α2antiplasmin and disseminated intravascular coagulation in liver cirrhosis

THROMBOSIS RESEARCH 37; 287-294, 1985 0049-3848/85 $3.00 + .OO Printed in the USA. Copyright (c) 1985 Pergamon Press Ltd. All rights reserved.

QANTIPLASMIN

F. Marongiu,

AND

DISSEMINATED INTRAVASCULAR IN LIVER CIRRHOSIS

A.M.

Mamusa, G. Mameli, G. Mulas, L. Demelia and L. Contu Institute of Internal Medicine, University of Cagliari,Italy

COAGULATION

A. Solinas.

(Received 26.4.1984; Accepted in revised form 20.10.1984 by Editor G. De Gaetano)

ABSTRACT a,Antiplasmin (c&AP) Subnormal concentrations of in liver cirrhosis may be due to an impaired hepatic synthesis and/or to a fibrinolysis activation in disseminated intravascular coagulation (DIC).In order to clarify this problem, in 26 cirrhotic patients (15 compensated and 11 decompensated) @AP plasma activity and plasma Fibrinopeptide A (FPA) were measured.Serum albumin, p-Cholinesterase (pand Fibrinogen Degradation ProCHE), Fibrinogen ducts (FDP) were also carried out.Our data showthat 0~~ AP and FPA were equally abnormal in compensated and decompensated cirrhosis.The significant negati Q,AP and FPA as ve correlation obtained between well as the lack of correlation between a2AP and albumin, IY,*AP and p-CHE in both groups suggests that, in our patients, (X,AP decrease may be due to a fibrinolysis activation induced by a DICwhich appears chronic since Fibrinogen and FDP werenormal.These findings are in agreement with the resul ts obtained in the four subgroups a posteriorisel?c cted on the basis of FPA levels: Ct,AP in subgroups with high FPA was significantly di.fferent from con trols while it did not differ in subgroups withno-rma1 FPA.

INTRODUCTION a2 Antiplasmin (o(~ AP) is a single chain glycoprotein with a molecularweightof 65000-70000 d (l).It forms a very stakey words:

disseminated intravascular a2Antiplasmin, A, liver cirrhosis. lation, Fibrinopeptide 287

coagu-

288

a2 AP AND DIC IN LIVER CIRRHOSIS

Vo1.37, No.2

ble 1:l stoichiometric complex with plasmin and is the primary fi brinolysis inhibitor in blood (2,3).Antiplasmin activity has been reported to be associated with platelets (l-3%) (4), but the main part of this protein is in the plasma and the liver has been suggested as the side of synthesis of as AP (5,6).However it is uncertain if the low c(~AP values found in liver cirrhosis are due to an hepatic synthesis defect and/or to a fibrinolysis activation in disseminated intravascular coagulation (DIG) (7,8).ln order to clarify the mechanism underlying decreased aZAP in liver cirrhoCI,AP concomitantly with a sensitive indicator sis, we measured of thrombin activity in vivo such as Fibrinopeptide A (FPA).Moreover serum albumin and p-Cholinesterase (p-CHE) were chosen asindicators of hepatic synthetic activity.

MATERIALS

AND METHODS

Patients.Twenty-six patients (21 men and 5 women, mean age 53.1 + 11.2 years) with histologically proven liver cirrhosis (12 alcoho lit and 14 postnecrotic) and a group of 20 age-and sex-matchedhe althy subjects were studied.The patients were classified in twocirrhosis (group A, n=15) and decompensated groups: compensated cirrhosis (group 8. n=ll).Patients were included in the latter group on the basis of the following criteria: a) overtascites and/ or jaundice,b) serum albumin level ~33 g/l and c) p-CHE <2OOOU/l. All patients were free of bleeding for at least two months and no ne had active thrombotic or inflammatory disease which have been ruled out on the basis of clinical and conventional laboratory pa ramethers.Any drug influencing blood coagulation was given for at least two weeks prior to blood samp1ing.A posteriori the patients were classed into four subgroups as follows: compensated cirrhosis with normal and high FPA (subgroup 1 and 2), decompensated cirrhosis with normal and high FPA (subgroup 3 and 4). Blood samples.Blood was drawn with a minimal stasis intoaplastic syringe from the cubital vein and added to two plastic tubes containing sodium citrate solution 3.8% ( UzAP assay) and EDTA,aprotinine in physiological saline (FPA assay) respectively.The tubes for both assays were immediately inverted several times and COOled in an ice bath.The total time of blood sampling was 40-55 sec. The samples were centrifuged within 30 min. at 2000 g and 4 'C for was immediately separated,then 20 and 60 min. respectively.Plasma frozen until determination could be performed. &AP plasma acivity was measured byami Laboratory investigation. dolytic method (chromogenic substrate S-2251, Kabi, Sweden) (9). The coefficient of variation (CV) was 3.5%.FPA was performed by a radioimmunoassay technique (BYK Mallinckrodt) (lO,ll), CV: 4.8 %. In addition serum albumin by electrophoresis technique (Gelman llf according to Clauss (121, CV: 3%, struments), CV: 53, Fibrinogen meFibrinogen degradation products (FDP) by a semiquantitative Wellcome) and p-CHE (Boehringer Biochethod (Thrombo-Wellcotest, mia Robin), CV: 2.1 %, were carried out.

Vo1.37,

a2

No.2

289

AP AND DIC IN LIVER CIRRHOSIS

statistical analgsis.Because of the non normal distributionof FPA values for cirrhotic patients, correlation coefficient r was perc%,AP values and the log of FPA concentration. In formed between addition distributions are described by median values and total range.For FPA values Wilcoxon's rank sum test was used to ebramine the significance of group differences. OlzAP was normally distribu ted and thus vralues are expressed as mean and SD;analysis of \!a-riance and the test of Newman-Keuls (1.3) were used in the r~aluation of azAP data.

RESULTS

c(, AP and are shown

FPA values in Fig. 1.

A

in

compensated

and

decompensated

.

A A

0

.G

A.0

A

.o

cirrhosis

@A

A

0 0

A

0

0.

A.

A 0. A

?? A A A

FIG. QzAP and FPA in (A) cirrhosis.FPA rithmic scale.

compensated values are

1 (01 and plotted

decompensated using a loga

AP AND DIC IN LIVER CIRRHOSIS

a2

290

Vo1.37,

No.2

According to the analysis of variance method azAP in group A and B was lower than controls (our reference range:80-119%)(F= was not a significant difference 20.4, df 2,43, pcO.OOl);there between group A and B (Table 1).

TABLE a, AP Values (mean + SD) Normal Subjects. -

1

in Cirrhotic

Patients

AP %

group

A Compensated

cirrhosis

B Decompensated C Normal

and

cirrhosis

2

14.6

67.8

2

19.1

101.1

subjects

A vs C peO.01

A vs B ns

79.6

-+ 11.6

B vs C ~~0.01

FPA levels in group A as well as in group B were significantly different from that of the controls:the difference was more pronunced between group A (pcO.001) than between group B (~~0.05). Group A was not different from group B (Table 2).

TABLE FPA Values (median and total tients and Normal Subjects.

2 range)

FPA

group A Compensated

cirrhosis

B Decompensated C Normal

A vs B ns

The

in Cirrhotic

cirrhosis

subjects

A vs C pcO.001

Pa-

(nglml)

5.2

(1.4-80)

5.3

(1.3-25.5)

2.6

(0.6-4.3)

B vs C ~~0.05

levels of FPA exceeded the upper limit of normal range in 7 of the patients in group A and in 5 (46%) in group B. decreaAmong the compensated cirrhosis 7 subjects (46%) exibited (46%)

Vo1.37,

a2

No.2

AP AND DIC IN LIVER CIRRHOSIS

291

sed aa.AP levels as compared with 6 subjects (54%) among the decompensated cirrh0sis.A significant inverse correlation was foa,AP and the log of FPA in group A and B (Table 3). und between

TABLE

Correlation FPA Values

Coefficients in Cirrhotic

3

between Patients.

ti2AP and the

Log of

r

group

Compensated

cirrhosis

Decompensated

cirrhosis

P

-0.719

CO.01

-0.773

CO.01

‘-

Fibrinogen and those of serum

FDP were normal in both groups.These albumin and p-CHE are shown in Table

TABLE

IS. albumin

(g/l)

(U/l) Fibrinogen FDP

with

4

Values (mean + SD) of Serum Albumin, gen and FDP in Cirrhotic Patients. reference values

data 4.

p-CHE,

group A

37-52

35.513.1

3800-11000

3441.2~1556.3

Fibrino-

group B 27.423.1

(g/l)

2-4.5

2.8+0.6 -

2.6+0.? -

@g/ml)

c 10

c 10

c 10

There was not a significant correlation between a,AP and albumin, a:!AP and p-CHE or between FPA and albumin, FPA andp-CHE in both groups.Subgroups with high FPA were significantly different from controls while subgroups with normal FPA did not.Subgroups I and 2 did not differ from subgroups 3 and 4 respectively.Moreover a significant difference was found between subgroups 1 and 2 as well as between subgroups 3 and 4. d2AP in subgroup 2 was lower than in subgroup 3 but this difference was not signifi cant.This fact may be due to the use of a statistical conservati ve test (Newman-Keuls) which has been chosen for a more reliable evaluation of the data (Table 5).

11x2AP AND DIC IN LIVER CIRRHOSIS

292

TABLE Evaluation died

of

d.,AP

Vol.37,

5

Differences

in

All

Subgroups

1 Compensated with normal 2 Compensated with high

1 vs

1 vs

68.8

-t

10.5

80.1

-+

16.1

cirrhosis (n=5)

53.1

f

7.3

(n=20)

101.1

cirrhosis FPA (n=6)

subjects

2 pco.05 3 ns 4 pco.01 5 ns

-+ 10.9

cirrhosis FPA (n=7)

4 Decompensated with high FPA 5 Normal

89.1

cirrhosis FPA (n=8)

3 Decompensated with normal

1 vs

Stu -

AP % (M+sD)

group

1 vs

No.?

2 vs 2 vs 2 vs

3 ns 4 ns 5 pco.01

3 vs 3 vs 4 vs

-+ 11.6

4 pco.05 5 ns 5 pco.01

DISCUSSION

Our data show that cxIAP and FPA were equally abnormal in compensated and decompensated cirrhosis.The latter finding confirms the that in liver cirrhosis data of Coccheri et al (14) who noticed -high FPA levels were not related to the severity of disease.Inad m2AP is inversely related to FPA levels in in our work, dition, if one accepts that FPA that, group A and B. This fact suggests is a marker of thrombin activity in vivo, the decrease in (X*AP seems to be a secondary phenomenon since this protein is a sensitive indicator of fibrinolysis activation (15).In fact enhanced has been postulated in fibrinolysis may be due to DIC (16) which liver cirrhosis (17).0n the other hand fibrinolysis activation may be also due to an impaired clearance by the liver of plasmino gen activators (18) aswellasto a diminished synthesis of fibrin5 lysis inhibitors (19);particularly a decreased level of histidine rich glycoprotein might play a role in this regard (20).Hence the that DIC and fibrinolysis activation are linked in possibility our patientsis strong.Furthermore these our results are in agreement with the data obtained in the four subgroups a posteriori se lected:aPAP was not different from controls in compensated or decompensated cirrhosis without evidence of DIC;but when CiaAP in compensated and decompensated cirrhosis with high FPA was compaits levels significantly fell in both groups. red with controls, Thus it could be tought that C(,AP decrease in liver cirrhosis may

~12 AP AND DIC IN LIVER CIRRHOSIS

vo1.37, No.2

293

be due to DIG more than to an hepatic synthesis defect.?loreover a2AP levels in compensated and decompensated cirrhotic patients with high FPA significantly differed from those with normal FPA while any difference was found between subgroups with normal or high FPA.These findings as well as the lack of a significantcorCI,AP and p-CHEmay azAP and albumin or between relation between that DIC could play an important role in defurther demonstrate creasing d,AP.However DIC appears to be mild in our patients since Fibrinogen and FDP were normal.In contrast Samama et al (8) fo-und a decrease in (X~AP activity in cirrhotic patients with normal distinction between cirrhosis with o I' increased FDP.However the and without DIC does not appear satisfactory because FDP values and indicators of thrombin activity in vivo were not given.In the since the role of DIC in liver cirrhosis has been considered end. important in the appearance of bleeding (21)? c(,AP and FPA may be two useful diagnostic tools for the assessment of the bleedingha aard in these patients.

REFERENCES I .

COLLEN, D. and WIMAN, B.Fast-acting plasma.Blood, 51, 563-569, 1978.

2.

MOROI, M. and AOKI, 2-plaemin inhibitor bitor which inhibits Chem.. 251, 5956-5965,

plasmin

inhibitor

in human

N.Isolation and characterization of alpha from human p1asma.A novel proteinase inhi activator-induced clot 1ysis.J.-Biol.

1976.

,? .

MULLERTZ, S. min in human

4.

JOIST, H.Platelet 962, 1978.

5.

HOGSTORP, H. and SALDEEN? liver cells.Thromb. Res.,

6.

AOKI, liver

/.

KNOT, E.A.R., DRIYFHOUT, H.R.,KAHLE', L.H.,ten CATE. J.W., TIJTGAT, G.N.Kinetic studies of radiolabelled a,-antiplasmin and antithrombin III simultaneously in patients with liver sease.Haemostasis, 110, 1984. 1411,

and CLEMMENSEN, plasma.Biochem. and

I.The primary inhibitor J..159, 545-553. 1976.

fibrinolysis.Thromb.

T.Synthesis 28, 19-25.

Haemost..

of

of

plas-

3,

?55-

a,-antiplasmin

by rat

1982.

N. and YAMANAKA, T.The &,-plasmin inhibitor levels disease.Clin. Chim. Acta, 84, 99-106. 1978.

in

di -

8.

SAMAMA, M., SCHLEGEL, N., CAZENANE, B., HORRELOU, !l.H.,COSARD. J ., CASTEL, M. and DOUENIAS, R. szantiplasmin assa!.:amidolytic and immunological method. Critical evaluation.Results in a clinical material.In:Synthetic substrate in clinical blood coagulation assays.H.R. Lijnen, D. Collen and M. \'estraete Martinus Nijhoff publishers. (eds.).The Hague, Boston. London. 1980, pp 93-101.

0.

TEGEZR-NILSSON, of a new rapid

A.C., FRIBERGER, plasmin inhibitor

P., in

GYZANDER, E.Determination human blood by means of

a

a 2 AP AND DIC IN LIVER CIRRHOSIS

294

plasma specific tripeptide 37, 403-404, 1977. 10.

substrate.Scand.

Vol.37,

J.

Clin.

No.2

Lab.Inv.,

NOSSEL, H.L., YULDEMAN, I., CANFIELD, R.E., BUTLER, V.P.jr, SPANONDIS, K., WILNER, G.D. and QURESHI, G.D.Measurement of fibrinopeptide A in human b1ood.J. Clin. Inv., 54, 43-53,

1974. 11.

KOCKUM, C. and fibrinopeptide bentonite.Thromb.

12.

CLAUSS, A.Geriunungphysiologische Schnellmethode mung des Fibrinogens.Acta Haematol., 17, 237-246,

FREBELIUS, S.Rapid radioimmunoassay of A-Removal of cross-reacting fibrinogen 589-598, 1980. Res., 2, zur

human with

Bestim-

1957.

13.

SNEDECOR, tion.The

14.

COCCHERI, S., MANNUCCI, P.M., PALARETI, G., GERVASONI, W., POGGI, M.and VIGANO' S.Significance of plasma fibrinopeptide A and high molecular weight fibrinogen in patients with liver cirrhosis.Br. J. Haematol., 2, 503-509, 1982.

15.

MUSUMECI, V., ZAPPACOSTA, B., CARUSO, A., ZUPPI, C., TUTINEL LI,F.Laboratory control of urokinase therapy at low or mode316-321,1983. rate dosages.An in-vitro study.Haemostasis,u,

16.

BILAND, L., DUCKERT, F., PRISENDER, S.and tive estimation of coagulation factors in diagnostic and prognostic value of factor plasminogen.Thromb. Haemost., 3, 646-656,

17.

G.W. and COCHRAN, W.G.Statistical Methods.6th Iowa State University Press, Ames.1967.

NYMAN, liver XIII,

Edi-

D.Quantitadisease.The factor Vand

1978.

TIJTGAT,G.N., COLLEN, D., VESTRAETE, M.Metabolism of fibrin0 gen in cirrhosis of the 1iver.J. Clin. Inv., 50,1690-1701,

-

1971. 18.

O'CONNELL, R.A., GROSSI, C.E. and bitors of fibrinolysis in hepatic 991, 1964.

19.

FLETCHER, A.P., BIEDERMAN, O., MOORE, D.,ALKJAERSIG, N. and SHERRY, S.Abnormal plasminogen-plasmin system activity (fibrinolysis) in patients with hepatic cirrhosis:its cause and c0nsequences.J. Clin. Inv., 43, 681-695, 1964.

20.

SAITO, H., GOODNOUGH,

L.T., BOYLE, J.M., HEIMBURGER, N.Reduted histidine-rich glycoprotein levels in plasma of patients with advanced liver cirrhosis.Possible implications for enhanted fibrinolysis.Am. J. Med., 73, 179-182, 1982.

21.

BERTAGLIA, E.,BELMONTE,P.,VERTOLLI, D.Bleeding in cirrhotic patients:a to intravascular coagulation or to *, 13, 328-334, 1983.

ROUSSELOT, L.M.Roleofinhi cirrhosis.Lancet, 1, 990-

U., AZZURRO, M, MARTINEZ, precipitating factor due hepatic failure?Haemosta-