Patterns of regional sympathetic nerve traffic in preascitic and ascitic cirrhosis

Patterns of Regional Sympathetic Nerve Traffic in Preascitic and Ascitic Cirrhosis MASSIMO POZZI,1 GUIDO GRASSI,1 ELENA REDAELLI,1 RAFFAELLA DELL’ORO,1 LAURA RATTI,1 ALESSANDRO REDAELLI,1 GERARDO FOGLIA,1 ALESSANDRO DI LELIO,2 AND GIUSEPPE MANCIA1,3

Overactivity of the sympathetic nervous system and portal hypertension are key factors in the development of ascites in cirrhosis. The sympathoexcitation that characterizes the more advanced stages of liver diseases is less clearly defined in preascitic cirrhosis. We measured sympathetic nerve traffic to skeletal muscle (peroneal nerve) and to skin districts by microneurography in (1) 12 Child class A cirrhotic patients with clinically significant portal hypertension (portal pressure gradient > 10 mm Hg, 14.8 ⴞ 1.2 mm Hg, mean ⴞ SEM) but without actual or previous ascites, (2) 16 Child class C cirrhotic patients with tense ascites, and (3) 10 patients with mild congestive heart failure, a condition paradigmatic of a marked sympathetic activation. Muscle sympathetic nerve traffic was markedly increased in Child class C subjects as compared with controls (23.9 ⴞ 1.6 bursts/min, P < .01) and superimposable to that recorded in heart failure patients (52.9 ⴞ 4.7 vs. 60.3 ⴞ 2 bursts/min, P ⴝ not significant). Muscle sympathetic nerve traffic was also increased in Child class A subjects (41.6 ⴞ 2 bursts/min, P < .01 vs. controls) although to a lesser extent (P < .05 vs. Child class C patients). Skin sympathetic nerve traffic was within the normal range in all patients. Neurohormones were all markedly increased in Child class C subjects. Only norepinephrine was increased in Child class A patients. Our data show that sympathetic nerve traffic activation (1) is already detectable in Child class A cirrhosis when clinically significant portal hypertension is present but ascites never developed and (2) is not generalized because although muscle traffic is increased, skin traffic is within normal range. The role of drugs modulating sympathoactivation should be investigated in preascitic cirrhosis. (HEPATOLOGY 2001;34:1113-1118.) Cirrhosis is accompanied by an increase in sympathetic activity1 as shown by the observation that in this condition plasma norepinephrine,2-5 norepinephrine spillover from

Abbreviations: HCV, hepatitis C virus; Ab, antibody; NE, norepinephrine; MSNA, muscle sympathetic nerve activity; SSNA, skin sympathetic nerve activity. From the 1Cattedra di Medicina Interna, Universita` degli Studi di Milano-Bicocca, Divisione di Medicina 1, and 2Dipartimento di Radiologia, Ospedale San Gerardo dei Tintori, Monza, Milano, Italy; 3Istituto Auxologico Italiano, Milano, Italy. Received May 7, 2001; accepted September 24, 2001. Address reprint requests to: Massimo Pozzi, M.D., Cattedra di Medicina Interna, Universita` degli Studi di Milano-Bicocca, Divisione di Medicina 1, Ospedale San Gerardo dei Tintori, Via Donizetti 106-20052 Monza, Milano, Italy. E-mail: [email protected]; fax: (39) 039 322274. Copyright © 2001 by the American Association for the Study of Liver Diseases. 0270-9139/01/3406-0009$35.00/0 doi:10.1053/jhep.2001.29198

neuroeffector junctions,6 and muscle sympathetic nerve traffic as directly quantified by microneurography7-9 are all increased. The sympathoactivation accompanying cirrhosis has not yet been thoroughly characterized, however. For example, it is not yet clear whether the sympathoactivation is a pathophysiologic feature of cirrhosis throughout its course or if it just appears when fluid retention, ascites, or the hepatorenal syndrome dramatically alter circulating blood volume.10 It is also not clear whether the sympathoactivation is generalized to the entire cardiovascular system or if it involves some vascular beds but not others. The aim of the present study has been to provide information on these two issues; that is, whether sympathetic activity is increased in earlier stages of cirrhosis in which portal pressure is increased but ascites has never developed and whether in either preascitic and more advanced cirrhosis the sympathetic activation has a generalized or more regional distribution. PATIENTS AND METHODS Study Population. This study was performed on a total of 48 subjects. Twenty-eight patients had cirrhosis and portal hypertension with different degrees of liver impairment as assessed by the ChildPugh classification. Twelve patients were Child class A, histologically proven, untreated cirrhotic patients (age 57.1 ⫾ 7.2 years, mean ⫾ SD; 3 were positive for hepatitis B surface antigen, 8 were positive for antibodies for hepatitis C virus [HCV Ab⫹], HCV RNA⫹, 1 was positive for HCV Ab plus alcohol in the past, 10 men, and 2 women) who displayed indirect (both biochemical and instrumental) and direct (hepatic venous pressure gradient ⬎10 mm Hg in all, 14.8 ⫾ 1.2 mm Hg, mean ⫾ SEM) signs of clinically significant portal hypertension in the absence of actual and previous episodes of ascites and fluid retention. The remaining 16 Child class C cirrhotic patients (age 59.5 ⫾ 8.4 years, mean ⫾ SD, 3 positive for hepatitis B surface antigen, 11 HCV Ab⫹, 2 HCV Ab⫹ plus alcohol in the past, 14 men, and 2 women) were hospitalized for an episode of tense ascites and scheduled for total paracentesis. Diagnosis of cirrhosis and portal hypertension were based on clinical, biochemical, instrumental (diagnostic imaging, upper gastrointestinal endoscopy) and histologic criteria (when available). Exclusion criteria were arterial hypertension, diabetes, terminal hepatic failure, hepatic encephalopathy, liver cancer, heart and pulmonary diseases, major arrhythmias, use of vasoactive drugs, recent digestive hemorrhage, severe anemia, fever, spontaneous bacterial peritonitis, hyponatremia (⬍125 mEq Na⫹), serum creatinine ⬎2 mg%, and recent alcohol intake. Twenty subjects were used for comparison. Ten subjects (age 57.7 ⫾ 9.8 years, mean ⫾ SD, 8 men, and 2 women) had a mild congestive heart failure (New York Heart Association class II) caused by ischemic heart disease (n ⫽ 7) or idiopathic dilated cardiomiopathy (n ⫽ 3), i.e., a condition known to be characterized by a clear-cut increase in sympathetic activity to muscle11,12 but not to skin tis-

1113

1114 POZZI ET AL. sues.13 The other 10 subjects were healthy normotensive lean volunteers and thus reflected a group with normal sympathetic activity. In both heart failure and control subjects, hepatic and renal function were normal and so were blood chemistry values. The study protocol was approved by the Ethics Committee of our institution, and the procedures performed were in accordance with institutional guidelines. The subjects agreed to participate in the study after being informed of its nature and purpose. Measurements. Biochemical measurements consisted of standard liver function and blood chemistry variables (serum albumin, prothrombin time, hematocrit, platelet count, leukocyte count, bilirubin, blood ammonia, serum electrolytes, serum creatinine, blood urea nitrogen, liver function tests, etc.) all obtained from antecubital vein blood samples. Plasma norepinephrine (NE) concentration was assayed by high performance liquid chromatography14 after the antecubital vein blood was collected into ice-chilled tubes containing EGTA glutathione, immediately centrifuged at 3,500 rpm at ⫺3°C for 10 minutes and stored at ⫺80°C, until assayed. The same samples served for assaying plasma renin activity and plasma aldosterone concentration (RENCTK; Sorin Biomedica, Saluggia, Vercelli, Italy; Coat-A-count Aldosterone Kit; Diagnostic Products Corporation, Los Angeles, CA) as previously reported.15 Normal values for our laboratory are as follows: NE, 125 to 250 pg/mL; plasma renin activity, 0.1 to 2.0 ng/mL/h; and aldosterone, 10 to 60 pg/mL. Portal pressure measurement was obtained by catheterization of the suprahepatic veins. Briefly a 6F venous introducer (Cordis Corporation, Miami, FL) was inserted in the right jugular vein under local anesthesia and a 5F catheter (Cordis Europa N.V., LJ Roden, The Netherlands) was positioned under fluoroscopic control in a hepatic vein. The catheter was then exchanged with a 6F balloon catheter (Meditech; Boston Scientific Corporation, Watertown, MA), which was used for measurements. Measurement of hepatic venous pressure was obtained by means of a pressure recorder (Sirecust 1260; Siemens Medical Electronics, Danvers, MA) in the occluded position, filling the baloon with diluted contrast medium and then deflating the baloon in the free position, after controlling that the tip of the catheter was freely floating in the middle of the hepatic vein. Hepatic venous pressure gradient was calculated as the difference between occluded and free hepatic venous pressure (mm Hg). Three measurements were always performed and results were given as arithmetic means of these measurements. Portal pressure measurement was obtained only in Child class A cirrhotic patients. In Child class C cirrhotic patients with ascites, portal pressure was assumed to be markedly increased according to clinical, instrumental, and laboratory variables. Arterial blood pressure was measured beat-to-beat by a finger photoplethysmographic device (Finapres; Ohmeda 2300, Louisville, CO) capable of providing accurate and reproducible beat-to-beat systolic and diastolic values.16,17 Heart rate was continuously monitored by a cardiotachometer triggered by the R wave of an electrocardiogram lead. Muscle sympathetic nerve activity (MSNA) was obtained by the microneurographic technique, described in detail elsewhere.11,18 Briefly, multiunit recording of efferent MSNA was obtained, after a weak electrical stimulation (1 to 3 V, 0.2 ms, 1 Hz) aimed at inducing involuntary muscle twitches, through a tungsten microelectrode inserted in a muscle fascicle at the right or left peroneal nerve, posterior to the fibular head, as previously described.12,18 The fascicles impaled usually supply the anterior tibial or peroneal muscles of the leg. The obtained neurogram revealed spontaneous pulse-synchronous bursts characteristic of MSNA. Each burst represents a bulk of nerve action potentials from multiple fibers. The nerve signal was amplified ⫻70,000, fed through a band-pass filter (700 to 2,000 Hz), and integrated with a custom nerve traffic analysis system (Bioengineering Department, University of Iowa, Iowa City, IA). Integrated nerve activity was continuously monitored by a loudspeaker, displayed on a storage oscilloscope (Model 511A, Tektronix) (Model 511A, Tektronics, Heerenveen, the Netherlands) and was recorded along with heart rate and arterial blood pressure on a termic paper by an ink

HEPATOLOGY December 2001

polygraph (Gould 3800 ink recorder; Gould Inc., Cleveland, OH) with a paper speed of 5 mm/s. Repeated minor electrode adjustments were often required before the characteristic loud-speaker sound of pulse-grouped discharge indicated that the tip of the electrode had come close to a bundle of active sympathetic fibers. MSNA was assessed according to published criteria.12,18 The muscle (MSNA) or the skin (SSNA) nature of sympathetic nerve activity was assessed by the criteria detailed in previous studies.19-22 For MSNA the criteria were that (1) a weak electrical stimulation through the microelectrode elicited an involuntary muscle contraction but not paresthesias, (2) tapping or passive stretching of the muscle supplied by the nerve gave rise to afferent mechanoreceptive impulses, and (3) the recording consisted of spontaneous pulsesynchronous bursts that increased during held respiration. For SSNA criteria were that (1) electrical stimulation through the microelectrode induced skin paresthesias without concomitant muscle contraction, (2) light skin touching evoked afferent nerve impulses, and (3) tapping or passive stretching of the muscle supplied by the nerve did not cause afferent mechanoreceptive impulses. The recorded neurograms were accepted only if they did not show simultaneous SSNA and MSNA activity and if the signal-to-noise ratio exceeded the value of 3. Because bursts are pulse-synchronous, burst frequency may vary with heart rate. Although MSNA was quantified as burst incidence over time (bursts/min) and as burst incidence corrected for heart rate values (bursts/100 heart beats), SSNA was quantified as bursts per minute. The SSNA response to an acoustic stimulus (see below) was quantified by calculating the percent change in the amplitude of the bursts after the stimulus as compared with the mean amplitude of the spontaneous bursts occurring over the 3 minutes preceding the stimulus. Previous studies by our group23 and others24 have shown these measures to be highly reproducible at both shortterm and more long-term intervals. In our laboratory the within operator variation coefficient of the measurements (bursts/min) taken at 2 different sessions was 3.8%. This was confirmed in the 10 healthy normotensive control subjects in whom MSNA values were 33.5 ⫾ 2.8 and 34.6 ⫾ 2.7 bursts per minute and 48.9 ⫾ 3.4 and 50.1 ⫾ 3.2 bursts per 100 heart beats, when assessed in 2 different sessions spaced by an interval of 16.4 ⫾ 1.3 months. Protocol and Data Analysis. In Child class C cirrhotic patients diuretic treatment was withdrawn for a period of 4 days before the study whereas in heart failure patients oral furosemide was allowed but other drugs (ACE-inhibitors and vasodilators) were withdrawn with the same time schedule. All subjects were studied in the morning after a light breakfast in a semidark and quiet room kept at the constant temperature of 20°C to 21°C. The protocol of the study was as follows. (1) The subjects were placed in the supine position and were fitted with the antecubital vein catheter to allow blood sampling and with devices to measure finger arterial blood pressure and heart rate. (2) A blood sample for neurohormone assay was obtained 30 minutes thereafter. (3) The microelectrode was inserted into the peroneal nerve to obtain MSNA or SSNA, which was recorded along with finger blood pressure and heart rate for 30 minutes. (4) The microelectrode was then repositioned in the peroneal nerve fascicles to obtain the sympathetic nerve activity (MSNA or SSNA), which had not been obtained in the previous recording period again along with finger blood pressure and heart rate for 30 minutes. (5) At the end of the SSNA recording period, a 5-second unforeseen acoustic signal provided by an alarm clock was delivered to check the SSNA ability to increase.20 Each protocol step was spaced from the following one by a 15- to 20-minute interval. The echocardiographic (heart failure patients) and portal pressure (cirrhotic patients) data were obtained within the month before enrollment. Calculations of sympathetic nerve traffic were made by a single independent observer. Values from individual subjects were averaged for each group and expressed as means ⫾ SEM. The differences in mean values between groups were assessed by 2-way ANOVA. The Student’s t-test for unpaired observations was used to locate the statistical significance of the differences, following the Bonferroni

HEPATOLOGY Vol. 34, No. 6, 2001

POZZI ET AL.

1115

TABLE 1. Demographic, Clinical, Hemodynamic, Neurohumoral, and Microneurographic Data of Cirrhotic Patients and Heart Failure Patients

Child class A cirrhosis Child class C cirrhosis Heart failure

Patients

Age

Child/ NYHA

MAP

HR

12

58

A5-A6

99.6 ⫾ 4.6

67.3 ⫾ 2.8

16 10

60 58

C10-C13 2

98.3 ⫾ 4 96.5 ⫾ 2.1

75.8 ⫾ 3.1 85.2 ⫾ 3.7

MSNA Corr

SSNA

NE

PRA

ALDO

HVPG

41.6 ⫾ 2

62.1 ⫾ 3.8

11.8 ⫾ 3.3

476.6 ⫾ 45.5

2.5 ⫾ 1.1

62.5 ⫾ 11.7

14.8 ⫾ 1.2

52.9 ⫾ 4.7 60.3 ⫾ 2

70 ⫾ 5.1 71.4 ⫾ 2.5

11.3 ⫾ 2.5 16.8 ⫾ 2

564.3 ⫾ 77.8 456.6 ⫾ 44.4

10.1 ⫾ 1.5 1.8 ⫾ 0.5

518.4 ⫾ 128.4 276 ⫾ 30

MSNA

NOTE. Data are expressed as means ⫾ SEM. Abbreviations: Child, Child-Pugh score; NYHA, New York Heart Association classification; MAP, mean arterial pressure (mm Hg); HR, heart rate (beats/ min); MSNA, muscle sympathetic nerve activity (bursts/min); MSNA Corr, corrected MSNA (bursts/100 heart beats); SSNA, skin sympathetic nerve activity (bursts/min); NE, plasma norepinephrine (pg/mL); PRA, plasma renin activity (ng/mL/h); ALDO, plasma aldosterone (pg/mL); HVPG, hepatic venous pressure gradient (mm Hg).

correction for multiple comparisons. A value of P ⬍ .05 was taken as the level of statistical significance. RESULTS

Hemodynamic, clinical, neurohumoral and microneurographic data of cirrhotic and heart failure patients are presented in Table 1. As shown in Fig. 1 mean arterial pressure (diastolic plus one third of pulse pressure) was not significantly different in healthy subjects, Child class A and C cirrhotic patients, and heart failure patients, whereas heart rate was significantly greater in the latter 2 as compared with the former 2 groups. Compared with healthy subjects, Child class A cirrhotic patients showed no change in plasma renin activity and aldosterone but a marked increase in NE was seen. NE was similarly increased in Child class C cirrhotic patients with ascites who also showed, however, a marked increase in plasma renin activity and aldosterone. MSNA was obtained in all 48 subjects, whereas SSNA was obtained in all healthy, all heart failure, and in 13 of the 28 cirrhotic patients, because of technical difficulties to obtain adequate tracings. As shown in the original examples of Fig. 2 and in the average data of Fig. 3, compared with healthy subjects MSNA was increased in Child class A and more so in Child class C cirrhotic patients with ascites in whom it was only slightly less than in heart failure patients. SSNA in contrast, showed no increase in either cirrhotic subgroup, and indeed its value was not significantly different in all 4 groups studied. This was not caused by the inability of SSNA to increase, because following the acoustic stimulus integrated SSNA increased by 112.9 ⫾ 28% in healthy controls and by 115.7 ⫾ 27% in cirrhotic patients.

FIG. 1. Hemodynamic (upper panel) and neurohumoral variables (lower panel) in control subjects, cirrhotic (Child class A and C), and congestive heart failure (CHF) patients. HR, heart rate (beats per minute); MAP, mean arterial pressure (mm Hg). Data are mean ⫾ SEM. *P ⬍ .05; **P ⬍ .01 between groups.

DISCUSSION

Our study confirms that sympathetic activity is increased in patients in whom an advanced cirrhosis leads to ascites25 and further shows the increase to be so marked as to match the one seen in a condition paradigmatic of a marked sympathetic activation such as congestive heart failure.11,12 The most important new findings, however, are that (1) sympathetic activity is also increased, albeit to a smaller extent, in a less advanced stage of cirrhosis, i.e., a stage in which portal pressure is increased but there is no history or evidence of ascitic fluid formation; and (2) in both the advanced and the less advanced stages of cirrhosis the sympathetic activation is not generalized to the entire cardiovascular system because an increase in sympathetic nerve traffic to skeletal muscle tissues is accompanied by skin traffic comparable with that seen in control subjects. Our study was not designed to elucidate the factors responsible for the early but heterogeneous alteration in sympathetic activity seen in cirrhosis. We can speculate that in patients with ascites the MSNA increase mainly originates from the reduction in effective blood volume26 because this reduction (1) decreases the pronounced inhibitory influence cardiopulmonary volume receptors tonically exert on sympathetic nerve activity to skeletal muscle tissues and (2) stimulates (through a decrease in the reflex inhibition tonically elicited by cardiopulmonary receptors and activation of the macula densa mechanism) the secretion of renin from the kidney with a resulting increase in the sympathostimulatory effects of angiotensin II.27 However, this cannot entirely account for the MSNA increase seen in patients without ascites in whom

1116 POZZI ET AL.

HEPATOLOGY December 2001

FIG. 2. Original recordings of muscle (MSNA) and skin (SSNA) sympathetic nerve activity in healthy subjects (controls), Child class A and C patients, and heart failure patients.

blood volume was unlikely to be substantially modified and no stimulation of the renin-angiotensin system occurred. This is in favor of the hypothesis that the sympathetic activation precedes the renin-angiotensin-aldosterone system activation in cirrhosis, being thus theoretically responsible for the late activation of the renin-angiotensin-aldosterone system observed in the more advanced stages of liver disease, a phenomenon already described in mild and severe heart failure.12 Thus other possibilities should be considered. One of them is that an increase in portal vein pressure stimulates receptors with sympathetic afferents that cause reflex sympathetic excitation following integration of their signal at a spinal level.28 Another is that the MSNA activation is caused by the increase in plasma insulin concentration occurring early and late in cirrhosis29 as a result of reduced hepatic degradation of this substance due to parenchimal damage and/or portal-systemic shunting,30 because in humans insulin has a pronounced stimulating effect on sympathetic traffic supplying skeletal muscles.31 Indeed this possibility may also account for the unchanged skin sympathetic nerve traffic because in humans skin sympathetic nerve traffic is not altered by an increase in circulating levels of insulin.31 Several other points are worthy of a mention. First, in our cirrhotic patients with ascites the MSNA values were comparable with those reported by Floras et al. in similar patients.7 However, while we observed an MSNA increase also in preascitic cirrhosis, Floras et al. did not, probably because their study population was more heterogeneous including patients with postviral and mostly alcohol-induced cirrhosis. This may depend on the stringent selection criteria used in our study, that is, on the fact that portal pressure measurement allowed us to select cirrhotic patients without ascites only if they had clinically significant portal hypertension.32 This allows us to conclude that, regardless of the development of ascites, when portal hypertension is present, sympathetic activation already does occur. Second, cirrhosis was characterized by an increase in NE, which thus reflected the sympathetic activation occurring in

this condition. As in other diseases, however, the ability of NE to discriminate between different degrees of activation was not optimal because, at variance with MSNA, the increase in NE was not progressive from controls to preascitic and ascitic stages of cirrhosis. This may be, at least in part, because of the fact that NE has a lower reproducibility than MSNA, which makes it more vulnerable to chance alterations.23 Third, because in humans sympathetic nerve recording is limited to muscle and skin fibers, our study cannot answer the important question of whether in early and advanced cirrhosis sympathetic activity is increased also in cardiovascular districts such as the heart, the splanchnic area, the kidney, and the brain. This holds true also for possible differences in SSNA in different skin districts, although no microneurographic study has so far reported different regional effects on skin sympathetic nerve traffic of a given clinical condition or intervention.20,21 In humans, measurements of cardiac, renal, visceral, and cerebral sympathetic activity can only be obtained through catheterization of their arteries and veins to allow quantification of the organ NE spillover rate.33 It should be emphasized, however, that these invasive procedures may not be strictly necessary under the present circumstance, because a marked increase in NE, such as the one we have seen in our patients, suggests that sympathetic activity could be increased in many other organs in addition to muscles. Fourth, on the basis of the evidence that cutaneous vasodilatation characterizes cirrhosis, we would expect a decrease in SSNA in cirrhotic patients as compared with healthy controls. Our results, by showing that this is not the case, imply that skin vasodilatation is triggered, to a major extent, by an augmented secretion (or an augmented vascular responsiveness) of vasodilating substances and not by a reduced skin sympathetic nerve traffic. The sympathetic activation shown in preascitic cirrhosis has clinical implications; that is, through muscle constriction in the vessel walls, this activation may (1) raise venous pressure thereby favoring loss of fluid from the intravascular to the extravascular compartment, (2) constrict the portal vein sys-

HEPATOLOGY Vol. 34, No. 6, 2001

POZZI ET AL.

1117

and those Child class A patients worsening after any precipitating factor). The relatively small difference in sympathetic activation raises the question of whether this difference was to some degree masked by treatment. This is a reasonable possibility because chronic aldosterone antagonism, primarily used in the long-term control of ascites in all of our Child class C cirrhotic patients, has the potential to induce both portal pressure34 and total blood volume35 reduction and hence to favor sympathetic deactivation. This leads to the possibility that also in the preascitic stage of liver disease judicious use of drugs moderating sympathetic activity may be therapeutically useful in delaying fluid retention, the event that marks the change of prognosis in previously compensated cirrhotic patients.36 REFERENCES

FIG. 3. Sympathetic nerve activity to skeletal muscle (MSNA) and skin (SSNA) in control subjects, cirrhotic (Child class A and C) patients, and congestive heart failure (CHF) patients. MSNA is expressed both as bursts per minute and as bursts per 100 heart beats. SSNA is expressed only as bursts per minute. Data are mean ⫾ SEM. *P ⬍ .05; **P ⬍ .01 between groups.

tem thereby hampering preexisting portal hypertension, and (3) reduce renal blood flow and directly enhance tubular sodium and water reabsorption, thereby favoring fluid retention. All this may contribute to ending the compensated phase of disease and further aggravate the decompensated one. We expected to find a more marked difference in sympathetic activation in Child class A and Child class C patients (we intentionally did not enroll Child class B cirrhotic patients because of the heterogeneous nature of this class, which may include Child class C patients improving after treatment

1. Shrier 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;8:11511157. 2. Bichet DG, Van Putten VJ, Schrier RW. Potential role of increased sympathetic activity in impaired sodium and water excretion in cirrhosis. N Engl J Med 1982;307:1552-1557. 3. Henriksen JH, Ring-Larsen H, Kanstrup IL, Christensen NJ. Splanchnic and renal elimination and release of catecholamines in cirrhosis. Evidence of enhanced sympathetic nervous activity in patients with decompensated cirrhosis. Gut 1984;25:1034-1043. 4. Macgilchrist AJ, Howes LG, Hawksby C, Reid JL. Plasma noradrenaline in cirrhosis: a study of kinetics and temporal relationship to ascites formation. Eur J Clin Invest 1991;21:238-243. 5. Henriksen JH, Ring-Larsen H, Christensen NJ. Circulating noradrenaline and central hemodynamics in patients with cirrhosis. Scand J Gastroenterol 1985;20:1185-1190. 6. Nicholls KM, Shapiro MD, Van Putten MJ, Kluge R, Chung HM, Bichet DG, Shrier RW. Elevated plasma norepinephrine concentration in decompensated cirrhosis. Association with increased secretion rates, normal clearance rates, and suppressibility by central volume expansion. Circ Res 1985;56:457-461. 7. Floras JS, Legault L, Morali GA, Hara K, Blendis LM. Increased sympathetic outflow in cirrhosis and ascites: direct evidence from intraneural recordings. Ann Int Med 1991;114:373-380. 8. Morali GA, Floras JS, Legault L, Tobe S, Skorecki KL, Blendis LM. Muscle sympathetic nerve activity and renal responsiveness to atrial natriuretic factor during the development of hepatic ascites Am J Med 1991;91:383392. 9. Pozzi M, Grassi G, Pecci V, Turri C, Boari G, Bolla GB, Dell’Oro R, et al. Early effects of total paracentesis and albumin infusion on muscle sympathetic nerve activity in cirrhotic patients with tense ascites. J Hepatol 1999;30:95-100. 10. Arroyo V, Bernardi M, Epstein M, Henricksen JH, Schrier RW, Rode´s J. Pathophysiology of ascites and functional renal failure in cirrhosis. J Hepatol 1988;6:239-257. 11. Leimbach WN Jr, Wallin BG, Victor RG, Aylward PE, Sundlof G, and Mark AL. Direct evidence from intraneural recordings for increased central sympathetic outflow in patients with heart failure. Circulation 1986; 73:913-919. 12. Grassi G, Seravalle G, Cattaneo BM, Lanfranchi A, Vailati S, Giannattasio C, Del Bo A, et al. Sympathetic activation and loss of reflex sympathetic control in mild congestive heart failure. Circulation 1995;92:3206-3211. 13. Grassi G, Colombo M, Seravalle G, Spaziani D, Mancia G. Dissociation between muscle and sympathetic nerve activity in essential hypertension, obesity and congestive heart failure. Hypertension 1998;31:64-67. 14. Hijemdal P, Daleskog M, Kanan T. Determination of plasma catecholamines by high performance liquid chromatography with electrochemical detection: comparison with a radioenzymatic method. Life Sci 1979;25:131-138. 15. Pozzi M, Osculati G, Boari B, Serboli P, Colombo P, Lambrughi C, De Ceglia S, et al. Time course of circulatory and humoral effects of rapid total paracentesis in cirrhotic patients with tense, refractory ascites. Gastroenterology 1994;106:709-719. 16. Wesserling KH, De Witt B, Settles JJ, Klaver WH. On the indirect regis-

1118 POZZI ET AL.

17. 18. 19. 20. 21. 22. 23.

24. 25. 26. 27.

tration of finger blood pressure after Penaz. Funct Biol Med 1982;1:245250. Parati G, Casadei R, Groppelli A, Di Rienzo M, Mancia G. Comparison of finger and intra-arterial blood pressure monitoring at rest and during laboratory testing. Hypertension 1989;13:647-655. Vallbo AB, Hagbarth KE, Torebjork HE, Wallin BG. Somatosensory, proprioceptive and sympathetic activity from peripheral nerves. Physiol Rev 1979;59:919-957. Delius W, Hagbarth KE, Hongell A, Wallin GB. Manoeuvres affecting sympathetic outflow in human muscle nerves. Acta Physiol Scand 1972; 84:82-94. Delius W, Hagbarth KE, Hongell A, Wallin BG. Manoeuvers affecting sympathetic outflow in human skin nerves. Acta Physiol Scand 1972;84: 177-186. Hagbarth KE, Hallin RG, Hongell A, Torebjork HE, Wallin BG. General characteristics of sympathetic activity in human skin nerves. Acta Physiol Scand 1972;84:164-176. Delius W, Hagbarth A, Hongell A, Wallin BG. General characteristics of sympathetic activity in human muscle nerves. Acta Physiol Scand 1972; 84:65-81. Grassi G, Bolla GB, Seravalle G, Cattaneo BM, Lanfranchi A, Gennari A, Mancia G. Comparison between reproducibility and sensitivity of muscle sympathetic nerve traffic and plasma noradrenaline in man. Clin Sci 1997;92:285-289. Mark AL, Victor RH, Nerhed C, Wallin BG. Microneurographic studies of the mechanisms of sympathetic nerve responses to static exercise in humans. Circ Res 1985;57:461-469. Henriksen JH, Moller S, Ring-Larsen E and Christensen NJ. The sympathetic nervous system in liver disease. J Hepatol 1998;29:328-341. Henriksen JH, Bendtsen F, Sorensen TIA, Stadeager C and Ring-Larsen E. Reduced central blood volume in cirrhosis. Gastroenterology 1989;97: 1506-1513. Mancia G, Mark AL. Arterial baroreflexes in humans. In: Sheperd JT, Abboud FM, eds. Handbook of Physiology- Section 2: The Cardiovascu-

HEPATOLOGY December 2001

28.

29. 30. 31. 32.

33. 34. 35.

36.

lar System. Vol III. Bethesda: American Physiology Society 1983:755795. Mark A, Mancia G. Cardiopulmonary baroreflexes in humans. In: Shepherd JT, Abboud FM, eds. Handbook of Physiology-Section 2: The Cardiovascular System. Vol III. Bethesda: American Physiology Society 1983:795-813. Creutzfeld W, Frerichs H, Sickinger K. Liver diseases and diabetes mellitus. In: Popper H, Schaffner F, eds. Progress in Liver Diseases. Vol 3. New York: Grune and Stratton. 1970;380-407. Johnston DG, Alberti KG, Faber OK, Binder C, Wright R. Hyperinsulinism of hepatic cirrhosis: diminished degradation or hypersecretion? Lancet 1977,i :10-13. Berne C, Fagius J, Pollare T, Hjemdal P. The sympathetic response to euglycaemic hyperinsulinemia: evidence from microelectrode recordings in healthy subjects. Diabetologia 1992;35:873-879. Feu F, Garcia-Pagan JC, Bosch J, Luca A, Tere`s J, Escorsell A, Rode`s J. Relation between portal pressure response to pharmacotherapy and risk of recurrent variceal hemorrage in patients with cirrhosis. Lancet 1995; 346:1056-1059. Henricksen JH, Christensen NJ, Ring-Larsen H. Noradrenaline and adrenaline in various vascular beds in patients with cirrhosis. Relation to haemodynamics. Clin Physiol 1981;1:293-304. Garcia-Pagan JC, Salmero`n JM, Feu F, Luca A, Gine`s P, Pizcueta P, Claria J, et al. Effects of low sodium diet and spironolactone on portal pressure in patients with compensated cirrhosis. HEPATOLOGY 1994;19:1095-1099. Okumura H, Arakami T, Katsuta Y, Satomura K, Akaike M, Sekiyama T, Terada H, et al. Reduction in hepatic venous pressure gradient as a consequence of volume contraction due to chronic administration of spironolactone in patients with cirrhosis and no ascites. Am J Gastroenterol 1991;86:46-52. Llach J, Gine`s P, Arroyo V, Rimola A, Tito` LL, Badalamenti S, Jime`nez W, et al. Prognostic value of arterial pressure, endogenous vasoactive systems and renal function in cirrhosis with ascites. Gastroenterology 1988; 94:482-487.