Defining Characteristics and Outcomes for Patients with Non-Alcoholic Fatty Liver Disease Admitted to Hospital with Decompensated Cirrhosis

Global incidence of non-alcoholic fatty liver disease: A systematic review and meta-analysis of 63 studies and 1,201,807 personsJournal of HepatologyVol. 79Issue 2PreviewThe prevalence of non-alcoholic fatty liver disease (NAFLD) is increasing. We aimed to estimate the pooled global NAFLD incidence. Full-Text PDF We read with interest the systematic review and meta-analysis by Le et al. which demonstrates the rapidly increasing worldwide incidence of non-alcoholic fatty liver disease (NAFLD) likely associated with the obesity epidemic.[1]Le MH Le DM Baez TC Wu Y Ito T Lee EY Global incidence of non-alcoholic fatty liver disease: a systematic review and meta-analysis of 63 studies and 1,201,807 persons.J Hepatol. 2023; 79: 287-295Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar Recently, we have demonstrated that NAFLD is the second most common cause of liver disease in patients admitted to hospital with decompensated cirrhosis in the UK (14.4% of admissions).[2]The Trainee Collaborative for Research and Audit in Hepatology UKRegional variation in characteristics of patients with decompensated cirrhosis admitted to hospitals in the UK.Lancet Gastroenterol Hepatol. 2023; 8: 604-606Abstract Full Text Full Text PDF Scopus (2) Google Scholar We strongly agree that targeted public health interventions to reduce the incidence and prevalence of NAFLD combined with optimal outpatient management strategies are needed to mitigate against NAFLD-related complications.[3]McPherson S Armstrong MJ Cobbold JF Corless L Anstee QM Aspinall RJ et al.Quality standards for the management of non-alcoholic fatty liver disease (NAFLD): consensus recommendations from the British Association for the Study of the Liver and British Society of Gastroenterology NAFLD Special Interest Group.Lancet Gastroenterol Hepatol. 2022; 7: 755-769Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar However, efforts to better characterise and stratify this cohort are similarly required. Utilising more robust phenotypic data should allow for the development of NAFLD-specific risk prediction models and may encourage implementation of precision approaches to improve overall patient outcomes. We analysed data from a UK multicentre, retrospective observational cohort study, including patients admitted to hospitals with decompensated cirrhosis in November 2019[4]The Trainee Collaborative for Research and Audit in Hepatology UKAdmission care bundles for decompensated cirrhosis are poorly utilised across the UK: results from a multi-centre retrospective study.Clin Med (Lond). 2023; 23: 193-200Crossref PubMed Scopus (4) Google Scholar (Table S1 provides regional submission data). We compared admissions for patients with NAFLD to the rest of the predominately alcohol-related liver disease (ARLD) cohort. Details of methods and statistical analyses are presented in the supplementary materials. The NAFLD cohort were significantly older (69.0 (IQR 62.3-77.0) vs. 55.5 (IQR 47.0-65.0), p <0.0001∗) and less likely to be male (50.6% vs. 63.5%, p = 0.001∗) (Table 1A). Whilst we note Le et al. demonstrated a higher incidence of NAFLD amongst male patients, this may reflect previous findings that mortality is comparable across male and female patients with NAFLD, reflecting a similar prevalence of advanced disease.[5]Simon T.G. Roelstraete B. Khalili H. Hagström H. Ludvigsson J.F. Mortality in biopsy-confirmed nonalcoholic fatty liver disease: results from a nationwide cohort.Gut. 2021; 70: 1375-1382Crossref PubMed Scopus (248) Google Scholar No differences were demonstrated in the proportion of admissions with a previous history of decompensation or known liver disease, or a history of hepatocellular carcinoma (HCC) between cohorts. Patients admitted with NAFLD were significantly less likely to regularly consume alcohol (18.0% vs. 61.8%, p <0.0001∗) than the rest of the predominant ARLD cohort (Table 1A). However, alcohol consumption has been shown to be underreported in previous cohorts of NAFLD and markers of alcohol use may have highlighted individual’s regularly consuming alcohol above recommended limits.[6]Staufer K Huber-Shonauer U Strebinger G Pimingstorfer P Suesse S Scherzer TM et al.Ethyl glucuronide in hair detects a high rate of harmful alcohol consumption in presumed non-alcoholic fatty liver disease.J Hepatol. 2022; 77: 918-930Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar Admissions for patients with NAFLD were predominantly related to the management of ascites (40.9% vs. 32.1%, p = 0.02∗) or encephalopathy (25.6% vs. 15.8%, p = 0.002∗), and less likely to be for jaundice (2.8% vs. 17.2%, p <0.0001∗) (Table 1A). Ascites is associated with the highest risk of readmission for patients with NAFLD.[7]Paik JM Eberly KE Kabbara K Harring M Younossi Y Henry L et al.Non-alcoholic fatty liver disease is associated with greater risk of 30-day hospital readmission in the United States (U.S.).Ann Hepatol. 2023; 28: 101108Crossref PubMed Scopus (2) Google Scholar Patient-centred elective outpatient paracentesis provision is therefore a requisite component of modern hepatology services. Recently, data has implicated the premature onset of encephalopathy in NAFLD, while the association of hyperammonaemia with deleterious outcomes is also well-described.[8]Thomsen KL Eriksen PL Kerbert AJC De Chiara F Jalan R Vilstrup H Role of ammonia in NAFLD: an unusual suspect.JHEP Reports. 2023; 5: 100780Abstract Full Text Full Text PDF PubMed Scopus (2) Google Scholar A high index of suspicion for encephalopathy is required in this cohort, in addition to a low threshold in initiating therapies for hyperammonaemia. The reduction in the proportion of patients admitted with jaundice likely reflects the cohort of patients from the predominately ARLD cohort presenting with alcohol-related hepatitis.Table 1Characterising UK NAFLD admissions.A) VariableNNAFLD, n = 176NAlternate aetiology of liver disease, n = 1,048p valueAge17669.0 (62.3-77.0)1,04855.5 (47.0-65.0)<0.0001∗Male sex17689 (50.6%)1,048665 (63.5%)0.001∗Current alcohol use12222 (18.0%)813502 (61.8%)<0.0001∗Previously known liver disease176151 (85.8%)1,048890 (84.9%)0.82Previous known decompensation176116 (65.9%)1,048706 (67.4%)0.73Known HCC17610 (5.7%)1,04854 (5.2%)0.72Reason for admission Ascites17672 (40.9%)1,048336 (32.1%)0.02∗ Encephalopathy17645 (25.6%)1,048166 (15.8%)0.002∗ Gastrointestinal bleeding17619 (10.8%)1,048161 (15.4%)0.13 Jaundice1765 (2.8%)1,048180 (17.2%)<0.0001∗ Sepsis17610 (5.7%)1,04869 (6.6%)0.74Prognostic scores MELD score14713.0 (11.0-18.0)95517.0 (12.0-21.0)<0.0001∗ UKELD score14753.0 (51.0-60.0)95557.0 (52.0-62.0)<0.0001∗ Child-Pugh score1398.0 (7.0-10.0)9329.0 (8.0-11.0)0.0005∗Post 24-hour care Managed by a specialist176125 (71.0%)1,048776 (74.1%)0.41 Predominately managed on a specialist ward17693 (52.8%)1,043597 (57.2%)0.29 Critical care admission during stay1769 (5.1%)1,048123 (11.7%)0.008∗ Transfer to another centre1765 (2.8%)1,04819 (1.8%)0.37 Admission mortality17127 (15.8%)1,029160 (15.6%)0.91 Length of stay1427.0 (4.0-13.0)8597.00 (3.0-13.0)0.60B) VariableNSurvivors, n = 144NNon-survivors, n = 27p valueAge14469.0 (62.0-76.0)2773.0 (65.0-78.0)0.17Male sex14472 (50.0%)2715 (55.6%)0.68Current alcohol use10220 (19.6%)172 (11.8%)0.74Previously known liver disease144125 (86.8%)2722 (81.5%)0.54Previous known decompensation14496 (66.7%)2717 (63.0%)0.83Known HCC1448 (5.6%)272 (7.4%)0.66Reason for admission Ascites14459 (41.0%)2711 (40.7%)>0.9999 Encephalopathy14438 (26.4%)275 (18.5%)0.47 Gastrointestinal bleeding14416 (11.1%)272 (7.4%)0.74 Jaundice1444 (2.8%)271 (3.7%)0.58 Sepsis14410 (6.9%)270 (0.0%)0.37Prognostic scores MELD score11913.0 (10.0-17.0)2321.0 (15.0-27.0)<0.0001∗ UKELD score11953.0 (50.0-57.0)2359.0 (54.0-61.0)0.0001∗ Child-Pugh score1138.0 (7.0-9.0)2210.0 (8.8-11.0)0.0002∗Post 24-hour care Managed by a specialist144100 (69.4%)2721 (77.8%)0.49 Predominately managed on a specialist ward14478 (54.2%)2712 (44.4%)0.40 Critical care admission during stay1445 (3.5%)274 (14.8%)0.04A) Comparison of admissions for patients with NAFLD compared to patients with alternate aetiologies of liver disease. B) Comparison of NAFLD admissions resulting in patient survival with those that did not. Non-normally continuous data were analysed using Mann-Whitney U tests and presented as median (IQR). Categorical data were analysed using Fisher’s exact tests and presented as number (%). ∗Statistical significance set as per Benjamini-Hochberg procedure with a false discovery rate of 0.05.HCC, hepatocellular carcinoma; MELD, model for end-stage liver disease; NAFLD, non-alcoholic fatty liver disease; UKELD, UK end-stage liver disease. Open table in a new tab A) Comparison of admissions for patients with NAFLD compared to patients with alternate aetiologies of liver disease. B) Comparison of NAFLD admissions resulting in patient survival with those that did not. Non-normally continuous data were analysed using Mann-Whitney U tests and presented as median (IQR). Categorical data were analysed using Fisher’s exact tests and presented as number (%). ∗Statistical significance set as per Benjamini-Hochberg procedure with a false discovery rate of 0.05. HCC, hepatocellular carcinoma; MELD, model for end-stage liver disease; NAFLD, non-alcoholic fatty liver disease; UKELD, UK end-stage liver disease. Following admission, no differences were noted between cohorts in the proportion of patients managed by a specialist Gastroenterologist/Hepatologist on specialist wards, or in patients transferred to specialist centres. No differences were demonstrated between patient cohorts for mortality during admission (15.8% vs. 15.6%, p = 0.91) despite admissions with NAFLD having significantly lower prognostic scores and being less likely to access critical care (5.1% vs. 11.7%, p = 0.008∗) (Table 1A). Whilst this may reflect the older age of this cohort and concomitant comorbidity, previous concerns have been raised regarding limited access to critical care for patients with ARLD with stigma amongst healthcare professionals suggested as a potential barrier for this cohort.[9]Mitchison H. Saksena S. Hudson M. NCEPOD and alcohol-related liver disease, what are the views of those who deliver the service? A survey of consultants and trainees in North Eastern England.J R Coll Physicians Edinb. 2018; 48: 293-298Crossref PubMed Google Scholar Understanding potential barriers to patients with NAFLD accessing critical care is required to optimise management. Admissions for patients with NAFLD were not more likely to result in mortality after adjustment for age, critical care admission or MELD score (adjusted odds ratio 1.16; 95% CI 0.65-2.00; Fig. S1). Whilst comparisons between non-survivors and survivors are likely underpowered, conventional prognostic models discriminated admission survival (Table 1B). However, no prognostic scores significantly outperformed other models (p = 0.41), with no model achieving an AUC of greater than 0.8 (Fig. S2). After exclusion of admissions resulting in mortality or critical care admission, length of stay was no different to the rest of the cohort (Table 1A). Limitations of these analyses are discussed in the supplementary materials and include the retrospective design, coverage of only a single month, incomplete coverage of the UK with potential selection bias, lack of data regarding comorbidities and the subjective nature of aetiology assignment. Whilst accepting these limitations, this study is representative of a large, real-world cohort. With the likely increased incidence of patients being admitted to hospital with decompensated NAFLD, further work to understand how to optimally manage this cohort are necessary. This cohort is older, often more co-morbid and will likely require a tailored approach to their care. Understanding barriers to providing best care, including access to critical care, is essential when developing the hepatology services of the future. We are grateful for the funding support provided by Guts-UK. We are grateful for the support and endorsement from the British Society of Gastroenterology, British Association for the Study of the Liver, Scottish Society of Gastroenterology and the Welsh Association for Gastroenterology and Endoscopy. The authors declare no conflicts of interest that pertain to this work. Please refer to the accompanying ICMJE disclosure forms for further details. The following are the supplementary data to this article. Download .pdf (.35 MB) Help with pdf files Multimedia component 1 Download .pdf (.31 MB) Help with pdf files Multimedia component 2 Download .docx (.03 MB) Help with docx files Multimedia component 3