Effect of Endotoxemia on Hepatic Portal and Sinusoidal Blood Flow in Rats

Journal of Surgical Research 89, 26 –30 (2000) doi:10.1006/jsre.1999.5811, available online at http://www.idealibrary.com on

Effect of Endotoxemia on Hepatic Portal and Sinusoidal Blood Flow in Rats A. Secchi, M.D., 1 J. M. Ortanderl, W. Schmidt, M. M. Gebhard, M.D.,* E. Martin, M.D., and H. Schmidt, M.D. Department of Anesthesiology and *Department of Experimental Surgery, University of Heidelberg, Im Neuenheimer Feld 110, Heidelberg 69120, Germany Submitted for publication August 6, 1999

Key Words: sepsis; endotoxin; liver; microcirculation; portal blood flow; sinusoidal blood flow.

A decrease in liver blood flow leads to dysfunction of hepatocytes and Kupffer cells, with subsequent local and systemic liberation of proinflammatory mediators that may maintain systemic inflammatory response syndrome (SIRS) and may lead to multiple organ dysfunction syndrome (MODS). There is only limited knowledge on the hepatic micro- and macrocirculation during sepsis or endotoxemia. Therefore, the aim of our study was to investigate alterations in hepatic portal blood flow (PBF) and sinusoidal blood flow (SBF) during endotoxemia. In male Wistar rats endotoxemia was induced by continuous infusion of 2 mg/ kg/h lipopolysaccharides from Escherichia coli 026:B6 immediately after baseline measurements (n ⴝ 8). The control group (n ⴝ 8) received an equivalent volume of Ringer’s solution. Mean arterial pressure (MAP), heart rate (HR), cardiac output (CO), PBF, and SBF were measured at baseline and 60 and 120 min after induction of endotoxemia. PBF was measured using an ultrasonic flow probe that was positioned around the portal vein. SBF was detected by in vivo videomicroscopy of the left liver lobe. In the LPS group MAP decreased, but CO remained at baseline values. During endotoxemia PBF decreased significantly from 23 ⴞ 3 to 15 ⴞ 4 mL/min (60 min) and 16 ⴞ 3 mL/min (120 min). SBF also significantly decreased to 68.5% (60 min) and 57.1% (120 min) of baseline value. Our results demonstrate that during early endotoxemia hepatic macroand microcirculatory perfusion is significantly decreased despite unchanged CO. This early reduction of hepatic perfusion might be caused by an increased hepatic vessel resistance as a consequence of liberation of vasoconstrictive mediators or/and by a decrease in intestinal perfusion. © 2000 Academic Press

INTRODUCTION

The liver plays an important role in the development of multiple organ failure (MOF) during sepsis. Whereas the gut is considered to be the “motor of MOF” the liver seems to be the modulator of it [1, 2]. Pathophysiologically, hypoperfusion of the gut during sepsis or endotoxemia may lead to disruption of the mucosal barrier with subsequent translocation of bacteria and endotoxin (ETX) from the gut to the portal vein [3, 4]. ETX is known to be a potent activator of macrophages and endothelial cells which release inflammatory mediators such as tumor necrosis factor ␣ (TNF-␣) and interleukins, as well as constrictive mediators like endothelin-1 and thromboxane [5, 6]. While these mediators may have beneficial effects, for example, regulation of the hepatic acute phase response [7], they also have a variety of toxic effects. In addition to alterations in macrophage function, microcirculatory disturbances and pathological leukocyte– endothelium interactions in liver sinusoids have been reported after ETX application [8]. In the isolated liver perfusion model it has been shown that a decrease in liver blood flow may occur during endotoxemia, leading to further dysfunction of hepatocytes and Kupffer cells [9]. Under normal conditions a decrease in portal venous blood flow leads to an increase in hepatic arterial blood flow. This arterial buffer response maintains hepatic blood flow [10]. During sepsis or endotoxemia, however, there is only limited knowledge about the hepatic micro- and macrocirculation. Therefore, the aim of our study was to investigate the effects of ETX on hepatic portal blood flow with an

1 To whom correspondence should be addressed. Fax: ⫹⫹496221565345. E-mail: [email protected].

0022-4804/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

26

SECCHI ET AL.: ENDOTOXEMIA AND HEPATIC MICROCIRCULATION

ultrasonic flow probe and sinusoidal blood flow with in vivo microscopy. To minimize systemic hemodynamic effects on regional blood flow in the liver we used a normodynamic model of endotoxemia. METHODS Animal preparation. The study protocol and all experimental procedures used in this investigation were approved by the Governmental Animal Protection Committee. Male Wistar rats (250 –350 g body wt) were used for the experiment. All animals were kept on a diet of standard rat chow until the day before the experiment. Twelve hours before the experiment the rats had only free access to water. Anesthesia was performed by intraperitoneal injection of 25 mg/kg body wt pentobarbital (Narcoren, Merial, Hallbergmoos, Germany) and intramuscular injection of 28 mg/kg body wt ketamine (Ketanest, Parke–Davis, Berlin, Germany). During the experimental period anesthesia was maintained by continuous infusion of 9.2 mg/kg body wt pentobarbital and 3.2 mg/kg body wt/h vecuronium (Organon, Boxtel, The Netherlands) to allow mechanical ventilation with room air, p aCO 2 between 32 and 42 mm Hg. The right jugular vein was cannulated with a polyethylene catheter (outer diameter 0.8 mm, inner diameter 0.5 mm) for drug administration. Then a thermistor probe was inserted into the left carotid artery for measurement of cardiac output using transpulmonary thermodilution method. For blood pressure measurement another polyethylene catheter was inserted into the femoral artery. A tracheotomy was performed for airway control and mechanical ventilation. The temperature of the animal was maintained at 37°C using a heating lamp. The abdomen was opened by midline laparotomy. The portal vein was cautiously liberated from perivascular tissue and a flow probe (2 S B, Transonic Systems, Ithaca, NY) was positioned around it. Then the left liver lobe was exteriorized upside down on a specially designed Plexiglas stage. FITC labeling of erythrocytes. Erythrocytes from separate donor rats were labeled with fluorescein isothiocyanate (FITC, Isomer I, No. F-7250, Sigma Chemical, Germany) using a modified procedure according to Butcher and Weissmann [11] and Sarelius and Duling [12]. Blood from donor rats was washed three times with Alsever’s buffer solution and one time with bicine–saline buffer solution to remove plasma. Then the washed erythrocytes were diluted 1:2 with bicine–saline buffer solution and incubated with FITC (9 mg/mL erythrocytes) for 180 min at 25°C. Labeled erythrocytes were further washed five times in bicine–saline buffer solution. Then the erythrocytes were diluted with saline until the hematocrit was 50%. CPD solution (citrate–phosphate– dextrose, No. C-7165, Sigma Chemical, Germany) was used for conservation. Thirty minutes prior to the first microscopy all animals received 0.5 mL/kg body wt FITC-labeled erythrocytes. Intravital microscopy. Intravital microscopy was performed using a method described by Marzi et al. [13]. The exposed liver surface was superfused with warmed Ringer’s solution (37°C). A microscope specially designed for intravital microscopy (Leica, Germany) and equipped with a 16-fold water immersion objective, a 10-fold eyepiece, and a transfer lens was used. The FITC-labeled erythrocytes were visualized using epifluorescence illumination. This was performed using an illuminator (Leitz, Germany) consisting of a XBO 100W/2 short-arc mercury lamp and a bypass filter (450 – 490 nm) for the excitation fluorescence wavelength. A dichroic mirror with a 510-nm cutoff wavelength was located in the body of the microscope. Further rejection of FITC emission was achieved using a barrier filter at 520 nm located in front of the eyepiece. The images from the microscope were transferred to a monitor (PVM/444QM; Sony, Japan) by a low-light camera (CF 8/1, Kappa, Germany) and simultaneously recorded on videotape using a video recorder (AG 7350, Panasonic, Japan) for off-line analysis.

27

Measurement of blood flow in the liver sinusoids. A modified method described by Marzi et al. [14] was used for measurement of blood flow in the liver sinusoids. At each time point the velocity (v) of 50 erythrocytes in five liver acini was assessed off-line with a computer-assisted measurement system (CAP IMAGE, Zeintl, Heidelberg, Germany). Also, the sinusoid diameters (d) were measured 90 ␮m around the central vein. Sinusoidal blood flow was then calculated as flow (␮m 3/s) ⫽ ␲ ⴱ d 2/4 (␮m 2) ⴱ v (␮m/s). Measurement of portal blood flow. Portal blood flow was measured using the flow probe of a small animal ultrasonic flowmeter (Transonic Systems). Cardiocirculatory monitoring. Mean arterial blood pressure (MAP) and heart rate (HR) were measured continuously using arterial catheterization and recorded every 60 min. Cardiac output (CO) was determined every 60 min using the transpulmonary thermodilution method with central venous injection of 100 ␮L cold saline solution. Hematocrit was also determined. Experimental protocol. The rats were randomized into two groups of eight animals each. Following administration of FITClabeled erythrocytes a stabilization period of 30 min was allowed. Animals in the test group (LPS group) were then challenged with a continuous intravenous infusion of endotoxin (2 mg/kg body wt/h lipopolysaccharide Escherichia coli 026:B6, Sigma Chemical, Germany) over a 120-min period. Each animal received a total amount of 25 mL/kg body wt/h fluids. The control group received equivalent volumes of Ringer’s solution throughout the study. Video microscopy for the measurement of sinusoidal blood flow and sinusoidal diameters was performed at baseline and 60 and 120 min later. Statistical analysis. For statistical analysis group means and standard deviation were calculated. Differences between the groups were determined with an unpaired Student t test. Differences within each group were analyzed by repeated-measures ANOVA. Differences were considered to be significant at P ⱕ 0.05.

RESULTS

None of the animals showed clinical signs of infection or sepsis prior to the experiments. There were no differences between the groups with respect to weight (304 ⫾ 25 g vs 290 ⫾ 34 g), hematocrit, or macro- and microhemodynamic parameters at the start of the experiment. The hematocrit remained stable during the entire experimental period in both groups (see Table 1). Macrohemodynamic Changes There were no changes in MAP during the study period in the control group. In the LPS-treated animals MAP decreased significantly at 60 min (P ⱕ 0.05) and 120 min (P ⱕ 0.01). At 120 min MAP in the LPS group was also significantly lower than in the control group. Heart rate remained unaltered in the animals that received Ringer’s solution, but increased significantly from 395.6 ⫾ 52.5 to 465.0 ⫾ 52 at 60 min (P ⱕ 0.05) and to 534.4 ⫾ 22.6 at 120 min in the LPS-treated animals (P ⱕ 0.01). At 120 min HR was significantly higher in the LPS-treated group than in the control group (P ⱕ 0.01). Cardiac output, however, remained at baseline values in both groups (see also Table 1).

28

JOURNAL OF SURGICAL RESEARCH: VOL. 89, NO. 1, MARCH 2000

TABLE 1 Hemodynamic Parameters and Hematocrit in Control Animals (Ringer’s Solution) and in Animals That Received LPS Ringer’s solution

LPS 2.0 mg/kg body wt/h

Variable

Baseline

60 min

120 min

Baseline

60 min

120 min

MAP (mm Hg) Heart rate (L/min) CO (mL/min) Hct (%)

106.5 ⫾ 12.2 423.8 ⫾ 43.7 116.0 ⫾ 15.6 43.8 ⫾ 1.2

107.9 ⫾ 8.0 442.5 ⫾ 33.1 127.5 ⫾ 21.0 42.3 ⫾ 2.9

105.6 ⫾ 6.8 459.4 ⫾ 39.2 118.8 ⫾ 13.3 43.0 ⫾ 2.8

112.4 ⫾ 10.1 395.6 ⫾ 52.5 109.4 ⫾ 21.1 42.5 ⫾ 3.2

101.4 ⫾ 7.0† 465.0 ⫾ 52.0† 121.6 ⫾ 32.3 42.1 ⫾ 3.7

96.0 ⫾ 8.1* ,‡ 534.4 ⫾ 22.6** ,‡ 106.3 ⫾ 14.5 41.4 ⫾ 3.3

Note. All data are means ⫾ SD. * P ⱕ 0.0.5 versus placebo. ** P ⱕ 0.01 versus placebo. † P ⱕ 0.05 versus baseline. ‡ P ⱕ 0.01 versus baseline.

Liver Hemodynamic Changes Portal blood flow remained stable in the control group, but significantly decreased from 22.8 ⫾ 4.3 to 15.8 ⫾ 4.3 mL/min at 60 min and to 16.0 ⫾ 2.9 mL/min at 120 min (P ⱕ 0.01) in the LPS group (Fig. 1). Sinusoidal blood flow, which was stable in the control group, decreased significantly to 67% of baseline value at 60 min and to 56% of baseline value at 120 min (P ⱕ 0.01) (Fig. 2). Both portal and sinusoidal blood flow were at 60 and 120 min significantly lower in the LPS-treated animals. Sinusoidal diameters also decreased significantly in the LPS group from 14.3 ⫾ 0.3 to 12.7 ⫾ 0.4 ␮m at 60 min and to 11.9 ⫾ 0.3 ␮m at 120 min (both P ⱕ 0.01 vs baseline) (Fig. 3). DISCUSSION

Intravital microscopy of the liver is a wellestablished method to study liver microcirculation. In addition to observation of leukocyte– endothelial cell interaction after liver transplantation [15] and during sepsis [16], it has been used to determine sinusoidal diameters and sinusoidal blood flow after hemorrhagic shock [14]. In our study, we evaluated the effects of a continuous systemic infusion of endotoxin on hepatic portal and sinusoidal blood flow. In earlier studies we demonstrated in a similar model that endotoxemic rats exhibited changes that were comparable to clinical findings observed during compensated sepsis in humans [3, 17], and they met the criteria for laboratory models of sepsis proposed by Fink and Heard [18]. In the control groups, all parameters we observed remained at baseline values throughout the entire experimental period. Human beings and animals often develop a hyperdynamic cardiovascular response during early septic shock [19, 20]. In our model, however, we tried to produce hemodynamic changes that were

comparable to those observed in humans during compensated sepsis [17]. Thus, a slight decrease in mean arterial pressure occurred and heart rate increased significantly, whereas cardiac output remained stable. It has been shown that hemodynamic effects of a slow infusion of endotoxin are dose dependent. So small doses (e.g., 0.01 mg/100 g rat wt) produced a hyperdynamic situation; with increasing doses, however, a hypodynamic response is seen [21]. The dose of endotoxin we used combined with high-volume therapy produced a nearly normodynamic situation in our experimental animals. Portal blood flow in the LPS group at baseline measurement and in the control group at each time point was approximately 20% of cardiac output. This is comparable to findings in healthy humans [22]. In the LPS-treated group, however, portal blood flow decreased significantly at 60 min and remained at this level until 120 min. Sinusoidal blood flow in endotoxemic rats constantly decreased to ca. 67% of baseline at 60 min and to ca. 56% of baseline at 120 min. This decrease in sinusoidal blood flow was accompanied by a decrease in sinusoidal diameters. Two reasons for the

FIG. 1. Portal blood flow in control animals (Ringer’s solution) and in animals receiving LPS (2 mg/kg body wt/h). All data are means ⫾ SD. *P ⱕ 0.05 versus placebo, **P ⱕ 0.01 versus placebo, ⫹⫹ P ⱕ 0.01 versus baseline.

SECCHI ET AL.: ENDOTOXEMIA AND HEPATIC MICROCIRCULATION

reduction in sinusoid diameters are discussed. First, swelling of Kupffer cells and endothelial cells might lead to a reduction in sinusoid diameters [23]. Second, stellate cells (Ito cells), the liver-specific microvascular pericytes covering the sinusoidal wall, might increase microvascular tone [24]. In the isolated liver perfusion model it has been shown that after endotoxin treatment sinusoidal diameters as well as sinusoidal volumetric flow are decreased [9]. Endothelin-1-mediated vasoconstriction is thought to play an important role in the increase in hepatic vessel resistance with subsequent decrease in hepatic perfusion. Thus, after endotoxin exposure an increase in the production of endothelin-1 by liver endothelial cells has been observed [6]. Pannen et al. [25] demonstrated that bosentan, an endothelin-A and endothelin-B receptor antagonist, was able to reduce increased total and sinusoid resistance after LPS administration. In contrast, Kaneda et al. [26] did not find sinusoidal or central venous vasoconstriction after endothelin-1 infusion using light and electronic microscopy, but described a constriction of the proximal and distal segments of preterminal portal venules. A possible reason for the lack of contraction of sinusoids in their study might be that the animals were not treated with endotoxin. It is known that LPS enhances portal venous contractile response to endothelin-1 at sinusoidal and presinusoidal levels [9] so that endothelin-1 might be a potent constrictor of sinusoids during endotoxemia. Thus, endotoxin-induced endothelin-1 might be an explanation for the observed hepatic microcirculatory alterations in our study. The role of angiotensin II, another important vasoconstrictor, is still unclear. On the one hand, treatment with the angiotensin-converting enzyme inhibitor enalapril ameliorated gastrointestinal structural and functional damage and decreased bacterial translocation in mice after thermal injury and bacterial challenge [27]. On the other hand, the microvascular response to angiotensin II is moderated in septic rats [28], and especially in hepatocytes, treatment with LPS decreased the angiotensin action on intracellular calcium levels [29].

FIG. 2. Sinusoidal blood flow in control animals (Ringer’s solution) and in animals receiving LPS (2 mg/kg body wt/h). All data are means ⫾ SD. **P ⱕ 0.01 versus placebo, ⫹⫹P ⱕ 0.01 versus baseline.

29

FIG. 3. Sinusoidal diameter in control animals (Ringer’s solution) and in animals receiving LPS (2 mg/kg body wt/h). All data are means ⫾ SD. *P ⱕ 0.05 versus placebo, **P ⱕ 0.01 versus placebo, ⫹⫹ P ⱕ 0.01 versus baseline.

Interestingly, we found that the reduction of sinusoidal blood flow was associated with a decrease in portal venous blood flow. Thus, besides an increased hepatic vascular resistance, a decreased influx from the portal vein that is not compensated by the hepatic arterial buffer response [30] might contribute to the impaired hepatic microvascular perfusion. Reduction of portal venous blood flow during sepsis has also been described in other experimental models [31] and might be the result of impaired intestinal perfusion during sepsis or endotoxemia [3, 32]. In summary, we found that the hepatic sinusoidal microcirculation is impaired during endotoxemia. Two major factors might contribute to this: an increased hepatic vessel resistance mediated by endothelin-1 and a decreased hepatic influx from the portal vein caused by an impairment of gut perfusion. REFERENCES 1.

Meakins, J. L., and Marshall, J. C. The gastrointestinal tract: The “motor” of MOF. Arch. Surg. 121: 197, 1986.

2.

Matuschak, G. M., and Rinaldo, J. E. Organ interactions in the adult respiratory distress syndrome during sepsis: Role of the liver in host defense. Chest 94: 400, 1988.

3.

Schmidt, H., Secchi, A., Wellmann, R., Bach, A., Bo¨hrer, H., Gebhard, M. M., and Martin, E. Effect of endotoxemia on intestinal villus microcirculation in rats. J. Surg. Res. 61: 521, 1996.

4.

Baron P., Trober, L. D., Nguyen, T., Hollyood, M., Heggers, J. P., and Herndon, D. N. Gut failure and translocation following burn and sepsis. J. Surg. Res. 57: 197, 1994.

5.

Nolan, J. P. Endotoxin, reticuloendothelial function, and liver injury. Hepatology 1: 458, 1981.

6.

Eakes, A. T., Howard, K. M., Miller, J. E., and Olson, M. S. Endothelin-1 production by hepatic endothelial cells: Characterization and augmentation by endotoxin exposure. Am. J. Physiol. 272: G605, 1997.

7.

Baumann, H., Prowse, K. R., Marinkovic, S., Won, K-A., and Jahreis, B. P. Stimulation of hepatic acute phase response by cytokines and glucocorticoids. Ann. NY Acad. Sci. 557: 280, 1989.

8.

Bauer, M., Marzi, I., Thuma, B., Bach, F., Bu¨hren, V., and Larsen, R. Dose related pattern of sinusoidal leukocyte adhe-

30

9.

10.

11.

12.

13.

14.

15.

16.

17.

18. 19.

20.

21.

JOURNAL OF SURGICAL RESEARCH: VOL. 89, NO. 1, MARCH 2000 sion in sublobular regions of the liver after systemic endotoxin challenge in the rat. Shock 1: 135, 1994. Pannen, B. H., Bauer, M., Zhang, J. X., Robotham, J. L., and Clemens, M. G. Endotoxin pretreatment enhances portal venous contractile response to endothelin-1. Am. J. Physiol. 270: H7, 1996. Lautt, W. W., Legare, D. J., and Ezzat, W. R. Quantitation of hepatic arterial buffer response to graded changes in portal blood flow. Gastroenterology 98: 1024, 1990. Butcher, E. C., and Weissmann, I. L. Direct fluorescent labelling of cells with fluorescein or rhodamin isocyanate. J. Immunol. Methods 37: 97, 1980. Sarelius, I. H., and Duling, B. R. Direct measurement of microvessel hematocrit, red cell flux, velocity and transit time. Am. J. Physiol. 243: H1018, 1982. Marzi, I., Takei, Y., Knee, J., Menger, M. D., Gores, G. J., Bu¨hren, V., Trentz, O., Lemasters, J. J., and Thurman, R. G. Assessment of reperfusion injury by intravital fluorescence microscopy following liver transplantation in the rat. Transplant. Proc. 22: 2004, 1990. Marzi, I., Bauer, M., Secchi, A., Bahrami, S., Redl, H., and Schlag, G. Effects of anti-tumor necrosis factor alpha on leukocyte adhesion in the liver after hemorrhagic shock: An intravital microscopic study in the rat. Shock 3: 27, 1995. Marzi, I., Walcher, F., and Bu¨hren, V. Macrophage activation and leukocyte adhesion after liver transplantation. Am. J. Physiol. 265: G172, 1993. McCuskey, R. S., Nishita, J., McDonnell, D., Baker, G. L., Urbaschek, R., and Urbaschek, B. Effect of immunoglobulin G on the hepatic inflammatory response during sepsis. Shock 5: 28, 1996. Bone, R. C., Balk, R. A., Cerra, F. B., Dellinger, R. P., Fein, A. M., Knaus, W. A., Schein, R. M. H., and Sibbald, W. J. Definition for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis: ACCP/SCCM consensus conference. Chest 101: 1644, 1992. Fink, M. P., and Heard, S. O. Laboratory models of sepsis and septic shock. J. Surg. Res. 49: 186, 1990. Dahn, M. S., Lange, P., Lobdell, K., Hans, B., Jacobs, A. L., and Mitchell, R. A. Splanchnic and total body oxygen consumption differences in septic and injured patients. Surgery 101: 69, 1987. Lang, C. H., Bagby, G. J., Ferguson, J. L., and Spitzer, J. J. Cardiac output and redistribution of organ blood flow in hypermetabolic sepsis. Am. J. Physiol. 246: R331, 1984. Johnston, T. D., Hampton, W. W., and Fry, D. E. Septic hemo-

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

dynamics produced by infusion of endotoxin. Curr. Surg. 46: 101, 1989. Reiss, W. G., Bauer, L. A., Horn, J. R., Zierler, B. K., Easterling, T. R., and Strandness, D. E., Jr. The effects of oral nifedipine on hepatic blood flow in humans. Clin. Pharmacol. Ther. 50: 379, 1991. McCuskey, R. S., and Reilly, F. D. Hepatic microvasculature: Dynamic structure and its regulation. Semin. Liver Dis. 13: 1, 1993. Rockey, D. C., and Chung, J. J. Inducible nitric oxide synthase in rat hepatic lipocytes and the effect of nitric oxide on lipocyte contractility. J. Clin. Invest. 95: 1199, 1995. Pannen, B. H., Bauer, M., Zhang, J. X., Robotham, J. L., and Clemens, M. G. A time-dependent balance between endothelins and nitric oxide regulating portal resistance after endotoxin. Am. J. Physiol. 271: H1953, 1996. Kaneda, K., Ekataksin, W., Sogawa, M., Matsumura, A., Cho, A., and Kawada, N. Endothelin-1-induced vasoconstriction causes a significant increase in portal pressure of rat liver: Localized constrictive effect on the distal segment of preterminal portal venules as revealed by light and electron microscopy and serial reconstruction. Hepatology 27: 735, 1998. Gennari, R., Alexander, J. W., Boyce, S. T., Lilly, N., Babcock, G. F., and Cornaggia, M. Effects of the angiotensin converting enzyme inhibitor enalapril on bacterial translocation after thermal injury and bacterial challenge. Shock 6(2): 95, 1996. Hollenberg, S. M., Tangora, J. J., Piotrowski, M. J., Easington, C., and Parrillo, J. E. Impaired microvascular vasconstrictive responses to vasopressin in septic rats. Crit. Care Med. 25: 869, 1997. Vergani, G., Portol’es, M. T., and Pagani, R. Escherichia coli lipopolysaccharide effects on proliferating rat liver cells in culture: A morphological and functional study. Tissue Cell 31: 1, 1999. Jakab, F., Rath, Z., Schmal, F., Nagy, P., and Faller, J. The interaction between hepatic arterial and portal venous blood flows: Simultaneous measurement by transit time ultrasonic volume flowmetry. Hepatogastroenterology 42: 18, 1995. Tighe, D., Moss, R., Haywood, G., Webb, A., Al-Saady, N., Heath, F., and Bennett, D. Dopexamine hydrochloride maintains portal flow and attenuates hepatic ultrastructural changes in a porcine peritonitis model of system organ failure. Circ. Shock 39: 199, 1993. Navaratnam, R. L. N., Morris, S. E., Traber, D. L., Flynn, J., Woodson, L., Lineares, H., and Herndon, D. N. Endotoxin increases mesenteric vascular resistance and bacterial translocation. J. Trauma 30: 1104, 1990.