Article Text


Responses to endothelin-1 in patients with advanced cirrhosis before and after liver transplantation
  1. A Helmy
  1. Correspondence to:
    Dr A Helmy
    Gastroenterology and Tropical Medicine Department, 6th Floor, Assiut University Hospital, Assiut 71111, Egypt;

Statistics from

I read with interest the article of

and was pleased to see my novel studies1,2 partially reproduced in patients with decompensated cirrhosis. I disagree with some of the results as the study involves substantial design, methodology, and analysis problems.

The authors said that advanced cirrhotic patients have “generalised vasodilatation”. Vasodilatation does occur in these patients but only in the splanchnic and pulmonary beds.3 Indeed, studies by me and others have shown vasoconstriction in the brachial, femoral, cerebral, and renal territories, especially in advanced cirrhosis.4–6 Therefore, I would like to stress the point that with advancing cirrhosis, further activation of the neurohumoral systems occurs, with consequent peripheral vasoconstriction. However, blood pooling, particularly in the splanchnic bed, lowers systemic vascular resistance.

A major criticism of the study of Vaughan et al is that they measured forearm blood flow (FBF) in only one arm. Changing levels of alertness and external stimuli produce similar fluctuations in blood flow of both arms, and lead to significant misleading alterations in the measured responses if unilateral measurements are used. Thus responses to intra-arterial infusions should have been measured in both arms with the results expressed as ratios of concurrent FBF in the infused and non-infused arms, where the latter serves as a contemporaneous control for the drug effects in the former.7–9 Furthermore, FBF ratios are significantly more reproducible than unilateral FBF measurements both at rest and following infusion of vasoconstrictors.9

The authors demonstrated a surprising increase in FBF (~35–40%) in response to infusion of a locally active dose of the potent vasoconstrictor endothelin-1 (ET-1), which reached its maximum within five minutes from the start. They attributed their finding to enhanced ETB receptor mediated vasodilatation. This needs to be tested by selectively blocking ETB receptors, using BQ-788. To date, upregulation of ETB receptors has been reported in the splanchnic and pulmonary vasculature but not in the forearm.10,11 How can the maximum response to the slowly acting ET-1 be reached within five minutes? Also, dose-response curves of the effects of ET-1 and BQ-123 should have been performed.

According to the authors, ETA receptor mediated responses were unaltered while those mediated by the ETB receptor were enhanced in patients with decompensated cirrhosis. Thus one would expect that blocking ETA receptors with BQ-123 would allow ET-1 to act unopposed on ETB receptors and produce enhanced vasodilatation. However, this was not the case (fig 2 in the article). What adds to my surprise here is that BQ-123 infusion also increased FBF by ~35–40%. How can infusion of ET-1 produce the same per cent change in FBF as infusion of its selective ETA receptor antagonists?!

Many of the included patients were receiving diuretics, β-blockers, and immunosuppressive medications, which were withheld only on the day of testing. These medications affect circulating volume, vascular tone, and the activity of the neurohumoral systems. To eliminate these effects, drugs need to be stopped for at least five times their half life. Alternatively, control subjects on the same medications should be used (for example, renal transplant recipients with a normal liver).

Vaughan et al reported normal plasma ET-1 concentrations in decompensated cirrhosis, without measuring preproendothelin-1 mRNA or big ET-1, the biological precursor of ET-1. Due to its autocrine, paracrine, and endocrine nature, plasma concentrations of ET-1 alone do not reflect the activity of the endothelin system or the status of ET-1 production.12 This should have been stated by the authors. I also recommend collecting samples in tubes containing 1000 KIU aprotinin and EDTA.

In conclusion, the scientific contents of this article would have been greater if the authors had: (1) measured FBF in both arms; (2) presented their data as per cent change in the ratio of flows in both arms at every time point; (3) assessed plasma big ET-1 or preproendothelin mRNA concentrations; (4) examined the responses to an ETB receptor antagonist; (5) performed a dose-response curve; and (6) selected a comparable control group on similar medications as the patients.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles