Mineralocorticoid escape in patients with compensated cirrhosis and portal hypertension

GASTROENTEROLOGY

1992;102:2114-2119

Mineralocorticoid Escape in Patients With Compensated Cirrhosis and Portal Hypertension GIORGIO LA VILLA, JOAN MANUEL SALMERON, VICENTE JAUME BOSCH, PERE GINES, JOAN CARLES GARCIA-PAGAN, ANGELS GIN& MONICA ASBERT, WLADIMIRO JIMGNEZ, FRANCISCA

RIVERA,

ARROYO,

and JOAN RODfiS

Liver Unit and Hormonal Laboratory, Hospital Clinic0 y Provincial, University of Barcelona, Spain; and Clinica Medica II, University of Florence School of Medicine, Florence, Italy

Failure to escape from mineralocorticoids in compensated cirrhosis is considered a major argument supporting the overflow theory of ascites. To assess the frequency and mechanism of mineralocorticoid escape in cirrhosis, 9-a-fluorohydrocortisone (0.6 mg/day) was administered to 19 patients with compensated cirrhosis, portal hypertension, and no history of ascites who were able to maintain sodium balance on a 250 mmol Na+ diet. Fifteen patients (78.9%) escaped from mineralocorticoids, while 4 patients (21.1%) did not escape and developed ascites. Patients who did not escape had significantly higher cardiac index (4.97 f 0.42 vs 3.46 f 0.21 L - min-’ - m-‘) and lower peripheral vascular resistance (485.9 f 37.5 vs. 665.8 + 32.9 dyne - s cme5/m2) than those who escaped. Hepatic venous pressure gradient was not significantly different. The escape phenomenon was associated with a significant increase in mean arterial pressure, creatinine clearance, and atria1 natriuretic factor and suppression of plasma renin activity. All of these parameters showed minimal or no changes in patients who did not escape. These results indicate that failure to escape from mineralocorticoids is uncommon in patients with compensated cirrhosis, is related to an inadequate expansion of effective plasma volume due to the accumulation of ascites, and occurs in patients with marked peripheral arteriolar vasodilation. dministration of pharmacological doses of mineralocorticoids to healthy subjects induces a short period of sodium retention, followed by a return of sodium excretion to normal levels. The re-establishment of normal sodium excretion despite continuous action of mineralocorticoids on the distal nephron to enhance sodium reabsorption has been called mineralocorticoid escape.le3 The escape phe-

A

Barcelona,

nomenon is thought to be caused by several natriuretic mechanisms activated by mineralocorticoidinduced volume expansion, including increased renal perfusion pressure,3-5 glomerular filtration rate,” and cardiac release of atria1 natriuretic factor (ANF)‘-” and suppression of renin-aldosterone’3*14 and sympathetic nervous systems.15 In contrast, Denison et a1.16 showed that when mineralocorticoids are administered to patients with compensated cirrhosis and portal hypertension [hepatic venous pressure gradient (HVPG) of >lO mm Hg], most of them do not escape from mineralocorticoid-induced sodium retention and develop ascites. The finding that renal sodium retention following mineralocorticoid administration leads to plasma volume expansion and ascites formation in patients with compensated cirrhosis and portal hypertension is considered a main argument supporting the “overflow” theory of ascites formation.‘6*‘7 According to this theory, volume-independent renal sodium retention of unknown cause is the initiating step in the formation of ascites. Sodium retention leads to the expansion of plasma volume, and with portal hypertension as a localizing force in the splanchnic bed, ascites formation occurs on an overflow basis. Recent hemodynamic and hormonal investigations in human and experimental cirrhosis, indicating that effective plasma volume is decreased in cirrhosis with ascites” and that renal sodium and water retention in this condition is the result of a homeostatic activation of endogenous neurohumoral systems aimed at maintaining arterial pressure within normal or near normal levels,‘g-2’ militate strongly against the overflow theory of ascites. The discrepancy between these investigations*e-21 and the findings of Denison et al.” and other authors22-24 0 1992

by the American Gastroenterological 0016-5065/92/$3.00

Association

MINERALOCORTICOID

lune 1992

prompted us to investigate further mineralocorticoid escape in cirrhosis. The current study was therefore designed to assess the frequency and mechanisms of mineralocorticoid escape in patients with compensated cirrhosis vious episodes

and portal of sodium

hypertension retention.

without

pre-

Materials and Methods Twenty-four consecutive outpatients (15 men and 9 women; mean age, 57 f 2 years; age range, 37-67 years) with histologically proven compensated cirrhosis, esophageal varices, and an HVPG of >lO mm Hg were included in the study. Ascites was ruled out in all patients by ultrasound examination. Cirrhosis was cryptogenic in 15 patients, alcoholic in 8, and hepatitis B virus (HBV)-associated in 1. No patient had a history of ascites; gastrointestinal bleeding; hepatic encephalopathy; hepatocellular carcinoma; renal, pulmonary, or cardiovascular diseases; hypertension; or peptic ulcer or was taking drugs for at least 1 month before inclusion.

Hemodynamic

Study

Hemodynamic investigations were performed as previously described25~26 after an overnight fast. Under local anesthesia and continuous electrocardiographic (ECG) monitoring, a venous catheter introducer was placed in the right femoral vein by the Seldinger technique and used to advance under fluoroscopy a 7F balloon-tipped catheter (Medi Tech; Cooper Scientific Corp., Watertown, MA) into the right hepatic vein for the measurement of the wedged (occluded) and free hepatic venous pressures. Portal pressure was estimated by the HVPG, the difference between wedged and free hepatic venous pressures.25*2s After these measurements were completed, a Swan-Ganz catheter (Edwards Laboratory, Los Angeles, CA) was advanced into the pulmonary artery25*26 to measure cardiopulmonary pressures and cardiac output (thermal dilution). Mean arterial pressure (MAP) was measured noninvasively using an automatic vital sign monitor (Dinamap; Criticon, Tampa, FL). Peripheral vascular resistance (PVR) (dyne. s. cmw5/ m2) was calculated as [(MAP - RAP)/CI] X 80, in which RAP is the right atria1 pressure (mm Hg) and CI is cardiac green index (L . min? - m-‘). A solution of indocyanine (Serb, Paris, France) was infused intravenously at a constant rate of 0.2 mg/min. After an equilibration period of 40 minutes, four samples of peripheral and hepatic venous blood were obtained simultaneously at Z-minute intervals for the measurement of hepatic blood flow (HBF), hepatic clearance, and intrinsic clearance (sinusoidal model) following previously reported methods.26,27 Steady indocyanine green levels and hepatic extraction above 0.1 were required for the calculation of HBF.” Cardiac output, PVR, HBF, hepatic clearance, and intrinsic clearance were corrected for body surface and expressed as units per square meter. Protocol

ESCAPE IN CIRRHOSIS

2115

high-sodium intake was tested; a 24hour urine collection was obtained to measure urine volume and sodium excretion, and patients underwent abdominal ultrasound examination to detect ascites. Patients unable to excrete all the ingested sodium (urinary sodium excretion of 425 mmol/day) or showing ascites were excluded from the study. In the remaining patients, after 1 hour of bed rest in a quiet room an antecubital vein was catheterized. Fortyfive minutes later, blood samples were taken to measure standard liver function tests, serum electrolytes and creatinine, blood urea nitrogen, plasma renin activity (PRA), and plasma concentration of aldosterone and ANF. Subsequently, radioiodinated serum albumin (10 uCi) was injected intravenously and 10 minutes later a blood sample was obtained from the contralateral arm to measure plasma volume. 2gBody weight, arterial pressure, and heart rate were also recorded. An aliquot of the 24hour urine collection was used to measure urinary excretion of creatinine. Creatinine clearance and MAP were calculated using standard formulas. Patients were then given 9-a-fluorohydrocortisone (0.6 mg/day in a single oral dose after breakfast) until they escaped or developed ascites. A z&hour urine collection was obtained, and abdominal ultrasound was performed on the fourth and seventh days of %a-fluorohydrocortisone and then every day to recognize the occurrence of the escape phenomenon (defined as daily sodium excretion of >225 mmol) or the appearance of ascites. When patients escaped or developed ascites, blood and urine samples were obtained as previously described to measure serum electrolytes and creatinine, PRA, plasma concentration of aldosterone and ANF, and creatinine excretion. Plasma volume was again measured, and body weight, MAP, and heart rate were recorded. 9-a-Fluorohydrocortisone was then withdrawn. The high-sodium diet and the dosage of 9-a-fluorohydrocortisone used in this investigation were the same as in the study performed in healthy subjects by Miyamori et a130 Informed written consent was obtained from all patients included in the study, which was approved by the Ethic Committee of the Hospital Clinic0 y Provincial of Barcelona in May 1988. Analysis PRA was determined by the radioimmunoassay (Clinical Assay, Baxter, Cambridge, MA) of generated angiotensin I after a s-hour incubation at pH 7.4 and 37°C under conditions inhibiting further conversion ofangiotensin I. Plasma levels of aldosterone were measured using a commercial kit (Coat-A-Count Aldosterone; Diagnostic and Products Corporation, Los Angeles, CA). ANF was extracted on Cl8 Sep-Pack cartridges (Waters Associates, Milford, MA) and assayed using a previously described radioimmunoassay technique.3’ Statistical analysis of the results was performed using Student’s t test for paired or unpaired data as appropriate. All results are given as mean +- SEM.

of the Study

After the hemodynamic study, patients were given a 250-mmol/day sodium diet throughout the protocol. After a week on this diet, the patients’ tolerance to the

Results before

Five patients were excluded from receiving 9-a-fluorohydrocortisone

the study because

2116

LA VILLA ET AL.

GASTROENTEROLOGY

Table 1. Main Baseline Clinical and Laboratory Data in Patients Who Escaped and Patients Who Developed Ascites

Age(~4

Sex (M/F) MAP (mm Hg) HR (beats/min)

Bilirubin

Escape

Ascites

(n = 15)

(n = 4)

56 + 2

56 +4

9/6 89.5 ? 2 71 + 2

3/l 91.5 f 6.0 67 + 4

bmol/L 14 * 2

(mg/dUl Serum AST [IU/L/(pkat/L)]

(0.80 + 0.07) 97 + 13

Serum ALT [ILJ/L(,ukat/L)]

(1.62 + 0.22) 125 f 20

Serum AP [JU/L(pkat/L)]

(2.08 f 0.33) 227 + 26

Serum albumin (g/L) y-Globulins (g/L) Prothrombin activity (%) Serum Na+ (mmol/L) Serum K+ (mmol/L) Blood urea nitrogen [mmoJ/L urea (mg/dL)]

(3.78 42 15 80 140 4.0

+ + + + + f

0.43) 1 1 3 1 0.1

5.5 + 0.5 (14.5 + 0.8)

Serum creatinine FmollL (mg/dL)l

80 f 0 (0.9 + 0.03)

HR, heart rate; AST, aspartate aminotransferase; aminotransferase; AP, alkaline phosphatase.

18 + (1.07 + 73 * (1.22 + 80 + (1.33 + 208 f (3.47 -t 40 f 16+ 84 + 139 f 3.8 +

2 0.17) 7 0.12) 5 0.08) 5 0.08) 2 1 5 1 0.1

5 + 0.5 (13.5 + 0.9) 80 f 0 (0.9 f 0.04)

ALT, alanine

of a 24-hour sodium excretion of ~225 mmol 1 week after starting the 250-mmol Na+ diet (n = 3), gastrointestinal bleeding (n = l), and flank pain with ultrasound evidence of a ureteral stone (n = 1). The remaining 19 patients completed the study. Frequency

of Mineralocorticoid

Escape

Fifteen patients (78.9%) escaped from the sodium-retaining effects of the mineralocorticoid, as indicated by their ability to return to sodium balance during mineralocorticoid administration (baseline sodium excretion, 248 + 4 mmol/day; sodium excretion at the time of escape, 243 + 4 mmol/day). None of these patients had ultrasound evidence of fluid accumulation in the peritoneal cavity. The escape phenomenon was observed on the fourth day of 9-afluorohydrocortisone administration in 4 patients, on the seventh day in 1, on the eighth day in 1, on the ninth day in 4, and on the 10th day in the remaining 5. Four patients (21.1%) developed ascites during 9a-fluorohydrocortisone administration, as detected by physical and ultrasound examination. Ascites was observed on the fourth day in 2 patients, on the seventh day in 1, and on the ninth day in 1. All these patients were retaining sodium when ascites was de-

Vol. 102, No. 6

tected (baseline sodium excretion, 258 +- 7 mmol/ day; sodium excretion at the time of ascites development, 150 f 17 mmol/day; P < 0.001 vs. values observed in escaping patients at the time of the escape). Baseline Clinical, Laboratory, Hemodynamic Data

and

Table 1 shows the main baseline clinical and standard laboratory data of the patients who escaped and those who developed ascites. No significant differences were found between these two groups of patients in any of the measured parameters. Results of the hemodynamic measurements are given in Table 2. Patients who developed ascites had significantly higher CI and lower PVR than patients who escaped. Figure 1 shows individual HVPG, CI, and PVR values in these two groups of patients. Results of the hemodynamic measurements in the 3 patients not included in the study because of sodium retention before 9-a-fluorohydrocortisone administration were similar to those in patients who did not escape from mineralocorticoids (HVPG, 17.8 + 1.52 mm Hg; CI, 5.04 + 0.38 L - min-’ - m-‘; PVR, 455.8 + 44.9 dyne-s- cm-+/m’; HBF, 0.94 f 0.21 Lemin-‘m-‘; hepatic clearance, 0.14 f 0.02 L - min-* - m-‘; intrinsic clearance, 0.15 * 0.02 L - min-’ - m-‘). Effects of Mineralocorticoid Administration on Systemic Hemodynamics, Renal Function, and Hormonal Measurements Table 3 shows the changes in systemic hemodynamics, renal function, and hormonal measurements observed in patients who escaped from the sodium-retaining effects of 9+fluorohydrocortisone. The escape phenomenon was associated with a significant increase in plasma volume, body weight, MAP, creatinine clearance, and plasma levels of

Table

2. Results of Hemodynamic Measurements and lndocyanine Green Clearances in Patients Who

Escaped and Patients Who Developed Ascites Escape HVPG (mm Hg) CI (L . min-’ . m-‘) PVR (dyne +s - cm-“/m’) HBF (L . min-’ . m-‘) Hepatic clearance (L - min-’ . m-‘) Intrinsic clearance (L - min-’ . m-‘)

13.9 3.46 665.8 0.70

f f f f

0.6 0.21 32.9 0.07

Ascites 16.7 4.97 485.9 0.98

+ ? f f

2.5 0.42 37.5 0.21

0.18 + 0.03

0.16 + 0.02

0.27 + 0.03

0.17 f 0.02

P
Normal values (mean ? ZSD) are as follows: HVPG, <6 mm Hg; CI, 2.5 + 0.3 L - min-’ . m- ‘; PVR, >700 dyne. s - cme5/mz; HBF, 0.73 + 0.18 L-min-‘-m-Z; hepatic clearance, 0.23 f 0.11 L * min-’ * m-*; intrinsic clearance, 0.47 f 0.19 L * min-’ - m-‘.

MINERALOCORTICOID ESCAPE IN CIRRHOSIS 2117

June 1992

HVPG (111Ha)

25r

3.0

ALDOSTEROK (ng/dl) 30

2.5

25

2.0

20

1.5

15

1.o

10

20

500

0.5

5

10

400

0.0

0

0

PVR

CI (L/nin/f#)

dyne.sec,cm-5/m2

700

20%

600

a 15.

fj0 lo-

Figure 1. Individual values of HVPG, CI, and PVR observed under baseline conditions in the 15 patients who escaped mineralocorticoids (0) and the 4 who did not escape and developed ascites (0).

ANF

and a marked suppression of PRA and plasma aldosterone concentration. Figure 2 shows the individual changes in PRA, aldosterone, and ANF. In the 4 patients who did not escape and developed ascites, there was a clear tendency toward lower expansion of effective plasma volume. Despite a more marked increase in body weight (from 72.7 + 3.7 to 77.0 -t 4.1 kg), changes in MAP (from 91 + 6 to 92 ? 4 mm Hg), creatinine clearance (from 87 f 5 to 90 t- 3 mL/min), PRA (from 0.50 -+ 0.28 to 0.40 f 0.20 ng - mL-’ - h-‘) and plasma levels of ANF (from 10.39 + 1.90 to 13.67 k 3.96 fmol/mL) were minimal or absent. Plasma volume increased (from 1971 f 212 to 2169 k 185 mL/m’) and plasma aldosterone concentration decreased [from 9.23 f 3.70 to 4.77 + 0.91 ng/dL) to a lesser extent than in patients escaping from mineralocorticoid administration. As expected, mineralocorticoid administration in-

Table 3. Effects of Mineralocorticoid Administration on Body Weight, Systemic Hemodynamics, Renal Function, and Hormonal Measurements in the

15 Patients Who Escaped

Plasma volume (mL/mZ)n Body weight (kg) MAP (mm Hg) Creatinine clearance (mL/min) PRA [ng.mL?.h-’ (ng . L-’ 1s-‘)I” Aldosterone [pmol/L (ng/dL)l” ANF ( fmol/mL)’

AW (fmol/ml) 50

a00

.

PRA([email protected])

Baseline

Escape

P

1,732 + 60 71.7 + 3.1 89.7 + 2.2

2,054 * 131 73.7 + 3.6 99.3 f 2.7

10.01
91.1 + 3.3 0.60 f 0.19 (0.17 + 0.05)

99.6 + 4.8 0.13 + 0.03 (0.04 f 0.01)

310 !z 40 (11.01 + 1.60) 12.50 + 2.04

100 + 10 (3.49 + 0.50) 25.90 f 4.12



“Values observed in our laboratory in healthy subjects on a high sodium intake are as follows: plasma volume, 1550 f 25 mL/m’; PRA, 0.50 + 0.10 nge mL_‘. h-’ (0.14 + 0.03 ng. L-’ es-‘); plasma aldosterone, 270 f 50 pmol/L (9.50 f 1.75 ng/dL); ANF, 9.73 f 1.67 fmol/mL.

40 30

Figure 2. Individual changes in PRA and plasma concentration of aldosterone and ANF induced by mineralocorticoid administration in the 15 patients who escaped.

duced a reduction in serum potassium concentration (from 4.0 * 0.1 to 3.5 f 0.1 mmol/L; P < 0.005 in escapers: and from 3.8 + 0.1 to 3.3 t 0.3 mmol/L in nonescapers) without affecting serum Na+ levels. In nonescaping patients, ascites disappeared within 1 week of moderate sodium restriction (60 mmol/day), and they were then able to tolerate a normal-sodium diet without evidence of sodium retention. No other side effects of mineralocorticoid administration were seen. Discussion Three studies have investigated the renal response to mineralocorticoids in cirrhosis. Denison et al.‘” studied 10 patients with compensated alcoholic cirrhosis (n = 8) or severe diffuse hepatic fibrosis and portal hypertension who were able to eat a “normal”-sodium diet without sodium retention. Only 3 of these 10 patients were able to escape from mineralocorticoid administration. The remaining 7 patients did not escape and developed ascites despite a marked increase in plasma volume. Wilkinson et a1.23studied 12 patients with compensated cirrhosis, 6 of whom developed ascites during mineralocorticoid administration. Portal pressure was not measured in this study. Finally, Unikowsky et al.24 studied 9 dogs with experimental cirrhosis and sideto-side portocaval anastomosis. The 5 animals that did not escape had functionally ineffective portocaval anastomoses, leading to portal hypertension. These studies have led to the belief that failure to escape from mineralocorticoids is very common in compensated cirrhosis with portal hypertension. The results of the current study do not support this contention. Of the 19 patients studied, only 4 (21%) did not escape and developed ascites. There are two possible reasons for the higher rate of ascites formation after mineralocorticoid administration in the studies by Denison et al.16 and Wilkinson et al.23 First, the patients’ ability to excrete the high-sodium diet before mineralocorticoid administration was not tested in these investigations, in which patients were on “normal”‘” or 50-mmol/dayz3 sodium intake be-

2118

LA VILLA ET AL.

fore administration of a ZOO-250-mmol/day sodium diet and 9-a-fluorohydrocortisone. Second, whereas in the current study no patient had a history of ascites, this was not the case in the studies by Denison et all6 and Wilkinson et a1.,23 in which 6 and 5 cases, respectively, had previous episodes of ascites. Therefore, it is possible that most patients with a history of ascites and perhaps some without previous ascites were already unable to excrete sodium normally before mineralocorticoid administration, and this defect was not detected because sodium intake was “normal” or restricted. The HVPG in the 10 patients with cirrhosis or diffuse hepatic fibrosis studied by Denison et al.” was almost identical to that observed in our series (15.6 + 0.9 vs. 14.4 + 0.7 mm Hg). The characteristics of mineralocorticoid escape in compensated cirrhosis are similar to those reported in normal conditions. In healthy subjects,3*‘1*30*32-34 the escape phenomenon occurs within 4-10 days of mineralocorticoid administration, after an increase in body weight, MAP, and glomerular filtration rate of 2-3 kg, 7-14 mm Hg, and 9-15 mL/min, respectively. This is associated with a suppression of PRA and plasma aldosterone concentration of 70%-90% and a 2-3-fold increase in the plasma levels of ANF. These figures are comparable to those observed in our patients (see Table 3). Patients showing an “early” escape (within 4 days] did not differ from those with a “late” escape in any of the parameters measured under baseline conditions or in the changes in body weight, plasma volume, and hormones induced by mineralocorticoid administration (data not shown). According to the overflow hypothesis,‘6p’7 ascites formation during mineralocorticoid administration in patients with compensated cirrhosis is the consequence of a marked expansion of effective plasma volume which overflows in the peritoneal cavity because of portal hypertension. However, this was not the case in our nonescaping patients, in whom no changes in MAP, creatinine clearance, PRA, or plasma ANF concentration were observed. All of these parameters are sensitive markers of effective plasma volume expansion during mineralocorticoid administration. Plasma aldosterone concentration decreased by approximately 50% in these patients, but this decrease could be related to the decrease in serum potassium concentration, which has a powerful effect on aldosterone secretion.35 The most likely possibility is that these patients did not achieve an adequate expansion of effective plasma volume because of the extravasation of fluid in the peritoneal cavity. This would account not only for the formation of ascites but also for the failure to escape from mineralocorticoids.

GASTROENTEROLOGY

Vol. 102, No. 6

In this regard, recent studies by Caramel0 et a1.36 and Sanz et a1.37showing that vascular permeability and the compliance of the interstitial space is higher in rats with experimental cirrhosis than in control animals, thus facilitating the accumulation of fluid in the interstitial space, are of particular relevance. These abnormalities, which become more prominent after maneuvers aimed at expanding plasma volume,36s37 explain why saline administration is less effective in increasing plasma volume in cirrhosis than in normal conditions. These findings in experimental cirrhosis can be reproduced in normal animals by the administration of the arteriolar vasodilator minoxidil,38 suggesting that arteriolar vasodilation occurring in cirrhosis could be the main determinant of these abnormalities. The observation in the current study that patients with compensated cirrhosis who did not escape from mineralocorticoids were those with higher cardiac output and lower PVR is in agreement with this concept. The marked peripheral arteriolar vasodilation in these patients, by facilitating the extravasation of fluid to the interstitial compartment, would preclude an adequate expansion of effective circulating blood volume, the activation of natriuretic mechanisms, and the occurrence of mineralocorticoid escape. Because the splanchnic vasculature is a major site of peripheral arteriolar vasodilation in cirrhosis,‘8,39-43 it is not surprising that fluid extravasation occurred mainly in leading to the formation of this compartment, ascites. References 1. Clinton M, Thorn GW. Effect of desoxycorticosterone

2.

3.

4.

5.

6.

7.

8.

9.

acetate administration on plasma volume and electrolyte balance of normal human subjects. Johns Hopkins Med J 1943;72:255264. August JL, Nelson DH, Thorn GW. Response of normal subjects to large amounts of aldosterone. J Clin Invest 1958;37:1549-1555. Knox FG, Burnett JC Jr, Kohan DE, Spielman WS, Strand JC. Escape from the sodium retaining effects of mineralocorticoids. Kidney Int 1980;17:263-276. Higgins JR. Escape from sodium-retaining effects of deoxycorticosterone in hypotensive and hypertensive dogs. Proc Sot Exp Biol Med 1970;134:768-772. Hall JE, Granger JP, Smith MJ Jr, Preman AJ. Role of renal hemodynamics and arterial pressure in aldosterone escape. Hypertension 1984;6(Suppl 1):183-192. Sonnemberg H. Proximal and distal tubular function in salt deprived and salt-loaded deoxycorticosterone acetateescaped rats. J Clin Invest 1973;52:263-273. Ballerman BJ, Bloch KD, Seidman JG, Brenner BM. Atria1 natriuretic peptide transcription, secretion, and glomerular receptor activity during mineralocorticoid escape in the rat. J Clin Invest 1986;78:840-843. Granger JP, Burnett JC Jr. Romero JC, Opgenorth TJ, Salazar J, Joyce M. Elevated levels of atria1 natriuretic peptide during aldosterone escape. Am J Physiol 1987;252:R878-R882. Metzler CH, Gardner DG, Keil LC, Baxter JD, Ramsey DJ. Increased synthesis and release of atria1 natriuretic peptide dur-

June 1992

ing DOCA escape in conscious dogs. Am J Physiol 1987;252:R188-R192. RS, Edwards BS, Schwab TR, Heublein DM, 10. Zimmermann Burnett JC Jr. Atria1 natriuretic peptide during mineralocorticoid escape in the human. J Clin Endocrinol Metab 1987;64:624-627. ND, Buckley MG, Sagnella GA, 11. Cappuccio FP, Markandu Shore AC, MacGregor GA. Changes in the plasma levels of atria1 natriuretic peptides during mineralocorticoid escape in man. Clin Sci 1987;72:531-539, JM, Romero JC, Knox FG. Escape from the 12. Gonzales-Campoy sodium-retaining effects of mineralocorticoids: role of ANF and intrarenal hormone systems. Kidney Int 1989;35:767-777. system, al13. Gross F, Brunner H, Ziegler M. Renin-angiotensin dosterone and sodium balance. Ret Prog Hormone Res 1965;21:119-122. 14. Johnston Cl, Davis JO, Robb ChA, Mackensie JW. Plasma renin in chronic experimental heart failure and during renal sodium escape from mineralocorticoids. Circ Res 1968;22: 113-125. 15. Lungqvist A. The effect of angiotensin

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

infusion, sodium loading and sodium restriction on the renal and cardiac adrenergic nerves. Acta Path01 Microbial Stand 1975;83:661-668. Denison EK, Lieberman FL, Reynolds TB. 9-a-fluorohydrocortisone induced ascites in alcoholic liver disease. Gastroenterology 1971;61:497-503. Lieberman FL, Denison EK, Reynolds TB. The relationship of plasma volume, portal hypertension, ascites formation and renal sodium retention in cirrhosis: the overflow theory of ascites. Ann NY Acad Sci 1970;170:202-206. Schrier RW, Arroyo V, Bernardi M, Epstein M, Henriksen JH, Rodes J. Peripheral arterial vasodilation hypothesis: a proposal for the initiation of renal sodium and water retention in cirrhosis. Hepatology 1988;8:1151-1157. Schroeder ET, Anderson GH, Goldman SH, Streeter DH. Effect of blockade of angiotensin II on blood pressure, renin and aldosterone in cirrhosis. Kidney Int 1976;9:511-519. Arroyo V, Bosch J, Mauri M, Ribera F, Navarro-Lopez F, Rodes J. Effect of angiotensin II blockade on systemic and hepatic hemodynamics and on the renin-angiotensin-aldosterone system in cirrhosis with ascites. Eur J Clinlnvest 1981;11:221229. Willet JR, Jennings G, Esler M, Dudley FJ. Sympathetic tone modulates portal venous pressure in alcoholic cirrhosis. Lancet 1986;2:939-943. Claria J, Jimenez W, Arroyo V, la Villa G, Lopez C, Asbert M, Castro A, Gaya J, Rivera F, Rodes J. Effect of V, vasopressin receptor blockade on arterial pressure in conscious rats with cirrhosis and ascites. Gastroenterology 1991;100:494-501. Wilkinson SP, Smith IK, Moodie H, Poston L, Williams R. Studies on mineralocorticoid escape in cirrhosis. Clin Sci 1979;56:401-406. Unikowshy B, Wexler MJ, Levy M. Dogs with experimental cirrhosis of the liver but without intrahepatic hypertension do not retain sodium or form ascites. J Clin Invest 1983:72:1594-1604. Bosch J, Mastai R, Kravetz D, Navasa M, Rodes J. Hemodynamic evaluation of the patient with portal hypertension. Semin Liver Dis 1986:6:309-317. Navasa M, Chesta J, Bosch J, Rodes J. Reduction of portal pressure by isosorbide-5-mononitrate in patients with cirrhosis. Gastroenterology 1989;96:1110-1118. Navasa M, Bosch J, Reichen J, Bru C, Mastai R, Zysset T, Silva G, Chesta J, Rodes J. Effects of verapamil on hepatic and systemic hemodynamics and liver function in patients with cirrhosis and portal hypertension. Hepatology 1988;8:850-854.

MINERALQCORTICOID

ESCAPE IN CIRRHOSIS

2119

28. Bosch J, Kravetz D, Rodes J. Effects of somatostatin on hepatic and systemic hemodynamics in patients with cirrhosis of the liver. Comparison with vasopressin. Gastroenterology 1981; 80:518-525. 29. International Committee for Standardization in Haematology. Recommended methods for measurements of red cell and plasma volume. J Nucl Med 1980;21:793-800. 30. Miyamori I, Ikeda M, Matsubara T, Okamoto S, Koshida H, Morise T, Takeda K. Human atria1 natriuretic polypeptide during escape from mineralocorticoid excess in man. Clin Sci 1987;73:431-436.30. 31. Gin&s P, Jimenez W, Arroyo V, Navasa M, Lopez C, Tit6 Ll, Serra A, Bosch J, Sanz G, Rivera F, Rod&s J. Atria1 Natriuretic Factor in cirrhosis with ascites: plasma levels, cardiac release and splanchnic extraction. Hepatology 1988;8:636-642. 32. Wambach G, Gijtz S, Suckau G, Kaufmann W. Plasma levels of atria1 natriuretic peptide (ANP) during mineralocorticoid escape in normal man. Klin Wochenschr 1986;64(Suppl6):5357. 33. Boer WH, Koomans HA, Dorhout Mees EJ. Lithium clearance in mineralocorticoid escape in humans. Am J Physiol 1987;252:F382-F386. 34. Gaillard CA, Koomans HA, Rabelink TJ, Braam B, Boer P, Dorhout Mees EJ. Enhanced natriuretic effect of atria1 natriuretic factor during mineralocorticoid escape in humans. Hypertension 1988;12:450-456. 35. Laragh JH, Sealey JE. The renin-angiotensin-aldosterone hormonal system and regulation of sodium. potassium, and blood pressure homeostasis. In: Orloff J, Berliner RW, eds. Handbook of physiology. Section 8, renal physiology. Washington, DC: American Physiological Society, 1973:831-908. D, Santos JC, Blanchart A, 36. Caramel0 C, Fernandez-Muiioz Rodriguez-Puyol D, Lopez-Novoa JM, Hernando L. Effect of volume expansion on hemodynamics, capillary permeability and renal function in conscious, cirrhotic rats. Hepatology 1986;6:129-134. 37. Sanz E, Caramel0 C, Lopez-Novoa JM. Interstitial dynamics in rats with early stage experimental cirrhosis of the liver. Am J Physiol 1989;256:F497-F503. 38. Sanz E, Lopez-Novoa JM, Linares M, Digiuni E, Caramel0 C. Intravascular and interstitial fluid dynamics in rats treated with minoxidil. J Cardiovasc Pharmacol 1990;15:485-492. 39. Bosch J, Gines P, Arroyo V, Navasa M, Rodes J. Hepatic and systemic hemodynamics and the neurohumoral systems in cirrhosis, In: Epstein M, ed. The kidney in liver disease. 3rd ed. Baltimore: Williams & Wilkins, 1988:286-305. 40. Bosch J, Mastai R, Kravetz D, Bruix J, Rigau J, Rodes J. Measurement of azygous blood flow in the evaluation of portal hypertension in patients with cirrhosis: clinical and hemodynamic correlations in 100 patients. J Hepatol 1985;1:125-139. 41. Vorobioff J, Bredfelt JE, Groszmann RJ. Increased blood flow through the portal system in cirrhotic rats. Gastroenterology 1984;87:1120-1126. 42. Benoit JN, Granger N. Splanchnic hemodynamics in chronic portal hypertension. Semin Liver Dis 1986;6:287-298. J, Prieto J, Quiroga J. Zozaya JM, Corbos MA, 43. Fernandez-Seara Rodriguez-Eire JL, Garcia-Plaza A, Lea1 J. Systemic and regional hemodynamics in patients with liver cirrhosis and ascites with and without functional renal failure. Gastroenterology 989;97:1304-1312. Received March 20, 1991. Accepted December 12, 1991. Address requests for reprints to: Vicente Arroyo, M.D., Liver Unit, Hospital Clinic0 y Provincial, Villarroel 170, Barcelona 08036, Spain. The authors gratefully acknowledge M. Roman, E. Calvo, and C. Escofet for their skillful technical assistance.