Increased Procoagulant Activity of Peripheral Blood Monocytes in Human and Experimental Obstructive Jaundice

GASTROENTEROLOGY 1989;96:892-8

Increased Procoagulant Activity of Peripheral Blood Monocytes in Human and Experimental Obstructive Jaundice NICOLA SEMERARO, PASQUALINA MONTEMURRO, GIUSEPPE CHETTA, DONATO F. ALTOMARE, DOMENICO GIORDANO, and MARIO COLUCCI Istituto di Patologia Generale and Istituto di Clinica Chirurgica, University of Bari, Bari, Italy

We studied the procoagulant activity of peripheral blood monocytes in 41 patients with severe obstructive jaundice and in 27 nonjaundiced control patients using a one-stage clotting assay. Mononuclear cells from jaundiced patients, tested immediately after isolation, expressed low levels of procoagulant activity, which were, however, significantly higher than in cells from controls (p < 0.01). In addition, after incubation in short-term cultures with and without endotoxin, these cells generated more procoagulant activity than did the control ones (p < 0.001). No significant difference in procoagulant activity was found between patients with and without malignancy in either group. The relief of biliary obstruction resulted in the reduction of both serum bilirubin levels and monocyte procoagulant activity. Endotoxin-induced monocyte procoagulant activity was about threefold higher in the jaundiced patients who died than in the survivors (p < 0.001). In rabbits made icteric by bile duct ligation and separation (15 days), the endotoxin-induced monocyte procoagulant activity was markedly increased as compared with sham-operated animals (p < 0.005). In all instances, procoagulant activity was identified as tissue factor. The increased capacity of mononuclear phagocytes to produce procoagulant activity might help explain the activation of blood coagulation in severe obstructive jaundice.

A

ctivation of intravascular coagulation and subsequent microvascular thrombosis are frequent findings in severe obstructive jaundice (OJ) and are considered important pathogenetic factors in renal failure (hepatorenal syndrome) associated with this pathologic condition, particularly after surgery (1). The hemostatic abnormalities indicating intravascular coagulation that have been recognized in OJ include increased levels of fibrinogen/fibrin degrada-

tion products (FDPs), the presence of circulating soluble fibrin, and increased fibrinogen catabolism with reduced fibrinolysis (2-4). In some instances an overt disseminated intravascular coagulation could be documented (5). In addition, an increased deposition of radioactive fibrinogen in the kidneys has been reported in an animal model of OJ (6). Despite this evidence, precise sources of procoagulant activities triggering the coagulation cascade in OJ remain to be identified. Cells of the monocytemacrophage series, when exposed in vitro to a wide variety of stimuli, can respond, directly or indirectly, with the production of selected procoagulant activities (PCAs) whereby they initiate the coagulation pathways (7-11). The most commonly described PCA, particularly in human cells, was tissue factor, although under certain circumstances prothrombinases or factor X activators have been shown (7-15). Mononuclear phagocytes can also produce or assemble coagulation factors on their surface, or both (16-18). The generation of PCAs in vivo has been implicated in the activation of intravascular and extravascular coagulation occurring in some pathologic conditions such as endotoxemia, malignancy, and immunologic diseases (9,10,19-21). In this study we demonstrated that (a) the capacity of peripheral blood monocytes to produce PCA is significantly increased in patients with OJ of different origin as compared with control nonjaundiced patients and (b) increased monocyte PCA is associated with a poor prognosis. The association between

Abbreviations used in this paper: FDP, fibrinogen/fibrin degradation product; OJ, obstructive jaundice; peA, procoagulant activity. © 1989 by the American Gastroenterological Association 0016-5085/89/$3.50

March 1989

OJ and increased monocyte PCA was also found in a rabbit model.

Materials and Methods Patients Forty-one patients with OJ, 19 women and 22 men, aged 39-86 yr (mean age, 64 yr), were studied on admission to the University Department of Surgery, after written informed consent had been obtained. They included 27 patients with malignant disease (16 with carcinoma of the pancreas, 5 with hepatoma, 3 with carcinoma of the ampulla, 2 with carcinoma of the common bile duct, and 1 with carcinoma of the duodenum) and 14 patients with nonmalignant cholestasis (8 with stones of the common bile duct, 3 with bile duct stricture, 2 with chronic pancreatitis, and 1 with bile duct fistula). All patients had a serum bilirubin level of >8 mg/l00 ml (range, 8-34; mean, 18). Serum creatinine levels, at the time of study, were normal in all but 3 patients, who had levels of 6.8, 2.0, and 3.2 mg/l00 m!, respectively. Six patients with OJ were also studied after surgery when bilirubin levels had fallen to <3 mgl100 ml (11-15 days). A control group consisted of 27 nonjaundiced patients (15 women and 12 men; aged 32-80 yr; mean age, 59 yr), including 13 with cholelithiasis, 12 with malignancy (6 with carcinoma of the colon, 3 with carcinoma of the stomach, and 3 with carcinoma of the pancreas), and 2 with chronic pancreatitis. No severe bacterial infection was present at the time of investigation. Results of screening studies of the hemostatic system, including platelet count, activated partial thromboplastin time, and prothrombin time, were within the normal range in all patients and controls.

Animals Male New Zealand white rabbits (Charles River Laboratories, Calea, Italy), weighing 1.8-2.5 kg, were anesthetized by intramuscular injection of 0.4 ml/kg body wt of Hypnorm (Duphar, Amsterdam, the Netherlands), containing 10 mg/ml of fluanisone and 0.2 mg/ml of fentanil. The animals were then subjected to bile duct ligation and division or sham operation and kept in clean cages in a temperature-controlled environment. Blood samples (10 ml) were collected from the central ear artery immediately before and 2 wk after the operation. Serum bilirubin measurements were made to confirm jaundice in the bile duct-ligated rabbits.

Isolation of Mononuclear Cells Mononuclear cells from patients and rabbits were isolated from citrated blood by the Ficoll-Hypaque (Lymphoprep; Nyegaard, Oslo, Norway) gradient technique (22), as reported (12). Final cell preparations, suspended in RPMI 1640 medium (Flow Laboratories, Irvine, U.K.), contained >97% mononuclear cells and less than one platelet per nucleated cell. Monocytes, identified by ester-

MONOCYTE TISSUE FACTOR IN OBSTRUCTIVE JAUNDICE

893

ase staining, made up 12%-29% (mean, 17%) and 12%30% (mean, 19%) of the mononuclear cells in patients and controls, respectively. Rabbit cell preparations contained 14%-27% (mean, 20%) monocytes. There were no changes after operation. Cell viability (trypan blue test) was always >95%. Cells of jaundiced patients or rabbits were isolated simultaneously with those of controls. All reagents used were free of endotoxin «0.2 ng/ml) as determined by the limulus amebocyte lysate assay (Microbiological Associates, Walkersville, Md.).

Study of Procoagulant Activity Procoagulant activity was first measured in mononuclear cell suspensions immediately after isolation (basal PCA). To study the capacity of mononuclear cells to produce PCA in vitro, aliquots of cell suspensions (2 x 10 6 /ml) were incubated with endotoxin (Escherichia coli 0111:B4; lipopolysaccharide; final concentration, l/Lg/ml) or sterile saline for 4 h at 37°C before testing (12). All assays were performed in duplicate with cells disrupted by repeated (triple) freezing and thawing. Procoagulant activity was evaluated by a one-stage plasma recalcification time, using human or rabbit plasma as substrate, and expressed in arbitrary units (U)/10 5 monocytes, as previously described (12). To define the nature of PCA, in some clotting assays human plasma deficient in clotting factor VII, IX, or X was used as substrate instead of normal plasma. To further characterize PCA, a goat antiserum against the apoprotein III component of human brain thromboplastin (kind gift of Professor H. Prydz, Research Institute for Internal Medicine, Rikshospitalet. Oslo, Norway) was used.

Other Assays The levels of FDPs were measured by the staphylococcal clumping test (Boehringer Biochemia, Mannheim, F.R.G.); serum levels >10 /Lg/ml were considered elevated. Soluble fibrin was determined on citrated plasma samples by a sensitized red cell agglutination assay (23) using a commercial kit (FM test; Boehringer Biochemia). Endotoxin was detected in heparinized plasma samples by the limulus amebocyte lysate assay (Microbiological Associates) (24,25); firm gel formation within 1 h was recorded as positive.

Statistical Methods Results were always expressed as the mean ± SEM. The significance of the difference between means was estimated by Student's t-test after logarithmic conversion of the data. Differences between the frequency of abnormalities in groups and evidence of association between phenomena have been tested by the K test. The analysis of variance (split-plot design) was used for comparisons of results before and after induction of experimental jaundice.

894

GASTROENTEROLOGY Vol. 96, No. 3

SEMERARO ET AL.

,p
7

DCONTROLS [JPATIENTS

5

_1

en

,P
w

-,

~

>

r P
U

0

Z

0

~

In

'" 0

en

2

5

~

z

::;)

C

U

ICI..

SALINE

LPS

GENERATED IN VITRO

BASAL

Figure 1. Procoagulant activity of peripheral blood mononuclear cells from 41 patients with OJ and 27 control patients. Procoagulant activity was measured immediately after cell isolation (basal PCA) and after incubation for 4 h at 37°C with sterile saline or endotoxin [lipopolysaccharide (LPS)l.

Results Human Study Peripheral blood mononuclear cells, freshly isolated from both jaundiced and control patients, expressed low basal PCA (Figure 1). However, this was significantly higher in jaundiced patients (p < 0.01). Figure 1 also shows the peA generated in vitro by cell preparations after 4 h of incubation with and without endotoxin. In both conditions, cells from jaundiced patients produced significantly more peA than cells from controls (p < 0.001). No significant

difference in peA was found between patients with and without malignancy in either group (Table 1). In 6 patients, 3 with malignant and 3 with benign OJ, who were also studied 11-15 days after surgery, the drop in serum bilirubin was associated with a significant reduction in PCA generated in vitro by unstimulated and lipopolysaccharide-stimulated cells (Figure 2). In all instances, the peA produced by mononuclear cells was identified as tissue factor, as it required factor VII for its expression and was neutralized by a goat antiserum against the apoprotein III

Table 1. Mononuclear Cell Procoagulant Activity in Patients With Malignant and Benign Disease PCA (Ul10 5 monocytes) Generated in vitro Saline

Basal

LPS

Patients

Malignant

Benign

Malignant

Benign

Malignant

Benign

Nonjaundiced Jaundiced

2.1 ± 0.5 5.3 ± 1.5

2.2 ± 0.4 7.7 ± 3.7

89 ± 17 219 ± 36

54 ± 18 354 ± 136

227 ± 34 605 ± 125

179 ± 52 935 ± 275

LPS, lipopolysaccharide; PCA, procoagulant activity.

March 1989

MONOCYTE TISSUE FACTOR IN OBSTRUCTIVE JAUNDICE

20

T

......

'"' III

10

1000

5

500

895

r- P
r-- P
>-

o

c o

.....

E

a

g)

E

z:

10

..

III

C :::I

lOCI

= "'" .....

oI

iii

o

o

o

o

SALINE

BASAL

LPS

GENERATED IN VITRO

Figure 2. Serum bilirubin and PCA of mononuclear cells before (0) and after (_) surgery (n

=

6). LPS, lipopolysaccharide.

182 ± 34 D, p < 0.02 without endotoxin; 1183 ± 282 vs. 525 ± 119 D, p < 0.001 with endotoxin). As illustrated in Figure 3, 100% of the patients with a fatal evolution had >500 D of endotoxin-induced PCA, whereas such high levels of PCA were found only in 8 (28%) of the jaundiced patients and in 1 (4%) of the control patients (all without complications). Table 4 shows the relationship between mononuclear cell PCA, as compared with other relevant laboratory parameters, and lethal outcome. In patients in whom PCA was above 500 D at the time of admission, the mortality rate was about 60% as compared with 60%,50%, and 33% in those with elevated FDPs, positive soluble fibrin, and positive endotoxin, respectively. On the other hand, a value of PCA below 500 D at the time of admission was never associated with death, whereas normal values of FDP, negative soluble fibrin, and negative endotoxin were associated with a death rate of 13%-24%.

component of human brain thromboplastin (Table 2). Table 3 summarizes the abnormal results in assays of FDPs, soluble fibrin, and endotoxin. The percentage of abnormalities in FDPs and soluble fibrin was significantly higher in the jaundiced group. Levels of FDPs in this group were increased about three times as compared with nonjaundiced controls (15.9 ± 2.5 vs. 4.4 ± 1.0 JLg/ml, p < 0.01). No correlation was found between any of these parameters and monocyte PCA. In OJ, 12 of 41 patients died in the hospital during the first month after admission. They included 7 patients with malignant disease and 5 with nonmalignant cholestasis; in 8 of these patients death occurred after surgery. The causes of death were acute renal failure (7 patients), septicemia (2 patients), and cardiac failure (3 patients). Monocyte PCA generated in vitro was significantly higher in these patients than in the survivors (460 ± 147 vs.

Table 2. Characterization of Mononuclear Cell Procoagulant Activity in Patients With Obstructive Jaundice and Controls PCA (U/l0 5 monocytes) Patients (n

= 10)

Sample

Basal

LPS induced

Tested in normal plasma Tested in factor IX-deficient plasma Tested in factor VII-deficient plasma Tested in factor X-deficient plasma Treated with anti-apoprotein nrb Treated with normal goat serum b

7.5 ± 1.5 7.1 ± 1.4 <1 <1 <1 7.4 ± 1.3

829 ± 99 848 ± 94 11 ± 2 <1 2 ± 0.5 835 ± 102

Controls (n = 8) LPS induced

Tissue factorU (n = 4)

225 ± 38 219 ± 35 4±1 <1 <1 229 ± 41

800 789 ± 14 9±1 <1 4 ± 0.3 804 ± 12

LPS, lipopolysaccharide; PCA, procoagulant activity. Basal PCA indicates PCA of mononuclear cells tested immediately after isolation; LPS-induced PCA indicates PCA of mononuclear cells after incubation at 37°C for 4 h with endotoxin. UHuman brain thromboplastin diluted to achieve a PCA comparable to that of patients' cells stimulated with endotoxin. b Samples (n = 5) of mononuclear cells or thromboplastin were incubated with a goat antiserum to apoprotein III or normal goat serum (1: 20 final dilution) for 1 h at room temperature before testing in normal plasma. The sera were adsorbed with 10 mglml of BaS0 4 for 15 min before use; after adsorption they contained <1% of factors of the prothrombin complex.

896

GASTROENTEROLOGY Vol. 96, No.3

SEMERARO ET AL.

Table 3. Abnormal Assays in Patients With Obstructive Jaundice and Controls Jaundiced patients

Controls

Abnormal assay

n

Percent

n

Percent

i'

Positive endotoxin Elevated FDP Positive fibrin

9/23 15/31 14/31

39 48 45

4/16 3/21 2/21

25 14 10

NS P < 0.05 P < 0.02

FDP, fibrinogen/fibrin degradation product; NS, not significant.

Animal Study

Rabbit peripheral blood mononuclear cells, unlike human cells, had undetectable levels of basal PCA in both the icteric and the control group «1 UI 10 5 monocytes). When incubated with endotoxin, cells from icteric animals generated significantly higher PCA than cells from sham-operated animals and cells before operation (p < 0.005; n = 10) (Figure 4). None of the control animals had detectable endotoxin in plasma, whereas a positive limulus assay was found only in 3 of 10 icteric animals. Procoagulant activity was always identified as tissue factor, as it required factor VII for its expression (data not shown).

D

o D

-

Table 4. Relationship Between Relevant Laboratory Parameters and Death

o

Assay PCA >500 U <500 U FDP >10 JLg/ml <10 JLg/ml Soluble fibrin Positive Negative Endotoxemia Positive Negative

Deaths

Patients (n)

n

Percent

i'

20 21

12 0

60 0

P < 0.0005

15 16

9 2

60 13

P < 0.025

14 17

7 4

50 24

NS

9 13

3 3

33 23

NS

FDP, fibrinogenlfibrin degradation product; NS, not significant; PCA, procoagulant activity.

In Vitro Study Experiments were performed to evaluate whether plasmas from jaundiced patients who had increased monocyte PCA contain factors that enhance the in vitro generation of PCA. Isolated mononuclear cells from normal subjects were suspended in jaundiced or control plasma and incubated at 37°C with or without endotoxin. After 4 h the cells were washed (three times) with RPMI medium, resuspended at a concentration of 2 X 106/ml, and tested for PCA. In all experiments PCA generation was unaffected by the presence of jaundiced plasma (6'5 ± 14 vs. 63 ± 15 U without endotoxin; 245 ± 64 VS. 230 ± 57 U with endotoxin; n = 6).

DD

Ul

IU

I-

>

U

o z

o

~JAUNDICED

:I

0

... 1000

...

a

:;;-.

•

I-

D

I-

D D

• ••

~

a..

- - - - r - - - - - - - - - - - -- --- - - - - - - - - - -

o

••• • ••• ••••

•• • ••••

...•••-

J_ - - - -

o8g

z

o

:I

'0

'e" z

2500

::I

C

00 00

~

a..

o~o

000 0 00

ALIVE

CONTROLS

o SHAM-OPERATED

>

Iil

C

IU

g

0 00

Z ::I

Ul 5000

DEAD

PATIENTS

Figure 3. Endotoxin-induced mononuclear cell PCA in relation to the clinical outcome of the disease (observation period = 1 mol.

BEFORE

15 d. AFTER

Figure 4. Endotoxin-induced PCA in rabbit peripheral blood mononuclear cells harvested before and 15 days after bile duct ligation and separation or sham operation (10 animals per group).

March 1989

Discussion This study demonstrates that OJ is associated with an increased capacity of peripheral blood monocytes to produce PCA. Cells from jaundiced patients expressed significantly higher PCA than controls when tested immediately after rapid isolation, suggesting in vivo induction. In addition, after incubation in short-term culture with and without endotoxin, they generated significantly more PCA than did cells from control patients. Recent studies have shown that both in human and experimental malignancy, mononuclear phagocytes may have an increased ability to generate PCA (9,19,26-29). We did not find significant differences between patients with and without malignancy, suggesting that enhanced PCA is associated with OJ rather than with the underlying disease. This view is supported by the simultaneous decrease of bilirubin and monocyte PCA after relief of biliary obstruction, even in those neoplastic patients in whom the primary tumor mass could not be removed. Moreover, monocytes from jaundiced rabbits generated significantly higher PCA in response to ~ndotoxin than cells from controls. The lack of difference in basal PCA in our animal model may be due either to the sensitivity of PCA assay (freshly isolated rabbit mononuclear cells express <1 U/10 5 monocytes) or to the relatively short duration of the biliary obstruction. Pro coagulant activity of both human and rabbit ~ononuclear cells was identified as tissue factor by Immunologic or biologic criteria, or both. The exact mechanism responsible for the increased production of PCA by mononuclear cells remains to be established. Endotoxin is one of the most potent inducers of monocyte PCA (7-9,19-21) and, in agreement with previous data (1-3), it was detected in a high percentage of our patients with OJ. We did not find any relationship between positive limulus assay and monocyte PCA in patients and in rabbits with OJ. However, the possibility that lipopolysaccharide had contributed to the induction of monocyte PCA cannot be totally excluded, as its detection in plasma is difficult and the sensitivity may be poor also in view of the complex interactions of lipopolysaccharide with plasma lipoproteins. Moreover, the cells could have encountered the endotoxin in the mesenteric veins before clearance in the liver. In this case the PCA-positive cells, but not the eliciting lipopolysaccharide, would be present in the systemic circulation. Another possibility is that other factors present in plasma, including bilirubin and bile salts, might be responsible, at least in part, for the increased induction of monocyte PCA. In our in vitro experiments,

MONOCYTE TISSUE FACTOR IN OBSTRUCTIVE JAUNDICE

897

the jaundiced plasma did not affect the generation of PCA by normal mononuclear cells, suggesting that either in vivo interactions with other components not present in the test tube or much longer exposure to the patients' plasma environment is required for stimulation of monocyte PCA. It is unclear whether the hypothetical stimulating factor acts directly on monocytes or interferes with cells known to modulate PCA production (8,12). Whatever the activating mechanism, monocyte PCA could be one of the triggering factors for fibrin deposition in OJ. These findings, together with evidence from our laboratory that OJ is associated with impaired fibrinolysis (30), indicate that, in this pathological condition, the coagulation-fibrinolytic balance is shifted toward promotion of fibrin deposition. Analysis of the data in relation to the clinical outcome of the disease revealed a significant association between exceedingly high levels of endotoxininduced PCA and lethal outcome. Previous studies have shown that preoperative elevated serum FDP concentration, positive endotoxin, and positive soluble fibrin are associated with a poor postoperative recovery from OJ (2,3). Monocyte PCA may represent an additional parameter of clinical value, although adequately constructed studies are required to assess its precise prognostic significance.

References 1. Pain JA, Cahill q, Bailey ME. Peri operative complications in obstructive jaundice: therapeutic considerations. Br J Surg 1985;72:942-5. 2. Hunt DR, Allison MEM, Prentice CRM, Blumgart LH. Endotoxemia, disturbance of coagulation, and obstructive jaundice. Am J Surg 1982;144:325-9. 3. Allison MEM, Prentice CRM, Kennedy AC, Blumgart LH. Renal function and other factors in obstructive jaundice. Br J Surg 1979;66:392-7. 4. Wardle EN. Fibrinogen in liver disease. Arch Surg 1974;109: 741-4. 5. Takeda S, Takaki A, Ohsato K. Occurrence of disseminated intravascular coagulation (DIC) in obstructive jaundice and its relation to biliary tract infection. Jpn J Surg 1977;7:82-9. 6. Sagar S, Shields R. Fibrinogen in the 'hepato-renal' syndrome. Br J Surg 1980;67:562-4. 7. Niemetz J, Muhlfelder T, Chierego ME, Troy B. Procoagulant activity of leukocytes. Ann NY Acad Sci 1977;283:208-17. 8. Edgington TS. Recognition coupled responses of the monocyte: activation of coagulation pathways. Nouv Rev Fr HematoI1983;25:1-6. 9. Edwards RL, Rickles FR. Macrophage procoagulants. In: Spaet TH, ed. Progress in hemostasis and thrombosis. Volume 7. New York: Grune & Stratton, 1984:183-201. 10. Geczy CL. Induction of macrophage procoagulant by products of activated lymphocytes. Haemostasis 1984;14:400-11. 11. Lyberg T, Prydz H. The role of cellular cooperation in thromboplastin synthesis. Nouv Rev Fr Hematol 1983;25:291-3. 12. Semeraro N, Biondi A, Lorenzet R, Locati D, Mantovani A, Donati MB. Direct induction of tissue factor synthesis by

898 SEMERARO ET AL.

13.

14. 15.

16. 17.

18.

19.

20.

21.

22. 23.

endotoxin in human macro phages from diverse anatomical sites. Immunology 1983;50:529-35. Maier R, Ulevitch RJ. The induction of a unique procoagulant activity in rabbit hepatic macro phages by bacteriallipopolysaccharides. J Immunol 1981;127:1596-600. Shands JW. Macrophage procoagulants. Haemostasis 1984;14: 373-7. Schwartz BS, Levy GA, Fair DS, Edgington TS. Murine lymphoid procoagulant activity induced by bacterial lipopolysaccharide and immune complexes is a monocyte prothrombinase. J Exp Med 1982;155:1464-79. Osterud B, Lindahl U, Seljelid R. Macrophages produce coagulation factors. FEBS Lett 1980;120:41-3. Tracy PB, Rohrbach MS, Mann KG. Functional prothrombinase complex assembly on isolated monocytes and lymphocytes. J Bioi Chern 1983;258:7264-7. Tsao BP, Fair DS, Curtiss LK, Edgington TS. Monocytes can be induced by lipopolysaccharide-triggered T lymphocytes to express functional factor VIINIIa protease activity. J Exp Med 1984;159:1042-57. Semeraro N, Lattanzio A, Montemurro P, et al. Mechanisms of blood clotting activation in inflammation: role of mononuclear phagocytes. Int J Tissue React 1985;7:313-20. Thiagarajan P, Niemetz J. Procoagulant-tissue factor activity of circulating peripheral blood leukocytes. Results of in vivo studies. Thromb Res 1980;17:891-6. Osterud B, Flaegstad T. Increased tissue thromboplastin activity in monocytes of patients with meningococcal infection: related to unfavourable prognosis. Thromb Haemost 1983;49: 5-7. Boyum A. Isolation of lymphocytes, granulocytes and macrophages. Scand J Immunol 1976;5(Suppl):9-15. Largo R, Heller V, Straub PW. Detection of soluble interme-

GASTROENTEROLOGY Vol. 96, No.3

24. 25. 26. 27.

28. 29.

30.

diates of the fibrinogen-fibrin conversion using erythrocytes coated with fibrin monomers. Blood 1976;47:999-1002. Levin J, Tomasulo PA, Oser RS. Detection of endotoxin in human blood and demonstration of an inhibitor. J Lab Clin Med 1970;75:903-11. McLeod C, Katz W. A rapid method for the detection of gram-negative bacterial endotoxins in whole blood. J Bioi Stand 1981;9:299-302. Semeraro N, De Lucia 0, Lattanzio A, et al. Procoagulant activity of human alveolar macrophages: different expression in patients with lung cancer. Int J Cancer 1986;37:525-9. Guarini A, Acero R, Alessio G, Donati MB, Semeraro N, Mantovani A. Procoagulant activity of macrophages associated with different murine neoplasms. Int J Cancer 1984;34: 581-6. Edwards RL, Rickles FR, Cronlund M. Abnormalities of blood coagulation in patients with cancer: mononuclear cell tissue factor generation. J Lab Clin Med 1981;98:917-28. Lorenzet R, Peri G, Locati D, et al. Generation of procoagulant activity by mononuclear phagocytes: a possible mechanism contributing to blood clotting activation within malignant tissues. Blood 1983;62:271-3. Colucci M, Altomare DF, Chetta G, Triggiani R, Cavallo LG, Semeraro N. Impaired fibrinolysis in obstructive jaundice. Evidence from clinical and experimental studies. Thromb Haemost 1988;60:25-9.

Received November 23, 1987. Accepted October 10, 1988. Address requests for reprints to: Professor Nicola Semeraro, Istituto di Patologia Generale, Universita-Policlinico, Piazza G. Cesare, 70124 BarL Italy. This work was presented in part at the XIth International Congress on Thrombosis and Haemostasis, Brussels, Belgium, July 5-10,1987.