Tumor necrosis factor-α in liver transplantation and resection

376

Journal of Hepatology, 1992; 16:376-37t. © 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved. 0168-8278/92/$05.0~

HEPAT 01262

Tumor necrosis factor-a in liver transplantation and resection No evidence for a key role in ischemia-reperfusion injury Olivier Chazouill6res a, J6r6me G u 6 c h o t b, Pierre Balladur c, Jean-Pierre Masini d, Eric Delva d, A b d e r r h a m a n e Laribi b, Jacqueline G i b o u d e a u b, Andr6 Lienhart d, Rolland Parc c, Raoul P o u p o n a and Laurent H a n n o u n ~ aUnitb d'Hbpatologie, bService de Biochimie A, cService de Chirurgie Digestive and dDbpartement d'Anesth~)sie-Rbanimation, Hfpital Saint Antoine, Paris, France

(Received2 March 1992)

Experimental studies have shown that liver ischemia-reperfusion induces Kupffer cell activation and tumor necrosi: factor-~ (TNFct) release. The aim of this work was to determine whether severe hepatic ischemia and subsequen reperfusion triggers TNFct release in man. Serum TNFct was measured before and 3, 10, 30, 60, 120 min afte: revascularization and postoperatively at day 1 and 2 in 11 patients with orthotopic liver transplantation (group I and 4 patients with liver resection with vascular occlusion (group 2). In group 1, TNFct levels decreased during the first few minutes of reperfusion, then increased slightly to peak at 120min (40+13 pg/ml). Primary non-functiot occurred in 1 patient in whom low peroperative levels of TNFct levels were measured. In group 2, no significan changes in TNFct levels were observed. These data, in a small number of patients: (a) show that hepatic ischemia reperfusion does not result in major TNFct production; (b) do not support a primary pathogenic role for TNF0t it damage after ischemia-reperfusion in humans. K e y words: Tumor necrosing factor ct; Liver transplantation; Liver resection; Ischemia-reperfusion

Primary graft dysfunction following liver transplantation is common and the most severe form, primary nonfunction (PNF), still occurs in 2-12% of cases (1). P N F is a devasting complication with significant morbidity and mortality, even when early re-transplantation is performed (2). P N F is probably multifactorial in origin. However, liver injury related to prolonged hypothermic storage and reperfusion plays a key role. Although mechanisms of liver ischemia-reperfusion (I/R) injury are still poorly understood, recent experimental studies have shown that reperfusion injury after cold ischemic storage is characterized by a loss of endothelial cell viability associated with Kupffer cell activation (3). Kupffer cells, representing the largest fixed macrophage population in the body, have a capacity for cytokine production when activated. Tumor necrosis factor-~ (TNFct) is a cytokine involved in many types of inflammatory processes and

septic shock (4). TNFct is directly toxic for endothelia cells (5), and it has recently been shown that hepatic I/R in the rat results in TNFct production intimatel~ associated with pulmonary and hepatic injury (6). Taker together, these experimental data suggest that, during I/R, the human liver graft could release TNFct wit! deleterious actions on both the liver and other organs the administration of anti-human T N F antibody couk thus have beneficial effects. This study was designed tc determine whether severe hepatic ischemia (in both live: transplantation and resection) and subsequent reperfu sion triggers TNF~ release in man.

Patients and Methods

Between January and June 1991, I1 consecutiw orthotopic liver transplantations (OLT) (group 1) a n d ,

Correspondence to: J,~r6meGu6chot, Laboratoire de Biochimie-Hormonologie, H6pital Saint Antoine, 184 rue du Faubourg Saint Antoine, 7557

Paris Cedex 12, France.

377

TNF0t IN LIVER I S C H E M I A - R E P E R F U S I O N INJURY

liver resections with vascular occlusion (group 2) were studied. The indications for OLT were elective in 7 patients (biliary cirrhosis in 3 and post-necrotic cirrhosis in 4; emergency OLT was performed in 4 patients (fulminant hepatitis in 3 and PNF in 1). Donor hepatectomy was performed by means of standard techniques with simultaneous removal of the kidneys and heart (7). The livers were preserved for 813+79 min (mean + SE) (range 435-1215) in University of Wisconsin (UW) solution (8). Recipient hepatectomy and liver replacement were performed using classical techniques (9). A peroperative intravenous bolus of steroids (methylprednisoione, 10 mg/kg) was injected at the beginning of the anhepatic phase. Postoperative immunosuppression consisted of triple-drug therapy (cyclosporine, prednisolone and azathioprine). In group 2, 4 liver resections (2 right hepatectomies, 1 right extended hepatectomy and 1 left lateral segmentectomy) were performed with the use of hepatic vascular exclusion (HVE) which associates portal triad clamping (PTC) and occlusion of the inferior vena cava below and above the liver (10). Indications for liver resection were metastases in non-cirrhotic liver in 3 cases and hepatocellular carcinoma in cirrhotic liver in 1 case. Mean duration of normothermic liver ischemia was 46 min and ranged from 35 to 60 min. In both groups, systemic peroperative arterial blood samples were collected at the end of the anhepatic phase, then 3, 10, 30, 60, 120 min after revascularization and postoperatively at day 1 and day 2. Initial graft function was determined by clinical assessment, bile production, and laboratory parameters including ASAT, ALAT, prothrombin time and total serum bilirubin. For TNF~ assay, blood was collected in sterile tubes and serum was rapidly separated after coagulation and stored at - 80 °C until assay. At the end of storage in UW solution (group 1), the livers were perfused with Ringer's lactate solution and an aliquot of the effluent was used for TNF0t measurement. TNFct was measured using a specific immunoradiometric assay kit (TNFct-IRMA, Medgenix Diagnostics, Brussels, Belgium). This test is based on coated-tube separation and an oligoclonal system, in which several monoclonal antibodies directed against distinct epitopes of TNFct are used. The sensitivity is 5 pg/ml, intra-assay variation is less than 8%, and inter-assay variation is less than 9%. Samples from the same subject were tested in the same assay. The normal upper limit for serum TNFct is estimated as 15 pg/ml. The results are expressed as mean + SE unless

otherwise stated. Date were analyzed using the nonparametric Wilcoxon matched pairs signed-ranks test or Spearman rank correlation coefficient.

Results

In group 1, TNFct levels decreased significantly during the first few minutes of reperfusion relative to prereperfusion levels (p<0.01). Levels then progressively increased to peak at 120 min (p<0.01 vs. 3 min) (Fig. 1). TNFct was detectable in the perfusion effluent at the end of the reperfusion phase (16.8 + 5.8 pg/ml). Postoperative levels were low except in the patient with PNF (Fig. 2). PNF, early suspected in the operative room by a lack of bile secretion and confirmed by postoperative serum transaminases >3000UI/! and prothrombin time >50s, occurred in 1 patient; this patient was retransplanted at day 5 (Fig. 2). io° ....

°.,.,.°°

:Re p~e~us!on i ~

---0.-----

Transplantation Resection

A

40.

,..=

20.

d i

-iofo ab

8

/a

~

120'rain/'/

I)l

I)2

Time

Fig. i. Per- and post-operative serum levels of T N F a (pg/ml) in group I (liver transplantation) and in group 2 (liver resection) patients.

6o

~3

second graft

,o

F', t..

-10 10 30

60

120'mln//

D1

D2

Time

Fig. 2. Per- and post-operative serum levels of T N F a (pg/ml) in the patient with PNF of the first graft.

378 In group 2, no significant changes in TNFcx levels were observed (Fig. 1). In group 1, no obvious relationship was found between peroperative T N F ~ levels and the postoperative course. There was no relationship between postoperative serum ALAT peak level (range 283-3058 I U / I , on day 2 + 0 . 3 post-transplantation) and T N F ~ at the end of the anhepatic phase (rs=0.182) or 120 min after reperfusion (rs= 0.054); and between the time for serum transaminase normalization (9.4+1.7 days) before and 120 min after reperfusion (rs=0.067 and 0.238 respectively). In 2 patients who had a biopsy-proven acute early rejection in the 2 months post-transplantation, TNFcc was 2.7 and 23.9 pg/ml in one, and 10.4 and 3.3 pg/ml in the other (median value 17.9 and 23.9 pg/ml) before and 120 min after reperfusion respectively. Furthermore, in the patient with PNF, T N F ~ levels were very low peroperatively and increased only in the postoperative days. During retransplantation with a new graft which functioned well, initial peroperative T N F levels were high and fell quickly (Fig. 2).

Discussion These data show that hepatic I/R in humans does not result in major T N F a production. In the transplantation group, the significant decrease in T N F ~ levels just after reperfusion could be related to an increased volume of distribution (hepatic volume) or to hepatic clearance of cytokines, since the liver is the main cytokine scavenger (1 I). The moderate increase in T N F ~ level after reperfusion might be due to Kupffer cell activation by endotoxins which enter the circulation during surgery or, more likely, by the process of I/R itself as suggested by experimental data (6). The T N F ~ peak at 120 min is consistent with an hepatic origin, as suggested by the detection of T N F ~ in the effluent of stored livers, although an extrahepatic origin cannot be ruled out. Nevertheless, levels were very much lower than in patients with septic shock (12,13). Since the T N F ~ -

References I Todo S, Nery J, Yanaga K, Podesta L, Gordon RD, Starzl TE. Extended preservation of human liver grafts with UW solution. J Am Med Assoc 1989; 261:711-4. 2 Shaw Jr BW, Gordon RD, lwatsuki S, Starzl TE. Hepatic retransplantation. Transplant Proc 1985; 17: 264-71. 3 Caldwell-KenkelJC, Currin RT, Tanaka Y, Thurman RG, Lemasters JJ. Kupffer cell activation and endothelial cell damage after storage of rat livers: effects of reperfusion. Hepatology 1991; 13: 83-95.

o. CHAZOUILLI~RESet al. I R M A used in this study is highly specific and sensitive and since the blood samples were collected using very strict precautions, it can be assumed that results are reliable. The intravenous administration of steroids before reperfusion could have altered the release of cytokines since corticosteroids have been reported to inhibit lipopolysaccharide-induced production of TNFcc by Kupffer cells (14). In particular, the OKT3-induced release of cytokines is markedly decreased if corticosteroids are injected 1 h before (15). It is thus possible that post-reperfusion levels of T N F ~ were reduced in our patients. However, even in the absence of corticosteroids, high levels of T N F ~ were not observed in group 2 (normothermic ischemia of both cirrhotic or noncirrhotic livers). Several factors may explain this finding, including the type of ischemia (cold vs. normothermic) and, especially, the duration of ischemia. In addition, it should be remembered that in group 2 patients, the potential for hepatic cytokine production during reperfusion was reduced by the resection of a significant part of the liver (20-50%). Lastly, P N F was not associated with high peroperative levels of TNF~, suggesting that TNFc¢ and Kupffer cell activation are not involved in its pathogenesis, at least in some patients. In our patient, the postoperative increase in TNFc~ levels after the onset of P N F would appear to be a consequence rather than a cause of PNF. Although there is evidence of no relationship between peroperative T N F ~ and clinical outcome, this is not proof of a lack of relation between TNFc¢ and I/R injury, since it cannot be measured directly. In conclusion, our results in a small number of patients do not support a major role for TNFc¢ in the damage after hepatic I/R in humans. These findings do not form a basis for the study of anti-TNF antiserum in human liver transplantation, even though a beneficial effect has been reported in a rat model of hepatic I/R (6) and a recent study (16) showed that an intraoperative elevation of TNFc~ seems to precede early rejection. Nevertheless, only controlled studies can show the effect of anti-TNF in the prevention of P N F or rejection.

4 Beutler B, Cerami A. Cachectin: more than a tumor necrosis factor N Engl J Med 1987; 36: 379-85. 5 Sato N, Goto T, Haranaka K, Satomi N, et al. Actions of tumor necrosis factor on cultured vascular endothelial cells: morphologic modulation, growth inhibition and cytotoxicity.J Natl Cancer Inst 1986; 776: I 113-8. 6 Colletti LM, Remick DG, Burtch GD, Kunkel SL, Strieter RM. Campbell Jr DA. Role of tumor necrosis factor-:~in the pathophysiologic alterations after hepatic ischemia/reperfusion injury in the rat. J Clin Invest 1990; 85: 1936-43. 7 Starzl TE, Hakala TR, Shaw BW, et al. A flexible procedure for

TNF0t IN LIVER 1SCHEMIA-REPERFUSION INJURY

8 9 10

I1 12

multiple cadaveric organ procurement. Surg Gynecol Obstet 1984; 158: 223-30. Kalayoglu M, Sollinger HW, Stratta RJ, et al. Extended preservation of the liver for clinical transplantation. Lancet 1988; i: 617-9. Starzl TE, iwatsuki S, Van Thiel DH, et al. Evolution of liver transplantation. Hepatology 1982; 2: 613-36. Delva E, Camus Y, Nordlinger B, et al. Vascular occlusions for liver resections. Operative management and tolerance to hepatic ischemia: 142 cases. Ann Surg 1989; 209:211-8. Andus T, Bauer J, Gerok W. Effects of cytokines on the liver. Hepatology 1991; 13: 364-75. Girardin E, Grau GE, Dayer JM, Roux-Lombard, Lambert PH. Tumor necrosis factor and interleukin-I in the serum of children with severe infectious purpura. N Engl J Med 1988; 319: 397-400.

379 13 Pinsky MR, Vincent JL, Deviere J, et al. Serum cytokine levels in human septic shock: relation to multiple-systems organ failure and mortality. Crit Care Med 1989; 17: 975-83. 14 Kutteh WH, Rainey WE, Carr BR. Glucocorticoids inhibit lipopolysaccharide-induced production of tumor necrosis factor-or by human fetal Kupffer cells. J Clin Endocrinol Metab 1991; 73: 296-301. 15 Chatenoud L, Legendre C, Ferran C, Bach JF, Kreis H. Corticosteroid inhibition of the OKT3-induced cytokine-related syndrome. Dosage and kinetics prerequisites. Transplantation 1991; 51: 334-8. 16 Ffigger R, Hamilton G, Steininger R, Mirza D, Schulz F, Mfihlbacher F. lntraoperative estimation of endotoxin, TNFct, and IL6 in orthotopic liver transplantation and their relation to rejection and postoperative infection. Transplantation 1991; 52: 302-6.