Clinical Science (1985)68, 701-707
701
Fibrinogen derivatives and platelet activation products in
acute and chronic liver disease
R . D. HUGHES, D. A. LANE*, H. IRELAND*, P. G. LANGLEY,
A. E. S. GIMSON A N D R O G E R WILLIAMS
Liver Unit, King's College Hospital and School of Medicine and Dentistry, London, and *Department of Haematology,
Charing Cross Hospital Medical School, Hammersmith, London
(Received 4 October 1984110 January 1985; accepted I 6 January 1985)
Summary
1. The concentration in plasma of fibrinogen
derivatives fibrinopeptide A (FPA) and BPl-42
and the platelet release products 0-thromboglobulin (PTG) and platelet factor 4 (PF4) have
been determined in patients with acute and
chronic liver disease.
2. In 21 patients with fulminant hepatic failure
on admission in grade I11 or IV coma the plasma
FPA, Bpi-42, PTG and PF4 levels were significantly increased compared with those in normal
control subjects. On heparinization before haemoperfusion the FPA levels returned to the normal
range and during resin and charcoal haemoperfusion there were no significant changes in the
coagulation or platelet factors, except for a small
increase in FPA with charcoal haemoperfusion.
3. In ten patients with compensated chronic
liver disease there was a significant increase in
Bpl-42 and PTG levels but not FPA and PF4 as
compared with normal controls.
4. Interpretation of the results is complicated
by the possible reduced clearance of these proteins
as a result of renal failure in some of the patients
with fulminant hepatic failure and also by the
damaged liver itself. However, these results have
confirmed that disseminated intravascular coagulation can occur in both acute and chronic liver
disease.
Key words; fibrinogen, fibrinopeptide, haemoperfusion, liver disease, platelets.
Correspondence: Dr R. D. Hughes, Liver Unit,
King's College Hospital and School of Medicine
and Dentistry, Denmark Hill, London SE5 9RS.
Abbreviations: FPA, fibrinopeptide A; PF4,
platelet factor 4 ; PTG, 0-thromboglobulin.
Introduction
The liver plays a central role in haemostasis not
only because it is the site of synthesis of a
majority of the clotting factors, but also because
it clears many factors or their derivatives from the
circulation. Bleeding is a major complication of
both acute and chronic liver disease and many
studies have shown gross abnormalities in blood
clotting times, including prolongation of the
prothrombin time, as well as low levels of many
of the individual clotting factors [l-31. The main
natural inhibitors of the coagulation proteinases,
e.g. antithrombin I11 [4] and a,-antiplasmin [5],
are also found in reduced concentrations in liver
disease and, furthermore, abnormal proteins may
be produced, which leads to impaired function
[6]. Increased consumption of the clotting factors
is also found, probably as a result of the release of
thromboplastic material from the damaged liver. A
similar mechanism may be responsible for the reduced platelet count and impaired platelet aggregation [7] in liver disease. In acute liver failure,
where there is massive hepatic necrosis, these
changes are so marked that disseminated intravascular coagulation (DIC) has been reported with
thrombocytopenia, clotting factors <30% of
normal and the presence of fibrinogen degradation
products and fibrin strands in the circulation [8].
Recent developments have led to specific and
sensitive assays of activation products in vivo of
coagulation and fibrinolytic systems and also of
platelet a-granule proteins that are released during
the platelet release reaction in vivo. Radioimmuno-
R. D.Hughes et al.
702
assays of fibrinopeptide A (FPA), cleaved by the
action of thrombin on fibrinogen [9], and of
fibrinogen fragment Bpl-42 [lo], cleaved by
plasmin, can be used to detect any inbalance that
may occur in the coagulation and fibrinolytic
systems. The platelet a-granule proteins [l 1 , 121
/3-thromboglobulin (PTG) and platelet factor 4
(PF4) are sensitive markers of platelet activation.
The advantage of determining these activation
products is that they are normally cleared very
rapidly from the circulation and thus reflect ongoing dynamic events rather than those which have
occurred at an earlier stage. Hence in this study we
have assayed plasma levels of the fibrinogen derivatives and platelet release products in patients with
fulminant hepatic failure on admission and during
their treatment by extracorporeal haemoperfusion,
as well as in patients with compensated chronic
liver disease.
(Haemocol 100, Smith and Nephew Pharmaceuticals, Romford, Essex, U.K.). For resin haemoperfusion, heparin (mucous, Leo Labs. Ltd,Princes
Risborough, Bucks., U.K.) was administered
initially as an intravenous bolus (7500 i.u.),
followed by continuous infusion (E 2500 i.u./h)
into the extracorporeal circuit to maintain a whole
body activated clotting time (Haemachron International Technidyne Corp., NJ, U.S.A.) of 250300 s [13]. With charcoal haemoperfusion a bolus
(7500 or 5000 i.u.) of heparin was given with an
infusion (= 1000 i.u./h) as well as infusion of
epoprostenol (Wellcome Research Laboratories,
Beckenham, Kent, U.K.) in glycine buffer, pH
10.5, to achieve a blood concentration of 5 ng/ml
for platelet protection [ 141. Twenty-two normal
hospital and laboratory staff served as controls
with a mean age of 33 years (range 18-56, 13
females, nine males).
Patients
Methods
Twenty-nine patients with fulminant hepatic
failure were studied (20 as a result of paracetamol
overdose, seven after viral hepatitis, one after
halothane anaesthesia and one after antituberculous
drugs), of whom 21 were seen on admission in
either grade I11 or IV encephalopathy, and five
of these had renal failure as defined by a plasma
creatinine > 300 pmol/l. A further ten patients
had chronic liver disease (three alcoholic cirrhosis,
four primary biliary cirrhosis, one primary hepatocellular carcinoma, one cryptogenic cirrhosis, one
Wilson's disease). In every instance the diagnosis
was proven histologically and all this group were
well compensated biochemically. The results of
some of the routine tests of liver function and
haematology are given in Table 1.
Twelve of the fulminant hepatic failure patients
were also studied during treatment by haemoperfusion, in six by using albumin coated resin
columns (Immuno AG, Vienna, Austria) and six
by using polymer coated charcoal columns
Blood samples were collected into tubes on ice
either by clean venepuncture or else from the
arteriovenous shunt before haemoperfusion.
During haemoperfusion the blood samples were
collected from the inlet line to the columns, after
the entry of the heparin infusion. For the fibrinopeptide assays nine parts of blood were collected
into 1 part of heparin (1000 i.u./ml) and Trasylol
(1 000 unitslml) and platelet poor plasma was
obtained by centrifugation at 4OC.
FPA was determined by radioimmunoassay on
plasma from which fibrinogen had been removed
by bentonite, as already described in detail [15].
Fibrinogen fragment Bpl-42 was determined by
radioimmunoassay on ethanol precipitated plasma
[ 15, 161, essentially as described by Nossel et al.
[lo], except an unsolubilized second antibody
(donkey anti-rabbit, Sac Cel; Wellcome Reagents,
Dartford, Kent, U.K.) was used to separate free
from bound antigen [17]. Blood (2.5 ml) for
radioimmunoassay of the platelet release proteins
tests in patients studied
Values are given as means with ranges in parentheses.
TABLE 1. Results of biochemical and haematological
n
Fulminant hepatic
21
failure (on admission)
Chronic liver disease
10
Normal range
Age
(years)
35
(16-58)
51
(16-73)
Sex
121:
9M
5F
5M
AST
(i.u./l)
Bilirubin
(!.tmol/l)
Creatinine
(!.tmol/l)
(S)
Platelet
count
(x 109m
89
(31-195)
18
(14-27)
178
(87-438)
187
(46-400)
2406
(245-6060)
98
(40-300)
-
73
(16-210)
245
(67-937)
76
(53-89)
(12-14)
(200-3 5 0)
(7-40)
(3-20)
(45-105)
Prothroinbin
time
Fibrinopeptides in liver disease
was collected into a cold solution consisting of
100 pl of theophylline (5.4 mglml), 100 pl of
EDTA (100 mg/ml) and 100 pl of PGEl (1 pg/ml)
kept on melting ice. Plasma was recovered after
centrifugation at 15OOg for 30 min and assayed
for PTG as described by Ireland et al. [15]. PF4
was determined by minor modification of the
Abbott Laboratories radioimmunoassay (Basingstoke, Hants, U.K.).
Statistics
The data was analysed by using the Wilcoxon
Rank Sum Test and Test for Pair Differences and
also with linear regression as appropriate.
703
Results
Fulminant hepatic failure
The results for the different factors measured
are given in Table 2. In the 21 patients studied on
admission the plasma levels of both FPA and
Bol-42 were slightly but significantly increased, as
compared with the normal control values, with a
greater proportionate increase in Bpl-42 (Fig. 1).
The platelet release proteins PTG and PF4 were
also significantly increased in these patients and in
this case there was no significant proportionate
change compared with the controls (Fig. 2), as
indicated by the plasma /3TG/PF4 ratio (mean
TABLE 2. Values for the plasma levels of fibrinogen derivatives and platelet activation products in acute
and chronic liver disease
Values shown are means with ranges in parentheses. Significance: **P< 0.01 ; *P< 0.05.
Fulminant hepatic
failure
Chronic liver
disease
Normal subjects
21
*2.64 (0.39-8.44)
**10.24 (1.53-36.0)
**4.09 (0.86-10.0)
**3.51 (1.16-7.19)
10
1.01 (0.36-2.47)
**4.87 (1.07-11.7)
**2.31 (1.58-4.17)
0.16 (0.08-0.35)
22
1.05 (0.54-2.34)
1.57 (0.65-4.00)
0.69 (0.43-1.57)
0.23 (0.08-0.65)
0
:
0
IWF*
Control
0
ia.
CLD
FHF** Control
i.
CLD**
FIG. 1. ( a ) Plasma fibrinopeptide A levels in patients with fulminant hepatic failure
(FHF), control subjects and chronic liver disease (CLD).( b ) Plasma Bpl-42 antigen
levels in patients with fulminant hepatic failure (FHF), control subjects and chronic
liver disease (CLD).* P< 0.05,**P < 0.01 vs control.
I
R. D.Hughes et al.
704
1 0 (~a )
8t
3
I )
$
2
v
.
0
0
F W * * Control
CLD**
i
FHF** Control
CLD
FIG. 2 . ( a ) Plasma 0-thromboglobulin levels in patients with fulminant hepatic failure
(FHF), control subjects and chronic liver disease (CLD). ( b ) Plasma platelet factor 4
levels in patients with fulminant hepatic failure (FHF), control subjects and chronic
liver disease (CLD). * * P < 0.01 vs control.
3.51, range 1.16-7.19; cf. mean of control group
3.73, range 1.42-7.40).
There was a significant linear correlation
between plasma FPA and PTG (r = 0.72, P<O.Ol),
Bpl-42 and PF4 (r = 0.80, P < 0.001), and PTG
and PF4 (r = 0.63, P < 0.01). The only correlation
found between these parameters and the routine
biochemical and haematological measurements was
between FPA and the plasma creatinine (r = 0.65,
P < 0.01).
111 x 109/1 (range 50-213 x 109/l). With charcoal
haemoperfusion there was a gradual increase in
FPA over 4 h (Fig. 4) and the mean level at the
end of haemoperfusion was different from the
initial value (P=0.05). Plasma Bpl-42 rose for
the first 2 h and then fell, but there was considerable variation in these values. The plasma
Effect of haemoperfusion
l6I
The initial samples in these studies were taken
after the bolus administration of heparin (5000 or
7500i.u.) to the patients, and the plasma FPA
levels (mean 0.96 pmol/ml, range 0.37-3.53 pmol/
ml) in this group were significantly lower (P<0.02)
after heparin administration than in the patients
with fulminant hepatic failure on admission
(mean 2.64 pmol/ml, range 0.39-8.44 pmol/ml)
and were similar to the normal values. During resin
haemoperfusion there was no significant change
over 4 h in levels of FPA, Bol-42 or PTG (Fig. 3)
(PF4 was not measured in the haemoperfusion
studies). The mean initial platelet count was
118x 109/1 (range 74-210) and after 4 h was
I2I
0
1
I
T
I
1
2
3
4
Perfusion (h)
FIG. 3. Plasma fibrinopeptide A (o), BP1-42 antigen (m) and 0-thromboglobin (A) levels in patients
with fulminant hepatic failure during resin haemoperfusion; means f SE, n = 6 .
Fibrinopeptides in liver disease
0
1
2
Perfusion (h)
3
4
FIG. 4. Plasma fibrinopeptide A (o), Bpl-42 antigen )(.
and 0-thromboglobulin (A) levels in
patients with fulminant hepatic failure during
charcoal haemoperfusion; means SE, n = 7. * P =
0.05 vs initial value.
*
levels of PTG remained constant during haemoperfusion as did the platelet count, which was
163 x 109/1 (range 99-253 x 109/1) initially, and
174 x 109/1(range 97-268 x 10’11) after 4 h.
Chronic liver disease
In the ten patients with chronic liver disease
there was a significant increase in Bol-42 levels
as compared with the normal controls (Fig. l),
but not in FPA levels. The PTG level was also
significantly increased in these patients, as was the
PTG/PF4 ratio (mean 32.45, range 5.94-189,
P < 0.001) (Fig. 2), but no similar increase was
found in PF4 levels. No significant correlations
were found between the FPA, Bbl-42, PTG and
PF4 in these patients, though there was a significant correlation between PF4 and the blood
platelet count (r = 0.82, P < 0.01) but not with
the other routine biochemical and haematological
measurements.
Discussion
On first consideration it might seem that these
results with elevated plasma concentrations of the
activation products of coagulation and fibrinolysis
(FPA and Bpl-42) and platelet release products
(PTG and PF4) would support the view that acute
and chronic liver disease induces disseminated
intravascular coagulation. In several other clinical
situations such findings have been taken to be
good evidence for an imbalance in haemostasis.
However, the plasma concentration of a protein
or protein fragment depends on its rate of pro-
705
duction, * volume of distribution and rate of
elimination from the circulation, and it is likely
that in the present patients one or more of these
factors is altered. Thus, in renal failure, investigations both in patients [18] and in experimental
animals [19, 201 have suggested that the plasma
concentration of Bpi-42, PTG and to a lesser
extent FPA may be influenced by decreased
elimination through the kidneys, as well as an
increased rate of production. Renal failure is an
important complication of acute liver failure and
was present in five of the 21 patients studied on
admission, and it is possible, therefore, that in
these cases the elevation of the coagulation and
platelet products was caused in part by the renal
impairment. Some support for this is provided
by the significant correlation found between
plasma FPA and the serum creatinine levels. PF4
levels in plasma are not influenced by renal failure,
but the possibility should be considered that the
liver may be a site of catabolism of this protein
and thus the reduced hepatic function could lead
to accumulation. Impaired clearance of denatured
protein due to defective reticuloendothelial cell
function has been found in fulminant hepatic
failure[21].
In spite of the above reservations the results
indicate increased activity of both thrombin and
plasmin in the patients. FPA levels in plasma are
not so affected by renal failure as are Bol-42 and
PTG and an increased plasma level is likely to be
caused at least in part by increased thrombin
action on fibrinogen. Support for this is provided
by the haemoperfusion studies. Administration of
heparin at the initiation of haemoperfusion
resulted in a significant fall in mean FPAlevels to
the normal control levels and, because heparin can
rapidly accelerate thrombin inhibition, such a
decrease in FPA levels is usually taken to be good
evidence for increased thrombin activation in
plasma. Continuous heparin administration during
haemoperfusion using the resin column suppressed
thrombin activity and FPA generation, which is
reasonable as, during haemodialysis of patients
with renal failure, less heparin was used without
fibrinopeptide release [22, 231. However, a small
increase in FPA levels was observed during
charcoal haemoperfusion, where less heparin was
administered. Patients with fuliiinant hepatic
failure have reduced antithrombin 111 levels and
thus may require more heparin to obtain an
adequate response. Also the PGIz infusion may
have affected the whole blood activated clotting
time in the extracorporeal circuit [24], so that
the value did not reflect the level of anticoagulation in the patient. There was no evidence of
platelet activation during resin haemoperfusion
706
R. D.Hxghes et al.
due to the albumin-coating and the effect of PGIz
during charcoal haemoperfusion.
We were unable to demonstrate an elevated
FPA in chronic liver disease, a result that appears
to be at variance with that of the study by
Coccheri et al. [25],but this may be because our
patients were stable and compensated biochemically. However, in these patients Bpl-42and PTG
levels were significantly elevated. It is unlikely
that these increases were caused by impaired renal
clearance as the serum creatinine levels were
normal in these patients and thus increased
plasmin and platelet releasing activities are
indicated. It is perhaps surprising to find the
slightly increased Bpl-42 levels in the patients
without an apparent increase in FPA levels, as to
date there is not firm evidence to suggest that
plasmin acts directly on fibrinogen rather than
fibrin. There are two possible explanations for this
result. Firstly a greater molar rise of Bpl-42than
FPA has been observed in human disseminated
intravascular coagulation [lo] and thus small
amounts of FPA released may not be as readily
detected as the secondary rise in Bpi-42 resulting
from plasmin action. Secondly, the liver is the
major organ of catabolism of the fibrinolytic
system activator, tissue plasminogen activator, and
increased plasma levels of tissue plasminogen
activator due to reduced clearance in liver disease
[26] may increase the level of Bpl-42 in plasma
without the need for a prior increase in fibrin
formation. Our results are consistent with those of
Stein & Harker [27], who proposed that in
patients with cirrhosis there is a complex process
with platelet consumption on incompletely endothelialized surfaces in the liver, leading to local
fibrin deposition and secondary activation of the
fibrinolytic system.
In summary, interpretation of increased plasma
levels of fibrinogen and platelet activation
products in liver disease is complicated because the
clearance of these products may be impaired.
Nevertheless, our results have shown that during
both acute and chronic liver disease, there is
platelet activation and increased fibrinolysis,
which is likely to be the result of low-grade intravascular coagulation.
Acknowledgments
The work in the Liver Unit forms part of MRC
support for development of artificial liver support.
References
1. Roberts, H.R. & Cederbaum, A.I. (1972) The liver
and blood coagulation: physiology and pathology.
Gastroenterology, 63, 297-320.
2. Bloom, AL. (1975) Intravascular coagulation and the
liver. British Journal of Haematology, 30, 1-7.
3. Lechner, K., Niessner, H. & Thaler, E. (1977)
Coagulation abnormalities in liver disease. Seminars
in Thrombosis and Haemostasis, 4, 40-56.
4. Duckert, F. (1973) Behaviour of antithrombin I11 in
liver disease. Scandinavian Journal of Gastroenterology, 8 (Suppl. 19), 109-112.
5 . Aoki, N. & Yamanaka, T. (1978) The or,:plasmin
inhibitor levels in liver diseases. Clinica Chimica Acta,
84,88-105.
6. Lane, D.A., Scully, M.F., Thomas, D.P., Kakkar, V.V.,
Woolf, I.L. & Williams, R. (1977) Acquired dysfibrinogenaemia in acute and chronic liver disease.
British Journal of Haematology, 35,301-308.
7. Rubin, M.H., Weston, M.J., Langley, P.G., White, Y.S.
& Williams, R. (1979) Platelet function in chronic
liver disease: relationship to disease severity. Digestive
Diseases and Sciences, 24, 197-202.
8 . Rake, M.O., Flute, P.T., Pannell, G. & Williams, R.
(1970) Intravascular coagulation in acute hepatic
necrosis. Lancet, i, 5 3 3-5 3 7.
9. Nossel, H.L., Yudelman, I., Canfield, R.E., Butler,
V.P., Jr, Spanonidis, K., Wilner, G.D. & Qureshi,
G.D. (1974) Measurement of fibrinopeptide A in
human blood. Journal of Clinical Investigation, 54,
43-53.
10. Nossel, H.L., Wassner, J., Kaplan, K.L., La Gamma,
K.S., Yudelman, I. & Canfield, R.E. (1979) Sequence
of fibrinogen proteolysis and platelet release after
intrauterine infusion of hypertonic saline. Journal of
Clinical Investigation, 64, 137 1-1378.
11. Ludlam, C.A., Moore, S., Bolton, A.E., Pepper, D.S.
& Cash, J.D. (1975) The release of a human platelet
protein measured by radioimmunoassay. 7hrombosis
Research, 6,543-548.
12. Kaplan, K.L. & Owen, J. (1981) Plasma levels of pthromboglobulin and platelet factor 4 as indices of
platelet activation in vivo. Blood, 57, 199-202.
13. Bihari, D., Hughes, R.D., Gimson, A.E.S., Langley,
P.G., Ede, R.J., Eder, G. & Williams, R. (1983)
Effects of serial resin haemoperfusion in fulminant
hepatic failure. International Journal of Artificial
Organs, 6,299-302.
14. Gimson, A.E.S., Langley, P.G., Hughes, R.D.,
Canalese, J., Mellon, P.J., Williams, R., Woods, H.F.
& Weston, M.J. (1980) Prostacyclin to prevent
platelet activation during charcoal haemoperfusion in
fulminant hepatic failure. Lancet, i, 173-175.
15. Ireland, H., Lane, D.A., Wolff, S. & Foadi, M. (1982)
In vivo platelet release in myeloproliferative disorders.
Thrombosis and Haemostatis (Stuttgart), 48,41-45.
16. Lane, D.A., Wolff, S., Ireland, H., Gawel, M. & Foadi,
M. (1983) Activation of coagulation and fibrinolytic
system following stroke. British Journal of Haemotology, 53, 655-658.
17. Lane, D.A., Ireland, H., Wolff, S., Grant, R., Jennings,
S. & Allen-Mersh, T. (1982) Plasma concentrations of
fibrinopeptide A, fibrinogen Bpl-42 and p-thromboglobulin following total hip replacement. Thrombosis
Research, 26, 111-118.
18. Lane, D.A., Ireland, H., Knight, I., Wolff, S., Kyle,
P. & Curtis, J.R. (1984) The significance of fibrinogen
derivatives in plasma in human renal failure. British
Journal of Haematology, 56,25 1-260.
19. Lane, D.A., Siodlak, M., Thompson, E. & AllenMersh, T.G. (1982) Clearance of human desaminotyrosyl fibrinopeptide A from the rat circulation: role
Fibrinopeptides in liver disease
of kidney and proteolytic enzymes. Thrombosis
Research, 26,13-82.
20. Lane, D.A., Markwick, J., Thompson, E. & Ireland,
H. (1984) Experimental evidence for renal catabolism
of fibrin fragment p15-42. Thrombosis Research, 33,
57 1-5 8 1.
21. Canalese, J., Gove, C.D., Gimson, A.E.S., Wilkinson,
S.P., Wardle, E.N. & Williams, R. (1982) Reticuloendothelial system and hepatocyte function in fulminant hepatic failure. Gut, 23,265-269.
22. Wilhemsson, S., Asaba, H., Gunnersson, B., Kudryk,
B., Robinson, D. & Bergstrom, J. (1982) Measurement
of fibrinopeptide A in the evaluation of heparin
activity and fibrin formation during haemodialysis.
Clinical Nephrology, 15,252-258.
23. Ireland, H., Lane, D.A. & Curtis, J.R. (1984) Objective assessment of heparin requirements for haemodialysis in man. Journal of Laboratory and Clinical
Medicine, 103,643-652.
707
24. Bunting, S., O'Grady, J., Fabiani, J.N., Terrier, E.,
Moncada, S., Vane, J.R. & Dubost, C. (1981) Cardiopulmonary bypass in man: effect of prostacyclii.
In: Clinical Pharmacology of Prostacyclin, p. 181.
Ed. Lewis, P.J. & O'Grady, J. Raven Press, New York.
25. Coccheri, S., Mannucci, P.M., Palareti, G., Gervasoni,
W., Poggi, M. & Vigano, S. (1982) Significance of
plasma fibrinopeptide A and high molecular weight
fibrinogen in patients with liver cirrhosis. British
Journal of Haematology, 52,503-509.
26. Fletcher, A.P., Biederman, O., Moore, D., Alkjaersig,
N. & Sherry, S. (1964) Abnormal plasminogenplasmin system activity (fibrinolysis) in patients with
hepatic cirrhosis: its cause and consequences. Journal
of Clinical Investigation, 43,68 1-695.
21. Stein, S. & Harker, L.A. (1982) Kinetic and
functional studies of platelets, fibrinogen and
plasminogen in patients with hepatic cirrhosis. Journal
of Laboratory and Clinical Medicine, 99,217-230.