Effect of Volume Expansion on Systemic Hemodynamics and Central and Arterial Blood Volume in Cirrhosis

Transcription

1 GASTROENTEROLOGY 1995;109: Effect of Volume Expansion on Systemic Hemodynamics and Central and Arterial Blood Volume in Cirrhosis SOREN MOLLER,* FLEMMING BENDTSEN,* and JENS H. HENRIKSEN* Departments of *Clinical Physiology and *Gastroenterology, Hvidovre Hospital, University of Copenhagen, Copenhagen, Denmark Background & Aims: Systemic vasodilatation in cirrhosis may lead to hemodynamic alterations with reduced effective blood volume and decreased arterial blood pressure. This study investigates the response of acute volume expansion on hemodynamics and regional blood volumes in patients with cirrhosis and in controls. Methods: Thirty-nine patients with cirrhosis (12 patients with Child-Turcotte class A, 14 with class B, and 13 with class C) and 6 controls were studied. During hepatic vein catheterization, cardiac output, systemic vascular resistance, central and arterial blood volume, noncentral blood volume, and arterial pressure were determined before and during a volume expansion induced by infusion of a hyperosmotic galactose solution. Results: During volume expansion, the central and arterial blood volume increased significantly in patients with class A and controls, whereas no significant change was found in patients with either class B or class C. Conversely, the noncentral blood volume increased in patients with class B and C. In both patients and controls, the cardiac output increased and the systemic vascular resistance decreased, whereas the mean arterial blood pressure did not change significantly. Conclusions: Only in mild cirrhosis is the effective blood volume able to increase in response to volume expansion. Our results are consistent with the peripheral vasodilatation hypothesis and the circulatory hyporeactivity occurring in advanced cirrhosis. B esides portal hypertension, patients with cirrhosis present a spectrum of circulatory disturbances. 1-3 These include increased cardiac output and heart rate and decreased systemic vascular resistance and arterial blood pressure. 4-6 According to the "peripheral arterial vasodilatation hypothesis, ''2'7-9 systemic vasodilatation leads to arterial underfilling. Arterial underfilling, together with sequestration of fluid into the peritoneal cavity, activates Saltwater-retaining mechanisms and vasopressor systems to counteract the otherwise very low arterial blood pressure Plasma and blood volumes are increased in the presence of cirrhosis, ~'<s'9'~1 but there are signs of abnormal distributiont's; it is now generally accepted that patients with decompensated cirrhosis and ascites have a decreased "effective blood volume. ''6-s'12'~'15 However, in early cirrhosis, the size of the central vascular compartment where volume receptors and baroreceptors are located remains controversial. We and others have recently produced evidence for a central and arterial blood volume contraction, <1<~7 whereas others question this) However, the effects of an acute volume expansion on the size of the different regions of the vascular compartment have not hitherto been investigated. The present study was undertaken to determine the response of an acute volume expansion on systemic hemodynamics and central and noncentral blood volumes in patients with cirrhosis and different degrees of liver dysfunction and in a control group. Patients Study Population and Methods The study population comprised 39 patients with a biopsy-verified cirrhosis; their mean age was 53 years (range, years). All the patients had a history of alcohol abuse, i.e., a consumption exceeding 50 g/day for more than 5 years. They had abstained from alcohol for at least 1 week before the study and had shown no signs of withdrawal symptoms at the time of the study. According to the modified Child-Turcotte classification, patients were in class A, 14 in class B, and 13 in class C. Twenty-two patients had ascites; they were receiving diuretics and had been put on a sodium restricted diet of 40 mmol/day. Eight patients received a diuretic dose of 100 mg of spironolactone, 7 patients received 40 mg of furosemide, and 7 patients received 80 mg of furosemide. Additional cardiovascular medication was not prescribed for any of the patients. The presence of slight or moderate ascites was confirmed by ultrasonography or paracentesis. The diet of patients without fluid retention was not restricted. None of the patients had hepatic encephalopathy above grade I or had experienced recent gastrointestinal bleeding. The serum concentrations of albumin; bilirubin; aspartate aminotransferase; alkaline phosphatase; coagulation factors II, VII, and X; creatinine; sodium; and potassium were determined by routine methods in an autoanalyzer (SMAC, Technicon Instruments 1995 by the American Gastroenterological Association /95/S3.00

2 1918 MgLLER ET AL. GASTROENTEROLOGY Vol. 109, No. 6 Table 1. Clinical and Biochemical Characteristics of Patients in the Different Child-Turcotte Classes Child-Turcotte class A Child-Turcotte class B Child-Turcotte class C (n = 12) (n = 14) (n = 13) Patient characteristics Height (cm) 172 ± 2 Weight (kg) 74.1 ± 4.7 Age (yr) 54.9 _+ 2.4 Ascites (yes/no) 0/12 Diuretic treatment (yes/no) 5/7 Blood biochemistry Serum aspartate aminotransferase (U/L; 10-40) 64 _+ 19 Serum bilirubin (#mol/l; 2-17) 16 ± 2 Serum alkaline phosphatases (U/L; ) 238 _+ 26 Plasma coagulation factors II, Vii, and X (U; ) 0.81 ± 0.07 Serum albumin (#mol/l; ) Serum creatinine (#mol/l; ) 93 ± 6 Serum sodium (mmel/l; ) 138 ± 1 Creatinine clearance (ml/min) 86 ± 12 Splanchnic hemodynamics Wedged hepatic venous pressure (ram Hg < 15) 18.1 _+ 1.7 Free hepatic venous pressure (mm Hg < 7) Hepatic venous pressure gradient (mm Hg < 5) Postsinusoidal resistance (dyne. s -I. cm 5) 784 ± 152 Hepatic blood flow (L/min; ) 1.11 ± 0.09 ICG clearance (L/min; ) 0, Galactose elimination capacity (mmol/min; female, 1.87 _ >1.4; male, >1.7) _ _ /4 12/1 11/3 12/ a,b _+ 30 a'b a,b 529 ± ± 12 a,o ± I a 129 _+ 1.5 a'b a'b 6.4 _ _ ± a'~ _+ 340 a _ ± a,b NOTE. Mean + SEM. Reference intervals in parentheses. Significance at differences between patient classes: adifferent from class A, P < 0.05; bdifferent from class B, P < Corp., Tarrytown, NY). Clinical and biochemical characteristics of the three Child-Turcotte classes are summarized in Table 1. A control group served 6 patients with relatively minor disorders (abdominal pain, fatty liver, and arterial hypertension) but without cirrhosis of liver. They underwent a hemodynamic investigation to exclude intestinal ischemia (n = 3), renovascular hypertension (n = 1), and portal hypertension (n = 2). None of these suspected diagnoses were confirmed during the present hemodynamic or other subsequent examinations. Patients and controls participated after giving their informed consent according to the Helsinki II declaration, and the study was approved by the Ethics Committee for Medical Research in Copenhagen. No complications or side effects were encountered during the study. Catheterization Splanchnic hemodynamics. All subjects underwent liver vein catheterization for diagnosis and assessment of the degree of portal venous hypertension. They were studied in the morning after an overnight fast and after at least 1 hour of resting supine. None of the patients received diuretics or other medication in the 24 hours preceding the investigations. Catheterization of hepatic veins and the right atrium was performed as previously described. 19 A Cournand or Swan-Ganz 7F catheter was guided under local anesthesia to the above locations via the femoral route under fluoroscopy. Pressures were measured by a capacitance transducer (Simonsen og Wed, Copenhagen, Denmark) in the wedged and free hepatic vein position in at least three different vessels with the midaxillary line being zero pressure level. Mean values of repeated measurements were used. Hepatic blood flow was determined by the indocyanine green constant infusion technique. 2 The indocyanine green clearance was measured as the infusion rate divided by the arterial plasma concentration of indocyanine green. 2 Systemic and central hemodynamics. A small indwelling polyethylene catheter was placed in the femoral artery and advanced to the aortic bifurcation by the Seldinger technique, and the systolic, diastolic, mean arterial, and right atrial blood pressures were measured directly. Cardiac output was measured by the indicator dilution technique. 21 Systemic vascular resistance expressed in (dyne" s)/cm 5 was assessed as 80" (Mean Arterial Pressure - Right Atrial Pressure)/Cardiac Output (pressures expressed in millimeters of mercury and cardiac output in liters per minute). Heart rate was determined by electrocardiography. The central and arterial blood volume (the blood volume in the heart cavities, lungs, and central arterial tree) was assessed according to the kinetic theory. = This volume can be determined as the cardiac output multiplied by the mean indicator transit time, i.e. the central circulation time (Figure 1). <iv The central circulation time is the mean time of indicator sojourn in the central vascular bed that can be determined as the time-

3 December 1995 VOLUME EXPANSION IN CIRRHOSIS 1919 Volume = Flow. Mean circulation time F = 1 ml/sec t" = 6 sec Central and arterial blood volume Central circulation time CBV = Cardiac output. Central circulation time Figure 1. Principles of determination of regional blood volume and flow. The upper panel shows the relation between volume (V), flow (F), and mean circulation time (t~. The mean central circulation time is determined as the time-weighted area under the concentration time curve (Iowerpanel). The central and arterial blood volume (CBV) is the volume in the heart, lungs, and central arterial tree. It is determined according to the kinetic theory as the cardiac output multiplied by the central circulation time. weighted area under the arterial indicator curve, corrected for the arterial catheter transit time. <1r'22 The central and arterial blood volume, as determined by the present technique, is the intravascular volume from the tip of the injection catheter (right atrium) to the tip of the sampling catheter (abdominal aorta). This volume includes the central arterial tree (the volume of blood contained within the arterial tree up to points that are temporally equidistant from the heart to the aortic bifurcation). <17'22 The coefficient of variation of duplicate determination of the central and arterial blood volume is <9%. ~ Plasma volume was measured as the injected amount of 125Ilabeled serum albumin divided by the plasma concentration of radioactivity 10 minutes after injection. The total blood volume was determined from plasma volume and hematocrit. 23'24 The noncentral blood volume was calculated as the difference between the total blood volume and the central and arterial blood volume. The noncentral circulation time was assessed as noncentral blood volume divided by the cardiac 1 output. The ideal body weight in kilograms was estimated as (height {cm} - 100) - t/4 (height {cm} - 150). 6 Normal values of blood volumes, flow, and transit times were obtained from 16 controls matched for weight and height: 6 central and arterial blood volume, L (mean ~ SEM); noncentral blood volume, L; cardiac output, L/min; plasma volume, L; central circulation time, seconds; and noncentral circulation time, seconds. Volume Expansion Baseline values of cardiac output, central circularion time, central and arterial blood volume, noncentral blood volume and circulation time, systemic vascular resistance, arterial blood pressure, and plasma volume were determined in a control period. Each patient then received a hyperosmotic load of galactose (50%; 0.5 g of galactose/kg body wt) intravenously during a period of 5 minutes. The average galactose load was 44 g (range, g). Infusion of this amount is equivalent to an average volume expansion of 0.8 L, as estimated from the infused mosmols of galactose and the plasma osmolality. Determination of the hemodynamic variables was repeated within 8 minutes of the start of the galactose infusion. The measurements were repeated at the presumed time of maximal volume expansion. The galactose elimination capacity, an estimate of the functional liver ceil mass, was subsequently measured as described by Tygstrup. 25 Statistics Data are expressed as mean and SE. The Mann-Whitney and Kruskal-Wallis tests were used to compare differences between Child-Turcotte classes and controls. The Wilcoxon test was used to compare differences within the patient and control groups. The two-tailed significance level of the type 1 error was fixed at 5%. Results Table 1 gives the characteristics and Table 2 the systemic hemodynamic variables before and during volume expansion. Changes in hemodynamics, volumes, and circulation times are shown in Figures 2-4. In the 6 control subjects, the central and arterial blood volume increased significantly after volume expansion (1.46 to 1.72 L, +18%; P < 0.05), whereas the noncentral blood volume increased less (2.99 to 3.21 L, +7%; NS). Both cardiac output (5.28 to 7.31 L/min, +38%; P < 0.05) and stroke volume (74 to 101 ml, +37%; P < 0.05) increased significant.ly after volume expansion. The systemic vascular resistance decreased significantly (1389 to 1043 dynes's'cm -5, -25%; P < 0.05), whereas the heart rate (72 to 74 minutes, +3%, NS) and mean arterial pressure (96 to 94 mm Hg, -2 %; NS) were unaltered. The central circulation time (16.9

4 1920 MgLLER ET AL. GASTROENTEROLOGY Vol. 109, No. 6 fable 2, Systemic Hemodynamics of Patients With Cirrhosis in the Different Child-Turcotte Classes Before and After Volume Expansion Child-Turcotte class A ChilcI-Turcotte class B Child-Turcotte class C (n = 12) (n = 14) (n = 13) Volumes Before After Before After Before After Plasma volume (L) 3.92 ± " ± 0.30 a 3.93 ± ± 0.51 a Central and arterial blood volume (L) ± 0.11 b 1.36 ± _ ± 0.10 Noncentral blood volume (L) 4.43 ± ± ± , ± 0.46 b Total blood volume (L) 5.88 ± ± 0.40" 5.73 ± ± 0.44 a ,61 ± 0.48 a Flow and resistance Cardiac output (L/min) ± 0.44 a 6.31 ± a 7.48 ± ± 0.52 b Stroke volume (ml) 83 ± 7 98 ± 7 ~ 88 _ ± 10" 90 ± _+ 8 b Heart rate (rain-1) 77 ± ± 4 77 ± 4 85 ± 3 'a Systematic vascular resistance (dyne, s -1. cm 5) 1161 ± ± 62 a 1187 ± ± 125" 905 ± 79 c 795 _+ 69 e Blood pressures Systolic blood pressure (ram Hg) 129 ± ± ± ± ± ± 4 Diastolic blood pressure (mm Hg) ± 3 58 ± ± Mean arterial blood pressure (mm Hg) 87 ± ± 2 82 ± 3 83 ± 4 80 ± 4 Right atrial pressure (mm Hg) 3.6 ± ± ± ± _+ 0.5 Circulation times Central circulation time (s) 14.8 ± ± 0.5" 13.4 ± ± 0.9 b 10.5 _+ 0.6.a 8.8 ± 0.5 b Noncentral circulation time (s) 44.2 ± ± ± ± 1.5 ~ 35.7 ± 2.0,d NOTE. Mean ± SEM. Significance of changes after volume expansion: "P < 0.001, bp < 0.01 compared with values before volume expansion. Significance of differences between patient classes: Cdifferent from class A, P < 0.05; adifferent from class B, P < to 14.8 seconds, -12%; P < 0.05) and the noncentral circulation time (35.2 to 26.3 seconds, -25%; P < 0.05) decreased both significantly after volume expansion. As expected, the plasma volume increased in all patient groups (+17%; P < 0.001; Table 2)[ The average measured increase in plasma volume following volume expansion was L, with no significant difference between the Child-Turcotte classes. For comparison, the calculated maximal volume expansion was estimated to be L. The central and arterial blood volume was significantly decreased in the patients with cirrhosis as compared with controls (1.36 vs L; P < 0.001), and the lowest values tended to be found in patients with the most advanced disease (Table 2). The central and arterial blood volume increased significantly in class A patients after volume expansion (1.46 to 1.63 L, +12%; P < 0.01; Table 2 and Figure 2). However, no significant change occurred either in class B (1.36 to 1.40 L, +3%; NS) or in class C patients (1.27 to 1.28 L, + 1%; NS). A significant negative correlation was found between the change in central and arterial blood volume after volume expansion, on the one hand, and the Child- Turcotte score on the other (r = -0.34; P < 0.03). In contrast, the noncentral blood volume was significantly increased in all Child-Turcotte classes (4.38 vs L; P < 0.001) and increased further during volume expansion both in class B patients (4.37 to 4.62 L, +6%; P < 0.01) and in class C patients (4.34 to 5.32 L, +23%; P < 0.01; Table 2 and Figure 2). The mean serum albumin concentration was 500 ~tmol/l. The changes in central and arterial blood volume above and below this value were 131 and 12 ml, respectively (P = 0.08). After volume expansion, a significant increase in cardiac output occurred in all Child-Turcotte classes (A, +22%; B, +18%; C, +16%; all P < 0.01; Table 2 and Figure 3). A comparable decrease in systemic vascular resistance was observed in all classes (A, -20%; B, -16%; C, -12%; all P < 0.01; Figure 3), and the systemic vascular resistance remained significantly lower in class C patients than in class A patients (P < 0.05). No significant changes in arterial blood pressure or heart rate were found following volume expansion (all P > 0.05; Table 2). Before volume expansion, the central circulation time was significantly lower in class C patients than in either class A or class B patients (P < 0.01), and this difference was accentuated during volume expansion (P < 0.001), which significantly decreased the central circulation time in all groups (class A, -12%; B, -13%; C, -16%; all P < 0.01; Table 2 and Figure 4). The noncentral circulation time is shown in Figure 4. No decrease was found in class C patients during volume expansion.

5 December 1995 VOLUME EXPANSION IN CIRRHOSIS O 2O > 15 o m 30 o) E m> 5o 0! i- o Z Controls A B C -N M Figure 2. Central and arterial blood volume and noncentral volume in control subjects and in Chiid-Turcotte class A, B, and C patients. *P < 0.05, **P < Discussion The main findings of the present study are the following. (1) Like control subjects, patients with mild cirrhosis can expand their central blood volume in response to a hyperosmotic volume expansion, whereas patients with advanced disease are unable to do so. Accordingly, the latter patients expanded their noncentral blood volume. (2) Volume expansion results in increased cardiac output in all patients and controls, most markedly in controls and Child-Turcotte class A patients, but with no significant change in arterial blood pressure or heart rate. (3) During volume expansion, the systemic vascular resistance decreased further in all Child-Turcotte classes. Our findings that the central and arterial blood volume is decreased in the presence of cirrhosis agree with the results of previous studies from our laboratory with other patients. <17 In the 6 control subjects who underwent an acute volume expansion, the central and arterial blood volume was relatively low and not significantly different from the cirrhotic class A patients. This may mainly be due to the fact that some of these controls were wasted and received antihypertensive treatment. Despite this, the controls were considered suitable for the evaluation of the relative changes in central hemodynamics. According to "the peripheral vasodilatation hypothesis," the initial event is reduced systemic vascular resistance) '11 The outcome is an abnormal distribution of the blood volume with reduced arterial blood volume. 8 Unloaded baroreceptors lead to enhanced sympathetic activity, activation of the renin-angiotensin-aldosterone and vasopressin systems with increased cardiac output and heart rate, and renal sodium-water retention as the outcome. <26 Increased circulating concentrations of catecholamines, renin, and vasopressin have been previously described by us and by others. 17'26'27 Because of the nature of this study that describes an acute volume expansion with a short duration, no attempts were made to detect a (minor) decrease in indicators of these chronically activated systems. In contrast, the "overflow theory" suggests primary renal sodium-water retention with overfilling of the vascular compartment and overflow of fluid into the peritoneal cavity. 3'28 A main feature of this theory is the increased plasma and blood volume. The present finding of a decreased central and arterial blood volume is in keeping with "the peripheral vasodilatation hypothesis." The distribution of the blood volume is governed by the complex interaction of cardiac output, vascular compliance, and hemodynamic resistance. 29 Even in the presence of increased cardiac output, cirrhotic patients seem unable to maintain a normal-sized central and arterial blood volume. However, our results show that in control subjects and in patients with less severe disease, it is possible to expand the central and arterial blood volume by about 18% and 12%, respectively, with a hyperosmotic galactose load. This elevation may be brought about by the 35% and 22% increase in cardiac output in controls and class A patients, respectively. In contrast, the central blood volume only increased by 3% and 1% in class B and C patients, respectively, and the average increase in cardiac output in these groups was 17 %. Our results are in agreement with a recent report by Wong et al., 3 who studied hemodynamic effects of transjugular intrahepatic portosystemic shunt. These investigators found in seven cirrhotic patients with severe disease no change in central blood volume, despite an increase in cardiac output and a decrease in systemic vascular resistance. There may be several explanations for these findings. It is possible that the patients with advanced disease may have reached the limit in their ability to increase

6 1922 MgLLER ET AL. GASTROENTEROLOGY Vol. 109, No. 6 CO (L/min) 10 A / Controls A B C SVR (dyn. sec/cm ) O Controls A B C Figure 3. (A) Cardiac output and (B) systemic vascular resistance in control subjects and in Child-Turcotte classes A, B, and C before and after completion of volume expansion. *P < 0.05, **P < 0.01, ***P < k./t cardiac output. 3~ Plasma volume expansion, which increases the preload, may produce an adequate increase in cardiac output only in controls and patients with a less disturbed circulation. This agrees with the association between the hyperkinetic circulation and cardiac dysfunction described in cirrhosis. 32 Reduced left ventricular afterload, because of the diminished systemic blood pressure, is often present in cirrhosis) 2'33 Attempts to increase the low arterial blood pressure with vasopressors, thereby increasing afterload, or with exercise may precipitate cardiac insufficiency and unmask a latent cardiac dysfunction in cirrhosis. 31'33'34 The relatively slight difference between the increase in cardiac output in Child class A patients and class B and C patients may only partially account for the differences in the change in the central and arterial blood volume. It is likely that the lack of an increase in central and arterial blood volume, achieved in patients with advanced disease, also may be explained by a higher venous vascular compliance, as recently proposed) 5 About 70% of the patients received diuretics that could influence the hemodynamic response by an increase in the vascular compliance. However, we found no statistically significant difference in the change in central and arterial blood volume in the patients who received diuretics as compared with those who did not. Thus, administration of diuretics may not have a major impact on vascular responsiveness. Another possibility could be a different sensitivity to vasoactive substances in the various vascular regions (see below). In none of our patient groups or in the control subjects was volume expansion accompanied by significant changes in the arterial blood pressure. The increase in cardiac output was succeeded by a further decrease in systemic vascular resistance in controls and all patient groups. This indicates that the systemic response to a hyperosmotic volume expansion in cirrhotic patients and in controls is a further decrease in systemic vascular resistance instead of an increase in the arterial blood pressure with unchanged resistance) During volume expansion, class A patients showed an equal increase in their central and arterial blood volume and total blood volume: both of 12%. Conversely, class C patients expanded their noncentral blood volume by about 23%. The overfilling of the peripheral circulation was enhanced by volume expansion with some improvement in the preload to the volumetrically reduced right heart) 6 This may produce some increment in cardiac output, and a transitory expansion of the central and arterial blood volume may take place (Figure 5) but with a subsequent reduction in central and arterial blood volume in the patients with advanced disease, most likely

7 December 1995 VOLUME EXPANSION IN CIRRHOSIS 1923 because of increased systemic and splanchnic venous compliance. In addition, differential reactivity to vasoconstrictors and vasodilators in the various vascular beds, and in different patient classes, may contribute to the abnormal distribution of the blood volume. A postreceptor defect has been postulated in cirrhosis and may be explained by involvement of nitric oxide) Moreover, the balance between vasoconstrictors and vasodilators is disturbed and sensitivity to vasoconstrictors, such as norepinephrine, vasopressin, angiotensin II, and endothelin 1, is reduced in the presence of cirrhosis) v This could also contribute to the abnormal distribution of the blood volume and explain, at least in part, why some patients are unable to maintain a normal-sized central and arterial blood volume. Recently, Bernardi et al) 8 suggested that volume expansion itself may account for the increased cardiac output by translocation of volume to the central area. However, our data do not support this view because we found no changes in the central and arterial blood volume of the most severely ill patients with the highest cardiac output in absolute terms. A primarily reduced systemic vascular resistance combined with low arterial blood pressure, enhanced sympathetic nervous activity, and increased noncentral blood volume are most probably the main elements in the elevated cardiac output in cirrhosis. 8 With respect to the importance of the plasma oncotic pressure to the hemodynamic responsiveness, we looked at the expansion of the central and arterial blood volume in relation to albumin concentrations. Patients with low serum albumin had a lower increase (of borderline sig- A CBV: + 12% CBV:1.46 L 20 3 L/min ). o 15 V p Non-CBV: 4.43 L III Non-CBV: +11% b Y 10 Controls A B C B CBV:1.27 L Volume expansion CBV:+1% CO: "-" 20 O c- O 30 Z 40 1 Non-CBV: 4.34 L CBV Volume expansion r"] Non-CBV Non-CBV: +23% 50 Figure 4. Central and noncentral circulation times in control subjects and in Child-Turcotte classes A, B, and C before and after completion of volume expansion. *P < 0.05, **P < 0.01, ***P < Figure 5. Illustration of acute volume expansion in (A) mild and (B) advanced cirrhosis. In patients with mild cirrhosis, cardiac output (CO), central and arterial blood volume (CBV), and noncentral blood volume (Non-CBV) increased after volume expansion. In patients with advanced cirrhosis, CBV was almost unaltered, whereas non-central blood volume increased further through a proposed intermediate state.

8 1924 MOLLER ET AL. GASTROENTEROLOGY Vol. 109, No. 6 nificance) in central and arterial blood volume than patients with high serum albumin levels. This suggests that a low plasma oncotic pressure may contribute to the inability to increase the central and arterial blood volume in severely ill patients. In comparison with vascular resistance (pressure relative to flow), the central circulation time may be regarded as the central and arterial blood volume relative to cardiac output. The central circulation time was decreased in all patients after volume expansion, which indicates that flow changes more than volume. This was also found with respect to the noncentral circulation time in controls and class A and class B patients. The unchanged or prolonged noncentral circulation time in class C patients may further add to the concept of insufficient cardiovascular response to volume load in patients with advanced disease. Because of the nature of hyperosmotic solutions, the galactose was infused during a period of 5 minutes. As the volume commences to expand immediately after the start of the galactose infusion, the central and arterial blood volume and cardiac output were determined within 3 minutes of termination of the infusion. However, despite this short time elapse, the maximal effect may have faded somewhat and the maximal response in cardiac output and central and arterial blood volume could be rather larger than that measured. Similarly, the plasma volume expansion measured was only about two thirds of the calculated maximal expansion. In conclusion, this study shows that in patients with advanced cirrhosis, the central and arterial blood volume is unable to expand in response to an acute volume expansion. These findings are consistent with the peripheral vasodilatation hypothesis and the circulatory hyporeactivity occurring in cirrhosis. Future research should focus on pharmacological methods to counteract the central hypovolemia and thereby improve cardiovascular dysfunction in cirrhosis. References 1. Sherlock S. Vasodilatation associated with hepatocellular disease: relation to functional organ failure. Gut 1990;31: Groszmann RJ. Hyperdynamic state in chronic liver diseases. J Hepatol 1993; 17:$38-$ Levy M. Pathogenesis of sodium retention in early cirrhosis of the liver: evidence for vascular overfilling. Semin Liver Dis 1994; 14: Henriksen JH. Systemic haemodynamic alterations in hepatic cirrhosis. Eur J Gastroenterol Hepatol 1991;3: Groszmann RJ. Vasodiiatation and hyperdynamic circulatory state in chronic liver disease. In: Bosch J, Groszmann R J, eds. Portal hypertension. Pathophysiology and treatment. Oxford: Blackwetl, 1994: Henriksen JH, Bendtsen F, S~rensen TIA, Stadeager C, Ring- Larsen H. Reduced central blood volume in cirrhosis. Gastroenterology 1989; 97: Epstein FH. Underfilling versus overfilling in hepatic ascites. N Engl J Med 1982;307: Schrier RW, Arroyo V, Bernardi M, Epstein M, Henriksen JH, Rod6s J. Peripheral artery vasodilatation hypothesis: A proposal for the initiation of renal sodium and water retention in cirrhosis. Hepatology 1988; 5: Schrier RW, Niederberger M, Weigert A, Gines P. Peripheral arterial vasodilatation: determinant of functional spectrum of cirrhosis. Semin Liver Dis 1994; 14: Jimenez W, Arroyo V. Pathogenesis of sodium retention in cirrhosis. J Hepatol 1993;18: Colombato LA, Albillos A, Groszmann RJ. Temporal relationship of peripheral vasodilatation, plasma volume expansion and the hyperdynamic circulatory state in portal-hypertensive rats. Hepatology 1992; 15: Dudley FJ. Pathophysiology of sodium retention in cirrhosis. In: Bosch J, Groszmann R J, eds. Portal hypertension. Pathophysiology and treatment. Oxford: Blackwell, 1994: Arroyo V, Gines P. Mechanism of sodium retention and ascites formation in cirrhosis. J Hepatol 1993;17:S24-$ Henriksen JH, Ring-Larsen H. Ascites formation in liver cirrhosis: the how and the why. Dig Dis 1990;8: Gentilini P, Arias IM, Arroyo V, Schrier RW. Liver disease and renal complications. New York: Raven, Colombato LA, Albillos A, Groszmann RJ. Role of central blood volume in the development of sodium retention in portal hypertensive rats (abstr). Hepatolog~ 1993;18:100A. 17. Henriksen JH, Bendtsen F, Gerbes AL, Christensen N J, Ring- Larsen H, S~rensen TIA. Estimated central blood volume in cirrhosis-relationship to sympathetic nervous activity, beta-adrenergic blockade and atrial natriuretic factor. Hepatoiogy 1992;16: Gluud C, The Copenhagen Study Group for Liver Diseases. Serum testosterone concentration in men with alcoholic cirrhosis: background for variation. Metabolism 1987; 36: Henriksen JH, Ring-Larsen H, Kanstrup I-L, Christensen NJ. Splanchnic and renal elimination and release of catecholamines in cirrhosis. Evidence of enhanced sympathetic nervous activity in patients with cirrhosis. Gut 1984;25: Henriksen JH, Winkler W. Hepatic blood flow determination. A comparison of 99-Tc-diethyl-IDA and indocyanine green as hepatic blood flow indicators in man. J Hepatel 1987;4: Henriksen JH, Ring-Larsen H, Christensen NJ. Circulating noradrenaline and central haemodynamics in patients with cirrhosis. Scand J Gastroenterol 1985;20: Lassen NA, Perl W. Tracer kinetics methods in medical physiology. New York: Raven, Larsen OA, Winkler K, Tygstrup N. Extra plasma in the liver calculated from the hepatic hematocrit in patients with portacaval anastomosis. Ctin Sci 1963;25: Lieberman FL, Reynolds TB. Plasma volume in cirrhosis of the liver. Its relations to portal hypertension, ascites, and renal failure. J Clin Invest 1967;47: Tygstrup N. Determination of the hepatic galactose elimination capacity after a single intravenous injection in man. Acta Physiol Scand 1963; 58: Henriksen JH, Ring-Larsen H. Hepatorenal disorders: role of the sympathetic nervous system. Semin Liver Dis 1994; 14: Bernardi M, Trevisani F, Gasbarrini A, Gasbarrini G. Hepatorenal disorders. Role of the renin-angiotensin-aldosterone system. Semin Liver Dis 1994;14: Rector WG, Robertson AD, Lewis FW, Adair OV. Arterial underfill-

9 December 1995 VOLUME EXPANSION IN CIRRHOSIS 1925 ing does not cause sodium retention in cirrhosis. Am J Med 1993;95: Mancia G, Mark AL. Arterial baroreflexes in humans. In: Berne RM, Sperelakis N, Geiger SR, eds. Handbook of Physiology, vol Ill. The cardiovascular system. Peripheral circulation and organ blood flow. Baltimore: Waverly, 1983: Wong F, Sniderman K, Liu P, Allidina Y, Sherman M, Blendis L. Transjugular intrahepatic portosystemic stent shunt: effects on hemodynamics and sodium homeostasis in cirrhosis and refractory ascites. Ann Intern Med 1995;122: Kelbaek H, Rab#l A, Brynjolf I, Eriksen J, Bonnevie O, Godtfredsen J, Munck O, Lund JO. Haemodynamic response to exercise in patients with alcoholic liver cirrhosis. Clin Physiol 1993;7: Green J, Better OS. Circulatory disturbance and renal dysfunction in liver disease and in obstructive jaundice. Isr J Med Sci 1994; 30: Limas C J, Guiha NH, Lekagul O. Impaired left ventricular function in alcoholic cirrhosis with ascites. J Lab Clin Med 1977;89: Gould L, Shariff M, Zahir M. Cardiac hemodynamics in alcoholic patients with chronic liver disease and a presystolic gallop. J Clin Invest 1969;48: Hadengue A, Moreau R, Gaudin C, Bacq Y, Champigneulle B, Lebrec D. Total effective vascular compliance in patients with cirrhosis: a study of the response to acute blood volume expansion. Hepatology 1992; 15: M#ller S, S~ndergaard L, M~gelvang J, Henriksen O, Henriksen JH. Decreased right heart blood volume determined by magnetic resonance imaging: evidence of central underfilling in cirrhosis. Hepatology 1995; 22: Sieber CC, Lopez-Talavera JC, Groszmann R J. Role of nitric oxide in the in vitro splanchnic vascular hyporeactivity in ascitic cirrhotic rats. Gastroenterology 1993; 104: Bernardi M, Dimarco C, Trevisani F, Fornale L, Andreone P, Cursaro C, Baraldini M, Ligabue A, Tame MR, Gasbarrini G. Renal sodium retention during upright posture in preascitic cirrhosis. Gastroenterology 1993; 105: Received January 19, Accepted August 21, Address requests for reprints to: Seren M~ller, M.D., Department of Clinical Physiology 239, Hvidovre Hospital, DK-2650 Copenhagen, Denmark. Fax: (45) Supported by The John and Birthe Meyer Foundation and The Danish Hospital Foundation for Medical Research, Region of Copenhagen, The Faroe Islands, and Greenland. Seren M~ller holds a research fellowship at the University of Copenhagen, Copenhagen, Denmark.