Over the past 20 years we have moved from a situation in which we had no therapy for alcoholic liver disease, through a period when any therapy we had was purely empirical, to an era where we have ...specific therapies for different aspects of this disease based upon sound pathogenic principles. In this short review, an attempt has been made to summarize these advances in the understanding of the pathogenesis of alcoholic liver disease. In particular, they explain why patients with severe acute alcoholic hepatitis continue to deteriorate in hospital despite withdrawal from alcohol, why they respond to corticosteroids, why only a small percentage of patients develop cirrhosis, and why propylthiouracil may offer protection.
It is possible that abnormalities in atrial natriuretic peptide may be involved in the pathogenesis of sodium retention in edema states. We performed a study in a group of 12 sodium-retaining ...cirrhotic subjects to determine the role of this peptide in mediating differences in the natriuretic response to central volume expansion induced by head-out water immersion.
Each patient was maintained for seven days on a 20-mmol sodium intake, and then studied on both control and immersion days. On each day, measurements of the following were obtained: plasma atrial natriuretic peptide, hematocrit, electrolytes, creatinine, plasma renin activity, serum aldosterone, urinary cyclic guanosine monophosphate (cGMP), blood pressure, and pulse rate.
In six subjects, immersion resulted in a marked natriuresis sufficient to induce negative sodium balance by the third hour, and these subjects were termed "responders." In these six patients, baseline pre-immersion levels of plasma renin activity and serum aldosterone were all below 3 ng/liter/second and 4 nmol/liter, respectively. In the other six subjects, the natriuretic response to immersion was markedly blunted and insufficient to induce negative sodium balance, and these subjects were termed "non-responders." In these subjects, baseline pre-immersion levels of plasma renin activity and aldosterone were all above 3.5 ng/liter/second and 5 nmol/liter, respectively, and were significantly elevated compared with the responders, and compared with the normal range for control subjects consuming the same sodium intake. In both groups of cirrhotic subjects, baseline levels of plasma atrial natriuretic peptide and cGMP excretion were significantly and comparably elevated compared with the normal range for control subjects ingesting the same sodium intake. Despite the marked difference in the natriuretic response to immersion in both responders and non-responders, there was a significant and comparable further elevation of plasma atrial natriuretic peptide and urinary cGMP excretion during immersion, compared with the control day.
These results suggest that the relative resistance to the natriuretic action of atrial natriuretic peptide in the non-responders compared with the responders is mediated by anti-natriuretic factors acting at a level parallel with or beyond atrial natriuretic peptide release or coupling to its cGMP-linked receptors.
Despite intensive investigation, the pathogenesis of sodium retention in patients with chronic liver disease is not fully known. We have studied 19 chronic liver disease patients, 13 without (group ...1) and six with (group 2) histories of clinical sodium retention (ascites or edema) by varying dietary sodium intake. The patients were placed on a 20 mmol/day constant diet for 1 wk, followed by a constant 100 mmol/day sodium diet for 1 wk under strict metabolic conditions. After 5 days of equilibration on each diet, blood and urine samples were collected for plasma atrial natriuretic factor levels and urinary sodium excretion. Group 1 patients (n = 6) achieved near sodium balance in 5 days on both a 20-mmol (urinary sodium output = 17 +/- 3 mmol/day) and a 100-mmol sodium diet (urinary sodium output = 80 +/- 5 mmol/day). Atrial natriuretic factor levels in these patients tended to be elevated, but the increase was not significantly greater than that in normal control subjects (10 +/- 4 pg/ml to 19 +/- 4 pg/ml) on the same diets. In contrast, group 2 patients (n = 5) were in significant positive sodium balance on both the 20 mmol/day sodium diet (mean urinary sodium output = 9.5 +/- 3.3 mol/day) and the 100 mmol/day sodium diet (urinary sodium output = 37 +/- 13 mmol/day). This occurred despite significantly elevated baseline atrial natriuretic factor levels and a significant increase in plasma atrial natriuretic factor levels after sodium challenge (62 +/- 9 pg/ml, p less than 0.05) on a 100 mmol/day sodium diet.
We have previously shown that unresponsiveness to atrial natriuretic factor is a marker of the severity of ascites. The tubular mechanisms are unknown, but it seems that increased reabsorption of ...sodium proximal to the main site of action of atrial natriuretic factor (i.e., the inner medullary collecting duct) plays an important role. We attempted to decrease the proximal reabsorption of sodium with mannitol in patients unresponsive to atrial natriuretic factor. The results of mannitol in such a group of patients has previously been conflicting. We studied 10 patients with massive, resistant ascites who were off diuretics and on a 20-mmol/day sodium diet for 7 days. Atrial natriuretic factor unresponsiveness was confirmed by failure of a 2-hr atrial natriuretic factor infusion to induce a natriuresis. The next day all patients received an infusion of 40 gm of mannitol and subsequently a combined infusion of mannitol and atrial natriuretic factor. Proximal reabsorption of sodium and water were evaluated by lithium clearance, and glomerular filtration rate and renal blood flow were evaluated by inulin clearance and p-aminohippurate clearances, respectively. Six patients responded to mannitol alone with an increased diuresis (from 39 +/- 7 to 148 +/- 35 ml/hr) and natriuresis (from 0.27 +/- 0.05 mmol/hr to 1.65 +/- 0.53 mmol/hr; p less than 0.05) (responders), whereas four did not (nonresponders). The combination of atrial natriuretic factor and mannitol induced a further significant increase in sodium excretion (3.28 +/- 0.68 mmol/hr) but not in urine excretion, compared with mannitol alone.
Resistance to the natriuretic action of atrial natriuretic factor (ANF) in cirrhosis with ascites has been correlated with rising levels of antinatriuretic factors, such as renin, angiotensin II ...(AII), and aldosterone, as well as increased sympathetic nerve activity. To determine whether AII can serve as a mediator rather than only as a marker of the antinatriuresis, a nonpressor dose of AII (5 ng/kg/min) was given during an ANF infusion in eight patients with cirrhosis and ascites who responded to ANF infusion with a natriuresis. Patients were maintained in metabolic balance and measurements of para-aminohippuric acid, inulin, and lithium clearance were taken before and during infusion of ANF with or without AII. Atrial natriuretic factor infusion was associated with a natriuretic response accompanied by an increase in glomerular filtration rate, filtration fraction, and lithium clearance compared with baseline. The addition of AII was associated with a return of the glomerular filtration rate to baseline, with no change in filtration fraction. This was reversible on withdrawal of AII infusion. Natriuresis induced by ANF occurred despite baseline elevations of the renin angiotensin aldosterone system and was associated with an increase in distal delivery of sodium and a decrease in fractional reabsorption of distally delivered sodium as estimated by lithium clearance parameters. Angiotensin II infusion exerted effects on both proximal and distal nephron sites to abrogate ANF-induced natriuresis. These results suggest that AII may serve as a mediator as well as a marker of resistance to the natriuretic effect of ANF in patients with cirrhosis and ascites.
Sodium retention in cirrhosis has been attributed to an imbalance between vasoconstrictive antinatriuretic forces such as the sympathetic nervous system and vasodilatory natriuretic agents such as ...atrial natriuretic factor (ANF). With the development of refractory ascites, cirrhotic patients become unresponsive to the natriuretic effect of ANF. Animal data suggest that the sympathetic nervous system plays a key role in mediating the refractoriness to ANF. We therefore studied the relationship between sympathetic nerve activity (SNA) and the natriuretic response to ANF in normal subjects and cirrhotic patients. We also attempted to localize the intrarenal site of refractoriness to ANF by lithium clearance.
Twenty-six patients with biopsy-proven cirrhosis and seven age- and sex-matched normal volunteers were studied after a week of 20 mmol/day sodium intake and no diuretics. Muscle SNA was recorded from the peroneal nerve (microneurography) and correlated with responsiveness to a 2-hour ANF infusion. Lithium clearance was used as a marker of sodium reabsorption proximal to the intramedullary collecting duct, the main site of ANF action. Plasma norepinephrine, renin, and aldosterone levels were also determined. Patients were categorized into three groups: nine patients free of ascites (by ultrasonography), five ascitic patients who responded to a 2-hour ANF infusion (i.e., had a natriuretic response to ANF above 0.83 mmol/hour), and 12 ascitic patients who did not respond.
Muscle SNA was greatly increased in the ascitic nonresponder patients compared with the normal subjects (64 +/- 4 versus 27 +/- 7 bursts/minute, p less than 0.001), moderately increased in ascitic responders (47 +/- 6 bursts/minute, p less than 0.05), but not significantly increased in nonascitic patients with cirrhosis (34 +/- 5 bursts/minute). SNA was positively correlated with plasma norepinephrine levels (r = 0.69; p less than 0.005) and inversely correlated with peak sodium excretion during the ANF infusion (r = -0.63; p less than 0.001). Plasma renin activity and aldosterone were markedly elevated in ascitic nonresponders, and normal in ascitic responders and nonascitic patients. Lithium clearance was reduced in ascitic patients compared with nonascitic patients, did not change after the ANF infusion, and correlated inversely with SNA (r = -0.61; p less than 0.01).
These results support the concept that the sympathetic nervous system is a factor in renal sodium handling in cirrhosis, especially in the initiation of sodium retention and the development of refractory ascites. Refractoriness to ANF might be explained, at least in part, by increased neurally mediated sodium reabsorption proximal to the intramedullary collecting duct, the main site of ANF action.
It has been postulated that one of the mechanisms of hypotension associated with cirrhosis is an attenuated responsiveness to catecholamines despite the increased activity of the sympathetic nervous ...system and the elevated plasma concentrations of the sympathetic neurotransmitter, norepinephrine. This abnormality was studied in a dog model of portal hypertension and cirrhosis. Twelve weeks after bile duct ligation (n = 16), intrasplenic pressure rose significantly from 6.3 +/- 0.4 to 14.6 +/- 1.6 mmHg (p < 0.05), mean arterial pressure had fallen from 106 +/- 4 to 83 +/- 8 mmHg (p < 0.01), cardiac output had risen from 3.1 +/- 0.2 to 3.8 +/- 0.8 l/min (p < 0.05) and plasma norepinephrine concentrations rose from 0.22 +/- 0.12 to 1.17 +/- 0.52 nmol/l (p < 0.05). In 7 sham-operated dogs, the changes in these 4 variables over the same period were non-significant. In vivo pressor responsiveness was tested by studying the effects of intravenous and intra-arterial infusions of norepinephrine and the non-selective beta-adrenoceptor agonist, isoproterenol. In vitro responsiveness was tested by measuring the effects of isoproterenol on the isometric twitch of isolated ventricular strips and the effects of norepinephrine on femoral, mesenteric and renal arterial rings. There was no significant change in the in vivo responses of chronic bile-duct-ligated dogs at 12 weeks compared to the preoperative assessment, or to sham-operated dogs at 12 weeks. Furthermore, there was no significant difference between the in vitro responses of ventricular strips to isoproterenol or arterial rings to norepinephrine prepared from chronic bile-duct-ligated and sham-operated dogs.
Portal hypertension (PHT) is known to be associated with a hyperdynamic circulation, yet the pathogenesis of both remains unclear. Therefore, we have studied serially the relationship between portal ...pressure and in vitro peripheral vascular responsiveness in an animal model of presinusoidal PHT. In rats with partial portal vein stenosis (PPVS) or sham-operated (SO) controls, we studied contractile responses to cumulative doses of norepinephrine (NE) and to a single dose of 0.8 microM NE of 20-mm helically cut strips of thoracic aorta. At both 2 and 10 days postoperatively, the portal pressures (mean intrasplenic pressure) in PPVS, 14.3 +/- 1.5 mmHg and 14.1 +/- 1.3 mmHg were significantly elevated compared with SO controls, 7.6 +/- 0.6 mmHg (P less than 0.005) and 9.7 +/- 0.5 mmHg (P less than 0.01), respectively. Yet, there was no significant differences between the two groups in the Rmax and ED50 of the contractile response curves to cumulative doses of NE as well as in the fast (phase 1) and slow phase (phase 2) of the contraction to 0.8 microM NE. In contrast, at 21 days, portal pressure in the PPVS rats fell to 11.0 +/- 0.7 mmHg but remained significantly higher than that of the SO controls, 8.2 +/- 1.0 mmHg (P less than 0.05). This occurred in conjunction with a significant decrease in Rmax 698 +/- 87 mg (PPVS) vs. 1148 +/- 92 mg (SO); P less than 0.005 but no significant change in ED50, indicating a decreased sensitivity to NE due to an alteration in alpha-adrenoreceptor function.