- Email: [email protected]
Increased elastase-α1-antitrypsin complex in fulminant hepatic failure: Relationship to bacterial infection and activation of coagulation
Clinica Chimica Acta, 200 (1991) 211-220 0 1991 Elsevier Science Publishers B.V. All rights reserved ADONIS 000989819100169K
211 0009-8981/91/$03.50
CCA 05038
Increased elastase-a!,-antitrypsin complex in fu lminant hepatic failure: relationship to bacterial infection and activation of coagulation P.G. Langley, R.D. Hughes, N. Roland0 and Roger Williams Institute of Liver Studies, King’s College Hospital and King’s College School of Medicine and Dentistry, London (UK) (Received Key words: Elastase;
5 November Fulminant
1990; accepted hepatic
failure;
22 May 1991) Infection;
cut-Antitrypsin
Summary To study the effect of infection, a frequent complication of fulminant hepatic failure (FHF), on the release of elastase from polymorphonuclear leucocytes and its inhibition in circulation we have measured the concentrations of cr,-antitrypsin, which binds and inhibits elastase in the circulation, and of elastase-cur-antitrypsin complex, in 30 patients with FHF. Elastase-a,-antitrypsin complex was significantly increased in FHF as compared to controls (303 f 51 pg/l compared to 37.+ 5 pg/l; II = 10; P < 0.001) demonstrating activation of leucocytes in FHF. Infection caused greater release of leucocyte elastase, complex levels were significantly greater in patients who were infected when compared to those who were not (463 f 84 pg/l; n = 13 compared to 180 f 46 pg/l; IZ= 17; P < 0.01). Also patients who survived had significantly lower complex levels than those who did not (212 f 49 pg/l; n = 18 compared to 440 + 94 pg/l; n = 12; P < 0.02). cY,-Antitrypsin activity was not significantly different from control subjects (0.99 rt 0.06 U/ml compared to 0.97 + 0.05 U/ml). However cY,-antitrypsin activity was significantly higher in patients who survived (1.17 f 0.05 U/ml; II = 18) compared to those who did not (0.71 k 0.03 U/ml; II = 12; P < 0.001) and patients who died had significantly lower levels than control subjects (P < 0.01) indicating the importance of maintenance of normal inhibitor levels in patients with FHF. The leucocyte activation and release of elastase in FHF was linked to activation of the coagulation system; elastase-cur-antitrypsin complex levels correlated signifi-
Correspondence College School
to: Dr. Roger Williams, Institute of Liver Studies, King’s College Hospital of Medicine and Dentistry, Denmark Hill, London SE5 9RS, UK.
and King’s
212
cantly with thrombin-antithrombin III complex levels (r = 0.68; P < 0.001) and inversely with fibrinogen (r = - 0.71; P < 0.001).
Introduction Patients with fulminant hepatic failure (FHF) are particularly susceptible to bacterial infection [l] which, among other complications, contributes to the high mortality of this condition [2]. Septic shock can develop and this may aggravate the severity of the coagulation disturbances [3,4] and the final multi-organ failure. As one part of the systemic response to infection leucocyte elastase (EC 3.4.21.11, a neutral lysosomal proteinase stored in polymorphonuclear leucocytes, is released into the circulation [5]. However leucocyte elastase is thought to be involved in the tissue damage which occurs in septicaemia, endotoxic shock and emphysema [6,7]. In addition to its action to destroy opsonised infecting organisms and cell debris [8], it can have deleterious effects such as complement activation, release of toxic peptides, degradation of connective tissue and immunoglobulins [9]. It also inactivates coagulation factors and the inhibitor antithrombin III [lo] and activates the fibrinolytic system [ 111. The major physiological inhibitor of elastase, q-antitrypsin [6], is a member of the serpin family of protease inhibitors [12]. It is synthesised mostly in the liver and also in monocytes and macrophages but to a much lesser extent. Although low concentrations might be expected in FHF due to the profound loss of liver synthetic function, it is also an ‘acute phase’ protein and an increase in levels could occur as a result. Because of the rapid binding of elastase to a,-antitrypsin it is not possible to measure free elastase activity in the circulation. Measurement of the concentration of the inert complex gives an estimate of the release of elastase into the circulation during leucocyte activation and may be a marker of sepsis in these patients [13]. In this study we have measured plasma levels of elastase-cu,-antitrypsin complex, using a solid phase enzyme linked sandwich immunoassay, in 30 patients with FHF to study the relationship between leucocyte activation and the severity of the disease and to the presence or absence of bacterial infection. Studies were also carried out during haemodialysis which, by virtue of activation of leucocytes by the dialysis membrane, could affect the release of elastase into circulation. Material and methods The 30 patients studied had grade 3 or 4 encephalopathy due to FHF as defined by Trey and Davidson [14]. In 25 patients the cause of the FHF was paracetamol overdose and in 5, viral hepatitis (2 hepatitis A, 2 hepatitis B, 1 hepatitis non-A non-B). The measurements were carried out on admission. Blood was collected into citrate and immediately centrifuged for 10 min at 2,000 x g at 4 o C. The resulting plasma was stored in portions at - 70 o C until assayed. Infection was diagnosed as previously described 111. Infection was suspected when the axillary temperature exceeded 38°C on two successive 4 h
213
readings, a peripheral blood leucocyte count exceeded 11 x 109/1 or inflammation of a wound or catheter sites or radiological signs of pulmonary consolidation were present. Blood, urine and sputum were microbiologically examined. Infection was confirmed when clinically significant positive cultures were made from body fluids or sites of infection. Microbiological cultures were performed daily and when clinically indicated. Six further patients were studied during a single 4 h haemodialysis required on account of development of renal failure. Haemodialysis was carried out using a Biospal 2400s dialyser containing a polyacrylonitrile membrane AN69S (Hospal, Rugby, UK). Blood samples were taken before haemodialysis then at 10 min, 30 min and 1, 2, 3 and 4 h after the start of haemodialysis. The patients received a bolus of 5,000 units of heparin prior to haemodialysis followed by a continuous infusion of heparin so that the whole blood activated clotting time was maintained between 200-250 s. Assays
Elastase-cr,-antitrypsin complex was measured using a solid phase, enzyme-linked sandwich immunoassay (Merck Ltd., Darmstadt, Germany) in which the first antibody, specific for elastase, is bound to the solid phase and the labelled antibody is specific for cy,-antitrypsin. The final colour change was measured at a wavelength of 405 nm using an Ultraspec 4050 spectrophotometer (Pharmacia-LKB Ltd., South Croydon, UK). Each sample was assayed in duplicate. cu,-Antitrypsin activity was measured using an indirect chromogenic substrate assay (Kabivitrum, Uxbridge, UK). (Y,-Antitrypsin concentration was measured by radial immunodiffusion (Behringwerke Ag, Marburg, Germany). There was a significant correlation between cu,-antitrypsin activity as measured by chromogenic substrate assay and cY,-antitrypsin concentration as measured by radial immunodiffusion (R = 0.89, P < 0.001). Thrombin-antithrombin III complex was measured by an enzyme-linked immunosorbent assay using a kit donated by Behringwerke Ag, Marburg, Germany. Fibrinogen was measured by the method of Claus [15] using a Fibrometer (Becton Dickinson, UK). Whole blood activated clotting time was measured using Haemochron (Technidyne, Metuchen, USA). The control group consisted of 10 healthy laboratory volunteers. Statistics Significance was assessed between groups using Wilcoxon’s rank sum test. Correlation between parameters was assessed by Spearman test. Results are reported as either median and range or mean f SE as appropriate. Results Laboratory measurements taken on admission of the 30 patients are shown in Table I. Eighteen (60%) of the patients survived to leave hospital comprising 14 of
214 TABLE I Laboratory parameters of patients on admission, median values and ranges shown Sex (M/F)
Prothrombin Time (secsl
AST a (hi/l)
Bilirubin (~mol/l)
Creatinine (~mol/l)
POD a Median Range
15/10
65 25-240
1550 122-10000
140 55-264
311 69-828
Viral Median Range
2/3
30 20- 70
950 43- 4230
210 174-333
95 58-432
5- 17
45-105
Normal range
7-
12- 14
a AST, serum aspartate aminotransferase;
40
POD, Paracetamol overdose.
the 25 patients with paracetamol overdose and 4 of 5 patients with viral hepatitis. At the time of admission 13 (43%) patients were considered to be infected of whom 6 (46%) survived to leave hospital. A further 9 patients became infected later in the course of their illness of whom 6 survived to leave hospital. Fourteen patients had renal failure (creatinine > 300 pmol/l) on admission of whom 9 (64%) were clinically infected. Measurement of elastase-a,-antitrypsin
complex and a,-antitrypsin
Elastase-a,-antitrypsin complex levels on admission were significantly greater in patients with FHF than in control subjects (303 + 51 pg/l compared to 37 & 5 pg/l; P < O.OOl>.None of the patients had levels within the normal range (Fig. 1).
Control
Survivors
Non-survivors p(O.02
Fig. 1. Concentration
of elastase-cu,-antitrypsin complex, on admission, in patients with fulminant hepatic failure in relation to outcome.
215
Not infected
Infected p(O.01
Fig. 2. Concentration
of elastase-cY,-antitrypsin complex in patients with fulminant hepatic failure with and without infection.
The 18 patients who survived had significantly lower levels of the complex on admission than those who died (212 _t 49 pg/l compared to 440 f 94 pg/l; P < 0.02). Levels in those who were infected at the time of sampling had significantly greater concentration of complex compared to those who had no infection (486 + 96 pg/l compared to 197 + 44 pg/l; P < 0.01) (Fig. 2). Patients with and without renal failure had similar concentrations of the complex (365 f: 89 pg/l and 259 + 60 pg/l, respectively). Patients with viral hepatitis had significantly lower complex levels than those with paracetamol overdose (104 + 30 pug/l and 343 f 60 pg/l, respectively; P < 0.05). However there were only 5 patients in the viral group and the levels may be affected by the high proportion of these patients who survived. When the activity levels of a,-antitrypsin, the inhibitor of leucocyte elastase, were measured in the total group of patients with FHF on admission they were not significantly different to normal control subjects (0.99 + 0.06 U/ml and 0.97 + 0.05 U/ml, respectively) (Fig. 3). Values in the patients who survived were significantly greater than those who died (1.17 k 0.05 U/ml compared to 0.71 + 0.03 U/ml; P < 0.001) and patients who died had significantly lower activity levels than the healthy controls (P < 0.01). There was no significant difference between infected and non-infected patients (0.92 f 0.08 U/ml and 1.03 k 0.08 U/ml, respectively) patients with and without renal failure (0.99 f 0.08 U/ml and 0.98 f 0.09 U/ml, respectively) or patients with viral hepatitis and paracetamol overdose (0.98 f. 0.15 U/ml and 0.99 + 0.07 U/ml, respectively). Elastase-cY,-antitrypsin complex was significantly correlated with thrombin-antithrombin III complex (Fig. 4) and inversely correlated with fibrinogen concentration (r = -0.71; P < 0.001). a,-Antitrypsin was significantly correlated with fibrinogen concentration. Patients who did not survive had both low fibrinogen and low a,-antitrypsin activity levels on admission (Fig. 5). There was no significant correlation between elastase-cu,-antitrypsin complex and a,-antitrypsin activity.
216 2.0
5
: 3 .ca tf
1.5
1.0
F i
0.5
0.0
Control
Non-survivors
p(O.001
p(O.01
Fig. 3. Concentration
of a,-antitrypsin
Survivors
in patients with fulminant hepatic failure on admission in relation to outcome. r=0.66:
0
10
20
30
40
50
60
Thrombin-antithrombin
Fig. 4. Correlation
70
80
~(0.001
90
100
Ill &g/l)
of elastase-a,-antitrypsin complex with thrombin-antithrombin patients with fulminant hepatic failure.
III complex in
TABLE II Elastase-a,-antitrypsin
Complex (I.Lg/f) Mean +SE
complex and white cell numbers in 6 patients during 4 h of haemodialysis
PRE
10 min
30 min
lh
2h
3h
4h
224 28
285 67
307 60
316 59
392 98
541 214
660 30
81.2 17.8 P < 0.05
82.5 16.9 P < 0.05
108.9 18.8
107.9 19.7
119.5 25.9
White cells t% of initial) Mean &SE
87.1 11.0
(Wilcoxon’s paired rank sum test, compared to initial value.)
217 0
6.0 v0.70;
al -antitrypsin 0 0
p(O.001
(u/ml)
Survivors Non-survivors
Fig. 5. Correlation of fibrinogen with a,-antitrypsin
in patients with fulminant hepatic failure.
Effect of haemodialysis
Elastase-cr,-antitrypsin complex was raised at the start of haemodialysis and increased further throughout the course of treatment (Table II). When white cell counts were expressed as percentage of initial levels, they were significantly decreased at 1 and 2 hours after the start of haemodialysis, subsequently they returned to initial levels (Table II). Thrombin-antithrombin III complex was also raised at the start (30.7 + 8.5 pg/l) and increased further during haemodialysis (86.9 + 33.4 pg/l at 1 h and 66.0 + 12.1 pg/l at 4 h). Discussion
The high levels of the elastase-cl!,-antitrypsin complex found in plasma in the present study strongly suggest that there is leucocyte activation, degranulation and release of proteases in FHF. Loss of hepatocyte and Kupffer cell function [161 could cause some increase in complex level due to lack of clearance, but as the half-life of the complex is short in healthy subjects [13] the majority of the increase in the present subjects is more likely to have been due to an increased and continued elastase release. The leucocyte activation could be caused by tissue destruction as has been demonstrated in other conditions which involve tissue damage [12] and where increases in elastase-cw,-antitrypsin complex have also been demonstrated. The findings in the present study that patients who are infected on admission have higher levels of elastase-a,-antitrypsin complex, is consistent with infection causing leucocyte activation and release of elastase. Seventy three percent of the patients had episodes of infection during the course of their illness. Although renal failure has been linked to infection in FHF [l], in the present study there was no significant increase in the complex in those with renal failure. Thus, the increases
218
in elastase-cY,-antitrypsin complex are likely to be due to both tissue damage and sepsis. Although patients who survived had significantly lower levels of the complex there was much overlap in the data and this would make the test of little value in the prognosis of individual cases. If local concentrations of elastase exceed those of the inhibitor the effects of free elastase may contribute to some of the serious complications, such as multiorgan failure, seen in patients with FHF. a,-Antitrypsin is an acute phase protein and would be expected at increased level in a severe disease. The levels measured in FHF are probably a balance between loss of production due to poor liver function and the acute phase response. Whether the lower levels in patients who died is due to a failure of the liver to undergo the acute phase response or reflects a greater loss of hepatocyte function is unclear. It is unlikely to be due to a greater consumption of the inhibitor as there was no correlation between a,-antitrypsin activity and elastase-c-u,-antitrypsin complex concentration. The coagulation system is known to be activated in FHF with increased levels of fibrin degradation products [17] and the thrombin-antithrombin III complex Cl81 as measured in this study. The strong correlation between the elastase-ai-antitrypsin complex and the thrombin-antithrombin III complex suggest that the activation of leucocyte and coagulation systems are linked. It is possible that a single stimulus, such as endotoxin or cytokine release activates both systems simultaneously. Tumour necrosis factor production has been shown to be increased in FHF 1191,but this cytokine alone does not cause realease of elastase activity in vitro 1201.Elastase has been shown to inhibit antithrombin III activity in vitro [lo] and low levels of antithrombin III may allow increased production of thrombin and thus activate the coagulation system. The latter, in its turn, may activate release of leucocyte elastase as it has been shown that antithrombin III concentrate and plasma substitution reduces elastase-cr,-antitrypsin complex levels in patients with septic shock [21]. A further link between the two systems is the significant correlation between fibrinogen, also an acute phase protein, and a,-antitrypsin. We have shown previously [18] that if fibrinogen levels are low on admission in FHF the prognosis is poor. If both fibrinogen and a,-antitrypsin are present at low levels on admission the prognosis is very poor indeed (Fig. 5). The increases in elastase-cY,-antitrypsin complex and the loss of white cells during haemodialysis demonstrate that there is still activation of white cells by the procedure despite adequate heparinisation. Haemodialysis is known to effect leucocytes in circulation by activation of the complement system [22]. Although complement proteins are present at low levels in FHF, activated leucocytes may still release elastase when in contact with the dialysis membrane. Although the loss of white cells was small in this study (17.5% at 1 h), probably due to the good biocompatibility of the polyacrylonitrile membrane used, the effect on white cells was sufficient for elastase to be released into the circulation despite the return of white cell count to initial levels. The coagulation system is also activated during haemodialysis [18] as shown by the increase in thrombin-antithrombin III levels, which is further supporting evidence for a link between the two systems.
219
References 1 Roland0 N, Harvey F, Brahm J, Philpott-Howard J, Alexander GJM, Gimson AES, Casewell M, Fagan E, Williams R. Bacterial infection in acute liver failure. Hepatology 1990;11:49-53. 2 O’Grady JG, Alexander GJM, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989;97:439-445. 3 Lechner K, Niessner H, Thaler E. Coagulation abnormalities in liver disease. Semin Thromb Haemostas 1977;4:40-56. 4 O’Grady JG, Langley PG, Isola LM, Aledort LM, Williams R. Coagulopathy in fulminant hepatic failure. Semin Liver Disease 1986;6:159-163. 5 Ohlsson K, Ohlsson I. The neutral proteases of human granulocytes. Isolation and partial characterisation of granulocyte elastases. Eur J Biochem 1974:42:519-527. 6 Egbring R, Schmidt W, Fuchs G, Havemann K. Demonstration of granulocyte proteases in plasma of patients with acute leukemia and septicemia with coagulation defects. Blood 1977;49:219-231. 7 Crystal RG. a,-Antitrypsin deficiency, emphysema and liver disease. Genetic basis and strategies for therapy. J Clin Invest 1990;85:1343-1352. 8 Babior BM. Oxidants from phagocytes; agents of defense and destruction. Blood 1984$X959-966. 9 Havemann K, Janoff A. Neutral proteases of human polymorphonuclear leukocytes. Baltimore and Munich: Urban and Schwarzenberg, 1978. 10 Jordan RE, Nelson RM, Kilpatrick J, Newgren JO, Esmon PC, Fournel MA. Inactivation of human antithrombin by neutrophil elastase. J Biol Chem 1989;264:10493-10500. 11 Gilboa N, Neuman PH, Gutmann JM, Del Vecchio PJ, Gudewicz PW. Evidence for regulation of endothelial plasminogen-activating system by ‘polymorphonuclear leukocyte elastase. Thrombosis Res 1989;54:467-475. 12 Carrel1 RW. cr,-Antitrypsin: molecular pathology, leukocytes, and tissue damage. J Clin Invest 1986;78:1427-1431. 13 Harm K, Bartfeld KP, Mathew T. Plasma concentrations of granulocytic elastase-a,-proteinase inhibitor complex in patients with severe head injury, multiple trauma or cerebral bleeding. Clin Biochem 1989;22:149-153. 14 Trey C, Davidson CS. The management of fulminant hepatic failure. In; Popper H, Schaffner F, eds. Progress in liver disease, Vol 3. New York: Grune and Stratton, 1970;282-298. 15 Clauss A. Gerinnungsphysiologisches schnellmethode zur bestinnung des fibrinogens. Acta Haematologica 1957;17:237-246. 16 Canalese J, Gove CD, Gimson AES, Wilkinson SP, Wardle EN, Williams R. Reticuloendothelial system and hepatocyte function in fulminant hepatic failure. Gut 1982;23:265-269. 17 Hughes RD, Lane DA, Ireland H, Langley PG, Gimson AES, Williams R. Fibrinogen derivatives and platelet activation products in acute and chronic liver disease. Clin Sci 1985;68:701-707. 18 Langley PG, Forbes A, Hughes RD, Williams R. Thrombin-antithrombin III complex in fulminant hepatic failure: evidence for disseminated intravascular coagulation and relationship to outcome. Eur J Clin Invest 1991;in press. 19 Muto Y, Nouri-Aria KT, Meager A, Alexander GJM, Eddleston AWLF, Williams R. Enhanced tumour necrosis factor and interleukin,l in fulminant hepatic failure. Lancet 1988;ii:72-74. 20 Yonemaru M, Stephens KE, Ishikaza A, Zheng H, Hogue RS, Crawley JJ, Hatherill JR, Raffin TA. Effects of tumor necrosis factor on PMN chemotaxis, chemiluminescence, and elastase activity. J Lab Clin Med 1989;114:674-681. 21 Seitz R, Wolf M, Egbring R, Havemann K. The disturbance of haemostasis in septic shock: role of neutrophil elastase and thrombin, effects of antithrombin III and plasma substitution. Eur J Haematol 1989;43:22-28. 22 Hakim RM, Breillatt J, Lazarus JM, Port FK. Complement activation and hypersensitivity reactions to dialysis membranes. N Engl J Med 1984;311:878-882.
211 0009-8981/91/$03.50
CCA 05038
Increased elastase-a!,-antitrypsin complex in fu lminant hepatic failure: relationship to bacterial infection and activation of coagulation P.G. Langley, R.D. Hughes, N. Roland0 and Roger Williams Institute of Liver Studies, King’s College Hospital and King’s College School of Medicine and Dentistry, London (UK) (Received Key words: Elastase;
5 November Fulminant
1990; accepted hepatic
failure;
22 May 1991) Infection;
cut-Antitrypsin
Summary To study the effect of infection, a frequent complication of fulminant hepatic failure (FHF), on the release of elastase from polymorphonuclear leucocytes and its inhibition in circulation we have measured the concentrations of cr,-antitrypsin, which binds and inhibits elastase in the circulation, and of elastase-cur-antitrypsin complex, in 30 patients with FHF. Elastase-a,-antitrypsin complex was significantly increased in FHF as compared to controls (303 f 51 pg/l compared to 37.+ 5 pg/l; II = 10; P < 0.001) demonstrating activation of leucocytes in FHF. Infection caused greater release of leucocyte elastase, complex levels were significantly greater in patients who were infected when compared to those who were not (463 f 84 pg/l; n = 13 compared to 180 f 46 pg/l; IZ= 17; P < 0.01). Also patients who survived had significantly lower complex levels than those who did not (212 f 49 pg/l; n = 18 compared to 440 + 94 pg/l; n = 12; P < 0.02). cY,-Antitrypsin activity was not significantly different from control subjects (0.99 rt 0.06 U/ml compared to 0.97 + 0.05 U/ml). However cY,-antitrypsin activity was significantly higher in patients who survived (1.17 f 0.05 U/ml; II = 18) compared to those who did not (0.71 k 0.03 U/ml; II = 12; P < 0.001) and patients who died had significantly lower levels than control subjects (P < 0.01) indicating the importance of maintenance of normal inhibitor levels in patients with FHF. The leucocyte activation and release of elastase in FHF was linked to activation of the coagulation system; elastase-cur-antitrypsin complex levels correlated signifi-
Correspondence College School
to: Dr. Roger Williams, Institute of Liver Studies, King’s College Hospital of Medicine and Dentistry, Denmark Hill, London SE5 9RS, UK.
and King’s
212
cantly with thrombin-antithrombin III complex levels (r = 0.68; P < 0.001) and inversely with fibrinogen (r = - 0.71; P < 0.001).
Introduction Patients with fulminant hepatic failure (FHF) are particularly susceptible to bacterial infection [l] which, among other complications, contributes to the high mortality of this condition [2]. Septic shock can develop and this may aggravate the severity of the coagulation disturbances [3,4] and the final multi-organ failure. As one part of the systemic response to infection leucocyte elastase (EC 3.4.21.11, a neutral lysosomal proteinase stored in polymorphonuclear leucocytes, is released into the circulation [5]. However leucocyte elastase is thought to be involved in the tissue damage which occurs in septicaemia, endotoxic shock and emphysema [6,7]. In addition to its action to destroy opsonised infecting organisms and cell debris [8], it can have deleterious effects such as complement activation, release of toxic peptides, degradation of connective tissue and immunoglobulins [9]. It also inactivates coagulation factors and the inhibitor antithrombin III [lo] and activates the fibrinolytic system [ 111. The major physiological inhibitor of elastase, q-antitrypsin [6], is a member of the serpin family of protease inhibitors [12]. It is synthesised mostly in the liver and also in monocytes and macrophages but to a much lesser extent. Although low concentrations might be expected in FHF due to the profound loss of liver synthetic function, it is also an ‘acute phase’ protein and an increase in levels could occur as a result. Because of the rapid binding of elastase to a,-antitrypsin it is not possible to measure free elastase activity in the circulation. Measurement of the concentration of the inert complex gives an estimate of the release of elastase into the circulation during leucocyte activation and may be a marker of sepsis in these patients [13]. In this study we have measured plasma levels of elastase-cu,-antitrypsin complex, using a solid phase enzyme linked sandwich immunoassay, in 30 patients with FHF to study the relationship between leucocyte activation and the severity of the disease and to the presence or absence of bacterial infection. Studies were also carried out during haemodialysis which, by virtue of activation of leucocytes by the dialysis membrane, could affect the release of elastase into circulation. Material and methods The 30 patients studied had grade 3 or 4 encephalopathy due to FHF as defined by Trey and Davidson [14]. In 25 patients the cause of the FHF was paracetamol overdose and in 5, viral hepatitis (2 hepatitis A, 2 hepatitis B, 1 hepatitis non-A non-B). The measurements were carried out on admission. Blood was collected into citrate and immediately centrifuged for 10 min at 2,000 x g at 4 o C. The resulting plasma was stored in portions at - 70 o C until assayed. Infection was diagnosed as previously described 111. Infection was suspected when the axillary temperature exceeded 38°C on two successive 4 h
213
readings, a peripheral blood leucocyte count exceeded 11 x 109/1 or inflammation of a wound or catheter sites or radiological signs of pulmonary consolidation were present. Blood, urine and sputum were microbiologically examined. Infection was confirmed when clinically significant positive cultures were made from body fluids or sites of infection. Microbiological cultures were performed daily and when clinically indicated. Six further patients were studied during a single 4 h haemodialysis required on account of development of renal failure. Haemodialysis was carried out using a Biospal 2400s dialyser containing a polyacrylonitrile membrane AN69S (Hospal, Rugby, UK). Blood samples were taken before haemodialysis then at 10 min, 30 min and 1, 2, 3 and 4 h after the start of haemodialysis. The patients received a bolus of 5,000 units of heparin prior to haemodialysis followed by a continuous infusion of heparin so that the whole blood activated clotting time was maintained between 200-250 s. Assays
Elastase-cr,-antitrypsin complex was measured using a solid phase, enzyme-linked sandwich immunoassay (Merck Ltd., Darmstadt, Germany) in which the first antibody, specific for elastase, is bound to the solid phase and the labelled antibody is specific for cy,-antitrypsin. The final colour change was measured at a wavelength of 405 nm using an Ultraspec 4050 spectrophotometer (Pharmacia-LKB Ltd., South Croydon, UK). Each sample was assayed in duplicate. cu,-Antitrypsin activity was measured using an indirect chromogenic substrate assay (Kabivitrum, Uxbridge, UK). (Y,-Antitrypsin concentration was measured by radial immunodiffusion (Behringwerke Ag, Marburg, Germany). There was a significant correlation between cu,-antitrypsin activity as measured by chromogenic substrate assay and cY,-antitrypsin concentration as measured by radial immunodiffusion (R = 0.89, P < 0.001). Thrombin-antithrombin III complex was measured by an enzyme-linked immunosorbent assay using a kit donated by Behringwerke Ag, Marburg, Germany. Fibrinogen was measured by the method of Claus [15] using a Fibrometer (Becton Dickinson, UK). Whole blood activated clotting time was measured using Haemochron (Technidyne, Metuchen, USA). The control group consisted of 10 healthy laboratory volunteers. Statistics Significance was assessed between groups using Wilcoxon’s rank sum test. Correlation between parameters was assessed by Spearman test. Results are reported as either median and range or mean f SE as appropriate. Results Laboratory measurements taken on admission of the 30 patients are shown in Table I. Eighteen (60%) of the patients survived to leave hospital comprising 14 of
214 TABLE I Laboratory parameters of patients on admission, median values and ranges shown Sex (M/F)
Prothrombin Time (secsl
AST a (hi/l)
Bilirubin (~mol/l)
Creatinine (~mol/l)
POD a Median Range
15/10
65 25-240
1550 122-10000
140 55-264
311 69-828
Viral Median Range
2/3
30 20- 70
950 43- 4230
210 174-333
95 58-432
5- 17
45-105
Normal range
7-
12- 14
a AST, serum aspartate aminotransferase;
40
POD, Paracetamol overdose.
the 25 patients with paracetamol overdose and 4 of 5 patients with viral hepatitis. At the time of admission 13 (43%) patients were considered to be infected of whom 6 (46%) survived to leave hospital. A further 9 patients became infected later in the course of their illness of whom 6 survived to leave hospital. Fourteen patients had renal failure (creatinine > 300 pmol/l) on admission of whom 9 (64%) were clinically infected. Measurement of elastase-a,-antitrypsin
complex and a,-antitrypsin
Elastase-a,-antitrypsin complex levels on admission were significantly greater in patients with FHF than in control subjects (303 + 51 pg/l compared to 37 & 5 pg/l; P < O.OOl>.None of the patients had levels within the normal range (Fig. 1).
Control
Survivors
Non-survivors p(O.02
Fig. 1. Concentration
of elastase-cu,-antitrypsin complex, on admission, in patients with fulminant hepatic failure in relation to outcome.
215
Not infected
Infected p(O.01
Fig. 2. Concentration
of elastase-cY,-antitrypsin complex in patients with fulminant hepatic failure with and without infection.
The 18 patients who survived had significantly lower levels of the complex on admission than those who died (212 _t 49 pg/l compared to 440 f 94 pg/l; P < 0.02). Levels in those who were infected at the time of sampling had significantly greater concentration of complex compared to those who had no infection (486 + 96 pg/l compared to 197 + 44 pg/l; P < 0.01) (Fig. 2). Patients with and without renal failure had similar concentrations of the complex (365 f: 89 pg/l and 259 + 60 pg/l, respectively). Patients with viral hepatitis had significantly lower complex levels than those with paracetamol overdose (104 + 30 pug/l and 343 f 60 pg/l, respectively; P < 0.05). However there were only 5 patients in the viral group and the levels may be affected by the high proportion of these patients who survived. When the activity levels of a,-antitrypsin, the inhibitor of leucocyte elastase, were measured in the total group of patients with FHF on admission they were not significantly different to normal control subjects (0.99 + 0.06 U/ml and 0.97 + 0.05 U/ml, respectively) (Fig. 3). Values in the patients who survived were significantly greater than those who died (1.17 k 0.05 U/ml compared to 0.71 + 0.03 U/ml; P < 0.001) and patients who died had significantly lower activity levels than the healthy controls (P < 0.01). There was no significant difference between infected and non-infected patients (0.92 f 0.08 U/ml and 1.03 k 0.08 U/ml, respectively) patients with and without renal failure (0.99 f 0.08 U/ml and 0.98 f 0.09 U/ml, respectively) or patients with viral hepatitis and paracetamol overdose (0.98 f. 0.15 U/ml and 0.99 + 0.07 U/ml, respectively). Elastase-cY,-antitrypsin complex was significantly correlated with thrombin-antithrombin III complex (Fig. 4) and inversely correlated with fibrinogen concentration (r = -0.71; P < 0.001). a,-Antitrypsin was significantly correlated with fibrinogen concentration. Patients who did not survive had both low fibrinogen and low a,-antitrypsin activity levels on admission (Fig. 5). There was no significant correlation between elastase-cu,-antitrypsin complex and a,-antitrypsin activity.
216 2.0
5
: 3 .ca tf
1.5
1.0
F i
0.5
0.0
Control
Non-survivors
p(O.001
p(O.01
Fig. 3. Concentration
of a,-antitrypsin
Survivors
in patients with fulminant hepatic failure on admission in relation to outcome. r=0.66:
0
10
20
30
40
50
60
Thrombin-antithrombin
Fig. 4. Correlation
70
80
~(0.001
90
100
Ill &g/l)
of elastase-a,-antitrypsin complex with thrombin-antithrombin patients with fulminant hepatic failure.
III complex in
TABLE II Elastase-a,-antitrypsin
Complex (I.Lg/f) Mean +SE
complex and white cell numbers in 6 patients during 4 h of haemodialysis
PRE
10 min
30 min
lh
2h
3h
4h
224 28
285 67
307 60
316 59
392 98
541 214
660 30
81.2 17.8 P < 0.05
82.5 16.9 P < 0.05
108.9 18.8
107.9 19.7
119.5 25.9
White cells t% of initial) Mean &SE
87.1 11.0
(Wilcoxon’s paired rank sum test, compared to initial value.)
217 0
6.0 v0.70;
al -antitrypsin 0 0
p(O.001
(u/ml)
Survivors Non-survivors
Fig. 5. Correlation of fibrinogen with a,-antitrypsin
in patients with fulminant hepatic failure.
Effect of haemodialysis
Elastase-cr,-antitrypsin complex was raised at the start of haemodialysis and increased further throughout the course of treatment (Table II). When white cell counts were expressed as percentage of initial levels, they were significantly decreased at 1 and 2 hours after the start of haemodialysis, subsequently they returned to initial levels (Table II). Thrombin-antithrombin III complex was also raised at the start (30.7 + 8.5 pg/l) and increased further during haemodialysis (86.9 + 33.4 pg/l at 1 h and 66.0 + 12.1 pg/l at 4 h). Discussion
The high levels of the elastase-cl!,-antitrypsin complex found in plasma in the present study strongly suggest that there is leucocyte activation, degranulation and release of proteases in FHF. Loss of hepatocyte and Kupffer cell function [161 could cause some increase in complex level due to lack of clearance, but as the half-life of the complex is short in healthy subjects [13] the majority of the increase in the present subjects is more likely to have been due to an increased and continued elastase release. The leucocyte activation could be caused by tissue destruction as has been demonstrated in other conditions which involve tissue damage [12] and where increases in elastase-cw,-antitrypsin complex have also been demonstrated. The findings in the present study that patients who are infected on admission have higher levels of elastase-a,-antitrypsin complex, is consistent with infection causing leucocyte activation and release of elastase. Seventy three percent of the patients had episodes of infection during the course of their illness. Although renal failure has been linked to infection in FHF [l], in the present study there was no significant increase in the complex in those with renal failure. Thus, the increases
218
in elastase-cY,-antitrypsin complex are likely to be due to both tissue damage and sepsis. Although patients who survived had significantly lower levels of the complex there was much overlap in the data and this would make the test of little value in the prognosis of individual cases. If local concentrations of elastase exceed those of the inhibitor the effects of free elastase may contribute to some of the serious complications, such as multiorgan failure, seen in patients with FHF. a,-Antitrypsin is an acute phase protein and would be expected at increased level in a severe disease. The levels measured in FHF are probably a balance between loss of production due to poor liver function and the acute phase response. Whether the lower levels in patients who died is due to a failure of the liver to undergo the acute phase response or reflects a greater loss of hepatocyte function is unclear. It is unlikely to be due to a greater consumption of the inhibitor as there was no correlation between a,-antitrypsin activity and elastase-c-u,-antitrypsin complex concentration. The coagulation system is known to be activated in FHF with increased levels of fibrin degradation products [17] and the thrombin-antithrombin III complex Cl81 as measured in this study. The strong correlation between the elastase-ai-antitrypsin complex and the thrombin-antithrombin III complex suggest that the activation of leucocyte and coagulation systems are linked. It is possible that a single stimulus, such as endotoxin or cytokine release activates both systems simultaneously. Tumour necrosis factor production has been shown to be increased in FHF 1191,but this cytokine alone does not cause realease of elastase activity in vitro 1201.Elastase has been shown to inhibit antithrombin III activity in vitro [lo] and low levels of antithrombin III may allow increased production of thrombin and thus activate the coagulation system. The latter, in its turn, may activate release of leucocyte elastase as it has been shown that antithrombin III concentrate and plasma substitution reduces elastase-cr,-antitrypsin complex levels in patients with septic shock [21]. A further link between the two systems is the significant correlation between fibrinogen, also an acute phase protein, and a,-antitrypsin. We have shown previously [18] that if fibrinogen levels are low on admission in FHF the prognosis is poor. If both fibrinogen and a,-antitrypsin are present at low levels on admission the prognosis is very poor indeed (Fig. 5). The increases in elastase-cY,-antitrypsin complex and the loss of white cells during haemodialysis demonstrate that there is still activation of white cells by the procedure despite adequate heparinisation. Haemodialysis is known to effect leucocytes in circulation by activation of the complement system [22]. Although complement proteins are present at low levels in FHF, activated leucocytes may still release elastase when in contact with the dialysis membrane. Although the loss of white cells was small in this study (17.5% at 1 h), probably due to the good biocompatibility of the polyacrylonitrile membrane used, the effect on white cells was sufficient for elastase to be released into the circulation despite the return of white cell count to initial levels. The coagulation system is also activated during haemodialysis [18] as shown by the increase in thrombin-antithrombin III levels, which is further supporting evidence for a link between the two systems.
219
References 1 Roland0 N, Harvey F, Brahm J, Philpott-Howard J, Alexander GJM, Gimson AES, Casewell M, Fagan E, Williams R. Bacterial infection in acute liver failure. Hepatology 1990;11:49-53. 2 O’Grady JG, Alexander GJM, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology 1989;97:439-445. 3 Lechner K, Niessner H, Thaler E. Coagulation abnormalities in liver disease. Semin Thromb Haemostas 1977;4:40-56. 4 O’Grady JG, Langley PG, Isola LM, Aledort LM, Williams R. Coagulopathy in fulminant hepatic failure. Semin Liver Disease 1986;6:159-163. 5 Ohlsson K, Ohlsson I. The neutral proteases of human granulocytes. Isolation and partial characterisation of granulocyte elastases. Eur J Biochem 1974:42:519-527. 6 Egbring R, Schmidt W, Fuchs G, Havemann K. Demonstration of granulocyte proteases in plasma of patients with acute leukemia and septicemia with coagulation defects. Blood 1977;49:219-231. 7 Crystal RG. a,-Antitrypsin deficiency, emphysema and liver disease. Genetic basis and strategies for therapy. J Clin Invest 1990;85:1343-1352. 8 Babior BM. Oxidants from phagocytes; agents of defense and destruction. Blood 1984$X959-966. 9 Havemann K, Janoff A. Neutral proteases of human polymorphonuclear leukocytes. Baltimore and Munich: Urban and Schwarzenberg, 1978. 10 Jordan RE, Nelson RM, Kilpatrick J, Newgren JO, Esmon PC, Fournel MA. Inactivation of human antithrombin by neutrophil elastase. J Biol Chem 1989;264:10493-10500. 11 Gilboa N, Neuman PH, Gutmann JM, Del Vecchio PJ, Gudewicz PW. Evidence for regulation of endothelial plasminogen-activating system by ‘polymorphonuclear leukocyte elastase. Thrombosis Res 1989;54:467-475. 12 Carrel1 RW. cr,-Antitrypsin: molecular pathology, leukocytes, and tissue damage. J Clin Invest 1986;78:1427-1431. 13 Harm K, Bartfeld KP, Mathew T. Plasma concentrations of granulocytic elastase-a,-proteinase inhibitor complex in patients with severe head injury, multiple trauma or cerebral bleeding. Clin Biochem 1989;22:149-153. 14 Trey C, Davidson CS. The management of fulminant hepatic failure. In; Popper H, Schaffner F, eds. Progress in liver disease, Vol 3. New York: Grune and Stratton, 1970;282-298. 15 Clauss A. Gerinnungsphysiologisches schnellmethode zur bestinnung des fibrinogens. Acta Haematologica 1957;17:237-246. 16 Canalese J, Gove CD, Gimson AES, Wilkinson SP, Wardle EN, Williams R. Reticuloendothelial system and hepatocyte function in fulminant hepatic failure. Gut 1982;23:265-269. 17 Hughes RD, Lane DA, Ireland H, Langley PG, Gimson AES, Williams R. Fibrinogen derivatives and platelet activation products in acute and chronic liver disease. Clin Sci 1985;68:701-707. 18 Langley PG, Forbes A, Hughes RD, Williams R. Thrombin-antithrombin III complex in fulminant hepatic failure: evidence for disseminated intravascular coagulation and relationship to outcome. Eur J Clin Invest 1991;in press. 19 Muto Y, Nouri-Aria KT, Meager A, Alexander GJM, Eddleston AWLF, Williams R. Enhanced tumour necrosis factor and interleukin,l in fulminant hepatic failure. Lancet 1988;ii:72-74. 20 Yonemaru M, Stephens KE, Ishikaza A, Zheng H, Hogue RS, Crawley JJ, Hatherill JR, Raffin TA. Effects of tumor necrosis factor on PMN chemotaxis, chemiluminescence, and elastase activity. J Lab Clin Med 1989;114:674-681. 21 Seitz R, Wolf M, Egbring R, Havemann K. The disturbance of haemostasis in septic shock: role of neutrophil elastase and thrombin, effects of antithrombin III and plasma substitution. Eur J Haematol 1989;43:22-28. 22 Hakim RM, Breillatt J, Lazarus JM, Port FK. Complement activation and hypersensitivity reactions to dialysis membranes. N Engl J Med 1984;311:878-882.