| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1First Department of Surgery, Kagoshima University School of Medicine, Kagoshima 890-8520; 2Division 2, Department of Physiology, Shinshu University School of Medicine; 3Department of Cardiovascular Medicine, Shinshu University School of Health Sciences, Matsumoto 390-8621; and 4Division 2, Department of Physiology, Kanazawa Medical University, Uchinada 920-0293, Japan
Submitted 26 December 2002 ; accepted in final form 21 August 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
blood viscosity; hepatic circulation; hepatic vascular occlusion methods
The longitudinal distribution of pulmonary vascular resistance has been extensively studied using vascular occlusion techniques, and the pulmonary vasculature can be represented by a simple hydrodynamic model consisting of three segments in series, each with a characteristic resistance and compliance (10, 12). The middle segment, which contains the capillaries, has relatively low resistance and high compliance, and the other two segments have relatively low compliance and high resistance. The pulmonary arterial and venous occlusion technique (9–11) allows partitioning of the pulmonary vasculature into these three segments. Theoretically, if the flow is stopped across one segment, the pressure gradient across that segments becomes zero, and the total pressure gradient would decrease accordingly. In other words, the rapid changes in pulmonary arterial pressure and pulmonary venous pressure with inflow and outflow occlusion, respectively, represent the pressure drops across the arterial and venous relatively indistensible vessels (10). The total arteriovenous pressure difference minus the sum of these two pressure drops gives the pressure drop across the vessels in the middle that are much more distensible. Although hepatic circulation is analogous to pulmonary circulation in several aspects, as described above, no investigations have been performed to adopt this inflow and outflow occlusion technique to partition the hepatic vasculature into the three segments.
The responsiveness of the hepatic longitudinal vascular segments differs depending on vasoactive agents (3, 23, 24). The differences in the response of these vascular segments to a variety of stimuli are, in part, due to the intrinsic vasomotor properties of the blood vessels as well as their passive viscoelastic properties. In addition, the microrheological behavior of blood may vary in the segments of the hepatic vasculature. Blood apparent viscosity has been shown to change depending on the different size vessels in the systemic circulation (13, 14). As blood vessels decrease in diameter between 300 and 30 µm, blood apparent viscosity decreases due to Fahraeus-Lindqvist effect (1, 6). Hematocrit (Hct) is an important determinant of blood viscosity and can affect resistance to blood flow in the circulation (4, 17). In the liver, little is known about the effect of changes in blood Hct on the longitudinal distribution of vascular resistance. Thus we examined the effect of different Hct on the longitudinal vascular resistance distribution in the rat liver.
Norepinephrine constricts predominantly the presinusoidal vessels over the postsinusoidal vessels of hepatic veins (3, 15, 22, 24). Maass-Moreno and Rothe (15), by using the double-lumen catheter inserted through the caval wall into hepatic vein in anesthetized dogs, reported that an infusion of norepinephrine caused a large increase in portal venous pressure but little change in pressure gradient from the large hepatic vein to vena cava. However, the side port of the catheter that was used to measure the hepatic venous pressure was far downstream from the hepatic venules. Their subsequent study (22), which used the micropuncture method, revealed that norepinephrine caused significant increases in microhepatic venous (venule) pressure and the resistance between the sinusoids and the vena cava. However, a more detailed investigation on the constrictive site of norepinephrine has not been reported.
Recently, Hakim et al. (11) showed that the double occlusion conducted in isolated perfused lungs can be analyzed to provide three segments without performing the single occlusion of pulmonary arterial or venous occlusion independently. We examined whether resistance of the three segments of portal-hepatic venous vessels can be determined by one double occlusion maneuver lasting 3 to 4 s in isolated perfused rat livers. This approach has the advantage of being able to partition the vasculature into three segments when the vasculature is not in a steady state.
The first purpose of this study was to obtain the segmental resistances of portal veins and middle and hepatic veins by using both the portal (inflow) and hepatic venous (outflow) occlusion techniques in isolated rat livers perfused via the portal vein with the hepatic artery ligation. In the liver, the highest compliant and most distensible segment of the middle segment, which could be obtained by the present vascular occlusion methods, corresponds to the sinusoidal bed, but not the small portal and hepatic veins, because the sinusoids are the locus of the high compliance in the liver (2, 8). This is based on the evidence that the sinusoids comprise the majority of the vasculature of the liver and that the vascular compliance of the liver is 
10 times higher than that of the body as a whole (2). Thus one of the most important purposes of the present study is to measure the resistance of the middle segment of the sinusoids. The second purpose was to describe the effects of Hct and blood flow on the distribution of vascular resistance. The third goal was to determine the effect of norepinephrine on the vascular resistance distribution. The final purpose of the present study was to determine whether one double occlusion technique could provide portal or hepatic venous occlusion pressure, either of which was obtained individually by occlusion of the corresponding vessel.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The perfusing blood was obtained by exanguination of an intact donor rat that was anesthetized and heparinized, and this blood was diluted with 5% bovine serum albumin (Fraction V powder-A2153, Sigma) in a Krebs solution composed of (in mM) 118 NaCl, 5.9 KCl, 1.2 MgSO4, 2.5 CaCl2, 1.2 NaH2PO4, 25.5 NaHCO3, and 5.6 glucose at the following Hct levels: 30% (n = 8), 20% (n = 7), 10% (n = 8), and 0% (n = 6). The blood (50 ml) was recirculated at a constant flow rate using a Masterflex pump through a heat exchanger and a bubble trap in the portal line. The perfusing blood in the reservoir was continuously bubbled with 95% O2-5% CO2 at 37°C. The portal (Ppv) and hepatic venous (Phv) pressures were continuously measured with pressure transducers (Gould) attached to a sidearm placed just proximal to the perfusion cannula. The zero reference was set to the level of the hepatic hilus. The flow rate and height of the venous reservoir could be adjusted independently to maintain Phv at 0 to1 cmH2O, with Ppv a dependent variable. The perfusion flow rate (Q) was measured with an electromagnetic flow meter (model MFV 1200, Nihonkohden), and the flow probe was positioned in the portal inflow line. To occlude the portal or hepatic venous line instantaneously for measurement of the portal occlusion pressure (Ppo) or hepatic venous occlusion pressure (Phvo), two solenoid valves were placed around the perfusion tubes upstream from the Ppv sidearm cannula and downstream from the Phv sidearm cannula. The hemodynamic variables were continuously monitored and displayed on a thermal physiograph (model 8K23, NEC Sanei; Tokyo, Japan).
Measurement of Ppo and Phvo by single occlusion maneuver. When a steady state of a constant Ppv was reached, the single occlusion maneuver of portal or venous occlusion was performed, and the signals were sampled at 100 Hz and stored in a computer. Portal occlusion was accomplished for 3 s by closing the solenoid valve set in the portal line. Hepatic venous occlusion was accomplished for 1.5 s by closing the solenoid valve set in the hepatic venous line while inflow continued. Portal and hepatic venous pressure tracings were displayed and analyzed independently. An example of portal occlusion was shown in Fig. 1A. A stretch of data on Ppv between 0.3 and 1.8 s after portal occlusion for 3 s was fitted to a single exponential and extrapolated back to time 0 (instant of occlusion). This extrapolated pressure was then used as Ppo. An example of hepatic venous occlusion tracing was shown in Fig. 1B, and a stretch of data (0.3–1.0 s) on Phv was fitted to a straight line and extrapolated back to time 0 (instant of occlusion). This extrapolated pressure was used as Phvo.
|
Estimation of Ppo and Phvo from double occlusion. We estimated Ppo and Phvo from the double occlusion tracing, as shown in Fig. 1C. Double occlusion was accomplished for 3 s by closing both of the inflow and outflow valves simultaneously. Ppv during double occlusion was analyzed to determine Ppo in the same manner as the portal occlusion maneuver. The stretch of Ppv data between 0.3 and 1.8 s was selected and fitted to a monoexponential. The fitted curve was extrapolated back to the time of occlusion, and the time 0 pressure was designated as the double occlusion-derived Ppo (Ppo-do). Likewise, data between 0.3 and 1.8 s on Phv tracing during double occlusion were fitted to an exponential and extrapolated back to the time of occlusion. This time 0 pressure was designated as the double occlusion-derived Phvo (Phvo-do).
Experimental protocol. Hepatic hemodynamic parameters were observed for at least 30 min after the start of perfusion until an isogravimetric state (no weight gain or loss) was obtained by adjusting flow rate and the height of the reservoir at a Phv of 0 to 1 cmH2O, and at a highest Q. After this baseline measurement, the flow rate was increased or decreased by 5 ml/min ranging from 5 to 40 ml/min with keeping Phv constant at 0 to1 cmH2O. At each steady state of a given flow rate, all three occlusion maneuvers of portal occlusion, hepatic venous occlusion, and double occlusion were performed at a random order. The flow rate was then returned to 25 ml/min for the 30% Hct group, 30 ml/min for the 20% Hct group, 35 ml/min for the 10% Hct group, and 40 ml/min for the 0% Hct group. After stabilization of vascular pressures, the effect of two doses of norepinephrine on the vascular resistance distribution was studied. Norepinephrine (Sigma) was infused continuously at the low dose of 1 µg/min into the portal vein until Ppv increased and stabilized. Under this steady state, portal, hepatic venous, and double occlusions were performed. The effect of the high dose of 10 µg/min was then examined in a similar manner.
Experiments in the present study were carried out in livers perfused with blood of 30%, 20%, 10%, and 0% Hct. In each preparation, the effect of only one fixed Hct was examined: the perfusate Hct was not changed throughout the experimental period, once the perfusion started at a given Hct. Blood flow rate was expressed as ml·min–1·10 g liver wt–1, and the blood flow groups were assigned to eight groups from 5 ml/min (2.5–7.5 ml·min–1·10 g liver wt–1) to 40 ml/min (37.5–42.5 ml·min–1·10 g liver wt–1) group.
For determination of hepatic segmental vascular resistances, the total portal-hepatic venous (Rt), and portal venous (Rpv), sinusoidal (Rsinus), and hepatic venous (Rhv) resistances were calculated as follows
 | (1) |
 | (2) |
 | (3) |
 | (4) |
For calculation of these segmental vascular resistances, we adopted the mean value of Ppv for 5 s before the initial vascular occlusion maneuver as the Ppv in the equations. The Ppv before the subsequent occlusion maneuver did not differ by 0.2 cmH2O from the Ppv of the initial occlusion maneuver.
Statistics. All results are expressed as the means ± SD, unless mentioned otherwise. Comparisons of a given variable between the groups were performed using analysis of variance, followed by Bonferroni's test. A P value <0.05 was considered significant. For the correlation between Ppo and Ppo-do or between Phvo and Phvo-do, least-square linear regression analysis was used. Statistical significance of correlation of the linear regression was tested with analysis of variance. A paired Student's t-test was used to compare the mean pressures obtained with the single occlusion method and the double occlusion method.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Within each Hct group, when the flow rate increased at a constant outflow pressure of Phv, Ppv and Phvo increased, whereas Ppo did not changed significantly, as shown in Fig. 2, left. This results in the flow-dependent increases in both Ppv- to-Ppo and Phvo-to-Phv gradients, and the flow-dependent decreases in the Ppo-to-Phvo gradients. On the basis of these pressure gradient changes, Rsinus decreased, and either Rpv or Rhv did not change when the blood flow increased within each Hct group, as shown in the right panel of Fig. 2. Actually, Rsinus at 30% Hct comprised 60 ± 12% of Rt in the minimal flow group of 5 ml/min, 41 ± 11% in 10 ml/min, 36 ± 11% in 20 ml/min, 28 ± 8% in 15 ml/min, and 25 ± 6% in 25 ml/min. An interesting finding was that the sensitivity of the inflow pressure of Ppv to changes in flow rate was low: in the 30% Hct groups, only a twofold increase in Ppv from 6 to 12 cmH2O was observed as blood flow was increased fivefold from 5 to 25 ml/min. Representative recordings of hepatic vascular pressures in various flow rates of a liver perfused with 30% Hct were shown in Fig. 3. Ppv and Phv in this figure were not recorded at same time, but recorded during portal occlusion and during hepatic venous occlusion, respectively, and were compositely shown in the same frame at each blood flow rate.
|
Effects of Hct on hepatic vascular pressures and resistances. Figure 4 shows the hepatic vascular pressures (left panel) and segmental vascular resistances (right panel), as a function of perfusate Hct. At all blood flow rates studied, Ppv, Ppo, and Phvo were significantly greater in 30% Hct groups than those in 20% Hct groups. Thus increases in Hct from 20% to 30% resulted in significant increases in all three segmental resistances of Rpv, Rsinus, and Rhv at blood flow >15 ml/min. However, at the low blood flow of 5 and 10 ml/min, only Rsinus did not change significantly whereas Rpv and Rhv significantly increased as Hct increased from 20% to 30%. All three segmental vascular resistances of any blood flow became significantly greater at 30% Hct than at 10% Hct and 0%. Figure 5 shows representative recordings of livers perfused with blood of different Hct at the same blood flow of 25 ml/min. Ppv and Phv in this figure were not recorded at same time but were compositely shown in the same frame at each blood Hct, as shown in Fig. 3.
|
|
Effects of norepinephrine infusion on the hepatic vascular resistance distribution at various Hct and flow rates. Figure 6 shows the summary of the hepatic segmental vascular resistances in response to norepinephrine at the four levels of Hct. Although the basal levels of segmental vascular resistances differed depending on Hct, the trends of the response of segmental vessels to norepinephrine were similar; Rhv did not change significantly, whereas Rpv and Rsinus increased dose and Hct dependently. Actually, an infusion of 1 µg/min norepinephrine caused a 1.5- to 1.7-fold increase in Rpv and a 1.7- to 2.4-fold increase in Rsinus and caused no significant changes in Rhv among any Hct groups studied. Furthermore, the higher dose of 10 µg/min norepinephrine caused a 2.5-fold increase in Rpv and a 3.1- to 4.4-fold increase in Rsinus and again caused no significant changes in Rhv.
|
Estimation of Ppo and Phvo by the double occlusion maneuver. Figure 1 shows a representative recording during portal occlusion, hepatic venous occlusion, and double occlusion in an isolated rat liver perfused with blood of 30% Hct at 25 ml/min. Ppo and Phvo obtained via portal occlusion and hepatic venous occlusion, respectively, were almost identical with the pressures obtained by analysis of the tracings of Ppv and Phv during the double occlusion maneuver. The agreement between the two methods was tested using all data obtained from both normal livers and livers with vasoconstriction induced by norepinephrine, where Ppv ranged from 2.25 to 36.58 cmH2O. Actually 235 paired measurements of Phvo and Phvo-do, and 225 paired measurements of Ppo and Ppo-do were analyzed. The agreement between these two methods was evaluated in two different ways. First, the mean values of each pressure with the two methods were compared using a simple Student's t-test. There were no significant differences between Ppo and Ppo-do, and between Phvo and Phvo-do under all conditions studied. The agreement between the two methods was further tested by using regression analysis. Regression line equations for Ppo versus Ppo-do, and for Phvo versus Phvo-do are given in Fig. 7, A and B, respectively. There were strong positive relationships for Ppo and Ppo-do (r > 0.99). The slope for Ppo (1.009 ± 0.009, P = 0.99) and the y-intercept (–0.021 ± 0.042 cmH2O, P = 0.612), statistically, did not differ from the median slope and zero, respectively. Likewise, concerning the regression lines for Phvo versus Phvo-do, the slope for Phvo (0.996 ± 0.013, P = 0.62) and the y-intercept (0.009 ± 0.031 cmH2O, P = 0.775), statistically, did not differ from the median slope and zero, respectively (Fig. 7).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One of the major findings of the present study was that Rpv comprises 44% of Rt, and Rsinus 37%, and Rhv 19% in livers perfused with 30% Hct at physiological Ppv of 9.1 cmH2O. This finding indicates that almost one-half of Rt occurs in the presinusoidal vessels, one-third in the hepatic sinusoids, and only one-fifth of Rt in the postsinusoidal vessels in rat livers. These results on hepatic vascular resistance distribution are consistent with the findings from the hepatic micropuncture study (3, 16). Bohlen et al. (3), by puncturing surface hepatic venules (10–30 µm diameter), into which two adjacent acini drained, with servo-null micropipettes, reported that hepatic venule pressure was 5.1 ± 1.0 mmHg, and Ppv and vena caval pressure averaged 8.0 ± 1.4 and 3.4 ± 0.9 mmHg, respectively, indicating that the hepatic venule to vena caval pressure gradient comprised as small as 37% of the total Ppv-to-vena caval pressure gradient. Maass-Moreno and Rothe (16) demonstrated much more definitively that the Ppv-to-portal venule gradient comprises 53% of the total Ppv-to-Phv gradient, the portal venule-to-hepatic venule gradient, which may correspond to Rsinus of the present study, 25%, and the hepatic venule-to-Phv gradient, which may correspond to Rhv, 22%.
Flow rates and hepatic segmental vascular resistances. It is well known that an increase in blood flow decreases vascular resistance in systemic circulation. The flow dependence of vascular resistance was also observed in isolated perfused rat livers of the present study. As shown in Fig. 3, when the flow rate at any given Hct was increased at a constant Phv, Rt decreased. This decrease in Rt was ascribed exclusively to a decrease in Rsinus but not in Rhv or Rpv. The mechanism for this decrease in Rsinus may be mainly due to microvascular distension and recruitment.
An interesting finding was that the sensitivity of inflow pressure of Ppv to changes in flow rate is low: in the 30% Hct groups, only a twofold increase in Ppv from 6 to 12 cmH2O was observed as the flow is increased fivefold from 5 to 25 ml/min. This small increase in Ppv, hence Rt, in response to increased blood flow is ascribed to minimal changes in the pressure gradient of the sinusoids, that is Ppo-to-Phvo gradient. This indicates that hepatic sinusoids could accept more blood flow without greatly elevating Ppv probably through sinusoidal distension and recruitment.
Hct changes and hepatic segmental vascular resistances. In the present study, an increase in Hct from 0% to 30% resulted in a significant increase in Rt. This increase in Rt seems to be due to increases in all three segments of Rpv, Rsinus, and Rhv, as shown in Fig. 4. However, as perfusate Hct increased from 20% to 30% at low flow rate of 5 and 10 ml/min, Rpv and Rhv significantly increased, whereas Rsinus did not change significantly. This lack of change in Rsinus may be ascribed to the Fahraeus effect, where Fahraeus and Lindqvist (6) first demonstrated that, as tube diameter is reduced <300 µm, the apparent viscosity of red blood cell suspensions decreases as a result of an actual decrease in small tube Hct. In the systemic microcirculation, in vivo studies indicate that the Hct starts decreasing in vessels with diameters <70 µm and reaches a value in the capillaries of <25% of the systemic Hct (14). If this finding of the systemic circulation could be applied to the hepatic circulation, the changes in the perfusate Hct, hence the perfusate viscosity, would be so small in hepatic sinusoids due to the Fahraeus effect, compared with the large portal and hepatic veins, then the sinusoidal resistance may not change significantly.
In contrast, at blood flow >15 ml/min, Rsinus also increased as perfusate Hct increased from 20% to 30%. This seems to be contradictory to the Fahraeus effect. A possible explanation may be related to an increase in sinusoidal Hct due to rapid transcapillary fluid shifts, which could be caused by an increase in blood flow, hence the microvascular pressures (2). At a high blood flow rate, the microvascular pressures between Ppo and Phvo are high enough to cause enhanced transvascular cell-free perfusate filtration at the sinusoids, whose endothelium is not continuous and extremely leaky (8). During passage of blood through the sinusoids at high sinusoidal pressure, the apparent Hct might have increased due to enhanced transsinusoidal filtration. This possible increasing Hct effect might have counteract the decreasing Hct effect of the Fahraeus effect, resulting in an increase in Rsinus at high flow rate. However, we did not measure the difference in Hct between inflow and outflow blood. Further careful study is required in this respect.
Effect of norepinephrine on hepatic segmental vascular resistances. The present study shows that norepinephrine increased primarily Rpv and, to the lesser extent, Rsinus, whereas it did not affect Rhv. The absence of vasoreactivity to norepinephrine in the large hepatic veins is also reported in in vivo rats. Bohlen et al. (3) showed that an infusion of norepinephrine 2.9 ± 1.3 µg·min–1·kg–1 caused a significant increase in Ppv by 2.96 ± 1.00 mmHg but no change in hepatic venule pressure (–0.22 ± 1.26 mmHg) with the decrease in portal flow rate 88% of the control. This finding suggests that the large hepatic veins distal to the hepatic venules do not respond to norepinephrine, and this is consistent with the present study. However, their subsequent study revealed that norepinephrine caused significant increases in the hepatic venule pressure and the resistance between the sinusoids and the vena cava in rabbits (22). The discrepancy might be ascribed to the species difference between rat and rabbit.
An interesting finding of the present study is that norepinephrine increased significantly the sinusoidal resistance. This suggests that norepinephrine causes sinusoidal constriction. Reilly et al. (20) reported that the small but significant decrease in sinusoidal diameter was observed when 
-adrenergic receptors were stimulated. Although sinusoids do not contain smooth muscle cells, hepatic stellate cells, which are contractile and located around the sinusoidal endothelial cells, might reduce the diameter of sinusoids if it could contract in response to norepinephrine (25). However, there is no evidence that norepinephrine or -adrenergic agonist contracts stellate cells (21). Zhang et al. (28), using intravital microscopy, measured the diameter of sinusoids in isolated perfused rat livers. They found no changes in sinusoidal hemodynamics in response to phenylephrine, -adrenergic agonist, although the portal pressure increases. In addition, although activated stellate cells clearly exhibit enhanced contractility, the degree of contractility of normal stellate cells remains controversial (5, 21). Another possible mechanism for the norepinephrine-induced increase in Rsinus may be due to contraction of vascular smooth muscle. Anatomic studies show that in rat livers, presinusoidal vessels of preterminal portal venules as small as 40 µm diameter contain a significant amount of smooth muscle (5). Thus the middle segment in the present study might contain the portal venules.
Estimation of Ppo and Phvo by double occlusion maneuver. We have shown that Ppo and Phvo obtained via the single-occlusion technique of portal occlusion and hepatic venous occlusion, respectively, were identical with Ppo-do and Phvo-do that were obtained by the double occlusion maneuver. The agreement between the two methods was verified in the wide range of Ppv from 2.25 to 36.58 cmH2O. There were strong positive relationships between Ppo and Ppo-do, and between Phvo and Phvo-do. More importantly, the slope and y-intercept of their regression lines, statistically, did not differ from the median slope and zero, respectively.
The shortcoming of portal occlusion and hepatic venous occlusion techniques is that measurement should be done during steady state to obtain the segmental vascular resistance distribution because each technique cannot be done simultaneously. When two occlusions are being performed, it is necessary to wait for the pressure to return to a steady value before the next occlusion can be performed. This may cause problems because vascular constriction is not always stable. This problem is totally solved by one double occlusion technique. The double occlusion technique makes it possible to measure Ppo and Phvo during an unsteady state, when hepatic vascular pressures are changing in response to experimental maneuvers such as a bolus injection of vasoactive substances.
Limitation of present study. There are limitations of the current experiment. The first is related to the lack of hepatic artery flow and a good source of oxygen, especially in the groups with a Hct of 0% and 10%. However, even in the perfused livers with 0% Hct, perfusate PO2 could attain as high as 300 mmHg by bubbling the perfusate with 95% O2-5% CO2, as revealed by our previous study (26). Another criticism is that the tying off of the hepatic artery could be influencing the compliance and resistance characteristics of hepatic circulation, especially the middle vascular segment. However, we do not think that the ligation of hepatic artery influenced critically determination of Ppo and Phvo. An anatomic study revealed that the hepatic artery does not directly supply the sinusoids and that its primary vascular beds are formed not in the parenchymal but the stromal compartment, such as peribiliary vascular plexus, portal tract interstitium, portal vein vasocasorum, hepatic capsule, and central-sublobular-hepatic vaso vasorum (5). The final confluence of the hepatic artery occurs at the first third of the sinusoids and terminal portal venules (27). We assume that the ligated hepatic artery should be collapsed with closed ends because of absence of blood flow, and therefore they could not influence the vascular resistance of the overall portal to hepatic veins. The ligated hepatic arterial end might serve as a capacitor of the sinusoids and possibly affect the compliance of the sinusoids, only when the sinusoidal pressure could exceed the pressure high enough to open this collapsed flow-deficient blind sac. However, even if the compliance of the sinusoids could increase, these might not affect measurement of Ppo and Phvo because these pressures are primarily determined by the characteristics of the noncompliant portal and hepatic venous segments, in which the hepatic artery does not substantially terminate.
Finally, the vascular occlusion technique estimates microvascular pressure in relationship to a point of high vascular compliance (10). The boundary of the middle vascular segment could not be known anatomically in the present study. However, the literature about the microstructure of the liver and Greenway and Lautt (8) showed that the sinusoids comprise the majority of the vasculature of the liver and that sinusoids, not the small portal and hepatic veins, are the locus of the high compliance. Thus we believe that the compliant middle vascular segment of the present study corresponds to the sinusoids. In contrast, the micropuncture technique provides a direct measurement of vascular pressures within anatomically defined vessels although it is limited to the measurement of the surface vessels. In isolated lungs, micropuncture and occlusion technique was applied in the same preparations. It is reported that the inflow occlusion pressure or pulmonary arterial occlusion pressure was slightly higher than the micropipette pressure of 50–80 µm diameter arterioles and that the venous occlusion pressure was slightly lower than the micropipette pressure of 20–50 µm diameter venules (7, 9). Further study is required in which micropuncture and vascular occlusion techniques are applied simultaneously in the same liver preparation to identify the hepatic vessels corresponding to vascular occlusion pressures.
In summary, we provided the hepatic vascular occlusion methods in isolated perfused rat livers to measure Ppo and Phvo, which enabled us to assign the portal hepatic Rt to the portal Rpv, Rsinus, and Rhv. It was demonstrated that Rpv comprises 44% of Rt, 37% Rsinus, and 19% Rhv in livers perfused at physiological Ppv and 30% Hct. We determined the effect of changes in blood Hct or blood flow rate on segmental vascular resistance distribution. As Hct increased at a given blood flow, all three segmental vascular resistances of Rpv, Rsinus, and Rhv increased at flow >15 ml/min. Because blood flow increased at a given Hct, either Rpv or Rhv did not change, but only Rsinus decreased presumably due to distension or recruitment of sinusoids. We also determined the preferential vasoconstrictive site induced by norepinephrine. Norepinephrine increased predominantly Rpv over Rsinus, but it did not affect Rhv. Finally, we demonstrated that Ppo and Phvo can be obtained by the double occlusion method in isolated perfused rat livers.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by Collaborative Research Grant C2003-1 from Kanazawa Medical University and Grant-in-Aid No. 15591665 for Scientific Research from the Ministry of Education, Culture, Sports, Sciences and Technology of Japan.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
