, the three groups of patients had similar clinical, hematological, and biochemical characteristics. The
clinical data setting of the hematological patients is indicated in
In the patients affected by alcoholic cirrhosis, the mean percentage
of reticulated platelets was very similar to that found in normal
healthy subjects (5.8 ± 3.2% vs. 5.9 ± 2.2% p: NS). Given the reduced
platelet count (76.826 ± 26.020 109/L) the absolute levels of reticulated platelets were signiﬁcantly higher in normal subjects compared to patients with alcoholic cirrhosis (4.233 ± 2.367 109/L
vs. 14.666 ± 5.999 109/L p <0.0000000012). Similar results have
been found in HBV+ liver cirrhosis (percentage of reticulated platelets 5.5 ± 4.1% with absolute levels of 4.996 ± 3.143 109/L). In
HCV+ liver cirrhosis, despite a similar mean platelet count
(67.811 ± 19.181 109/L p: NS), the percentage of reticulated
platelets was signiﬁcantly higher (11.3 ± 4.7%) than in patients with
alcoholic cirrhosis (5.8 ± 3.2% p <0.043) and HBV+ cirrhosis
(5.5 ± 4.1% p <0.05). Consequently, the absolute counts of reticulated platelets were higher in HCV patients (6.629 ± 7.409 109/
L) than in either patients of alcoholic cirrhosis (4.233 ± 2.367
109/L) or HBV (4.966 ± 3.143 109/L); given the high standard deviation of the HCV patients, the difference (with HBV of alcoholic cirrhosis) does not reach statistical signiﬁcance (p = 0.068). In the HCV
patients, two groups can be identiﬁed, the ﬁrst one (27 cases) with
low reticulated platelet count (4.3 ± 1.7% i.e. 3.171 ± 1.572 109/L)
and second one (nine cases) with normal-high reticulated platelet
count (31.6 ± 17.6% i.e. 16.640 ± 9.377 109/L p <0.0000000027).
The patients affected by ITP showed very high levels of reticulated platelets (25.7 ± 11.2%), while very low levels were found
in the patients with aplastic anemia (1.0 ± 0.9%). The results
obtained from these hematological cases are in line with the
pathophysiology of the underlying diseases: increased platelet
production for accelerated destruction in the former and very
low production in the latter.
In the normal healthy subjects, given the slow rate of platelet
turnover, the glycocalicin index was within normal limits
(0.9 ± 0.2), while in the patients with liver cirrhosis, independent
of etiology, the glycocalicin index was high: 1.96 ± 1.40 in HCV+
cirrhosis (p <0.0005 vs. healthy controls), 1.79 ± 1.51 in the alcoholics (p <0.006 vs. healthy controls) and 1.71 ± 1.69 in HBV+ cirrhosis (p <0.006 vs. healthy controls). In the HCV patients, even
for the glycocalicin index, two groups can be identiﬁed: the ﬁrst
one (27 cases) with a value comparable to alcoholics and HBV+
cirrhosis (1.76 ± 1.48) and a second one (nine cases) with high
glycocalicin index (2.46 ± 0.84; p <0.02). The mean glycocalicin
level and index were signiﬁcantly greater in patients affected
by ITP (12.9 ± 4.4, p <0.000002 vs. healthy controls). In the
patients with aplastic anemia, the glycocalicin index was normal
(0.8 ± 0.1), in accordance with the normal platelet turnover.
The serum thrombopoietin levels were signiﬁcantly (p <0.0003)
lower in patients with cirrhosis (29.9 ± 18.1 pg/ml) compared to
healthy controls (94.7 ± 35.9 pg/ml). No differences were found
on the basis of the etiology of the liver cirrhosis (29.2 ± 16.2 in
alcoholics, 30.6 ± 22.0 in HCV+ and 31.1 ± 19.9 in HBV+). On the
contrary, levels above the normal limits were found in the
patients with aplastic anemia (507.7 ± 86.1 pg/ml) and in
patients affected by ITP (155 ± 76 pg/ml). The thrombopoietin
serum levels were inversely correlated to platelet counts in ITP
patients (r: 0.87), while this correlation was not found in
patients with liver cirrhosis. Moreover, no difference was
observed in TPO serum levels between patients with or without
splenomegaly, and no correlation was observed between TPO
serum levels and either spleen longitudinal diameter or surface
area, or liver histological inﬂammatory activity (in both HCV
and HBV subjects) (data not shown).
Seventy percent of the cirrhosis patients presented splenomegaly
(i.e. a spleen longitudinal diameter over 12 cm). The level of
thrombocytopenia was not correlated to either spleen size (considering both spleen longitudinal diameter and spleen surface
area) or liver histological necro-inﬂammatory activity (in both
HCV and HBV patients). The spleen size was normal in all patients
with ITP and aplastic anemia.
Normal healthy subjects were not found to express any detectable levels of anti-platelet antibodies As expected, most patients
(85%) affected by ITP presented platelet-associated antibodies,
and a large fraction of them (65%) also expressed serum circulating anti-platelet antibodies. None of the patients with aplastic
anemia expressed anti-platelet antibodies. The frequencies of
Journal of Hepatology 2011 vol. 54 j 894–900
JOURNAL OF HEPATOLOGY
Table 1. Demographic, biochemical, and virological proﬁles of the patients with HCV-, HVB-, and alcoholic cirrhosis.
ALT, alanine-aminotransferase; AFP, a-fetoprotein; HBV, hepatitis B virus; HCV, hepatitis C virus.
platelet-associated antibodies in patients with liver cirrhosis (all
etiologies combines) and patients with ITP were signiﬁcantly
higher compared to healthy controls (p <0.005 and 0.0001,
respectively). Among the patients with HCV-related cirrhosis,
platelet-associated antibodies were found in seven cases
(19.4%) and serum circulating anti-platelet antibodies in nine
cases (25.0%). However, ﬁve patients had both platelet-associated
and circulating anti-platelet antibodies, while four patients only
expressed serum circulating antibodies and two cases expressed
platelet-associated antibodies only. In alcoholic cirrhosis, only
one case was found to have platelet-associated antibodies
(2.5%) and two cases had circulating anti-platelet antibodies
(5.0%); even in HBV-related cirrhosis, only one patient showed
serum anti-platelet antibodies (6.6%), while no cases presenting
platelet-associated antibodies were found. Statistical analysis
showed that the prevalence of platelet-associated antibodies
was signiﬁcantly higher in HCV+ cirrhosis compared to healthy
subjects (p <0.0064), alcoholic cirrhosis (p <0.018), and HBV+ cirrhosis (p <0.0001). Even the prevalence of circulating anti-platelet antibodies was signiﬁcantly higher in HCV+ cirrhosis
compared to normal controls (p <0.0021), alcoholic cirrhosis
(p <0.0152), and HBV+ cirrhosis (p <0.006).
B-cell monoclonality was found in eight (27%) of the HCV-positive patients, whereas no monoclonality was found in HBV
(p <0.004) or alcoholic patients (p <0.003). The ﬁve HCV+ patients
with both serum and platelet-associated antibodies and one out
of the four patients with serum circulating antibodies showed
B-cell monoclonality, while the remaining two HCV+ cases with
B-cell monoclonality also suffered from other autoimmune diseases i.e. Coombs-positive hemolytic anemia and autoimmune
thyroiditis. In all these cases, a skewed immunoglobulin usage
was found; in fact, these cases presented a preferential usage of
the monoclonal 51p1 gene. The sequence of the variable regions
(Complementary Determinant Regions) of the immunoglobulins
showed a region of a different length with a different nucleotide
and amino acid sequence in each case.
Table 2. Clinical and hematological proﬁles of the patients with ITP or aplastic
ALT, alanine-aminotransferase. n.d., not determined. All patients were HCV, HBV,
and HIV negative.
The possibility to compare hematological and hepatological
patients allowed us to better understand the complex mechanisms of platelet turnover. Thrombocytopenia is the main hematological disorder observed in liver cirrhosis, and this condition
has been traditionally attributed to hypersplenism , a condition accompanied by an increase in the sequestration and an
accelerated destruction of platelets. However, improvements in
thrombocytopenia have not been achieved through the use of
portal decompression, including portosystemic shunting, or even
after splenectomy. Alternatively, the insufﬁcient production of
thrombopoietin has also been proposed to be involved in the
pathogenesis of thrombocytopenia in patients with LC. Indeed,
thrombopoietin, the principal stimulating factor of megakaryothrombopoiesis, is produced by the liver and its production is
impaired in advanced liver failure [25–28]. The circulating level
of thrombopoietin is known to be regulated by a ‘‘sponge effect’’;
this refers to the fact that free levels are controlled through its
Journal of Hepatology 2011 vol. 54 j 894–900
binding to the thrombopoietin receptor that is mainly expressed
on bone marrow megakaryocytes and circulating platelets, while
thrombopoietin continues to be produced at constant rate by the
liver . Thus, the circulating levels of thrombopoietin depend
upon total receptor numbers (at least in subjects with normal
liver function). Indeed, it has been observed that following liver
transplantation, serum thrombopoietin concentrations return to
normal levels in conjunction with an increase in platelet count
[30–32]. However, several studies have showed that some
patients with liver cirrhosis have normal serum thrombopoietin
levels despite low platelet counts. Our data support the hypothesis of a role of low thrombopoietin levels in cirrhosis thrombocytopenia, since its serum levels were found to be below normal
levels in most cases. On the contrary, in a recent paper, Kajihara
et al.  did not ﬁnd any signiﬁcant differences in the thrombopoietin levels between patients with liver cirrhosis and controls.
These different ﬁndings could be explained by the different cutoff limit used for the platelets (below 100 109/L in our cases
and 150 109/L in the others) and by the different prevalence
of hepatocellular carcinoma (1% in our cases vs. 63% in those of
Kajihara et al.). There are also some reports indicating that
thrombopoietin receptors are expressed in fetal liver and in some
hepatocellular carcinoma (HCC) cell lines ; however, other
authors have observed increased platelet levels in HCC , thus
such a high prevalence of HCC should be avoided in physiological
Some rather old observations [36–39] exist indicating
increased levels of immunoglobulin G (IgG) bound to platelets
in patients with chronic liver disease that suggest the presence
of autoantibodies reactive to platelets. Immunoglobulin bound
to platelets is also found in idiopathic thrombocytopenic purpura
(ITP) [40,1], a typical autoimmune disease characterized by
increased platelet destruction mediated by autoantibodies
against several platelet surface antigens, including GPIIb-IIIa
(being the most common) and GPIb-IX [42,43]. ITP is characterized by accelerated platelet turnover (high glycocalicin levels)
and thus, in turn, by an increased platelet production (high reticulated platelet count). Glycocalicin is a fragment of the platelet
membrane glycoprotein Ib, and the glycocalicin index (GC index:
the plasma GC level corrected for the platelet count) is considered
to be a parameter of peripheral platelet turnover . Since it
was recently found that most patients with either liver cirrhosis
(99% of cases) or ITP (94% of cases) present circulating anti-GPIIbIIIa antibody-producing B cells , the immune-mediated
destruction of platelets could be considered as the main factor
determining thrombocytopenia in cirrhosis. Accordingly, we also
found that mean GC levels and GC indices were abnormally high
in patients with liver cirrhosis, suggesting an accelerated platelet
turnover, though not at the same levels as in ITP. In patients subgrouped according to the etiology of the cirrhosis, the glycocalicin
index was signiﬁcantly higher in the patients affected by HCV
compared to patients with other aetiologies. In line with these
data, the prevalence of anti-platelet antibodies was signiﬁcantly
higher in HCV+ cirrhosis than in HBV+ or alcoholic cirrhosis. This
difference is not surprising as it has been known for many years
that chronic HCV infection determines several immunological
(thyroid dysfunctions, Sjogren disease, and lichen ruber planus)
[46,47] and hematological disorders (monoclonal gammopathies
of undetermined signiﬁcance, non-Hodgkin’s lymphoma, Waldenström macroglobulinemia) [48–50]; although the strongest
association has been found with mixed cryoglobulinemia
[51,52]. The interaction of HCV envelope proteins and CD81, a
tetraspanin mainly expressed on B-lymphocytes, seems to be
the major mechanism of naïve B-cell activation . Note, the
absence of any difference between HCV-positive and negative
cases in auto-antibody production in Japanese patients can be
explained by the ethnical difference in the immunological
response. In fact, while HCV infection determines B-cell monoclonality in about 25% of chronically infected Caucasian patients,
this does not occur in Japanese subjects . Though the presence of detectable anti-platelet antibodies does not necessarily
imply an accelerated platelet destruction, in this series of cases
the subjects expressing anti-platelet antibodies (16 cases: seven
with detectable platelet-associated antibodies and nine cases
with serum circulating anti-platelet antibodies) showed a very
high GC index (2.51 ± 1.26, p <0.06 vs. HCV+ cirrhosis without
anti-platelet antibodies) and a very high level of reticulated
platelets (11,962 ± 10,258/mmc, p <0.05 vs. HCV+ cirrhosis without anti-platelet antibodies). On the basis of these results, as previously indicated in the ‘‘Results’’ section, in the HCV+ cirrhosis,
two groups can be identiﬁed: in the ﬁrst one (75% of the cases)
the platelet production and the platelet turnover overlaps HBV+
and alcoholic cirrhosis, while the second one (25% of the cases)
is characterized by high reticulated platelet, high GC index, high
incidence of anti-platelet antibodies, and high incidence of B-cell
monoclonality. Given the comparable severity of liver disease, the
serum thrombopoietin levels were low in both groups
(31.1 ± 16.7 vs. 30.5 ± 23.8 pg/ml; p: NS), this determines a platelet count signiﬁcantly (p <0.02) lower in the cases with high
platelet turnover (53,889 ± 11.241 109/L) than in the cases with
standard platelet turnover (72,631 ± 19,131 109/L).
A quantitative analysis of megakaryocyte concentrations in
bone marrow was not performed in this study. The identiﬁcation
of megakaryocytic cells based on morphological studies at the
light microscopy is difﬁcult and results are therefore often unreliable. In fact, while the identiﬁcation of the mature megakaryocytes at different stages of development (given their large size,
complex nuclear appearance, and typical cytoplasmic staining)
is easy, even using light microscopy, the identiﬁcation of immature megakaryocytic cells, which are small and do not have the
classical and obvious features of megakaryocytes, generally
requires ultrastructural studies and the use of electron microscopy for quantiﬁcation. Alternatively, dual-color immunoﬂuorescence staining and ﬂow cytometry seems to be a reliable method
to quantify bone marrow megakaryocytes at any stage of differentiation . On the basis of these considerations, megakaryocyte quantiﬁcation is even more difﬁcult when the platelet
turnover is elevated and, as a consequence, the immature fraction
of small megakaryocytes is likely to be larger than in normal subjects. Therefore, the analysis of marrow megakaryocytes by the
conventional light microscopy method is not reliable and the
comparison of marrow megakaryocyte density in liver cirrhosis
to that in ITP is unreliable.
In conclusion, patients with liver cirrhosis presented a normal
or decreased plasma TPO, an accelerated platelet turnover (on
the basis of a high glycocalicin index), and low or normal platelet
production (on the basis of the absolute reticulated platelet count).
Taken together, these ﬁndings indicate that cirrhotic thrombocytopenia is a multifactorial condition, involving both increased
platelet clearance in the periphery and impaired thrombopoiesis.
This is similar to what may happen in ITP since recent studies have
shown that anti-GPIIb/IIIa autoantibodies were able to suppress
Journal of Hepatology 2011 vol. 54 j 894–900
JOURNAL OF HEPATOLOGY
the production and maturation of megakaryocytes ‘‘in vitro’’, suggesting that the thrombocytopenia in ITP may not only result from
Fcc receptor-mediated platelet clearance but also from the suppression of megakaryogenesis [56,57]. This could explain the
unexceptional ﬁnding of low levels of reticulated platelets and normal glycocalicin indices in ‘‘classical’’ ITP subjects. In these cases,
the growth factor-mediated stimulation of megakaryopoiesis
might be expected to increase the platelet count. Indeed, initial
clinical trials of the investigational thrombopoietic agent
Eltrombopag (SB-497115, GlaxoSmithKline) a small-molecule,
non-peptide that acts as a thrombopoietin-receptor agonist inducing proliferation and differentiation of megakaryocytes, were
found to show promising responses in adults with chronic ITP
refractory to other treatments [58,59]. In liver cirrhosis, this agent
was found to increase the platelet count in HCV-related cirrhosis
and to facilitate antiviral treatment with fairly good results .
This treatment might also be effective in advanced liver failure
exhibiting thrombocytopenia and bleeding.
Finally, at least in Caucasians, a fraction of HCV-related liver
cirrhosis is characterized by an increased prevalence of autoimmune phenomena, including detectable levels of anti-platelet
antibodies  and this should be taken into account during
the follow-up of these patients.
Conﬂict of interest
The authors who have taken part in this study declared that they
do not have anything to disclose regarding funding or conﬂict of
interest with respect to this manuscript.
 Toghill PJ, Green S, Ferguson F. Platelet dynamics in chronic liver disease
with special reference to the role of the spleen. J Clin Pathol
 Aster RH. Pooling of platelets in the spleen: role in the pathogenesis of
‘hypersplenic’ thrombocytopenia. J Clin Invest 1966;45 (5):645–657.
 McCormick PA, Murphy KM. Splenomegaly, hypersplenism and coagulation
abnormalities in liver disease. Baillières Best Pract Res Clin Gastroenterol
 Mutchnik MG, Lerner E, Conn HO. Effect of portacaval anastomosis on
hypersplenism. Dig Dis Sci 1980;25 (12):929–938.
 McAllister E, Goode S, Cordista AG, Rosemurgy A. Partial portal decompression alleviates thrombocytopenia of portal hypertension. Am Surg 1995;61
 Alvarez OA, Lopera GA, Patel V, Encarnacion CE, Palmaz JC, Lee M.
Improvement of thrombocytopenia due to hypersplenism after transjugular
intrahepatic portosystemic shunt placement in cirrhotic patients. Am J
Gastroenterol 1996;91 (1):134–137.
 Sanyal AJ, Freedman AM, Purdum PP, Shiffman ML, Luketic VA. The
hematologic consequences of transjugular intrahepatic portosystemic
shunts. Hepatology 1996;23 (1):32–39.
 Gschwantler M, Vavrik J, Gebauer A, Kriwanek S, Schrutka-Kölbl C, Fleischer
J, et al. Course of platelet counts in cirrhotic patients after implantation of a
transjugular intrahepatic portosystemic shunt – a prospective, controlled
study. J Hepatol 1999;30 (2):254–259.
 Jabbour N, Zajiko 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 (11):2459–2465.
 Yanaga K, Tzakis AG, Shimada M, Campbell WE, Marsh JW, Stieber AC, et al.
Reversal of hypersplenism following orthotopic liver transplantation. Ann
Surg 1989;210 (2):180–183.
 Kaushansky K. Thrombopoietin: the primary regulator of megakaryocyte
and platelet production. Thromb Haemost 1995;74 (1):521–525.
 Eaton DL, de Sauvage EJ. Thrombopoietin. The primary regulator of
megakaryocytopoiesis and thrombopoieis. Exp Hematol 1995;25 (1):
 Jelkmann W. The role of liver in the production of thrombopoietin compared
with erythropoietin. Eur J Gastroenterol Hepatol 2001;13 (7):791–801.
 Radosavljevic M. Thrombocytopenia in liver disease. Can J Gastroenterol
2000;14 (Suppl. D):60–66.
 Ballard HS. Hematological complications of alcoholism. Alcohol Clin Exp Res
 Drews RE. Critical issues in hematology: anemia, thrombocytopenia, coagulopathy, and blood product transfusions in critically ill patients. Clin Chest
Med 2003;24 (4):607–622.
 Bedossa P, Poynard T. An algorithm for the grading of activity in chronic
hepatitis C The METAVIR Cooperative Study Group. Hepatology
 British Committee for Standards in Haematology General Haematology Task
Force. Guidelines for the investigation and management of idiopathic
thrombocytopenic purpura in adults, children and in pregnancy. Br J
 Camitta BM, Thomas ED, Nathan DG, Gale RP, Kopecky KJ, Rappeport JM,
et al. A prospective study of androgens and bone marrow transplantation for
treatment of severe aplastic anemia. Blood 1979;53 (3):504–514.
 Vicente V, Kostel PJ, Ruggeri ZM. Isolation and functional characterization of
the von Willebrand factor-binding domain located between residues His1Arg293 of the alpha-chain of glycoprotein Ib. J Biol Chem 1988;263
 Kunishima S, Hayashi K, Kobayashi S, Naoe T, Ohno R. New enzyme-linked
immunosorbent assay for glycocalinicin in plasma. Clin Chem 1991;37
 Beer JH, Büchi L, Steiner B. Glycocalicin: a new assay the normal plasma
levels and its potential usefulness in selected diseases. Blood 1994;83
 Chomczynski P, Sacchi N. The single-step method of RNA isolation by acid
guanidinium thiocyanate–phenol–chloroform extraction. Anal Biochem
 Peck-Radosavljevic M. Hypersplenism. Eur J Gastroenterol Hepatol 2001;13
 Peck-Radosavljevic M, Zacherl J, Meng YG, Pidlich J, Lipinski E, Langle F, et al.
Is inadequate thrombopoietin production a major cause of thrombocytopenia in cirrhosis of the liver? J Hepatol 1997;27 (1):127–131.
 Tanikawa K. Thrombopoietin and liver disease. J Gastroenterol Hepatol
 Kawasaki T, Souda Takeshita A, Kobayashi Y, Kikuyama M, Kikuyama F, et al.
Serum thrombopoietin levels in patients with chronic hepatitis and liver
cirrhosis. Am J Gastroenterol 1999;94 (7):1918–1922.
 Espanol I, Gallego A, Enriquez J, Rabella N, Lerma E, Hernández A, et al.
Thrombocytopenia associated with liver cirrhosis and hepatitis C viral
infection: role of thrombopoietin. Hepatogastroenterology 2000;47
 Wolber E-M, Jelkmann W. Thrombopoietin: the novel hepatic hormone.
News Physiol Sci 2002;17:6–10.
 Goulis J, Chau TN, Jordan S, Mehta AB, Watkinson A, Rolles K, et al.
Thrombopoietin concentrations are low in patients with cirrhosis and
thrombocytopenia and are restored after orthotopic liver transplantation.
Gut 1999;44 (5):754–758.
 Peck-Radosavljevic M, Zacherl J, Wichlas M, Sims P, Meng YG, Panzer S, et al.
Thrombopoietic cytokines and reversal of thrombocytopenia after liver
transplantation. Eur J Gastroenterol Hepatol 1999;11 (2):151–156.
 Peck-Radosavljevic M, Wichlas M, Zacherl J, Stiegler G, Stohlawetz P,
Fuchsjager M, et al. Thrombopoietin induces rapid resolution of thrombocytopenia after orthotopic liver transplantation through increased platelet
production. Blood 2000;95 (3):795–801.
 Kajihara M, Okazaki Y, Kato S, Ishii H, Kawakami Y, Ikeda Y, et al. Evaluation
of platelet kinetics in patients with liver cirrhosis: similarity to idiopathic
thrombocytopenic purpura. J Gastroenterol Hepatol 2007;22 (1):112–118.
 Columbyova L, Loda M, Scadden DT. Thrombopoietin receptor expression in
human cancer cell lines and primary tissues. Cancer Res 1995;55
 Hwang SJ, Luo JC, Li CP, Chu CW, Wu JC, Lai CR, et al. Thrombocytosis: a
paraneoplastic syndrome in patients with hepatocellular carcinoma. World J
Gastroenterol 2004;10 (17):2472–2477.
 Barrison IG, Knight ID, Viola L, Boots MA, Murray-Lion IM, Mitchell TR.
Platelet associated immunoglobulins on chronic liver disease. Br J Haematol
 Leone G, Agostini A, Mango G, Landolﬁ R, Valori VM, Bizzi B. Megathrombocytes, platelet regeneration time and platelet associated IgG in idiopathic
thrombocytopenic purpura and in thrombocytopenia associated with
chronic liver disease. Acta Haematol 1981;65 (1):40–47.
Journal of Hepatology 2011 vol. 54 j 894–900
 Graber D, Giuliani D, Leevy CM, Morse BS. Platelet-associated IgG in hepatitis
and cirrhosis. J Clin Immunol 1984;4 (2):108–111.
 Mueller-Eckhardt C, Kayser W, Mersch-Baumert K, Mueller-Eckhardt G,
Breidenbach M, Kugel HG, et al. The clinical signiﬁcance of plateletassociated IgG: a study on 298 patients with various disorders. Br J
Haematol 1980;46 (1):123–131.
 Kelton JG, Powers PJ, Carter CJ. A prospective study of the usefulness of the
measurement of platelet-associated IgG for the diagnosis of idiopathic
thrombocytopenic purpura. Blood 1982;60 (4):1050–1053.
 Cines DB, Blanchette VS. Immune thrombocytopenic purpura. N Engl J Med
 Hürlimann-Forster M, Steiner B, von Felten A. Quantitation of plateletspeciﬁc autoantibodies in platelet eluates of ITP patients measured by a
novel ELISA using the puriﬁed glycoprotein complexes GPIIb/IIIa and GPIb/IX
as antigens. Br J Haematol 1997;98 (2):328–335.
 Fujisawa K, O’Toole TE, Tani P, Loftus JC, Plow EF, Ginsberg MH, et al.
Autoantibodies to the presumptive cytoplasmic domain of platelet glycoprotein IIIa in patients with chronic immune thrombocytopenic purpura.
Blood 1991;77 (10):2207–2213.
 Chong BH, Keng TB. Advances in the diagnosis of idiopathic thrombocytopenic purpura. Semin Hematol 2000;37 (3):249–260.
 Kajihara M, Kato S, Okazaki Y, Kawakami Y, Ishii H, Ikeda Y, et al. A role of
autoantibody-mediated platelet destruction in thrombocytopenia in
patients with cirrhosis. Hepatology 2003;37 (6):1267–1276.
 Clifford BD, Donahue D, Smith L, Cable E, Luttig B, Manns M, et al. High
prevalence of serological markers of autoimmunity in patients with chronic
hepatitis C. Hepatology 1995;21 (3):613–619.
 Pawlotsky JM, Ben Yahia M, Andre C, Voisin MC, Intrator L, Roudot-Thoraval
F, et al. Immunological disorders in C virus chronic active hepatitis: a
prospective case-control study. Hepatology 1994;19 (4):841–848.
 Mazzaro C, Zagonel V, Monfardini S, Tulissi S, Pussini E, Fanni-Canelles E,
et al. Hepatitis C virus and non-Hodgkin’s lymphomas. Br J Haematol
 Terrier B, Sène D, Saadoun D, Ghillani-Dalbin P, Thibault V, Delluc A, et al.
Serum free light chain assessment in hepatitis C virus-related lymphoproliferative disorders. Ann Rheum Dis 2009;68 (1):89–93.
 Fiorilli M, Mecucci C, Farci P, Casato M. HCV-associated lymphomas. Rev Clin
Exp Hematol 2003;7 (4):406–423.
 Agnello V, Chung RT, Kaplan LM. A role for hepatitis C virus infection in type
II cryoglobulinemia. N Engl J Med 1992;327 (21):1490–1495.
 Pozzato G, Mazzaro C, Crovatto M, Modolo ML, Ceselli S, Mazzi G, et al. Lowgrade malignant lymphoma, hepatitis C virus infection, and mixed cryoglobulinemia. Blood 1994;84 (9):3047–3053.
 Rosa D, Saletti G, De Gregorio E, Zorat F, Comar C, D’Oro U, et al. Activation of
naïve B lymphocytes via CD81, a pathogenetic mechanism for hepatitis C
virus-associated B lymphocyte disorders. Proc Natl Acad Sci USA 2005;102
 Pozzato G, Burrone O, Baba K, Matsumoto M, Hijiiata M, Ota Y, et al. Ethnic
difference in the prevalence of monoclonal B-cell proliferation in patients
affected by hepatitis C virus chronic liver disease. J Hepatol 1999;30
 Law HK, Bol SJ, Palatsides M, Williams NT. Analysis of human megakaryocytic cells using dual-color immunoﬂuorescence labeling. Cytometry
 McMillan R, Wang L, Tomer A, Nichol J, Pistillo J. Suppression of in vitro
megakaryocyte production by antiplatelet autoantibodies from adult
patients with chronic ITP. Blood 2004;103 (4):1364–1369.
 Chang M, Nakagawa PA, Williams SA, Schwartz MR, Imfeld KL, Buzby JS,
et al. Immune thrombocytopenic purpura (ITP) plasma and puriﬁed ITP
monoclonal autoantibodies inhibit megakaryocytopoiesis in vitro. Blood
 Bussel JB, Cheng G, Saleh MN, Psaila B, Kovaleva L, Meddeb B, et al.
Eltrombopag for the treatment of chronic idiopathic thrombocytopenic
purpura. N Engl J Med 2007;357 (22):2237–2247.
 Stasi R, Evangelista ML, Amadori S. Novel thrombopoietic agents: a review of
their use in idiopathic thrombocytopenic purpura. Drugs 2008;68
 McHutchison JG, Dusheiko G, Shiffman ML, Rodriguez-Torres M, Sigal S,
Bourliere M, et al. Eltrombopag for thrombocytopenia in patients with
cirrhosis associated with hepatitis C. N Engl J Med 2007;357
 Weksler BB. The pathophysiology of thrombocytopenia in hepatitis C virus
infection and chronic liver disease. Aliment Pharmacol Ther 2007;26 (Suppl.
Journal of Hepatology 2011 vol. 54 j 894–900