Mesenteric venous hypertension: importance after portal systemic shunting?

Hepatic dysfunction after portacaval shunting (PCS) has been attributed to loss of portal perfusion to the liver. Proponents of selective systemic shunting state that reduced encephalopathy and hepatic dysfunction with this procedure result from the maintenance of portal perfusion to the liver through the hypertensive mesenteric venous circulation. We questioned the importance of maintaining the diminished portal flow to the cirrhotic liver because hepatofugal flow is known to develop in many of these patients. We sought to further define mechanisms that may contribute to the maintenance of critical flow to the liver in compensated hepatic cirrhosis. We demonstrated a primary relationship between mesenteric venous hypertension (MVH) and increased hepatic arterial blood flow after diversion of portal blood flow. Fifteen dogs had vena caval stenosis above an end-to-side PCS to establish MVH and deprive the liver of portal blood flow. Another 15 dogs had end-to-side PCS alone. A half hour after shunting, hepatic arterial blood flow had increased significantly in all dogs. Hemodynamic parameters remained stable throughout. Six weeks later, mesenteric pressure increased 98% +/- 3% with intracaval stenosis (from 9.6 +/- 0.1 to 19.0 +/- 0.3 cm H2O). Mesenteric pressure was unchanged with PCS alone (9.0 +/- 0.1 cm H2O). Increased hepatic arterial flow was significantly elevated in all dogs above pre-shunt values by 6 weeks postshunt. With MVH, however, further augmentation in hepatic arterial flow was noted in the chronic state (1.5 +/- 0.1 vs 0.9 +/- 0.1 ml/min/gm, p less than 0.05). There was significant correlation between MVH and increased hepatic arterial flow in the chronic state (r = 0.79, p = 0.05). Hepatic arterial flow 6 weeks after PCS with MVH was associated with lower blood ammonia and improved hepatocellular function compared with animals with PCS alone. These results support the hypothesis that MVH is important in maintaining blood supply--beyond providing driving force for sustained portal flow to the liver. This is an important consideration in the medical and surgical management of portal hypertension, a condition in which profound reduction in portal pressure may negatively affect compensatory hepatic arterial blood flow.     

Blood flow in renal carcinomas evaluated by a dye dilution technique and angiography.

In 31 patients with renal carcinoma referred for angiography, rapid sequence angiography was performed in 25. In 18 of the patients determination of renal blood flow and related parameters was made in conjunction with the angiography. Carcinomas with high and medium-high differentiation comprised Group I, while those with medium and poor differentiation constituted Group II. In both groups a positive correlation was demonstrated between the renal blood flow and the tumour area. In group I, positive correlation (t greater than 2) was demonstrated between the vascular volume and the tumour area, between the vascular resistance and the emptying time of the tumour arteries and the normal renal arteries. There was some degree of shunting of blood in all tumours. Comparison of the arterial emptying time of the tumour arteries with the parenchymal arteries does not allow of staging of renal carcinomas. The degree of shunting may be an indication of the degree of malignancy and poor differentiation.     

The effect of halothane, isoflurane, and blood loss on hepatotoxicity and hepatic oxygen availability in phenobarbital-pretreated hypoxic rats

This study evaluated the role of ventilatory and circulatory depression in anesthesia-induced hepatotoxicity in rats. Male Sprague-Dawley rats (181 animals) were pretreated with phenobarbital and exposed to hypoxia (FIO2 = 0.14) for 2 hr. The animals were divided into four groups: group 1 received 1% inspired halothane in the hypoxic gas mixture; group 2 received 1.4% inspired isoflurane and hypoxia; group 3 had 25-30% of their blood volume removed 2 hr before exposure to hypoxia; and group 4 served as a control with no treatment other than hypoxia. Hepatic blood flow was studied using microspheres; oxygen availability to the liver was calculated using values of hepatic blood flow and oxygen content of arterial and portal venous blood; and liver injury was quantitatively evaluated. Ventilation was depressed in rats that received halothane and, to a lesser extent, isoflurane. The lowest portal blood flow was observed in groups 1 and 3. Hepatic arterial blood flow was lowest in group 1 and highest in group 3. There was an inverse relationship between hepatic oxygen availability and severity of histologic lesions. The most severe lesions and lowest oxygen availability was associated with halothane. Hemorrhage and isoflurane were associated with less diminution of oxygen availability and less severe hepatic lesions. The least decrease in oxygen availability and the least severe histologic changes occurred in control rats subjected to hypoxia only.     

Evaluation of microcirculation in the tumor-bearing liver of rabbits by laser-Doppler flowmetry.

Hemodynamic changes in normal liver tissue and in intrahepatic tumors (Vx2 carcinoma) after occlusion of the hepatic arterial branch or the portal branch (ex. 1), and with intrahepatic arterial infusion of vasoactive agents (ex. 2) were studied in rabbits by a laser-Doppler flowmeter. Ex. 1: After occlusion of the hepatic arterial branch to the main lobe, laser-Doppler flow (LDF) in main lobe normal tissue decreased by 11 +/- 11% in the control group, 37 +/- 37% in group Ia (tumors were 10-20 mm in diameter) and 49 +/- 37% in group Ib (tumors were 25-50 mm), so it seemed that the proportion of portal blood flow in the normal tissue microcirculation decreased with tumor growth. The tumor LDF decreased by 88 +/- 13%. After occlusion of the portal branch, the normal tissue LDF in the main lobe decreased and then recovered slightly (most evident in the control group and least in group Ib). This recovery was probably due to the hepatic arterial buffer response. The tumor LDF decreased by 36 +/- 10% in group Ia and 11 +/- 17% in group Ib. There was no difference between group I (tumors were implanted directly) and group II (tumors were implanted via portal vein). Ex. 2: Adenosine and prostaglandin E1 increased blood flow in the normal tissue and decreased the tumor blood flow, while angiotensin II had the opposite effect. Vasoactive agents can be used to selectively increase or decrease tumor blood flow and are available as adjuvants for the treatment of liver tumors. Adenosine may enhance the selective tumor heating in local hyperthermia.     

A comparison of passive and active shunting for bypass of the retrohepatic IVC

In order to provide improved shunting of caval blood around the liver for major juxtahepatic venous injuries, a modification of the venovenous bypass (active shunt) used in liver transplantation was developed. Using a porcine model, hemodynamic comparisons of active shunting with an interposed Bio-Medicus pump (group I: n = 6) and passive shunting (group II: n = 4) around the liver for 60 minutes were made. One end of the shunt was placed in the infrahepatic cava and the other end was inserted into the right atrium. Systolic blood pressure (sBP) and cardiac output (CO) were well maintained in group I. However, with passive shunting (group II), sBP fell from 134 +/- 28 to 83 +/- 28 mm Hg (p less than 0.05) and CO fell from 4.1 +/- 0.07 to 1.3 +/- 0.5 L/min (p less than 0.001) after 1 hour. The well-maintained sBP and CO in group I were associated with much better shunt flow rates than in group II (31 +/- 7 vs. 11 +/- 3 mL/kg/min) (p less than 0.001). The cause of the fall in sBP and CO with the passive shunt (group II) in spite of a well-maintained PAWP is unclear at this time. Thus, it appears that active shunting of blood around the liver using a venovenous bypass with a pump is much superior hemodynamically to passive shunting, which relies only on hydrostatic pressure.     

Portal hyperperfusion causes disturbance of microcirculation and increased rate of hepatocellular apoptosis: investigations in heterotopic rat liver transplantation with portal vein arterialization

Clinical results of portal vein arterialization (PVA) in liver transplantation are controversial. One reason for this is the lack of a standardized flow regulation. Our experiments in rats compared PVA with blood-flow regulation to PVA with hyperperfusion in heterotopic auxiliary liver transplantation (HALT). In group I (n = 19), the graft's portal vein was completely arterialized via the right renal artery in-stent technique, using a 0.3-mm stent, leading to a physiological average portal blood flow. In group II (n = 19), a 0.5-mm stent was used. In group II, the average portal blood flow after reperfusion was significantly elevated (group II: 6.4 +/- 1.5; group I: 1.7 +/- 0.4 mL/min/g of liver weight; P < .001). The sinusoidal diameter after reperfusion was significantly greater in group II (9.8 +/- 0.5 microm) than in group I (5.5 +/- 0.2 microm; P < .001). Red blood cell velocity in the dilated sinusoids was significantly lower in group II (171 +/- 18 microm/s) than in group I (252 +/- 13 microm/s). Stasis of erythrocytes occurred; consequently, the functional sinusoidal density was significantly reduced in group II (38 +/- 7%) compared with group I (50 +/- 3%; P < .01). Two hours after reperfusion of the portal vein, the number of apoptotic hepatocytes was significantly higher in group II than in group I (I: 0 +/- 0 vs II: 7 +/- 9 M30-positive hepatocytes/10 high-power fields). The 6-week survival rate was 9 of 11 in both groups. In group II, 6 of 9 grafts showed massive hepatocellular necroses after 6 weeks, whereas in group I, only 1 of 9 presented a slight hepatocellular necrosis. Finally, our results demonstrate negative effects of portal hyperperfusion in transplanted livers, which are correctable by adequate flow regulation.     

Intraoperative measurement of graft blood flow - a necessity in liver transplantation

Abstract Portal venous and hepatic arterial flow was measured intraop-eratively in the 70 most recent patients undergoing liver transplantation in our institution. Impaired graft flow due to vascular abnormalities was detected in six patients. One patient suffered from arterial steal due to stenosis of the recipient celiac trunk with blood shunting from the hepatic to the splenic artery. Ligation of the recipient hepatic artery restored the arterial graft flow. In two patients we found reduced portal venous flow due to large portosystemic collaterals. The collaterals accountable for the impaired portal flow were identified and ligated, which restored portal venous graft flow. Excessive sensitivity of the portal venous flow to the position of the graft was found in a 6-month-old boy. Portal venous flow varied considerably, depending upon the position of the graft, and intraoperative flow measurement allowed the best position of the graft to be identified. Two patients developed arterial thrombosis in the early postoperative course. Immediate laparatomy with thrombectomy resulted in good, palpable pulsation in the graft artery in both patients. Intraoperative flow measurement demonstrated satisfactory arterial flow in one patient, whereas there was no net flow in the other patient's graft artery. Pulsation in this patient was caused by blood oscillating in and out of the liver. In conclusion, we find that causes of primary graft dysfunction due to technically flawed reperfusion of the graft can be identified and alleviated by intraoperative measurement of the flow in the graft vessels.     

Intraoperative measurement of graft blood flow — a necessity in liver transplantation

Portal venous and hepatic arterial flow was measured intraoperatively in the 70 most recent patients undergoing liver transplantation in our institution. Impaired graft flow due to vascular abnormalities was detected in six patients. One patient suffered from arterial steal due to stenosis of the recipient celiac trunk with blood shunting from the hepatic to the splenic artery. Ligation of the recipient hepatic artery restored the arterial graft flow. In two patients we found reduced portal venous flow due to large portosystemic collaterals. The collaterals accountable for the impaired portal flow were identified and ligated, which restored portal venous graft flow. Excessive sensitivity of the portal venous flow to the position of the graft was found in a 6-month-old boy. Portal venous flow varied considerably, depending upon the position of the graft, and intraoperative flow measurement allowed the best position of the graft to be identified. Two patients developed arterial thrombosis in the early postoperative course. Immediate laparatomy with thrombectomy resulted in good, palpable pulsation in the graft artery in both patients. Intraoperative flow measurement demonstrated satisfactory arterial flow in one patient, whereas there was no net flow in the other patient's graft artery. Pulsation in this patient was caused by blood oscillating in and out of the liver. In conclusion, we find that causes of primary graft dysfunction due to technically flawed reperfusion of the graft can be identified and alleviated by intraoperative measurement of the flow in the graft vessels.     

Hypoxia-induced bacterial translocation in the puppy.

Because hypoxia is one of the most common major stresses to which a neonate is exposed, we postulated that it alone might be the cause of intestinal bacterial translocation, which could be the underlying etiology of neonatal sepsis. An animal model, in which hypoxia is the sole stress, was developed in our laboratory and tested in 18 puppies to determine the effect of hypoxia and reoxygenation on intestinal bacterial translocation. In group I (n = 8), following laparotomy and cannulation of the superior mesenteric vein (SMV), the FIO2 was decreased from 21% to 9% for 90 minutes followed by reoxygenation at 21% for 120 minutes. The abdomen was closed and the animals were allowed to recover. After 24 hours the mesenteric lymph nodes (MLNs), spleen, and liver were harvested for bacterial determination and the ileum and jejunum for histological evaluation. Group II (n = 7) was treated the same as group I with the FIO2 maintained at 21%. Group III (n = 3) animals were killed, without intervention, for bacterial analysis. In group I, the systemic PO2 decreased by 75%, SMV PO2 decreased by 64%, and oxygen delivery to the small bowel decreased by 80% in comparison with group II. The mean arterial pressure and cardiac output were not significantly different between group I and group II; however, the mucosal blood flow was decreased by 60% (P less than .001) in group I. Arterial and SMV blood lactic acid levels were unchanged in group I in comparison with group II, suggesting that anaerobic metabolism was not initiated in the splanchnic circulation during hypoxia.(ABSTRACT TRUNCATED AT 250 WORDS)     

The influence of hypoxia and hyperoxia on the kinetics of propofol emulsion

PURPOSE: To study the effect of hypoxia and hyperoxia on the pharmacokinetics of propofol emulsion, hepatic blood flow and arterial ketone body ratio in the rabbit. METHODS: Twenty four male rabbits were anesthetized with isoflurane (1.5-2%) in oxygen. After the surgical procedure, isoflurane administration was discontinued and intravenous propofol infusion (30 mg x kg(-1) x hr(-1)) was started. The infusion rate of propofol was maintained throughout the study. After an initial 90 min period of propofol infusion, rabbits were randomly allocated to one of three groups: hypoxia (F(I)O2 = 0.1), normoxia (F(I)O2 = 0.21), and hyperoxia (F(I)O2 = 1.0). Propofol infusion was continued under the allocated F(I)O2 for 60 min. Propofol concentrations in arterial blood, total body clearance of propofol, hepatic blood flow and arterial ketone body ratio were measured. RESULTS: The mean arterial propofol concentration at the end of infusion was higher in the hypoxia group (15.2 +/- 2.8 microg x mL(-1), mean +/- SD) than in the normoxia (7.4 +/- 1.7) and hyperoxia (8.0 +/- 1.9) groups (P < 0.05). Total body clearance of propofol, hepatic blood flow and arterial ketone body ratio were all reduced in the hypoxia group (P < 0.05). Total ketone body concentration in arterial blood increased in the hyperoxia group (P < 0.01). CONCLUSION: Hypoxia produced an accumulation of propofol in blood and reduced propofol clearance. These changes could result from decreased hepatic blood flow and low cellular energy charge in the liver. Hyperoxia, on the other hand, increased total ketone body in arterial blood.     

Effects of ketamine on liver blood flow and hepatic oxygen consumption. Studies in the anaesthetised greyhound

The effects of three increasing doses of ketamine on the blood flow to, and oxygen consumption of, the liver, were studied in seven anaesthetised greyhounds. Hepatic arterial and portal venous blood flows were measured continuously using electromagnetic flow probes, and mean arterial pressure and cardiac output monitored as appropriate. Ketamine, even at the highest dose, had little effect on the blood flow to the liver: hepatic arterial blood flow and portal venous blood flow did not differ significantly from their baseline values. However, the oxygen delivery to the liver decreased due, probably, to an increase in oxygen consumption by the pre-portal organs.     

The status of splanchnic blood circulation in patients with varicose veins of the esophagus and stomach in liver cirrhosis]

Complex hemodynamical investigations were done in 166 patients with liver cirrhosis and the portal hypertension syndrome. Patients with varicose veins of the esophagus and stomach versus patients with isolated varicose veins of the esophagus have significantly higher resistance of vessels of the a. hepatica and v. porta systems, more pronounced losses of portal perfusion at the expense of varicose veins of stomach, gastro- and splenorenal shunts and lower volumetric blood flow in v. lienalis. While varicose veins of the esophagus and stomach occur an absolute values of the arterial and portal blood flow in liver are lowering, common hepatic blood flow reduces. The varicose veins of the stomach existence testifies high degree of the portosystemic shunting development with subsequent lowering of volumetric blood flow in v. lienalis in comparison with such in isolated varicose veins of the esophagus.     

Peculiarities of splanchnic hemodynamics after performance of partial portosystemic shunting

Complex hemodynamical investigations were conducted in 1985-2004 yrs in 60 patients with hepatic diseases before the operation, in 6-8 and 12-24 months after performance of portosystemic shunting operation (in 26 patients was formed H-like mesentericocaval anastomosis, in 10--central splenorenal anastomosis, in 14--splenorenal anastomosis side-to-side, in 10--H-like splenorenal anastomosis). Performance of shunting operation had promoted the lowering of volumetric speed of blood flow and pressure in portal vein, her diameter reduction, the general hepatic blood flow lowering. The lowering of blood flow in the portal vein system after the shunting have caused enhancement of arterial hepatic blood flow.     

肝缺血再灌注后肝内血流动力学的变化

目的探讨肝脏缺血再灌注（I／R）后肝内分流（IHSF）和功能性肝血流（FHBF）的变化。方法健康雄性SD大鼠l2只，作右侧颈动脉、颈静脉插管；开腹后，经回结肠静脉作门静脉插管，分别用以输血、输液、给药、留样、检测等。大鼠经部分肝I／R制模后，随机分为2组（每组6只）：（1）正常对照组（对照组），术中只分离肝周围韧带，不作肝门阻断及再灌注。（2）缺血再灌注组（1／R组，实验组），进行45min的肝门阻断及60min的再灌注。然后两组均经门静脉输注D一山梨醇（10mmol／L，0．2mL／min），2min后同时取颈动脉、门静脉、肝静脉血各1mL。测定门静脉血流量（PVF）、肝动脉血流量（HAF）。计算肝脏山梨醇摄取率、FHBF和IHSF。结果两组PVF，HAF及总肝血流量（THBF）无明显统计学差异；与对照组比较，I／R组肝脏山梨醇摄取率和FHBF减少，IHSF增加（P〈0．01）。结论肝I／R后，肝内血流动力学变化表现为肝内门一体分流开放，功能性肝血流减少。     

Changes in hepatic haemodynamics and hepatic perfusion index during the growth and development of hypovascular HSN sarcoma in rats

Experimental liver tumours were induced in the Hooded Lister rat by the intraportal inoculation of 10(6) HSN sarcoma cells. The hepatic perfusion index was raised 10 days after the inoculation of cells (at the micrometastatic stage) and when overt tumour was present 20 days after inoculation. Overt tumours were hypovascular compared with normal liver. Portal venous flow and portal venous inflow fell significantly when the hepatic perfusion index was increased, but hepatic arterial flow did not alter. Portal vascular resistance and splanchnic vascular resistance were both increased in tumour-bearing animals but portal pressure, arteriosystemic shunting and portosystemic shunting did not increase significantly at any stage during the growth of hepatic tumour. These findings confirm that the hepatic perfusion index can be elevated in the presence of both micrometastic and overt hepatic tumour and that the changes are not due to either arteriosystemic shunting or mechanical portal venous obstruction.     

An experimental study on the pathophysiology of stenosis and occlusion in portal vein reconstruction in resection of cancer of the hepatic hilus

An experimental study using mature mongrel dogs was performed to clarify the pathophysiology of stenosis and occlusion of portal vein reconstruction accompanied with hepatectomy. All the animals underwent partial (53%) hepatectomy. They were arbitrarily divided into three groups: Non-stenosis Group I with hepatectomy only, Stenosis Group II with partial '70%) stenosis of the portal vein, and Occlusion Group III with ligation of the portal vein. All cases of Group III died within about 122 minutes. The blood flow and pressure of the portal vein, portography, ICG Rmax and the residual liver weight were serially examined until the fourth week following the operation in Group I and Group II. Two principal results were derived: 1) In Group I, portal circulation was sufficiently restored and the residual liver showed adequate regeneration. 2) In Group II, hepatofugal collateral vessels developed. However, the portal pressure remained significantly high (p less than 0.002) and, the portal blood flow and liver tissue blood flow were markedly reduced (p less than 0.001) for 1 week after operation. The residual liver weight and liver function (ICG Rmax) were significantly decreased even in the fourth week. Recently, portal vein resection accompanied with hepatectomy has been accepted as a procedure for advanced carcinoma of the hepatic hilus. This study suggests that stenosis or occlusion of the portal vein should be avoided in the procedure.     

Protective effect of agiotensin II type I receptor antagonist, CV-11974, on ischemia and reperfusion injury of the liver

BACKGROUND: Microcirculatory disturbance has been shown to play a critical role in hepatic ischemia and reperfusion (I/R) injury. Angiotensin II (AngII) is one of the most potent endogenous vasoconstrictors. Angiotensin II type I (AT1) receptor antagonist has been reported to have protective effects on I/R injury of the heart and kidney. However, effect on hepatic I/R injury has not been determined. In this study, we investigate our hypothesis that AT1 receptor antagonist, CV-11974, attenuates hepatic I/R injury. METHODS: Twelve beagle dogs underwent a 2-hr total hepatic vascular exclusion with veno-venous bypass. CV-11974 was given to animals at a dose of 0.002 mg/ kg/min for 5 min followed by 0.001 mg/kg/min for 25 min via portal vein before ischemia (group II, n=6). Nontreated animals were used as the control (group I, n=6). Animal survival, hemodynamics, hepatic tissue blood flow (HTBF), liver function, platelet count, renin activity, and AngII concentration of hepatic vein, energy metabolism, and histopathology were analyzed. RESULTS: Two-week survival was 33% in group I, in contrast, 100% in group II. Mean arterial blood pressure during early reperfusion was maintained, and HTBF after reperfusion was significantly higher in group II. Treatment attenuated liver enzyme release and decrease of platelet count, increased renin and AngII, suppressed ATP degradation during ischemia and enhanced ATP resynthesis after reperfusion. Neutrophil infiltration and histopathological damages were lessened in group II. CONCLUSIONS: Our data demonstrated that the local renin-angiotensin system might play a role in hepatic microcirculation. AT1 receptor blockade with CV-11974 attenuated hepatic microcirculatory disturbance and ameliorated I/R injury.     

The feasibility of in vivo resection of the left lobe of the liver and its use for transplantation

The anatomical possibility of resecting the left lobe of the liver (segments II and III) in living subjects and using it for transplantation was evaluated. A group of 60 cadaveric livers were dissected at autopsy. The vascular and biliary elements of the left lobe were isolated and the lobe was resected and evaluated for possible grafting. The left lobe was 12-28% (mean 19.4%) of the liver mass. An extrahepatic segment of the left hepatic vein was isolated in 95% of specimens. Arterial blood supply to the left lobe consisted of a single artery (92%) or two arteries (8%). A single portal vein segment to the left lobe (type I) was found in 35% livers. Portal vein branches originated from a common orifice (type II, 35%) or separately (type III, 30%) from the left portal vein, and in these instances, preparation of a portal segment necessitated partial section of the left portal vein wall. Biliary drainage was extrahepatic in 56 livers and consisted of a single duct (type I, 78%), or two ducts (type II, 15%). The resected left lobe was evaluated as satisfactory (single hepatic vein and artery, types I or II portal vein, type I bile duct) in 48% of cases, while a less-satisfactory lobe (type III portal vein or type II bile duct) was obtained in 33%. It was found anatomically difficult or impossible to resect the left lobe for possible transplantation in 11 (19%) liver specimens.     

Hepatic oxygen supply during halothane or isoflurane anesthesia in guinea pigs

The present study was designed to determine changes in hepatic oxygen supply in guinea pigs during halothane or isoflurane anesthesia. Twenty-seven guinea pigs were randomly divided into three equal groups: control (no anesthesia) group, and animals anesthetized with halothane or isoflurane to decrease mean arterial pressure (MAP) by 50%. Hepatic arterial blood flow (HABF) and portal blood flow (PBF), as well as arterial and portal venous blood oxygen content, were determined in awake animals (stage I, baseline values), and during anesthesia (stage II). HABF was found to be extremely low (0.04 ml.min-1.g-1) during both stages of observation in the control (no anesthesia) group, as well as during stage I (awake) in animals treated with halothane or isoflurane. Equal degrees of arterial hypotension during halothane and isoflurane anesthesia were accompanied by decreased HABF during halothane (37%), but no significant change in HABF during isoflurane anesthesia. PBF decreased significantly in both experimental groups; however, the decrease was more prominent during halothane than during isoflurane anesthesia (57% vs. 23%). The observed hepatic circulatory changes led to a 65% decrease in hepatic oxygen delivery during halothane, but only a 34% decrease during isoflurane anesthesia. The present study does not exclude the possibility that liver damage in the guinea pig model is related to the reductive metabolism of halothane or any other mechanism. However, the extremely low HABF and a prominent reduction in both HABF and PBF during halothane anesthesia may be responsible for hepatic damage observed in the guinea pig model.