- Email: [email protected]
Sinusoidal Flow Block after Warm Ischemia in Rats with Diet-Induced Fatty Liver
JOURNAL OF SURGICAL RESEARCH ARTICLE NO.
70, 12–20 (1997)
JR975077
Sinusoidal Flow Block after Warm Ischemia in Rats with Diet-Induced Fatty Liver1 Kenichi Hakamada, M.D., Mutsuo Sasaki, M.D., Katsuro Takahashi, M.D., Yutaka Umehara, M.D., and Mitsuru Konn, M.D. Second Department of Surgery, Hirosaki University School of Medicine, 5 Zaifu-cho, Hirosaki 036, Japan Submitted for publication July 5, 1996
vation [2–5], 1–17% of recipients still experience PNF [2–4, 6, 7]. Immediate retransplantation is mandatory and results in survival for 70% of patients, but such waste of time, money, and manpower [7], as well as the increased morbidity itself, mandate avoidance of PNF by all means. Fatty liver in donors has been associated with PNF [8–13]. Portmann and Wight [8] first reported a case with unexpected severe fatty infiltration of the donor liver in which the graft never functioned. The outcomes of steatotic allografts then were evaluated using timezero biopsy at other liver transplant centers [11, 12]. Now donor livers with massive fatty infiltrates are thought to be at great risk for PNF and are discarded based on frozen-section examination [13]. However, the processes by which steatotic grafts undergo PNF are incompletely known. Todo et al. [9] have reported two examples of PNF with severe fatty infiltration, in which they observed extracellular fatty droplets with disruption of the sinusoidal microvasculature in autopsy specimens. They speculated that a degree of hepatocellular death occurred in the course of graft preservation and that rupture of some steatotic hepatocytes released fat droplets into sinusoids, compromising hepatic microcirculation. However, PNF can occur despite shorter preservation times. A previous report found no significant difference in warm and cold ischemic times between PNF cases and the patients without this complication [14]. Another hypothesis is that sinusoidal microcirculatory changes are the initial event after liver transplantation. Teramoto et al. [15] observed increased blood cell adhesion to sinusoidal lining cells after ischemia and reperfusion under in vivo microscopy using a rat fatty-liver model. They speculated that the microcirculatory disturbances caused liver cell death and contributed to primary graft nonfunction in fatty livers. Ischemia/reperfusion injury is thought to be of major clinical importance as a cause of early graft dysfunction. At the time of reperfusion, lymphocyte-functionassociated antigen-1 (LFA-1), which is expressed on the surface of leukocytes, adheres to intercellular adhesion molecule-1 (ICAM-1) on the endothelial cell membrane. Hepatic microcirculation is injured by the formation of
Donor livers with massive fatty infiltration reportedly are susceptible to ischemia/reperfusion injury after transplantation, which contributes to risk of primary nonfunction. We investigated the effect of warm ischemia and reperfusion on sinusoidal microcirculation in rats with fatty livers from a choline-deficient diet. Rats were subjected to partial hepatic warm ischemia for 30, 60, or 90 min. In a second study, an antiICAM-1 monoclonal antibody was injected intraportally 2 min after a 60-min ischemic period. In both studies, injury was assessed by liver histology 6 hr after vascular clamp release and by animal survival. After 30 min of hepatic warm ischemia, almost all control and fatty-liver rats survived 7 days. After 60-min ischemia, however, survival was significantly less in rats with fatty livers than in controls with normal livers (10% vs 90%, P õ 0.0001). Histologically, rats with fatty livers showed marked sinusoidal congestion, especially in the midzone of the acinus, while control rats showed no disturbance in microcirculation. In rats with fatty livers treated with intraportal injection of an anti-ICAM-1 antibody, sinusoidal microcirculation was well preserved and the 7-day survival rate after warm ischemia was improved (50% vs no antibody 10%; P Å 0.0112). In fatty livers, midzonal sinusoidal flow block occurs after hepatic warm ischemia and reperfusion. Although intraportal injection of an anti-ICAM-1 monoclonal antibody corrected this microcirculatory failure, animal survival was not as good as for controls without fatty livers. These results suggest that both sinusoidal microcirculatory failure and ischemic hepatocellular damage contribute to warm ischemia/reperfusion injury in fatty livers. q 1997 Academic Press
INTRODUCTION
Primary nonfunction (PNF) following liver transplantation is a serious complication, which increases postoperative morbidity and mortality. In spite of improvements in surgical technique [1] and organ preser1 This work was supported partly by Research Grants from the Ministry of Education of Japan.
0022-4804/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
JSR 5077
/
6n1c$$$221
12
06-13-97 13:07:29
srga
HAKAMADA ET AL.: SINUSOIDAL FLOW BLOCK AFTER WARM ISHCEMIA
intrasinusoidal thrombi, which now are stressed as a causative factor in ischemia/reperfusion injury of hepatic allografts [16–18]. In this paper, the authors introduced fatty changes in rats by feeding a choline-deficient diet and then investigated the effect of warm ischemia and reperfusion on hepatic microcirculation. Intraportal treatment with an anti-ICAM-1 monoclonal antibody also was examined for possible protective effect against sinusoidal microcirculatory disturbances [18]. MATERIALS AND METHODS Introduction of fatty liver in rats. Male Sprague–Dawley rats (CLEA Japan, Inc., Tokyo, Japan), weighing 170–190 g, were housed in metal cages with controlled light and dark cycles and given free access to water. To introduce fatty changes in livers, rats were fed a chow deficient in choline (Oriental Yeast Co., Tokyo, Japan) for 7, 14, 21, 28, 35, or 42 days (for each, n Å 10). Ten control rats were fed a standard commercial pelleted diet. Development of fatty liver did not affect body weight gain (data not shown). After completion of observation periods, all rats were anesthetized by either inhalation and the liver was removed, washed in cold saline, frozen in liquid nitrogen, and kept at 0707C until analyzed. Whole lipids in freezedried liver homogenates were extracted by chloroform:methanol (2:1, v/v) [19] and contents of triglyceride, phospholipids, and total cholesterol were determined enzymatically in a autoanalyzer (Hitachi 736, Tokyo, Japan). In 10 rats on the choline-deficient diet for 14 or 28 days (n Å 5 each), relative molar fatty acid fractions also were measured as percentages by gas chromatography using the same extracts. Extracts from experimental rats were compared with those from 5 control rats. Hepatic histology under light microscopy was studied in all rats after specimens were fixed in 10% Formalin, embedded in paraffin, and stained with hematoxylin and eosin. All slides were evaluated in a blinded manner by one of the authors, and the degree of fatty change was expressed as the proportion of hepatocytes with fatty infiltration in 10 randomly chosen fields from each slide. First study. Rats were fed a chow deficient in choline for 21 days to induce fatty livers and were compared with control rats given a standard pelleted diet for 21 days. All rats underwent hepatic warm ischemia. Under clean conditions, the abdomen was entered through a transverse incision after induction of ether anesthesia. The liver was dissected from peritoneal connections and the hilum of the median and left lobes was exposed. Segmental ischemia was induced by occluding the blood vessels and the bile duct to these lobes with an atraumatic vascular clamp. Then the abdomen was closed. After warm ischemia the clamp was released and the nonischemic right lateral and caudate lobes were resected. This technique avoided portal congestion, and endotoxin levels in portal vein and vena cava were not elevated (data not shown). Survival time and mortality in groups of 20 rats each after 30, 60, and 90 min of warm ischemia were recorded for both fatty-liver and control rats every 6 hr for 7 days postoperatively. Animals alive 7 days after operation were considered survivors and were killed for morphological study. Necropsy was performed on all animals to confirm the absence of surgical complications. Blood samples and liver tissue were taken 6 hr after the end of 30- and 60-min hepatic warm ischemia in both fatty-liver and control rats (n Å 10 each). Alanine aminotransferase (ALT), total bilirubin, prothrombin time (PT), fibrinogen, fibrin–fibrinogen degradation products (FDP), and antithrombin III (AT-III) were determined by routine methods. Liver specimens stained with hematoxylin and eosin were evaluated under light microscopy. Lobular distribution of congestion and necrosis was noted specifically. For electron microscopic examination, other portions of the same livers were fixed with 2.5% glutaraldehyde in 50 mM cacodylate buffer (pH 7.4). Ten blocks with a volume of about 8 mm3 were postfixed in 1.0% OsO4 in 0.1 M
AID
JSR 5077
/
6n1c$$$222
06-13-97 13:07:29
13
phosphate buffer, pH 7.4, for 2 hr, dehydrated in ethanol, and prepared with gold coating. Lobular structures were observed under a JSM-5310 scanning electron microscope (Nippon Denshi Co. Tokyo, Japan). Second study. Anti-rat ICAM-1 monoclonal antibody (1A29, Seikagaku Corp., Tokyo, Japan) was administered intraportally to evaluate its ability to maintain sinusoidal microcirculation after warm ischemia and reperfusion in rats with fatty livers. Rats on the choline-deficient diet for 21 days underwent partial hepatic warm ischemia and reperfusion in a manner identical to the first study. In the treatment group, 0.8 mg/kg of anti-ICAM-1 antibody was administered directly into the branch of superior mesenteric vein 2 min before unclamping [18], while in sham-operated rats, the same volume of normal saline was administered via the same vein at the same time. Seven-day survival rates (n Å 20, each group) and histologic changes (n Å 10 each) 6 hr after the 60-min warm ischemia were observed similarly to the first study. Statistical analysis. Statistical analysis was performed using the computer program Statview 4.02 (Abacus Concepts Inc., Berkeley, CA). Results were expressed as means { standard error of the mean (SEM). Overall statistical differences were determined according to analysis of variance (ANOVA), and Fisher’s test was used as a post hoc test. The unpaired Student’s t test was used to compare groups for ALT, total bilirubin, and other coagulation parameters. Cumulative survival rates after warm ischemia and reperfusion were obtained by the Kaplan–Meier method and were compared by the logrank test. P values less than 0.05 were considered significant.
RESULTS
Characterization of the Rat Fatty-Liver Model Triglyceride was the main lipid constituent in the livers of rats given the choline-deficient diet. Its content had increased by Day 14 and then remained relatively constant (no significant difference, or NS, from Days 14 to 42). Total amount of phospholipids had decreased by Day 7, remaining at the same level subsequently (NS from Days 7 to 42). Total cholesterol levels increased slightly at first, but were constant from Days 7 to 42 (NS; Fig. 1). Gas chromatography of the same liver tissue extracts demonstrated increased relative molar percentages of palmitic acid (C16:0) and oleic acid (C18:1; for both, control vs Days 14 and 28, P õ 0.01; Fig. 2). Under light microscopy, the extent of visible fat accumulation varied from minute droplets scattered in the cytoplasm of ballooning hepatocytes to distention of the entire cytoplasm of most cells by coalesced droplets. The proportion of hepatocytes with macro- or microvesicular fat vacuoles increased until the liver showed uniform macrovesicular fatty changes by Day 21. The proportion of fatty hepatocytes remained 75.9 { 5.3% from Days 21 to 42. In our models, fatty droplets appeared in the mid- to centrilobular zone first, but the entire lobular architecture was involved after Day 14. Hepatocyte necrosis was rarely seen even in rats on the diet for 42 days. First Study Cumulative survival rates after 30-, 60-, and 90-min warm liver ischemia were 100, 90, and 35% in control rats and 95, 10, and 5% in fatty-liver rats respectively at the 7th postoperative day (Fig. 3). There was no significant difference in survival between control and
srga
14
JOURNAL OF SURGICAL RESEARCH: VOL. 70, NO. 1, JUNE 1997
FIG. 1. Lipid content of rat liver after feeding with choline-deficient diet. The mean and SEM are shown for triglyceride, phospholipids, and total cholesterol content as mg/g liver tissue for rats given control and choline-deficient diets for 7 to 42 days. C, control; D x, rats given choline-deficient diets for x days; NS, no significant difference.
fatty-liver rats after 30-min hepatic ischemia, while the survival rate was significantly worse in fatty-liver rats after 60-min ischemia (P õ 0.0001). After 90-min ischemia, few rats survived in the fatty-liver group (vs control, P Å 0.0043). At 6 hr following 60-min warm ischemia, fatty-liver rats showed significant decreases in fibrinogen and ATIII levels compared with controls (both, P õ 0.01). These differences were greater between fatty-liver rats and control rats after 30-min ischemia, when fibrinogen and AT-III levels already had decreased significantly and PT was markedly prolonged in fatty-liver rats (P õ 0.01). ALT, total bilirubin, and FDP all increased significantly compared with controls as well. The latter parameters were not different between rats with 30- and 60-min hepatic warm ischemia (Table 1). The overall lobular architecture of the liver was well maintained, and no fatty droplets was seen in hepatocytes of control rats. Although sinusoidal lining cells and hepatocytes appeared normal 6 hr after 30-min ischemia, hepatocyte necrosis was seen diffusely scattered through the lobules after 60-min warm ischemia in control rats. In fatty livers, on the other hand, the most prominent finding after warm ischemia was sinusoidal congestion (Fig. 4). Its severity varied from localized congestion in the midzonal sinusoid to sinusoidal distention in the peripheral zone. The entire acinus showed marked congestion in some areas. This zonal distribution did not change with degree of fatty change; the midzone was
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
more vulnerable to sinusoidal congestion in lobules of both mildly and severely fatty livers. The extent of congestion was diffuse and similar between rats after 30and 60-min ischemia. Hepatocyte necrosis, however, was found throughout the lobular architecture in fattyliver rats after 60-min ischemia; this was less evident after only 30 min of ischemia. Under scanning electron microscopy, irregular, narrow sinusoids were plugged by thrombus formation in congested areas in rats with 30- or 60-min ischemia (Fig. 5). Second Study Survival rates were improved significantly, but not fully normalized, by intraportal administration of antiICAM-1 monoclonal antibody (50% vs 10%, P Å 0.0112; Fig. 6). Histologic findings indicated that sinusoidal congestion was markedly reduced in the treated group; no sinusoidal congestion was evident and both sinusoidal lining cells and hepatocytes were well preserved in 7 of 10 treated fatty-liver rats. In the other 3 rats, localized sinusoidal congestion was present in some areas. Extensive hepatocyte necrosis and congestion, which could be seen in fatty-liver rats treated with saline injection, were not seen in the anti-ICAM-1 antibody treatment group. DISCUSSION
Primary nonfunction following liver transplantation contributes significantly to postoperative morbidity
srga
HAKAMADA ET AL.: SINUSOIDAL FLOW BLOCK AFTER WARM ISHCEMIA
15
FIG. 2. Changes in fatty acid fractions in rat liver after feeding with choline-deficient diet. The mean and SEM are shown for relative molar percentage of fatty acid fractions in liver for rats on control and on choline-deficient diets for 14 and 28 days. Proportions of palmitic acid (C16:0) and oleic acid (C18:1) increased significantly in fatty liver models. *P õ 0.01: control (black bars) vs 14 days (white) and 28 days (stripes).
and mortality [2, 4, 6, 7]. Massive fatty infiltration in donor livers reportedly is associated with PNF [8–13], and most liver transplant centers tend to avoid such a steatotic allograft. The reason why fatty-liver grafts are susceptible to PNF is not known. The present study used rats with fatty livers from a choline-deficient diet to analyze survival and pathologic changes in the liver after warm ischemia and reperfusion. A choline-deficient diet is well known to reduce phosphatidylcholine synthesis in rats [20]. Phosphatidylcholine is one of the essential components of very low density lipoprotein (VLDL), which carries triglyceride from the liver to other organs. Choline deficiency consequently induces triglyceride accumulation in the liver [20–22]. Triglyceride was the main constituent of the increased lipid droplets in our model. In rats given a choline-deficient diet for more than 21 days, about 75% of all hepatocytes contained macrovesicular fat vacuoles under light microscopy, and in extracts the contents of triglyceride, phospholipids, and total cholesterol were relatively constant from Days 21 to 42. Gas chromatography of the extracts revealed increased molar percentages of palmitic and oleic acids. Although the mechanism by which fatty changes occur in human liver is not well known, triglyceride is the main component of accumulated fatty droplets when macro- or microvesicular steatosis occurs in either humans or rats. Moreover, palmitic acid is the main in-
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
creased fatty acid in humans with excessive dietary intake of carbohydrate [21]. Because of these biochemical and pathologic similarities between human fatty livers and the rat fatty-liver model, we used rats on a choline-deficient diet as an experimental model to analyze the mechanism of warm ischemia/reperfusion injury in fatty livers. In massive fatty livers, it has been reported that intracytoplasmic fat droplets bulging into the sinusoidal lumen narrow and distort hepatic microcirculatory pathways [15, 23, 24]. Consequently, intrahepatic portal resistance increases, as reflected by elevated portal vein pressure [23, 24]. In vivo microscopy has disclosed sluggish intrasinusoidal blood flow in rats with fatty livers [15]. In our model, where some 75% of hepatocytes have fatty droplets in the cytoplasm after Day 21, the sinusoidal lumen was compressed and distorted by ballooned hepatocytes, indicating that morphologically the microcirculation was impaired by fatty change per se. The most striking change in fatty livers after warm ischemia was sinusoidal congestion. The midzone was more vulnerable to congestion than other areas. This distribution of congestion in the hepatic acinus did not differ between lobules where fatty infiltration was localized to the mid- to centrilobular zone and lobules where the entire acinus was involved by fatty change. After hepatic ischemia and reperfusion, inflammatory
srga
16
JOURNAL OF SURGICAL RESEARCH: VOL. 70, NO. 1, JUNE 1997
FIG. 3. Cumulative-survival curves after hepatic warm ischemia in control and fatty-liver rats (Kaplan–Meier method). The survival rate after 60-min ischemia was significantly worse in fatty-liver rats than in controls. Numbers at right of curves designate minutes of hepatic warm ischemia for control (C, solid line) and fatty-liver (F, dotted line) rats. NS, no significant difference. *P õ 0.0001, **P Å 0.0003, ***P Å 0.0043 (log-rank test).
cytokines released from Kupffer cells [16, 25] activate endothelial lining cells which then express ICAM-1 on their cell membranes; they also induce LFA-1 expression on white cells [17, 18]. Thrombi may then form hepatic sinusoids as a result of adhesion of these molecules, causing hepatic microcirculatory failure [26]. In our studies, progressive accumulation of blood cells was seen as early as 6 hr following reperfusion. Increased resistance to portal blood flow through sinusoids in fatty livers can exaggerate sinusoidal flow block after warm ischemia.
Our data suggest that an even shorter warm ischemic time of 30 min caused hepatic microcirculatory failure, and that 60-min warm ischemia, which was compatible with survival for normal rats, resulted in death for rats with fatty livers. Prolonged PT, decreased levels of fibrinogen and AT-III, and elevated FDP in fatty-liver rats indicated that intravascular consumption coagulopathy occurred in the fatty-liver rats even after only 30 min of ischemia. Fibrinogen and AT-III levels decreased significantly in fatty-liver rats after 60-min warm ischemia. Because relatively little
TABLE 1 Effect of Hepatic Warm Ischemia on Liver-Function and Coagulation Indices 30-min warm ischemia Fatty-liver rat (n Å 10) ALT (U/L) Total bilirubin (mg/dl) Prothrombin time (sec) Fibrinogen (mg/dl) FDP (mg/ml) Antithrombin III (%)
1956.6 0.66 19.9 45.0 10.7 55.3
{ 151.3* { 0.08* { 0.8* { 7.6* { 1.6** { 2.3*
60-min warm ischemia
Control rat (n Å 10) 1196.3 0.14 16.5 123.4 6.2 85.8
{ 133.1 { 0.03 { 0.6 { 6.6 { 0.8 { 3.1
Fatty-liver rat (n Å 10) 1851.4 0.67 21.0 38.4 14.8 45.7
{ { { { { {
66.2 0.10* 0.9 7.6* 1.9 5.9*
Control rat (n Å 10) 1861.5 0.14 18.4 98.8 10.3 72.9
{ 212.6 { 0.02 { 1.3 { 7.1 { 1.8 { 2.8
Note. Values are means { SEM of 10 rats in each group with 6-hr reperfusion following 30- or 60-min hepatic warm ischemia. ALT, alanine aminotransferase; FDP, fibrin–fibrinogen degradation products. * P õ 0.01 and **P õ 0.05, compared with control rats with the same ischemic time.
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
srga
HAKAMADA ET AL.: SINUSOIDAL FLOW BLOCK AFTER WARM ISHCEMIA
17
FIG. 4. Photomicrograph (a) of rat liver in the fatty-liver group with 6-hr reperfusion following 30-min warm ischemia. Sinusoidal congestion is seen from the mid- to peripheral zone (hematoxylin and eosin, H & E; original magnification, 120). Higher magnification (b) demonstrates diffuse marked deposition of large droplets in hepatocytes, which compress the sinusoidal lumina. CV, central vein; PC, portal canal (H & E; original magnification, 1100).
change occurred in these variables in control rats after 30-min ischemia, differences between controls and fatty-liver rats were more prominent after 30-min than 60-min ischemia. Koneru et al. [27] also found in genet-
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
ically fat or lean Zucker strain rats that fatty livers were more sensitive to warm ischemia/reperfusion injury than nonfatty livers. In this study, sinusoidal congestion was seen pre-
srga
18
JOURNAL OF SURGICAL RESEARCH: VOL. 70, NO. 1, JUNE 1997
FIG. 5. Scanning electron micrographs of the rat liver sinusoid with 6-hr reperfusion following 30-min warm ischemia. Thrombus formation is seen in a narrow and distorted sinusoidal lumen in the fatty-liver group (a). Intact perfused rat liver with normal lobular structure is seen in a control (b). (original magnification for both, 13500.)
dominantly in the midzone of the acinus. A similar distribution is observed for oxidative stress after ischemia and reperfusion in rats. Oxidant stress and resulting cell death during low-flow hypoxia reportedly are spatially restricted in the midzone of the hepatic
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
lobule [28, 29]. At reperfusion, activated Kupffer cells release various cytokines and free radicals, and also express ICAM-1 on their surfaces. The heterogeneity of oxidative stress in the lobular structure may relate to the zonal distribution of sinusoidal congestion as
srga
HAKAMADA ET AL.: SINUSOIDAL FLOW BLOCK AFTER WARM ISHCEMIA
19
FIG. 6. Cumulative survival curves after hepatic warm ischemia in fatty-liver rats treated with anti-ICAM-1 monoclonal antibody (1A29) or sham-treated (Kaplan–Meier method). Animal survival was significantly better in rats treated with intraportal injection of 1A29 (solid line) than with saline (dotted line; 50% vs 10%, P Å 0.0112, log-rank test).
observed in this study. Indeed, the degree of ICAM-1 expression and the release of free radicals and cytokines should be investigated to elucidate the basis for the zonal distribution of sinusoidal congestion. Monoclonal antibodies specific for ICAM-1 now are available [30], and have been reported to suppress ischemia/reperfusion injury in rat liver transplantation. A previous study found efficacy of intraportally injected anti-ICAM-1 monoclonal antibody against liver cell injury following warm ischemia in normal rats [18]. An anti-ICAM-1 monoclonal antibody (1A29) greatly diminished sinusoidal congestion after warm ischemia and reperfusion in our fatty-liver rats. In 7 of 10 rats, no congestion could be seen in the entire specimen, while congestion was present in some lobules in 3 rats. Despite improved microcirculation, however, survival rates after 60-min ischemia and reperfusion in treated fatty-liver rats did not equal survival in normal rats. Fifty percent of the treated rats died within 36 hr after unclamping. Although sinusoidal flow block, which occurs primarily in the midzone of the lobule, is a factor in warm ischemia/reperfusion injury, warm ischemia was seen to contribute directly to hepatocellular damage because survival significantly decreased between 30- and 60-min ischemia and also because correction of sinusoidal circulation by intraportally injected antiICAM-1 monoclonal antibody did not fully normalize survival. In summary, both sinusoidal flow block, which occurs
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
primarily in the midzone of the acinus after hepatic ischemia and reperfusion, and warm ischemic hepatocellular damage probably cause warm ischemia/reperfusion injury in fatty liver. We believe that these are the mechanisms by which massively fatty-liver grafts fail from PNF after liver transplantation. REFERENCES 1. Starzl, T. E., Iwatsuki, S., Esquivel, C. O., Todo, S., Kam, I., Lynch, S., Gordon, R. D., and Shaw, B. W., Jr. Refinements in the surgical technique of liver transplantation. Semin. Liver Dis. 5: 349, 1985. 2. Todo, S., Nery, J., Yanaga, K., Podesta, L., Gordon, R. D., and Starzl, T. E. Extended preservation of human liver grafts with UW solution. JAMA 261: 711, 1989. 3. Stratta, R. J., Wood, R. P., Langnas, A. N., Marujo, W., Duckworth, R. M., Williams, L., Saito, S., Pillen, T. J., and Shaw, B. W., Jr. Effect of extended preservation and reduced-size grafting on organ availability in pediatric liver transplantation. Transplant. Proc. 22: 482, 1990. 4. D’Alessandro, A. M., Kalayoglu, M., Sollinger, H. W., Hoffmann, R. M., Pirsch, J. D., Lorentzen, J. S., Meizer, J. S., and Belzer, F. O. Experience with Belzer UW cold storage solution in human liver transplantation. Transplant. Proc. 22: 474, 1990. 5. Belzer, F. O., and Southard, J. H. Principles of solid-organ preservation by cold storage. Transplantation 45: 673, 1988. 6. Greig, P. D., Woolf, G. M., Sinclair, S. B., Abecassis, M., Strasberg, S. M., Taylor, B. R., Blendis, L. M., Superina, R. A., Glynn, M. F. X., Langer, B., and Levy, G. A. Treatment of primary liver
srga
20
7. 8.
9.
10.
11.
12.
13.
14.
15.
16. 17.
18.
AID
JOURNAL OF SURGICAL RESEARCH: VOL. 70, NO. 1, JUNE 1997 graft nonfunction with prostaglandin E1 . Transplantation 48: 447, 1989. Shaw, B. W., Jr., and Wood, R. P. Improved results with retransplantation of the liver. Transplant. Proc. 21: 2407, 1989. Portmann, B., and Wight, D. G. D. Pathology of liver transplantation (excluding rejection). In Calne, R. Y. (Ed.), Liver Transplantation. London: Grune & Stratton, 1987. Pp. 437– 470. Todo, S., Demetris, A. J., Makowka, L., Teperman, L., Podesta, L., Shaver, T., Tzakis, A., and Starzl, T. E. Primary nonfunction of hepatic allografts with preexisting fatty infiltration. Transplantation 47: 903, 1989. Kakizoe, S., Yanaga, K., Starzl, T. E., and Demetris, A. J. Evaluation of protocol before transplantation and after reperfusion biopsies from human orthotopic liver allografts: Considerations of preservation and early immunological injury. Hepatology 11: 932, 1990. D’Alessandro, A. M., Kalayoglu, M., Sollinger, H. W., Hoffman, R. M., Reed, A., Knechtle, S. J., Pirsch, J. D., Hafez, G. R., Lorentzen, D., and Belzer, F. O. The predictive value of donor liver biopsies for the development of primary nonfunction after orthotopic liver transplantation. Transplantation 51: 157, 1991. Adam, R., Reynes, M., Johann, M., Morino, M., Astarcioglu, I., Kafetzis, I., Castaing, D., and Bismuth, H. The outcome of steatotic grafts in liver transplantation. Transplant. Proc. 23: 1538, 1991. Markin, R. S., Wisecarver, J. L., Radio, S. J., Stratta, R. J., Langnas, A. N., Hirst, K., and Shaw, B. W., Jr. Frozen section evaluation of donor livers before transplantation. Transplantation 56: 1403, 1993. Colina, F., Lopez Carreira, M., Moreno, E., Garcia, I., Corral, M. A., Fiano, C., Garcia Munoz, H., and Gomez Sanz, R. A clinopathologic review of 8 liver graft primary nonfunctions. Hepatogastroenterology 42: 212, 1995. Teramoto, K., Bowers, J. L., Kruskal, J. B., and Clouse, M. E. Hepatic microcirculatory changes after reperfusion in fatty and normal liver transplantation in the rat. Transplantation 56: 1076, 1993. Arii, S., and Imamura, M. Liver transplantation and hepatic sinusoidal cells. J. Gastroenterol. Hepatol. 10: S92, 1995. Nishimura, Y., Takei, Y., Kawano, S., Goto, M., Nagano, K., Tsuji, S., Nagai, H., Ohmae, A., Fusamoto, H., and Kamada, T. The F(ab*)2 fragment of an anti-ICAM-1 monoclonal antibody attenuates liver injury after orthotopic liver transplantation. Transplantation 61: 99, 1996. Nakano, H., Kuzume, M., Namatame, K., Yamaguchi, M., and
JSR 5077
/
6n1c$$$224
06-13-97 13:07:29
19.
20.
21.
22. 23.
24.
25.
26.
27.
28.
29.
30.
Kumada, K. Efficacy of intraportal injection of anti-ICAM-1 monoclonal antibody against liver cell injury following warm ischemia in the rat. Am. J. Surg. 170: 64, 1995. Folch, J., Lees, M., and Stanley, G. H. S. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226: 497, 1957. Yao, Z., and Vance, D. E. The active synthesis of phosphatidylcholine is required for very low density lipoprotein secretion from rat hepatocytes. J. Biol. Chem. 263: 2998, 1988. Hoyumpa, A. M. Jr., Greene, H. L., Dunn, G. D., and Schenker, S. D. D. Fatty liver: Biochemical and clinical considerations. Dig. Dis. 20: 1142, 1975. Zeisel, S. H. Choline deficiency. J. Nutr. Biochem. 1: 332, 1990. Rappaport, A. M., Knoblauch, M., Black, R. G., and Ohira, S. Hepatic microcirculatory changes leading to portal hypertension. Ann. N.Y. Acad. Sci. 170: 48, 1970. Wada, K., Fujimoto, K., Fujikawa, Y., Shibayama, Y., Mitsui, H., and Nakata, K. Sinusoidal stenosis as the cause of portal hypertension in choline deficient diet induced fatty cirrhosis of the rat liver. Acta Pathol. Jpn. 24: 207, 1974. Colletti, L. M., Burtch, G. D., Remick, D. G., Kunkel, S. L., Strieter, R. M., Guice, K. S., Oldham, K. T., and Campbell, D. A. Jr. The production of tumor necrosis factor alpha and the development of a pulmonary capillary injury following hepatic ischemia/reperfusion. Transplantation 49: 268, 1990. Takei, Y., Marzi, I., Gao, W., Gores, G. J., Lemasters, J. J., and Thurman, R. G. Leukocyte adhesion and cell death following orthotopic liver transplantation in the rat. Transplantation 51: 959, 1991. Koneru, B., Reddy, M. C., dela Torre, A. N., Patel, D., Ippolito, T., and Ferrante, R. J. Studies of hepatic warm ischemia in the obese Zucker rat. Transplantation 59: 942, 1995. Suematsu, M., Suzuki, H., Ishii, H., Kato, S., Yanagisawa, T., Asako, H., Suzuki, M., and Tsuchiya, M. Early midzonal oxidative stress preceding cell death in hypoperfused rat liver. Gastroenterology 103: 994, 1992. Suzuki, H., Suematsu, M., Ishii, H., Kato, S., Miki, H., Mori, M., Ishimura, Y., Nishino, T., and Tsuchiya, M. Prostaglandin E1 abrogates early reductive stress and zone-specific paradoxical oxidative injury in hypoperfused rat liver. J. Clin. Invest. 93: 155, 1994. Tamatani, T., and Miyasaka, M. Identification of monoclonal antibodies reactive with the rat homolog of ICAM-1, and evidence for a differential involvement of ICAM-1 in the adherence of resting versus activated lymphocytes to high endothelial cells. Int. Immunol. 2: 165, 1990.
srga
70, 12–20 (1997)
JR975077
Sinusoidal Flow Block after Warm Ischemia in Rats with Diet-Induced Fatty Liver1 Kenichi Hakamada, M.D., Mutsuo Sasaki, M.D., Katsuro Takahashi, M.D., Yutaka Umehara, M.D., and Mitsuru Konn, M.D. Second Department of Surgery, Hirosaki University School of Medicine, 5 Zaifu-cho, Hirosaki 036, Japan Submitted for publication July 5, 1996
vation [2–5], 1–17% of recipients still experience PNF [2–4, 6, 7]. Immediate retransplantation is mandatory and results in survival for 70% of patients, but such waste of time, money, and manpower [7], as well as the increased morbidity itself, mandate avoidance of PNF by all means. Fatty liver in donors has been associated with PNF [8–13]. Portmann and Wight [8] first reported a case with unexpected severe fatty infiltration of the donor liver in which the graft never functioned. The outcomes of steatotic allografts then were evaluated using timezero biopsy at other liver transplant centers [11, 12]. Now donor livers with massive fatty infiltrates are thought to be at great risk for PNF and are discarded based on frozen-section examination [13]. However, the processes by which steatotic grafts undergo PNF are incompletely known. Todo et al. [9] have reported two examples of PNF with severe fatty infiltration, in which they observed extracellular fatty droplets with disruption of the sinusoidal microvasculature in autopsy specimens. They speculated that a degree of hepatocellular death occurred in the course of graft preservation and that rupture of some steatotic hepatocytes released fat droplets into sinusoids, compromising hepatic microcirculation. However, PNF can occur despite shorter preservation times. A previous report found no significant difference in warm and cold ischemic times between PNF cases and the patients without this complication [14]. Another hypothesis is that sinusoidal microcirculatory changes are the initial event after liver transplantation. Teramoto et al. [15] observed increased blood cell adhesion to sinusoidal lining cells after ischemia and reperfusion under in vivo microscopy using a rat fatty-liver model. They speculated that the microcirculatory disturbances caused liver cell death and contributed to primary graft nonfunction in fatty livers. Ischemia/reperfusion injury is thought to be of major clinical importance as a cause of early graft dysfunction. At the time of reperfusion, lymphocyte-functionassociated antigen-1 (LFA-1), which is expressed on the surface of leukocytes, adheres to intercellular adhesion molecule-1 (ICAM-1) on the endothelial cell membrane. Hepatic microcirculation is injured by the formation of
Donor livers with massive fatty infiltration reportedly are susceptible to ischemia/reperfusion injury after transplantation, which contributes to risk of primary nonfunction. We investigated the effect of warm ischemia and reperfusion on sinusoidal microcirculation in rats with fatty livers from a choline-deficient diet. Rats were subjected to partial hepatic warm ischemia for 30, 60, or 90 min. In a second study, an antiICAM-1 monoclonal antibody was injected intraportally 2 min after a 60-min ischemic period. In both studies, injury was assessed by liver histology 6 hr after vascular clamp release and by animal survival. After 30 min of hepatic warm ischemia, almost all control and fatty-liver rats survived 7 days. After 60-min ischemia, however, survival was significantly less in rats with fatty livers than in controls with normal livers (10% vs 90%, P õ 0.0001). Histologically, rats with fatty livers showed marked sinusoidal congestion, especially in the midzone of the acinus, while control rats showed no disturbance in microcirculation. In rats with fatty livers treated with intraportal injection of an anti-ICAM-1 antibody, sinusoidal microcirculation was well preserved and the 7-day survival rate after warm ischemia was improved (50% vs no antibody 10%; P Å 0.0112). In fatty livers, midzonal sinusoidal flow block occurs after hepatic warm ischemia and reperfusion. Although intraportal injection of an anti-ICAM-1 monoclonal antibody corrected this microcirculatory failure, animal survival was not as good as for controls without fatty livers. These results suggest that both sinusoidal microcirculatory failure and ischemic hepatocellular damage contribute to warm ischemia/reperfusion injury in fatty livers. q 1997 Academic Press
INTRODUCTION
Primary nonfunction (PNF) following liver transplantation is a serious complication, which increases postoperative morbidity and mortality. In spite of improvements in surgical technique [1] and organ preser1 This work was supported partly by Research Grants from the Ministry of Education of Japan.
0022-4804/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
JSR 5077
/
6n1c$$$221
12
06-13-97 13:07:29
srga
HAKAMADA ET AL.: SINUSOIDAL FLOW BLOCK AFTER WARM ISHCEMIA
intrasinusoidal thrombi, which now are stressed as a causative factor in ischemia/reperfusion injury of hepatic allografts [16–18]. In this paper, the authors introduced fatty changes in rats by feeding a choline-deficient diet and then investigated the effect of warm ischemia and reperfusion on hepatic microcirculation. Intraportal treatment with an anti-ICAM-1 monoclonal antibody also was examined for possible protective effect against sinusoidal microcirculatory disturbances [18]. MATERIALS AND METHODS Introduction of fatty liver in rats. Male Sprague–Dawley rats (CLEA Japan, Inc., Tokyo, Japan), weighing 170–190 g, were housed in metal cages with controlled light and dark cycles and given free access to water. To introduce fatty changes in livers, rats were fed a chow deficient in choline (Oriental Yeast Co., Tokyo, Japan) for 7, 14, 21, 28, 35, or 42 days (for each, n Å 10). Ten control rats were fed a standard commercial pelleted diet. Development of fatty liver did not affect body weight gain (data not shown). After completion of observation periods, all rats were anesthetized by either inhalation and the liver was removed, washed in cold saline, frozen in liquid nitrogen, and kept at 0707C until analyzed. Whole lipids in freezedried liver homogenates were extracted by chloroform:methanol (2:1, v/v) [19] and contents of triglyceride, phospholipids, and total cholesterol were determined enzymatically in a autoanalyzer (Hitachi 736, Tokyo, Japan). In 10 rats on the choline-deficient diet for 14 or 28 days (n Å 5 each), relative molar fatty acid fractions also were measured as percentages by gas chromatography using the same extracts. Extracts from experimental rats were compared with those from 5 control rats. Hepatic histology under light microscopy was studied in all rats after specimens were fixed in 10% Formalin, embedded in paraffin, and stained with hematoxylin and eosin. All slides were evaluated in a blinded manner by one of the authors, and the degree of fatty change was expressed as the proportion of hepatocytes with fatty infiltration in 10 randomly chosen fields from each slide. First study. Rats were fed a chow deficient in choline for 21 days to induce fatty livers and were compared with control rats given a standard pelleted diet for 21 days. All rats underwent hepatic warm ischemia. Under clean conditions, the abdomen was entered through a transverse incision after induction of ether anesthesia. The liver was dissected from peritoneal connections and the hilum of the median and left lobes was exposed. Segmental ischemia was induced by occluding the blood vessels and the bile duct to these lobes with an atraumatic vascular clamp. Then the abdomen was closed. After warm ischemia the clamp was released and the nonischemic right lateral and caudate lobes were resected. This technique avoided portal congestion, and endotoxin levels in portal vein and vena cava were not elevated (data not shown). Survival time and mortality in groups of 20 rats each after 30, 60, and 90 min of warm ischemia were recorded for both fatty-liver and control rats every 6 hr for 7 days postoperatively. Animals alive 7 days after operation were considered survivors and were killed for morphological study. Necropsy was performed on all animals to confirm the absence of surgical complications. Blood samples and liver tissue were taken 6 hr after the end of 30- and 60-min hepatic warm ischemia in both fatty-liver and control rats (n Å 10 each). Alanine aminotransferase (ALT), total bilirubin, prothrombin time (PT), fibrinogen, fibrin–fibrinogen degradation products (FDP), and antithrombin III (AT-III) were determined by routine methods. Liver specimens stained with hematoxylin and eosin were evaluated under light microscopy. Lobular distribution of congestion and necrosis was noted specifically. For electron microscopic examination, other portions of the same livers were fixed with 2.5% glutaraldehyde in 50 mM cacodylate buffer (pH 7.4). Ten blocks with a volume of about 8 mm3 were postfixed in 1.0% OsO4 in 0.1 M
AID
JSR 5077
/
6n1c$$$222
06-13-97 13:07:29
13
phosphate buffer, pH 7.4, for 2 hr, dehydrated in ethanol, and prepared with gold coating. Lobular structures were observed under a JSM-5310 scanning electron microscope (Nippon Denshi Co. Tokyo, Japan). Second study. Anti-rat ICAM-1 monoclonal antibody (1A29, Seikagaku Corp., Tokyo, Japan) was administered intraportally to evaluate its ability to maintain sinusoidal microcirculation after warm ischemia and reperfusion in rats with fatty livers. Rats on the choline-deficient diet for 21 days underwent partial hepatic warm ischemia and reperfusion in a manner identical to the first study. In the treatment group, 0.8 mg/kg of anti-ICAM-1 antibody was administered directly into the branch of superior mesenteric vein 2 min before unclamping [18], while in sham-operated rats, the same volume of normal saline was administered via the same vein at the same time. Seven-day survival rates (n Å 20, each group) and histologic changes (n Å 10 each) 6 hr after the 60-min warm ischemia were observed similarly to the first study. Statistical analysis. Statistical analysis was performed using the computer program Statview 4.02 (Abacus Concepts Inc., Berkeley, CA). Results were expressed as means { standard error of the mean (SEM). Overall statistical differences were determined according to analysis of variance (ANOVA), and Fisher’s test was used as a post hoc test. The unpaired Student’s t test was used to compare groups for ALT, total bilirubin, and other coagulation parameters. Cumulative survival rates after warm ischemia and reperfusion were obtained by the Kaplan–Meier method and were compared by the logrank test. P values less than 0.05 were considered significant.
RESULTS
Characterization of the Rat Fatty-Liver Model Triglyceride was the main lipid constituent in the livers of rats given the choline-deficient diet. Its content had increased by Day 14 and then remained relatively constant (no significant difference, or NS, from Days 14 to 42). Total amount of phospholipids had decreased by Day 7, remaining at the same level subsequently (NS from Days 7 to 42). Total cholesterol levels increased slightly at first, but were constant from Days 7 to 42 (NS; Fig. 1). Gas chromatography of the same liver tissue extracts demonstrated increased relative molar percentages of palmitic acid (C16:0) and oleic acid (C18:1; for both, control vs Days 14 and 28, P õ 0.01; Fig. 2). Under light microscopy, the extent of visible fat accumulation varied from minute droplets scattered in the cytoplasm of ballooning hepatocytes to distention of the entire cytoplasm of most cells by coalesced droplets. The proportion of hepatocytes with macro- or microvesicular fat vacuoles increased until the liver showed uniform macrovesicular fatty changes by Day 21. The proportion of fatty hepatocytes remained 75.9 { 5.3% from Days 21 to 42. In our models, fatty droplets appeared in the mid- to centrilobular zone first, but the entire lobular architecture was involved after Day 14. Hepatocyte necrosis was rarely seen even in rats on the diet for 42 days. First Study Cumulative survival rates after 30-, 60-, and 90-min warm liver ischemia were 100, 90, and 35% in control rats and 95, 10, and 5% in fatty-liver rats respectively at the 7th postoperative day (Fig. 3). There was no significant difference in survival between control and
srga
14
JOURNAL OF SURGICAL RESEARCH: VOL. 70, NO. 1, JUNE 1997
FIG. 1. Lipid content of rat liver after feeding with choline-deficient diet. The mean and SEM are shown for triglyceride, phospholipids, and total cholesterol content as mg/g liver tissue for rats given control and choline-deficient diets for 7 to 42 days. C, control; D x, rats given choline-deficient diets for x days; NS, no significant difference.
fatty-liver rats after 30-min hepatic ischemia, while the survival rate was significantly worse in fatty-liver rats after 60-min ischemia (P õ 0.0001). After 90-min ischemia, few rats survived in the fatty-liver group (vs control, P Å 0.0043). At 6 hr following 60-min warm ischemia, fatty-liver rats showed significant decreases in fibrinogen and ATIII levels compared with controls (both, P õ 0.01). These differences were greater between fatty-liver rats and control rats after 30-min ischemia, when fibrinogen and AT-III levels already had decreased significantly and PT was markedly prolonged in fatty-liver rats (P õ 0.01). ALT, total bilirubin, and FDP all increased significantly compared with controls as well. The latter parameters were not different between rats with 30- and 60-min hepatic warm ischemia (Table 1). The overall lobular architecture of the liver was well maintained, and no fatty droplets was seen in hepatocytes of control rats. Although sinusoidal lining cells and hepatocytes appeared normal 6 hr after 30-min ischemia, hepatocyte necrosis was seen diffusely scattered through the lobules after 60-min warm ischemia in control rats. In fatty livers, on the other hand, the most prominent finding after warm ischemia was sinusoidal congestion (Fig. 4). Its severity varied from localized congestion in the midzonal sinusoid to sinusoidal distention in the peripheral zone. The entire acinus showed marked congestion in some areas. This zonal distribution did not change with degree of fatty change; the midzone was
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
more vulnerable to sinusoidal congestion in lobules of both mildly and severely fatty livers. The extent of congestion was diffuse and similar between rats after 30and 60-min ischemia. Hepatocyte necrosis, however, was found throughout the lobular architecture in fattyliver rats after 60-min ischemia; this was less evident after only 30 min of ischemia. Under scanning electron microscopy, irregular, narrow sinusoids were plugged by thrombus formation in congested areas in rats with 30- or 60-min ischemia (Fig. 5). Second Study Survival rates were improved significantly, but not fully normalized, by intraportal administration of antiICAM-1 monoclonal antibody (50% vs 10%, P Å 0.0112; Fig. 6). Histologic findings indicated that sinusoidal congestion was markedly reduced in the treated group; no sinusoidal congestion was evident and both sinusoidal lining cells and hepatocytes were well preserved in 7 of 10 treated fatty-liver rats. In the other 3 rats, localized sinusoidal congestion was present in some areas. Extensive hepatocyte necrosis and congestion, which could be seen in fatty-liver rats treated with saline injection, were not seen in the anti-ICAM-1 antibody treatment group. DISCUSSION
Primary nonfunction following liver transplantation contributes significantly to postoperative morbidity
srga
HAKAMADA ET AL.: SINUSOIDAL FLOW BLOCK AFTER WARM ISHCEMIA
15
FIG. 2. Changes in fatty acid fractions in rat liver after feeding with choline-deficient diet. The mean and SEM are shown for relative molar percentage of fatty acid fractions in liver for rats on control and on choline-deficient diets for 14 and 28 days. Proportions of palmitic acid (C16:0) and oleic acid (C18:1) increased significantly in fatty liver models. *P õ 0.01: control (black bars) vs 14 days (white) and 28 days (stripes).
and mortality [2, 4, 6, 7]. Massive fatty infiltration in donor livers reportedly is associated with PNF [8–13], and most liver transplant centers tend to avoid such a steatotic allograft. The reason why fatty-liver grafts are susceptible to PNF is not known. The present study used rats with fatty livers from a choline-deficient diet to analyze survival and pathologic changes in the liver after warm ischemia and reperfusion. A choline-deficient diet is well known to reduce phosphatidylcholine synthesis in rats [20]. Phosphatidylcholine is one of the essential components of very low density lipoprotein (VLDL), which carries triglyceride from the liver to other organs. Choline deficiency consequently induces triglyceride accumulation in the liver [20–22]. Triglyceride was the main constituent of the increased lipid droplets in our model. In rats given a choline-deficient diet for more than 21 days, about 75% of all hepatocytes contained macrovesicular fat vacuoles under light microscopy, and in extracts the contents of triglyceride, phospholipids, and total cholesterol were relatively constant from Days 21 to 42. Gas chromatography of the extracts revealed increased molar percentages of palmitic and oleic acids. Although the mechanism by which fatty changes occur in human liver is not well known, triglyceride is the main component of accumulated fatty droplets when macro- or microvesicular steatosis occurs in either humans or rats. Moreover, palmitic acid is the main in-
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
creased fatty acid in humans with excessive dietary intake of carbohydrate [21]. Because of these biochemical and pathologic similarities between human fatty livers and the rat fatty-liver model, we used rats on a choline-deficient diet as an experimental model to analyze the mechanism of warm ischemia/reperfusion injury in fatty livers. In massive fatty livers, it has been reported that intracytoplasmic fat droplets bulging into the sinusoidal lumen narrow and distort hepatic microcirculatory pathways [15, 23, 24]. Consequently, intrahepatic portal resistance increases, as reflected by elevated portal vein pressure [23, 24]. In vivo microscopy has disclosed sluggish intrasinusoidal blood flow in rats with fatty livers [15]. In our model, where some 75% of hepatocytes have fatty droplets in the cytoplasm after Day 21, the sinusoidal lumen was compressed and distorted by ballooned hepatocytes, indicating that morphologically the microcirculation was impaired by fatty change per se. The most striking change in fatty livers after warm ischemia was sinusoidal congestion. The midzone was more vulnerable to congestion than other areas. This distribution of congestion in the hepatic acinus did not differ between lobules where fatty infiltration was localized to the mid- to centrilobular zone and lobules where the entire acinus was involved by fatty change. After hepatic ischemia and reperfusion, inflammatory
srga
16
JOURNAL OF SURGICAL RESEARCH: VOL. 70, NO. 1, JUNE 1997
FIG. 3. Cumulative-survival curves after hepatic warm ischemia in control and fatty-liver rats (Kaplan–Meier method). The survival rate after 60-min ischemia was significantly worse in fatty-liver rats than in controls. Numbers at right of curves designate minutes of hepatic warm ischemia for control (C, solid line) and fatty-liver (F, dotted line) rats. NS, no significant difference. *P õ 0.0001, **P Å 0.0003, ***P Å 0.0043 (log-rank test).
cytokines released from Kupffer cells [16, 25] activate endothelial lining cells which then express ICAM-1 on their cell membranes; they also induce LFA-1 expression on white cells [17, 18]. Thrombi may then form hepatic sinusoids as a result of adhesion of these molecules, causing hepatic microcirculatory failure [26]. In our studies, progressive accumulation of blood cells was seen as early as 6 hr following reperfusion. Increased resistance to portal blood flow through sinusoids in fatty livers can exaggerate sinusoidal flow block after warm ischemia.
Our data suggest that an even shorter warm ischemic time of 30 min caused hepatic microcirculatory failure, and that 60-min warm ischemia, which was compatible with survival for normal rats, resulted in death for rats with fatty livers. Prolonged PT, decreased levels of fibrinogen and AT-III, and elevated FDP in fatty-liver rats indicated that intravascular consumption coagulopathy occurred in the fatty-liver rats even after only 30 min of ischemia. Fibrinogen and AT-III levels decreased significantly in fatty-liver rats after 60-min warm ischemia. Because relatively little
TABLE 1 Effect of Hepatic Warm Ischemia on Liver-Function and Coagulation Indices 30-min warm ischemia Fatty-liver rat (n Å 10) ALT (U/L) Total bilirubin (mg/dl) Prothrombin time (sec) Fibrinogen (mg/dl) FDP (mg/ml) Antithrombin III (%)
1956.6 0.66 19.9 45.0 10.7 55.3
{ 151.3* { 0.08* { 0.8* { 7.6* { 1.6** { 2.3*
60-min warm ischemia
Control rat (n Å 10) 1196.3 0.14 16.5 123.4 6.2 85.8
{ 133.1 { 0.03 { 0.6 { 6.6 { 0.8 { 3.1
Fatty-liver rat (n Å 10) 1851.4 0.67 21.0 38.4 14.8 45.7
{ { { { { {
66.2 0.10* 0.9 7.6* 1.9 5.9*
Control rat (n Å 10) 1861.5 0.14 18.4 98.8 10.3 72.9
{ 212.6 { 0.02 { 1.3 { 7.1 { 1.8 { 2.8
Note. Values are means { SEM of 10 rats in each group with 6-hr reperfusion following 30- or 60-min hepatic warm ischemia. ALT, alanine aminotransferase; FDP, fibrin–fibrinogen degradation products. * P õ 0.01 and **P õ 0.05, compared with control rats with the same ischemic time.
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
srga
HAKAMADA ET AL.: SINUSOIDAL FLOW BLOCK AFTER WARM ISHCEMIA
17
FIG. 4. Photomicrograph (a) of rat liver in the fatty-liver group with 6-hr reperfusion following 30-min warm ischemia. Sinusoidal congestion is seen from the mid- to peripheral zone (hematoxylin and eosin, H & E; original magnification, 120). Higher magnification (b) demonstrates diffuse marked deposition of large droplets in hepatocytes, which compress the sinusoidal lumina. CV, central vein; PC, portal canal (H & E; original magnification, 1100).
change occurred in these variables in control rats after 30-min ischemia, differences between controls and fatty-liver rats were more prominent after 30-min than 60-min ischemia. Koneru et al. [27] also found in genet-
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
ically fat or lean Zucker strain rats that fatty livers were more sensitive to warm ischemia/reperfusion injury than nonfatty livers. In this study, sinusoidal congestion was seen pre-
srga
18
JOURNAL OF SURGICAL RESEARCH: VOL. 70, NO. 1, JUNE 1997
FIG. 5. Scanning electron micrographs of the rat liver sinusoid with 6-hr reperfusion following 30-min warm ischemia. Thrombus formation is seen in a narrow and distorted sinusoidal lumen in the fatty-liver group (a). Intact perfused rat liver with normal lobular structure is seen in a control (b). (original magnification for both, 13500.)
dominantly in the midzone of the acinus. A similar distribution is observed for oxidative stress after ischemia and reperfusion in rats. Oxidant stress and resulting cell death during low-flow hypoxia reportedly are spatially restricted in the midzone of the hepatic
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
lobule [28, 29]. At reperfusion, activated Kupffer cells release various cytokines and free radicals, and also express ICAM-1 on their surfaces. The heterogeneity of oxidative stress in the lobular structure may relate to the zonal distribution of sinusoidal congestion as
srga
HAKAMADA ET AL.: SINUSOIDAL FLOW BLOCK AFTER WARM ISHCEMIA
19
FIG. 6. Cumulative survival curves after hepatic warm ischemia in fatty-liver rats treated with anti-ICAM-1 monoclonal antibody (1A29) or sham-treated (Kaplan–Meier method). Animal survival was significantly better in rats treated with intraportal injection of 1A29 (solid line) than with saline (dotted line; 50% vs 10%, P Å 0.0112, log-rank test).
observed in this study. Indeed, the degree of ICAM-1 expression and the release of free radicals and cytokines should be investigated to elucidate the basis for the zonal distribution of sinusoidal congestion. Monoclonal antibodies specific for ICAM-1 now are available [30], and have been reported to suppress ischemia/reperfusion injury in rat liver transplantation. A previous study found efficacy of intraportally injected anti-ICAM-1 monoclonal antibody against liver cell injury following warm ischemia in normal rats [18]. An anti-ICAM-1 monoclonal antibody (1A29) greatly diminished sinusoidal congestion after warm ischemia and reperfusion in our fatty-liver rats. In 7 of 10 rats, no congestion could be seen in the entire specimen, while congestion was present in some lobules in 3 rats. Despite improved microcirculation, however, survival rates after 60-min ischemia and reperfusion in treated fatty-liver rats did not equal survival in normal rats. Fifty percent of the treated rats died within 36 hr after unclamping. Although sinusoidal flow block, which occurs primarily in the midzone of the lobule, is a factor in warm ischemia/reperfusion injury, warm ischemia was seen to contribute directly to hepatocellular damage because survival significantly decreased between 30- and 60-min ischemia and also because correction of sinusoidal circulation by intraportally injected antiICAM-1 monoclonal antibody did not fully normalize survival. In summary, both sinusoidal flow block, which occurs
AID
JSR 5077
/
6n1c$$$223
06-13-97 13:07:29
primarily in the midzone of the acinus after hepatic ischemia and reperfusion, and warm ischemic hepatocellular damage probably cause warm ischemia/reperfusion injury in fatty liver. We believe that these are the mechanisms by which massively fatty-liver grafts fail from PNF after liver transplantation. REFERENCES 1. Starzl, T. E., Iwatsuki, S., Esquivel, C. O., Todo, S., Kam, I., Lynch, S., Gordon, R. D., and Shaw, B. W., Jr. Refinements in the surgical technique of liver transplantation. Semin. Liver Dis. 5: 349, 1985. 2. Todo, S., Nery, J., Yanaga, K., Podesta, L., Gordon, R. D., and Starzl, T. E. Extended preservation of human liver grafts with UW solution. JAMA 261: 711, 1989. 3. Stratta, R. J., Wood, R. P., Langnas, A. N., Marujo, W., Duckworth, R. M., Williams, L., Saito, S., Pillen, T. J., and Shaw, B. W., Jr. Effect of extended preservation and reduced-size grafting on organ availability in pediatric liver transplantation. Transplant. Proc. 22: 482, 1990. 4. D’Alessandro, A. M., Kalayoglu, M., Sollinger, H. W., Hoffmann, R. M., Pirsch, J. D., Lorentzen, J. S., Meizer, J. S., and Belzer, F. O. Experience with Belzer UW cold storage solution in human liver transplantation. Transplant. Proc. 22: 474, 1990. 5. Belzer, F. O., and Southard, J. H. Principles of solid-organ preservation by cold storage. Transplantation 45: 673, 1988. 6. Greig, P. D., Woolf, G. M., Sinclair, S. B., Abecassis, M., Strasberg, S. M., Taylor, B. R., Blendis, L. M., Superina, R. A., Glynn, M. F. X., Langer, B., and Levy, G. A. Treatment of primary liver
srga
20
7. 8.
9.
10.
11.
12.
13.
14.
15.
16. 17.
18.
AID
JOURNAL OF SURGICAL RESEARCH: VOL. 70, NO. 1, JUNE 1997 graft nonfunction with prostaglandin E1 . Transplantation 48: 447, 1989. Shaw, B. W., Jr., and Wood, R. P. Improved results with retransplantation of the liver. Transplant. Proc. 21: 2407, 1989. Portmann, B., and Wight, D. G. D. Pathology of liver transplantation (excluding rejection). In Calne, R. Y. (Ed.), Liver Transplantation. London: Grune & Stratton, 1987. Pp. 437– 470. Todo, S., Demetris, A. J., Makowka, L., Teperman, L., Podesta, L., Shaver, T., Tzakis, A., and Starzl, T. E. Primary nonfunction of hepatic allografts with preexisting fatty infiltration. Transplantation 47: 903, 1989. Kakizoe, S., Yanaga, K., Starzl, T. E., and Demetris, A. J. Evaluation of protocol before transplantation and after reperfusion biopsies from human orthotopic liver allografts: Considerations of preservation and early immunological injury. Hepatology 11: 932, 1990. D’Alessandro, A. M., Kalayoglu, M., Sollinger, H. W., Hoffman, R. M., Reed, A., Knechtle, S. J., Pirsch, J. D., Hafez, G. R., Lorentzen, D., and Belzer, F. O. The predictive value of donor liver biopsies for the development of primary nonfunction after orthotopic liver transplantation. Transplantation 51: 157, 1991. Adam, R., Reynes, M., Johann, M., Morino, M., Astarcioglu, I., Kafetzis, I., Castaing, D., and Bismuth, H. The outcome of steatotic grafts in liver transplantation. Transplant. Proc. 23: 1538, 1991. Markin, R. S., Wisecarver, J. L., Radio, S. J., Stratta, R. J., Langnas, A. N., Hirst, K., and Shaw, B. W., Jr. Frozen section evaluation of donor livers before transplantation. Transplantation 56: 1403, 1993. Colina, F., Lopez Carreira, M., Moreno, E., Garcia, I., Corral, M. A., Fiano, C., Garcia Munoz, H., and Gomez Sanz, R. A clinopathologic review of 8 liver graft primary nonfunctions. Hepatogastroenterology 42: 212, 1995. Teramoto, K., Bowers, J. L., Kruskal, J. B., and Clouse, M. E. Hepatic microcirculatory changes after reperfusion in fatty and normal liver transplantation in the rat. Transplantation 56: 1076, 1993. Arii, S., and Imamura, M. Liver transplantation and hepatic sinusoidal cells. J. Gastroenterol. Hepatol. 10: S92, 1995. Nishimura, Y., Takei, Y., Kawano, S., Goto, M., Nagano, K., Tsuji, S., Nagai, H., Ohmae, A., Fusamoto, H., and Kamada, T. The F(ab*)2 fragment of an anti-ICAM-1 monoclonal antibody attenuates liver injury after orthotopic liver transplantation. Transplantation 61: 99, 1996. Nakano, H., Kuzume, M., Namatame, K., Yamaguchi, M., and
JSR 5077
/
6n1c$$$224
06-13-97 13:07:29
19.
20.
21.
22. 23.
24.
25.
26.
27.
28.
29.
30.
Kumada, K. Efficacy of intraportal injection of anti-ICAM-1 monoclonal antibody against liver cell injury following warm ischemia in the rat. Am. J. Surg. 170: 64, 1995. Folch, J., Lees, M., and Stanley, G. H. S. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226: 497, 1957. Yao, Z., and Vance, D. E. The active synthesis of phosphatidylcholine is required for very low density lipoprotein secretion from rat hepatocytes. J. Biol. Chem. 263: 2998, 1988. Hoyumpa, A. M. Jr., Greene, H. L., Dunn, G. D., and Schenker, S. D. D. Fatty liver: Biochemical and clinical considerations. Dig. Dis. 20: 1142, 1975. Zeisel, S. H. Choline deficiency. J. Nutr. Biochem. 1: 332, 1990. Rappaport, A. M., Knoblauch, M., Black, R. G., and Ohira, S. Hepatic microcirculatory changes leading to portal hypertension. Ann. N.Y. Acad. Sci. 170: 48, 1970. Wada, K., Fujimoto, K., Fujikawa, Y., Shibayama, Y., Mitsui, H., and Nakata, K. Sinusoidal stenosis as the cause of portal hypertension in choline deficient diet induced fatty cirrhosis of the rat liver. Acta Pathol. Jpn. 24: 207, 1974. Colletti, L. M., Burtch, G. D., Remick, D. G., Kunkel, S. L., Strieter, R. M., Guice, K. S., Oldham, K. T., and Campbell, D. A. Jr. The production of tumor necrosis factor alpha and the development of a pulmonary capillary injury following hepatic ischemia/reperfusion. Transplantation 49: 268, 1990. Takei, Y., Marzi, I., Gao, W., Gores, G. J., Lemasters, J. J., and Thurman, R. G. Leukocyte adhesion and cell death following orthotopic liver transplantation in the rat. Transplantation 51: 959, 1991. Koneru, B., Reddy, M. C., dela Torre, A. N., Patel, D., Ippolito, T., and Ferrante, R. J. Studies of hepatic warm ischemia in the obese Zucker rat. Transplantation 59: 942, 1995. Suematsu, M., Suzuki, H., Ishii, H., Kato, S., Yanagisawa, T., Asako, H., Suzuki, M., and Tsuchiya, M. Early midzonal oxidative stress preceding cell death in hypoperfused rat liver. Gastroenterology 103: 994, 1992. Suzuki, H., Suematsu, M., Ishii, H., Kato, S., Miki, H., Mori, M., Ishimura, Y., Nishino, T., and Tsuchiya, M. Prostaglandin E1 abrogates early reductive stress and zone-specific paradoxical oxidative injury in hypoperfused rat liver. J. Clin. Invest. 93: 155, 1994. Tamatani, T., and Miyasaka, M. Identification of monoclonal antibodies reactive with the rat homolog of ICAM-1, and evidence for a differential involvement of ICAM-1 in the adherence of resting versus activated lymphocytes to high endothelial cells. Int. Immunol. 2: 165, 1990.
srga