Decreasing hepatic vascular tone by liver-specific NO donors: wishful thinking or a promising reality?

Abstract

The past 20 years have witnessed two major steps forward in the understanding of the pathophysiology of portal hypertension. The first was the characterisation that the splanchnic blood flow was markedly increased in portal hypertension due to a pronounced splanchnic vasodilatation, and that such increased portal vein inflow contributed to the maintenance and aggravation of portal hypertension [1,2]. The second was that contrary to what was assumed, increased hepatic resistance in cirrhosis not was only a mechanical consequence of the architectural disruption of the hepatic vascular bed due to fibrosis, scarring and nodule formation, but that the increased hepatic vascular resistance was in part due to a high hepatic vascular tone, which could be decreased by vasodilators [3]. The finding that portal hypertension was supported by an increased portal blood inflow provided the rationale for therapies based on the use of splanchnic vasoconstrictors, among which the introduction of non-selective betablockers for the long-term treatment of portal hypertension is the best example of a successful translation into clinical practice of pathophysiological concepts derived from studies in animal models [4]. Its academic impact is demonstrated by over 800 published papers in this field since then. The finding that an active increase of the hepatic vascular tone contributed to the increased portal pressure in cirrhosis provided the rationale for testing different vasodilators [5,6]. It was expected that by increasing blood flow to the liver, vasodilators would not only decrease portal pressure but also improve liver function [7]. However, the clinical impact of this approach has been limited by the fact that all available vasodilators cause some degree of systemic and splanchnic vasodilatation and induce hypotension, which may cause or worsen sodium retention, ascites and renal function abnormalities in patients with cirrhosis [8,9]. In this context, the recent demonstration that an insufficient intrahepatic release of nitric oxide (NO) was a major player in the increased hepatic vascular tone of cirrhosis opened new perspectives [8,10,11]. The ideal situation would be to supplement the liver with enough NO without increasing NO in the splanchnic and/or systemic circulation, where its synthesis is already excessive and contributes to splanchnic and systemic vasodilatation [8,12]. Such an agent would be close to being the ideal drug for the treatment of portal hypertension. How close are we from that goal? The first attempt was the oral administration of isosorbide-5-mononitrate (ISMN) a NO-donor that is metabolised by the liver [13]. It was proposed that at low doses, ISMN would be able to sufficiently increase hepatic NO delivery as to reduce the hepatic vascular tone with minimal systemic spill over. In fact, ISMN does reduce portal pressure [13], but even at very low doses it induces a statistically significant reduction of arterial pressure [14], which makes this compound inadequate as single therapy, since achieving a significant fall in portal pressure would require the use of doses that cause marked systemic hypotension [13]. ISMN, however, may be used in association with non-selective beta-blockers, since it enhances the portal pressure reduction caused by the betablockers [14–16] while renal function and arterial pressure are not altered, due to the fact that these agents increase the peripheral vascular resistance and prevent the activation of the renin-angiotensin system [17]. This drug association has been used successfully in the prevention of recurrent variceal bleeding [18,19], but with controversial results in primary prophylaxis [20–22]. However, side-effects from ISMN cause withdrawal of therapy in a substantial number of patients. It was clear therefore, that conventional organic Journal of Hepatology 39 (2003) 1072–1075 www.elsevier.com/locate/jhep