Liver disease in cystic fibrosis presents as non-cirrhotic portal hypertension

Abstract

Cystic fibrosis-related liver disease (CFLD) is allegedly caused by small bile duct obstruction leading to focal biliary fibrosis and eventually to cirrhosis with portal hypertension (PHT). However, CFLD patients typically present with PHT, without cholestasis and with preserved liver function. End-stage liver disease is exceptional. We previously suggested that CFLD-related PHT could be a vascular disease, more specifically non-cirrhotic PHT (NCPH) [[1]Witters P. Libbrecht L. Roskams T. Boeck K.D. Dupont L. Proesmans M. et al.Noncirrhotic presinusoidal portal hypertension is common in cystic fibrosis-associated liver disease.Hepatology. 2011; 53: 1064-1065Crossref PubMed Scopus (35) Google Scholar]. Indeed, other groups have also reported that cirrhosis is present in only 27% of patients with PHT, even when performing dual pass biopsies [[2]Pereira T.N. Lewindon P.J. Greer R.M. Hoskins A.C. Williamson R.M. Shepherd R.W. et al.Transcriptional basis for hepatic fibrosis in cystic fibrosis-associated liver disease.J Pediatr Gastroenterol Nutr. 2012; 54: 328-335Crossref PubMed Scopus (22) Google Scholar]. In the present work, we studied 8 patients with CFLD and PHT in detail. In 4 patients we performed transjugular hepatic venous portal gradient (HVPG) measurements and biopsies. Furthermore, we studied 5 CFLD explant livers. Details on the materials and methods: see online Supplementary materials. In the 4 patients with HVPG measurements, PHT was diagnosed because of esophageal varices (with bleeding in 2 patients), splenomegaly, abdominal venous collaterals and/or ascites. None had hepatic encephalopathy and Model for End-Stage Liver Disease (MELD) scores were low (range 6–9), confirming the preserved liver function. All had normal bilirubin levels. In all 4 patients, HVPG was 4–9 mm Hg, clearly below the cut-off for clinically significant PHT. In the absence of a portal vein thrombosis, such low HVPG can only be explained by a presinusoidal type of portal hypertension [[3]Khanna R. Sarin S.K. Non-cirrhotic portal hypertension — diagnosis and management.J Hepatol. 2014; 60: 421-441Abstract Full Text Full Text PDF PubMed Scopus (210) Google Scholar]. On histology, none of them had cirrhosis, while vascular changes were found (Fig. 1; detailed results in online Supplementary Table 1). All of the above findings constitute NCPH [3Khanna R. Sarin S.K. Non-cirrhotic portal hypertension — diagnosis and management.J Hepatol. 2014; 60: 421-441Abstract Full Text Full Text PDF PubMed Scopus (210) Google Scholar, 4Verheij J. Schouten J.N.L. Komuta M. Nevens F. Hansen B.E. Janssen HLa et al.Histological features in western patients with idiopathic non-cirrhotic portal hypertension.Histopathology. 2013; 62: 1083-1091Crossref PubMed Scopus (50) Google Scholar]. Explant livers of 5 patients with PHT (MELD range 7–12) did not reveal cirrhosis either, only mild to moderate fibrosis (F2–F3) or incomplete septal cirrhosis (n = 2) (Fig. 1). The same vascular changes as seen in the biopsies were present. Obliterative venopathy with dense fibrosis within portal vein branches or the disappearance of portal vein branches was evident in all tissue samples (Fig. 1; detailed results in online Supplementary Table 1). In three patients, calcifications in the larger portal vein branches were present. They were also visible on abdominal imaging (Fig. 1). This obliterative venopathy can indeed be held accountable for the PHT, as is typically seen in NCPH [[4]Verheij J. Schouten J.N.L. Komuta M. Nevens F. Hansen B.E. Janssen HLa et al.Histological features in western patients with idiopathic non-cirrhotic portal hypertension.Histopathology. 2013; 62: 1083-1091Crossref PubMed Scopus (50) Google Scholar]. In our patient cohort, both haemodynamic (HVPG) measurements and histological findings are identical to those in NCPH. Hence, in at least a subset of CFLD patients, CFLD-related PHT is not due to the paradigmal “biliary” cirrhosis but corresponds to vascular damage as in NCPH. Although PHT clearly precedes the development of cirrhosis and end-stage liver failure in our patients, which percentage of PHT patients will go on to develop cirrhosis remains to be determined. The underlying pathogenesis of NCPH and CFLD remains to be elucidated. Recently, an association between gut inflammation and CFLD has been demonstrated [[5]Flass T. Tong S. Frank D.N. Wagner B.D. Robertson C.E. Kotter C.V. et al.Intestinal lesions are associated with altered intestinal microbiome and are more frequent in children and young adults with cystic fibrosis and cirrhosis.PLoS One. 2015; 10e0116967Crossref Scopus (56) Google Scholar] and this could contribute to the development of NCPH [[3]Khanna R. Sarin S.K. Non-cirrhotic portal hypertension — diagnosis and management.J Hepatol. 2014; 60: 421-441Abstract Full Text Full Text PDF PubMed Scopus (210) Google Scholar]. These observations may have important consequences for the management of CFLD. The efficacy of ursodeoxycholic acid as treatment of CF-related “cholestasis” has been questioned repeatedly, and indeed may target the wrong pathophysiological mechanism. For more advanced stages of CFLD, porto-systemic shunting procedures to alleviate the PHT, might be preferable over transplantation given the absence of cirrhosis and the preserved liver function, similar to the situation in NCPH. The data presented here open new avenues for research into new therapies which are desperately needed. The following are the supplementary data related to this article.