Abstract Background and Aims People with end stage kidney disease (ESKD) commonly co-exhibit multiple risk factors (type 2 diabetes mellitus, obesity and hypertension) for non-alcoholic fatty liver disease (NAFLD) and its progressive, fibroinflammatory form non-alcoholic steatohepatitis (NASH). NAFLD and NASH both associate with increased risk of fatal cardiovascular events. NASH will additionally soon become the leading cause of cirrhosis both in the UK and worldwide. The cause of systemic inflammation driving early mortality in patients with ESKD is unclear. It is thought that liver disease can contribute to inflammation due to reduced reticuloendothelial function and subsequent penetration of gut-derived toxins into the systemic circulation. Very little is currently known about prevalence of NAFLD and NASH in ESKD, or how these conditions affect patients. By identifying the scale of the risk, this study will help better understand the extent of NAFLD and NASH and their links to clinical outcomes in advanced kidney disease. We report interim results from this study looking non-invasively for liver disease in patients with ESKD. Method This prospective study involves prevalent patients with ESKD treated with dialysis (for >3 months) at five participating UK kidney centres. Results from this study are derived from analysis of the first 238 patients (final recruitment target 450). A FibroScan® (Echosens) device was used to measure both hepatic steatosis using controlled attenuation parametrography (CAP) and fibrosis using transient elastography. A fibrosis-4 index score was calculated to assess fibrosis risk. These results were supplemented by baseline clinical and radiological data, serum beta-D-glucan levels taken pre- and post-dialysis, clinical assessment of fluid status, and bioimpedance spectroscopy (Fresenius Medical Care). Survival analyses were performed using Kaplan-Meier estimates. Results The mean age of participants was 63 years (67% male). Mean body mass index was 27.8kg/m2. 97% of participants had hypertension, 53% diabetes mellitus and 65% hyperlipidaemia. 231 participant FibroScan scores (97%) were valid (interquartile range ≤30% over 10 consecutive readings) and available for analysis. 74 participants (31%) had suspected hepatic steatosis of grade S1-S3 and 72 participants (30%) had suspected hepatic fibrosis of grade F2-F4. 53 participants (22%) had suspected F2-F3 fibrosis (moderate fibrosis) and 19 (8%) had suspected F4 fibrosis (advanced fibrosis/cirrhosis). There was increased mortality associated with suspected hepatic fibrosis/steatosis (9.1% absolute risk of mortality with an abnormal CAP or fibrosis score compared to 1.6% in those with normal FibroScan imaging). The majority of mortality in people with suspected hepatic fibrosis was from cardiovascular disease. Overall mortality in participants with suspected hepatic fibrosis remained significant even after adjustment for univariate predictors of survival (median CRP, baseline BDG levels and age) (Figure 1). Other univariate predictors of survival were Charlson Comorbidity Score and diabetes status. Conclusion These results demonstrate a significant burden of suspected hepatic steatosis and hepatic fibrosis in people with ESKD. Suspected hepatic fibrosis assessed by FibroScan imaging is an independent risk factor for mortality based on this interim analysis and strategies to improve liver health in the setting of advanced kidney disease may be of benefit to this group of patients.
Read full abstract