Noninvasive predictors of fibrosis in NASH with and without cirrhosis, just as good as histology (and hepatic venous pressure gradient?)
C. Ripoll
DOI: https://doi.org/10.1002/hep.28273
IF: 17.298
2016-02-01
Hepatology
Abstract:Nonalcoholic fatty liver disease (NAFLD) is an increasingly common cause of end-stage liver disease. Although it was initially considered that patients with NAFLD without inflammation were not at risk of developing cirrhosis, recent studies emphasize that the sole presence of fibrosis is an indepedent predictor of death in these patients. Furthermore, patients with NAFLD may have portal hypertension, even though no cirrhosis or advanced fibrosis is observed on the liver biopsy. In this context, a recent study from Sebastiani et al. has evaluated the prognostic value of noninvasive tools to estimate fibrosis and steatosis to predict liver-related outcomes and death compared to hepatic venous pressure gradient (HVPG) and histology in patients with nonalcoholic steatohepatitis (NASH). One hundred and forty-eight compensated patients with suspected NASH who had undergone a transjugular liver biopsy were included in the study retrospectively. Interestingly, the transjugular route is the preferred one used in their clinical practice to obtain samples to diagnose and stage patients with suspected NASH, given the possibility of performing HVPG, which complements the information obtained by means of histological analysis. Patients were followed until death, liver transplantation, and endstage liver disease complications as defined by hepatocellular carcinoma, ascites, spontaneous bacterial peritonitis, hepatic encephalopathy, de novo varices, or significant worsening of varices (such as bleeding or high-risk stigmata on endoscopy). The clinical outcome variable was a composite variable combining these aforementioned endpoints. Advanced fibrosis (F3 or F4 by Brunt) and clinically significant portal hypertension (CSPH; HVPG 10 mm Hg) were observed in 34% and 18% of patients, respectively. Overall, after a median follow-up of 5 years, 24 patients (16%) developed the composite endpoint, with an incidence rate of 1.2/100 person-years. The authors describe that all noninvasive tools that estimate fibrosis (aspartate aminotransferase-to-platelet ratio index, FIB-4, and NAFLD fibrosis score), but not the ones that estimate steatosis, were associated with the composite clinical outcome. No differences were observed in the prognostic performance of the noninvasive fibrosis biomarkers compared to HVPG or histological fibrosis stage. For the analysis, each variable was divided into a high-risk and a low-risk category. Noninvasive tools that identify which patients with NAFLD are more likely to require specialized care would definitely be useful for physicians who handle these patients in order to see who should be maintained in, or referred to, specialized care and who could be followed in the primary care setting. This study demonstrates that these tools can be used in NASH patients, with the setback that these patients were preselected with an initial invasive procedure. Nevertheless, a previous study had shown that noninvasive markers of fibrosis could predict the development of liver-related complications and overall mortality in all NAFLD patients. Indeed, increasing evidence suggests that the presence of fibrosis in NAFLD/NASH is clinically relevant, which is not surprising as fibrosis could be considered the footprint of the previous injury that has been going on in the liver until that moment. The detection of fibrosis, by both invasive and noninvasive markers, indicates a greater likelihood of a negative outcome than in those without fibrosis. The main advantage of the noninvasive markers of fibrosis is the fact that they can be repeatedly measured and therefore could be used as surrogate markers in the management of these patients. However, until now, data supporting the use of changes in these markers in the clinical setting in order to monitor NAFLD/NASH have been lacking. An interesting aspect of this study is the routine use of the combination of histology and HVPG in clinical practice, although the patients did not necessarily have cirrhosis. This was done in order to overcome the underestimation of the liver injury that may happen with liver biopsy. Although not an aim of the study, if this assumption were true, the combination of histology and HVPG should have improved accuracy over each of the individual variables to predict clinical outcomes. Another remarkable aspect of the study from Sebastiani et al. is the prognostic relevance of HVPG. Previous studies have suggested that there may be portal hypertension without advanced fibrosis in patients with NAFLD, which would mainly be associated to