Responses to endothelin-1 in patients with advanced cirrhosis before and after liver transplantation

In response to Helmy’s comments in his recent letter (Gut 2004;53:470–1), we wish to emphasise the following points, many of which were clearly stated in our original paper.¹

In agreement with the comments on “generalised vasodilatation” in cirrhosis, we made it clear that basal forearm blood flow was normal in our patient cohort despite the presence of a vasodilated circulation, as evidenced by a reduced systemic vascular resistance index. As pointed out in our paper, this observation is consistent with findings of previous studies and suggests that beds other than the forearm circulation, such as those of the splanchnic and pulmonary circulation, were dilated in our patients.

With regard to the issues raised about the use of one arm plethysmography, our own results and those of others have shown that under well controlled circumstances the effects of external stimuli on results obtained using this approach are minimal.² Indeed, in our study, the results of forearm plethysmography were very consistent across both the control and cirrhotic patient groups.¹ However, we acknowledge that single forearm plethysmography could be affected by changes in systemic haemodynamics (due for example to the effects of drug infusions). However, as stated in the text, neither heart rate nor blood pressure altered significantly throughout the course of the experiment.

In forearm resistance arteries (and elsewhere), ET_B receptors on vascular smooth muscle and endothelial cells mediate opposing effects on vascular tone.³ Thus ET_B blockade could result in either vasodilatation or vasoconstriction, depending on which receptor subpopulation is most affected. In our hands, preliminary experiments with the ET_B blocker BQ788 yielded ambiguous responses, even in control subjects, causing vasoconstriction in some and vasodilatation in others. Until a selective ET_B receptor antagonist (for VSMC or endothelium) is available, interpretation of the results obtained using BQ788 remains difficult.

As for the concern that similar vasodilatation was observed with endothelin-1 (ET-1) and BQ123 (an ET_A antagonist), we wish to re-emphasise these were two very different experiments in two separate groups of patients, asking two different questions. We observed:

1. that ET-1 infusion in these advanced cirrhotics produced mild vasodilatation; and

2. in similar patients, there was no difference between cirrhotics and controls in the effects of BQ123 on vascular tone.

We put these two results together to propose that it is likely that the abnormal response to ET-1 infusion reflects alterations in ET_B mediated responses in cirrhotics (either via receptor changes or downstream pathways such as changes in nitric oxide synthesis, prostanoid production, or endothelium derived hyperpolarising factor).

As pointed out by Helmy, there was an early dilatory response following ET-1 infusion in cirrhotics. This is not totally unexpected as it has previously been demonstrated that ET-3 (an ET_B receptor agonist) causes early vasodilation in control subjects; similarly, a trend towards an early vasodilatory effect of ET-1 has been observed in healthy subjects.³

Regarding the use of concomitant drug therapy, all medications were ceased more than 24 hours prior to the experiments. While some residual effect of these agents is possible, more prolonged cessation of drug therapy in these decompensated patients was not considered safe or ethical. With regard to measurement of ET-1, as detailed in our paper, a commercially available assay with cross reactivity between big ET-1 and ET-1 was used. The study was not powered (nor was it designed to) to pick up small differences in brachial artery ET-1 levels.