Active and passive liver microvascular responses from angiotensin, endothelin, norepinephrine, and vasopressin.

Vasoconstrictor agents may induce a decrease in hepatic vascular volume passively, by decreasing distending pressure, or actively, by stimulating contractile elements of capacitance vessels. Hepatic venular resistance was estimated in anesthetized rabbits from hepatic venular pressure (P(mu hv); by servo-null micropipette), inferior vena cava pressure, and total hepatic blood flow (F(hv); by ultrasound flow probe). Changes in liver volume were estimated from measures of liver lobe thickness. Angiotensin (ANG) II, endothelin (ET)-1, norepinephrine (NE), and vasopressin (VP) were infused into the portal vein at a constant rate for 5 min. We conclude that ANG II and NE induced active constriction of hepatic capacitance vessels, because the liver lobe thickness decreased significantly even though P(mu hv) and portal venous distending pressure (P(pv)) increased. All four agents increased splanchnic and hepatic venous resistances in similar proportions. With VP, P(mu hv) and P(pv) decreased, but with ET-1, P(mu hv) and P(pv) increased. However, lobe thickness was not significantly changed by either drug during the infusion compared with the 2-min control period. Thus VP and ET-1 have only minor effects on hepatic capacitance vessels. ET-1, at 0.04 microg. min(-1). kg body wt(-1), caused an increase in systemic arterial blood pressure, but erythrocyte movement through the sinusoids in some animals stopped.

Hepatic venular resistance responses to norepinephrine, isoproterenol, adenosine, histamine, and ACh in rabbits.

Changes in hepatic venous resistance were estimated in rabbits from the hepatic venular-inferior vena caval pressure gradient [servo-null micropipettes in 49 +/- 15 (SD) microns vessels] and the total hepatic blood flow (ultrasound probe encircling the hepatic artery and the portal vein). Changes in liver volume, and thus vascular capacitance, were estimated from measures of the liver lobe thickness. Norepinephrine (NE), isoproterenol (Iso), adenosine (Ado), histamine (Hist), or acetylcholine (ACh) was infused into the portal vein at a constant rate for 5 min. NE, Hist, and Ado increased hepatic venular pressure, but only NE and Hist significantly increased hepatic venular resistance. NE reduced the liver thickness, but Hist and Ado caused engorgement. Hepatic blood flow was increased by NE and Ado and decreased by ACh. The influence of intraportal vein infusion of Iso on the liver vasculature, at doses similar to that of NE, was insignificant. We conclude that NE acted on all the hepatic microvasculature, increasing resistance and actively decreasing vascular volume. Hist passively induced engorgement by increasing outflow resistance, whereas the liver engorgement seen with Ado was passively related to the increased blood flow. ACh constricted the portal venules but did not change the liver volume.

Distribution of pressure gradients along hepatic vasculature.

To test the hypothesis that the hepatic sinusoidal pressure is virtually identical to the portal venous pressure (Ppv) or the abdominal vena caval pressure (Pave), microvascular pressures were measured in liver vascular networks supplied by a single portal venule near the liver surface of alpha-chloralose-urethan-anesthetized rabbits. With the use of a servo-null pressure-measuring system, the pressures in terminal portal venules, sinusoids, and initial hepatic venules averaged 5.7 +/- 0.8 (SD), 5.4 +/- 0.7, and 4.7 +/- 0.6 mmHg, respectively, relative to the middle of the right atrium. The mean Ppv was 7.7 +/- 1.1 mmHg and the Pave was 3.9 +/- 0.7 mmHg. The sinusoidal pressure (Psinu) averaged 2.33 +/- 1.04 mmHg lower than the Ppv, and 1.54 +/- 0.63 mmHg higher than the Pave. The pressure gradient across the sinusoids averaged < 1 mmHg. The fractional pressure gradient from the sinusoids to the vena cava averaged 41 +/- 15% of the total Ppv to Pave gradient. We conclude that most of the transhepatic resistance cannot be attributed to a specific vascular segment or vessel type and that sinusoidal pressures are not closely similar to either the Ppv or the Pave, as is often assumed.

Assessing microvascular volume change and filtration from venous hematocrit variation of canine liver and lung.

The volume increase of canine liver after 1 min of a 10 mmHg elevation in hepatic venous pressure has been reported as 251 ml/kg tissue. An analysis of the transient hematocrit variation in hepatic venous blood indicated that 16% of the volume change results from transcapillary filtration, 72% from microvascular expansion, and 12% from macrovascular expansion. In the analysis, we first used the temporal change of the liver volume to determine the time course of the filtration and microvascular and macrovascular volume change. We next deduced, for a permeable microcirculation with a microvascular hematocrit lower than the feed hematocrit (the Fahraeus effect), how the filtration and microvascular volume change (MVC) produce a hematocrit variation in the blood leaving microcirculation. By accounting for the dispersion of the blood flow, the analysis predicted a hematocrit variation in the hepatic venous blood that matched well with the measured variation over the 1-min course of experiment. A reasonable fit with the hematocrit variation of pulmonary blood also was obtained for experiment with an 8 mm/Hg increase in the arterial and venous pressure perfusing the canine left lower lung lobe. The tissue and vascular volume increase at 1 min was 149 ml/kg tissue with 4% as a result of filtration, 41% as a result of microvascular expansion, and 55% as a result of macrovascular expansion. The large MVCs from the hepatic and pulmonary circulation indicate their microcirculations function as a reservoir in controlling blood volume redistribution.

Nonlinear resistances in hepatic microcirculation.

The liver provides a reservoir available for mobilizing large amounts of blood, but if a change in downstream (outflow) pressure below a certain magnitude (break pressure) does not change upstream pressures, blood volume redistribution may be limited. For downstream pressures larger than the break pressure, the upstream pressures change proportionately. We tested the hypothesis that this nonlinear mode of pressure transmission could be found from the abdominal vena cava to the hepatic microcirculation and from the hepatic microcirculation to the portal vein. Using a servo-null micropipette technique, we measured microvascular pressures at the liver surface of rabbits. In 16 of 30 measurements, increasing the pressure at the liver outflow, by partially occluding the caudal thoracic vena cava, caused an increase in hepatic venular pressure only after the abdominal vena caval pressure exceeded a break pressure of 2.85 +/- 0.92 mmHg. In 13 of 31 measurements, portal venous pressure was not changed until the hepatic venular pressure exceeded a break pressure of 3.36 +/- 0.54 mmHg. Similar behavior and values were obtained for sinusoids and portal venules. When present, the sharp inflection in the upstream-downstream pressure plots suggests that this may be caused by a Starling resistor-type mechanism. When the break was absent, the downstream pressure may have been larger than the break pressure. We conclude that significant hepatic resistances with nonlinear characteristics exist upstream and downstream to the central venules, sinusoids, and portal venules.

Gastrointestinal hemodynamics during compensation for hemorrhage and measurement of Pmcf.

To quantify the degree of autonomic reflex control of the gastrointestinal vasculature, we studied the responses to a 10-ml/kg hemorrhage or transfusion and autonomic blockade in fentanyl- and pentobarbital-anesthetized dogs. The active total blood volume was estimated by indocyanine green dilution. Transfusion and hemorrhage did not significantly change gastrointestinal vascular compliance [1.82 +/- 0.68 (SD) ml/mmHg], but autonomic blockade with hexamethonium and atropine increased it by 0.57 +/- 0.37 ml/mmHg. Neither hemorrhage nor autonomic blockade significantly changed gastrointestinal vascular resistance from its control value of 10.8 +/- 4 mmHg.ml-1.min.kg body wt, but transfusion reduced it by 3.0 +/- 1.2 mmHg.ml-1.min.kg body wt. The ratio of gastrointestinal vascular resistance to total peripheral resistance was not significantly changed, however. We conclude that vascular compliance and resistance of the gastrointestinal bed are minimally influenced by the autonomic nervous system under the conditions studied. Portal pressure and flow measurements (transit-time ultrasound) during the above maneuvers were also combined with estimations of mean circulatory filling pressure (Pmcf) to test the hypothesis that, when the heart is stopped to measure Pmcf, portal pressure equals central venous pressure (Pcv) and hence that portal flow is zero. Seven seconds after the heart was stopped, portal venous pressure (Ppv) remained 0.83 +/- 0.78 mmHg higher than Pcv and portal flow decreased to only 25% of its control value. However, gastrointestinal compliance times (Ppv-Pcv), an estimate of the extra distending volume, was only 0.07 +/- 0.07 ml/kg body wt. Thus we conclude that the error in estimating Pmcf, given this (Ppv-Pcv) difference, is physiologically insignificant.

Mean circulatory filling pressure: its meaning and measurement.

The volume-pressure relationship of the vasculature of the body as a whole, its vascular capacitance, requires a measurement of the mean circulatory filling pressure (Pmcf). A change in vascular capacitance induced by reflexes, hormones, or drugs has physiological consequences similar to a rapid change in blood volume and thus strongly influences cardiac output. The Pmcf is defined as the mean vascular pressure that exists after a stop in cardiac output and redistribution of blood, so that all pressures are the same throughout the system. The Pmcf is thus related to the fullness of the circulatory system. A change in Pmcf provides a uniquely useful index of a change in overall venous smooth muscle tone if the blood volume is not concomitantly changed. The Pmcf also provides an estimate of the distending pressure in the small veins and venules, which contain most of the blood in the body and comprise most of the vascular compliance. Thus the Pmcf, which is normally independent of the magnitude of the cardiac output, provides an estimate of the upstream pressure that determines the rate of flow returning to the heart.

Ultrasonic crystal measurement of blood volume changes in liver and spleen.

Two methods of measurement of splenic and hepatic vascular capacitance responses were compared in cats anesthetized with pentobarbital. For the spleen, measurements using ultrasonic crystals were compared with recordings of splenic weight. The cube of the relative change in thickness at one point on the longitudinal axis of the spleen was closely correlated with the overall change in spleen weight. For the liver, measurements by ultrasonic crystals were compared with plethysmographic recordings of liver volume. Individual measurements of cubed relative thickness did not correlate well with plethysmographic volume, and two sets of crystals on the same liver lobe gave different estimates of relative liver volume. However, averaged measurements for a group of animals showed similar means by the two methods. We conclude that ultrasonic crystals, when appropriately attached, can reliably monitor changes in splenic volume, but that for the liver, the variability in thickness responses at different sites on different livers is high. Unless a large number of observations are made, the results will be unreliable.

Contribution of the large hepatic veins to postsinusoidal vascular resistance.

We tested the hypothesis that the larger (greater than 2 mm ID) hepatic veins are the primary site of the portal-to-caval venous pressure gradient in the dog. Double-lumen catheters were inserted through the caval wall into hepatic veins of pentobarbital sodium-anesthetized dogs. One lumen opened at the end, and the other to the side. Each catheter was advanced until stopped and then it was withdrawn. The pressure at either port dropped from 87 +/- 31 to 13 +/- 11% of the portal-to-caval pressure difference as each moved past a transition point (TP). The location of the TP depended on the catheter diameter. Intraportal histamine or norepinephrine, 4 and 2.6 micrograms.min-1.kg body wt-1 respectively, augmented only the pressure measured upstream to the TP. A mathematical model of flow through a vessel with a catheter inside predicted a marked increase in resistance when the ratio of catheter OD to vessel ID exceeded approximately 0.6. Autopsy revealed ratios greater than 0.6 upstream to the TP. A hydraulic model confirmed that this effect caused the appearance of the TP. Given the depth (11.7 cm) at which near caval pressures could be found, even during histamine administration, we conclude that the major pressure gradients in the canine liver must lie upstream to the large hepatic veins.

Hepatic venular pressures of rats, dogs, and rabbits.

We tested the hypotheses that the hepatic venule pressures (Phv), just downstream from the hepatic sinusoids, are closely similar (less than 2 mmHg) either to the portal venous pressure (Ppv), indicating a high hepatic venous resistance, or to the inferior vena cava (Pivc) pressure, indicating a high portal-sinusoidal venous resistance, as reported by previous investigators. A micropipette servo-null pressure measurement technique was used with rats, dogs, and rabbits. Phv, referred to the anatomic level of the vena cava, averaged 5.1 +/- 1.0, 6.4 +/- 1.1, and 5.4 +/- 1.0 (SD) mmHg in the rats, puppies, and rabbits, respectively. Ppv averaged 8.0 +/- 1.4, 10.8 +/- 2.2, and 7.4 +/- 1.5 mmHg, respectively. Norepinephrine infusion into the portal vein (1-5 micrograms.min-1.kg-1) caused Ppv to increase and the portal venous flow to decrease but did not significantly affect Phv. The hepatic venous circuit contributed 44 +/- 17% (rats) and 31 +/- 26% (dogs) of the total liver venous vascular resistance under control conditions. We conclude that the portal and sinusoidal vasculatures are the dominant, but not exclusive, resistance sites of the liver venous vasculature both at rest and during norepinephrine-induced vasoconstriction.

Carotid baroreceptor control of liver and spleen volume in cats.

We tested the hypothesis that the blood volumes of the spleen and liver of cats are reflexly controlled by the carotid sinus (CS) baroreceptors. In pentobarbital-anesthetized cats the CS area was isolated and perfused so that intracarotid pressure (Pcs) could be controlled while maintaining a normal brain blood perfusion. The volume changes of the liver and spleen were estimated by measuring their thickness using ultrasonic techniques. Cardiac output, systemic arterial blood pressure (Psa), central venous pressure, central blood volume, total peripheral resistance, and heart rate were also measured. In vagotomized cats, increasing Pcs by 100 mmHg caused a significant reduction in Psa (-67.8%), cardiac output (-26.6%), total peripheral resistance (-49.5%), and heart rate (-15%) and significantly increased spleen volume (9.7%, corresponding to a 2.1 +/- 0.5 mm increase in thickness). The liver volume decreased, but only by 1.6% (0.6 +/- 0.2 mm decrease in thickness), a change opposite that observed in the spleen. The changes in cardiovascular variables and in spleen volume suggest that the animals had functioning reflexes. These results indicate that in pentobarbital-anesthetized cats the carotid baroreceptors affect the volume of the spleen but not the liver and suggest that, although the spleen has an active role in the control of arterial blood pressure in the cat, the liver does not.

Effects of hypercapnia and hypoxia on the cardiovascular system: vascular capacitance and aortic chemoreceptors.

Aortic chemoreceptor influences on vascular capacitance after changes in blood carbon dioxide and oxygen were studied in mongrel dogs anesthetized with methoxyflurane and nitrous oxide. The mean circulatory filling pressure (Pmcf), measured during transient cardiac fibrillation, provided a measure of capacitance vessel tone. Hypercapnia, hypoxia, and hypoxic hypercapnia significantly increased most variables, except that hypercapnia caused the total peripheral resistance (TPR) to decrease. Hypocapnia caused a significant decrease in mean systemic (Psa) and pulmonary (Ppa) arterial blood pressures, cardiac output (CO), and central blood volume and an increase in TPR and heart rate. The changes in Pmcf on changing blood gas tensions could be described by the equation delta Pmcf = -1.60 + 0.036 (arterial PCO2) + 50.8/arterial PO2. Thus a 10 mmHg increase in arterial PCO2 caused a 0.36 mmHg increase in Pmcf with receptors intact. Cold block (2 degrees C) of the cervical vagosympathetic trunks did not significantly influence the measured variables at control. During severe hypercapnia, vagal cooling caused a small but significant decrease in Pmcf, Psa, Ppa, and CO but not TPR. During hypoxia, vagal cooling caused the Pmcf, Psa, and TPR to decrease. We conclude that although hypercapnia or hypoxia acts reflexly to increase the capacitance vessel tone (an increase in Pmcf), the aortic and cardiopulmonary chemoreceptors with afferents in the vagi have only a small influence on the capacitance system, accounting for only approximately 25% of the total body response.

Role of beta-adrenergic agonists in the control of vascular capacitance.

The role of beta-adrenergic agonists, such as isoproterenol, on vascular capacitance is unclear. Some investigators have suggested that isoproterenol causes a net transfer of blood to the chest from the splanchnic bed. We tested this hypothesis in dogs by measuring liver thickness, cardiac output, cardiopulmonary blood volume, mean circulatory filling pressure, portal venous, central venous, pulmonary arterial, and systemic arterial pressures while infusing norepinephrine (2.6 micrograms.min-1.kg-1), or isoproterenol (2.0 micrograms.min-1.kg-1), or histamine (4 micrograms.min-1.kg-1), or a combination of histamine and isoproterenol. Norepinephrine (an alpha- and beta 1-adrenergic agonist) decreased hepatic thickness and increased mean circulatory filling pressure, cardiac output, cardiopulmonary blood volume, total peripheral resistance, and systemic arterial and portal pressures. Isoproterenol increased cardiac output and decreased total peripheral resistance, but it had little effect on liver thickness or mean circulatory filling pressure and did not increase the cardiopulmonary blood volume or central venous pressure. Histamine caused a marked increase in portal pressure and liver thickness and decreased cardiac output, but it had little effect on the estimated mean circulatory filling pressure. Isoproterenol during histamine infusions reduced histamine-induced portal hypertension, reduced liver size, and increased cardiac output. We conclude that the beta-adrenergic agonist, isoproterenol, has little influence on vascular capacitance or liver volume of dogs, unless the hepatic outflow resistance is elevated by agents such as histamine.

Reflex control of vascular capacitance during hypoxia, hypercapnia, or hypoxic hypercapnia.

We tested the hypothesis that the changes in venous tone induced by changes in arterial blood oxygen or carbon dioxide require intact cardiovascular reflexes. Mongrel dogs were anesthetized with sodium pentobarbital and paralyzed with veruronium bromide. Cardiac output and central blood volume were measured by indocyanine green dilution. Mean circulatory filling pressure, an index of venous tone at constant blood volume, was estimated from the central venous pressure during transient electrical fibrillation of the heart. With intact reflexes, hypoxia (arterial PaO2 = 38 mmHg), hypercapnia (PaCO2 = 72 mmHg), or hypoxic hypercapnia (PaO2 = 41; PaCO2 = 69 mmHg) (1 mmHg = 133.32 Pa) significantly increased the mean circulatory filling pressure and cardiac output. Hypoxia, but not normoxic hypercapnia, increased the mean systemic arterial pressure and maintained the control level of total peripheral resistance. With reflexes blocked with hexamethonium and atropine, systemic arterial pressure supported with a constant infusion of norepinephrine, and the mean circulatory filling pressure restored toward control with 5 mL/kg blood, each experimental gas mixture caused a decrease in total peripheral resistance and arterial pressure, while the mean circulatory filling pressure and cardiac output were unchanged or increased slightly. We conclude that hypoxia, hypercapnia, and hypoxic hypercapnia have little direct influence on vascular capacitance, but with reflexes intact, there is a significant reflex increase in mean circulatory filling pressure.

Autoregulation of cardiac output by passive elastic characteristics of the vascular capacitance system.

After a change in cardiac output, the magnitude of potential blood volume redistribution was investigated in 10 dogs anesthetized with chloralose. All of the venous return was pumped into a reservoir, using servocontrolled pumps to maintain fixed superior and inferior vena cava pressures. The cardiac output was set at various levels by pumping from the reservoir into the right atrium. Changes in reservoir volume were assumed to reflect the changes in vascular blood volume. After measuring the control responses, cardiovascular reflexes were blocked with hexamethonium. Reducing the cardiac output, for example, from 110 to 80 ml/(min.kg) with reflexes intact, caused a 9.2-ml/kg transfer of blood from the dog to the reservoir. With reflexes blocked, the same change in cardiac output caused 6.8 ml/kg of the blood to be transferred. Under the control conditions, throughout the range of 50-140 ml/(min.kg), an increase or decrease of cardiac output of 1 ml/(min.kg) elicited a 0.304 +/- 0.086 (mean +/- SD) ml/kg change in dog blood volume; with reflexes blocked, the flow sensitivity was 0.239 +/- 0.062 ml/kg. Thus, only 21% of the total blood volume redistribution was attributable to active reflex responses. Deterioration of the preparation may have attenuated the magnitude of active reflex activity. Neither the systemic vascular compliance of 1.80 +/- 0.35 ml/mm Hg.kg nor the fraction of venous return from the superior vena cava of 26.5 +/- 4.6% was significantly changed by reflex blockade.(ABSTRACT TRUNCATED AT 250 WORDS)

Mean circulatory filling pressure: potential problems with measurement.

Three experimental series using 22 acutely splenectomized mongrel dogs were completed to 1) compare fibrillation (Fib) and acetylcholine (ACh) injection as methods to stop the heart for the mean circulatory filling pressure (Pmcf) maneuver, and 2) test whether Pmcf equals portal venous pressure 7 s after heart stoppage (Pportal7s). Blood volume changes of -10, -20, +10, or +20 ml/kg were imposed and Pmcf and Pportal measurements were obtained. Pportal7s and Pmcf were significantly different with volume depletion but were similar under control conditions. Pmcf with ACh and Pmcf with Fib were significantly different only after a volume change of -20 ml/kg. However, severe pulmonary congestion and atelectasis were detected in animals where Ach was used to stop the heart. In some cases (with injection directly into the pulmonary artery) the damage was severe enough to cause irreversible arterial hypoxia. Thus we conclude that the repeated use of ACh may exert a detrimental influence on pulmonary function, changing the physiological status of the experimental animal. Also, the central venous pressure at 7 s of heart stoppage (Pcv7s) is not a fully accurate estimate of the true mean circulatory filling pressure during the Pmcf maneuver, because Pcv7s did not equal the Pportal7s under all experimental conditions.

Vascular capacitance responses to hypercapnia of the vascularly isolated head.

Hypercapnic stimulation of the brain may account for some of the decrease in vascular capacitance (venoconstriction) seen with whole-body hypercapnia. Six mongrel dogs were anesthetized with alpha-chloralose and paralyzed with pancuronium bromide. The vagi were cut and the carotid bodies and sinuses were denervated. The head circulation was isolated and perfused with normoxic [arterial partial pressure of O2 (Pao2) = 112 mmHg], normocapnic (PaCO2 = 40 mmHg) blood, or one of three levels of normoxic, hypercapnic (PaCO2 = 56, 68, or 84 mmHg) blood. A membrane oxygenator was used to change gas tensions in the perfusate blood. The systemic circulation received normoxic, normocapnic blood (Pao2 = 107 mmHg; PaCO2 = 32 mmHg). Systemic arterial pressure increased from 111 to 134 mmHg, and heart rate decreased from 174 to 150 beats/min with a head blood PaCO2 of 84 mmHg. Central blood volume was not affected by head hypercapnia. Cardiac output significantly decreased only with a head blood PaCO2 of 84 mmHg. Mean circulatory filling pressure increased by 0.014 mmHg/1 mmHg increase in head PaCO2. The sensitivity of the total peripheral resistance to cephalic blood hypercapnia was 0.88%/mmHg, whereas that for the mean circulatory filling pressure was only 0.19%/mmHg. We conclude that stimulation of the brain, via perfusion of the head with hypercapnic blood, causes a small but significant increase in mean circulatory filling pressure, due to systemic venoconstriction.

Physiology of venous return. An unappreciated boost to the heart.

Adequate cardiovascular function depends on the control of venous tone as well as cardiac contractility, heart rate, vascular resistances, and an adequate blood volume. Venous tone is a major determinant of cardiac preload, a clinically important factor influencing cardiac function, especially during cardiac failure. In this review, vascular capacitance, venous tone, and venous return are discussed, and the concepts relating them to cardiovascular function are summarized. Active venoconstriction or dilation provides a rapid compensation, equivalent to a change in blood volume, for cardiovascular homeostasis.