Editorial: measuring inflammatory and fibrotic components of portal hypertension - a non-invasive hepatic venous pressure gradient? Authors' reply.

Abstract

We appreciate the editorial comments of Drs Snowdon and Fallowfield to our paper.1, 2 Non-invasive diagnosis of portal hypertension is advancing with the goal of a non-invasive test, which is precise and accurate, clinically easy to use and with a low cost. Our study shows that the goal is coming closer in terms of reliability and ease of use, and confirms that a combination of markers based on the pathobiology of portal hypertension is a viable approach. However, we expect and look forward to further developments aiming at all the criteria for the optimal test. There are several methods available for assessing the fibrous/structural component of portal hypertension, Enhanced Liver Fibrosis, Fibrotest, transient elastography, shear wave elastography, etc., and although none of these can serve as stand alone for the purpose they will undoubtedly be involved in future studies. For the dynamic component, focus is traditionally mostly on endothelial dysfunction as increased production of vasoconstrictors (e.g. endothelins, angiotensin-II, norepinephrine, thromboxane A2) and reduced release of vasodilators are thought to be instrumental. However, hepatic inflammation seems to be a major part of the dynamic component and therefore we selected the circulating macrophage activation marker sCD163 for our study. sCD163 has proved to be a very interesting biomarker in liver diseases.3-5 We and others have showed that blood sCD163 is independently linearly related to the degree of portal hypertension and with the development of decompensation including variceal bleeding.6-10 Likewise, we have showed that sCD163 is elevated in acute-on-chronic liver failure in relation to disease severity and mortality.11 However, the dynamic utility of sCD163 in cirrhosis patients remains unclear and as noted by Drs Snowdon and Fallowfield, sCD163 does not fall after TIPSS. This may indicate that the hepatic inflammation in cirrhosis is not caused by portal hypertension per se and that macrophages may become constitutively activated. This may impair the utility of sCD163 as a marker to evaluate treatment responses, such as to, e.g. beta-blockers. Thus, although our score for clinically significant portal hypertension has a tolerable predictive value for the cirrhosis aetiologies we studied, the pathogenic processes driving portal hypertension may be different in different liver diseases or stages, which may necessitate more detailed expansion of the biomarkers used for the purpose. Improvement over what we achieved of as well sensitivity as specificity of a score is highly desirable for clinical decision-making and of large potential consequences. In conclusion, the classical invasive hepatic venous pressure gradient measurements will remain a key procedure for diagnosis and treatment of portal hypertension in tertiary liver centres, but this does not broadly benefit the general population of patients with portal hypertension. Therefore, to harvest the potential for improvement of clinical management of all cirrhosis patients, the development, validation and implementation of non-invasive markers of portal hypertension is a prerequisite. We believe that the combined score we propose is one step down this road although the end of the road may be a long way. The authors’ declarations of personal and financial interests are unchanged from those in the original article.2