Prevalence of liver fibrosis in the general population (the LiverScreen

# Prevalence of liver fibrosis in the general population (the LiverScreen project): a multinational European cohort study *The Lancet 2026* Articles Prevalence of liver fibrosis in the general population (the LiverScreen project): a multinational European cohort study Isabel Graupera*, Maja Thiele*, Laurent Castera, Guillem Pera, Salvatore Piano, Anna Sòria, Núria Fabrellas, Pere Toran, Carla Chacon, Katrine T Bech, Helle Lindholm Schnefeld, Marta Tonon, Simone Incicco, Joël Moussy, Vincent Lévy, Anita Madir, Sandro Kukic, Daniel Jan Havaj, Svetlana Adamcova-Selcanova, Jesse Pustjens, Laurens A van Kleef, Alba Jiménez-Masip, Laura Pagès, Mirko Zoncapè, Susanne N Weber, Peter R Galle, Rebecca Harris, Luis Ibáñez-Samaniego, Rosa M Morillas, Alba Diaz, Sönke Detlefsen, Miquel Serra-Burriel, Anita Arslanow, Peter Andersen, Judit Pich, Eva Bonfill, Marko Korenjak, Céline Fournier-Poizat, Anne Llorca, Marie-Caroline Gourmelon, Harry J de Koning, Martina Pérez-Guasch, Ann Thu Ma, Adrià Juanola, Elisa Pose, Ingrid Arteaga, Ida Villesen, Johanne Kragh Hansen, Valeria Calvino, Roberta Gagliardi, Bahija Boutouria, Frane Pastrovic, Petra Dinjar Kujundzic, Daniela Zilincanova, Karolina Kristina Sulejova, Diego Rojo, Robert J de Knegt, Maykon Diego Melo, Antonio Torrejón, Rosario Hernández-Ibáñez, Jordi Hoyo, Laura Muñoz, Raquel López-Martos, Simon J Griffin, Michael Manns, Tom H Karlsen, Phil N Newsome, Patrick S Kamath, Rafael Bañares, Indra N Guha, Jörn M Schattenberg, Frank Lammert, Emmanuel Tsochatzis, Willem P Brouwer, Juan M Pericàs, Lubomir Skladany, Ivica Grgurevic, Dominique Roulot, Paolo Angeli, Aleksander Krag, Llorenç Caballeria, Pere Ginès† on behalf of the LiverScreen Consortium Investigators‡ Summary Lancet 2026; 407: 1448–58 Background Small scale, single-country studies suggest that undiagnosed liver fibrosis is prevalent in the general See Comment page 1402 population; however, its true burden and main risk factors remain unclear. We aimed to assess the prevalence of *Contributed equally as co-first undiagnosed liver fibrosis and its relationship with metabolic factors or alcohol consumption in a large prospective authors population-based multinational cohort study. †Senior author ‡Investigators listed at the end Methods We enrolled individuals from the general population who were aged 40 years and older across 35 sites, of the Article including primary health centres and screening units, and their 16 affiliated tertiary hospitals in nine European Liver Unit, Hospital Clínic of countries. Liver fibrosis was estimated using the liver stiffness measurement (LSM) by vibration-controlled transient Barcelona, Barcelona, Spain elastography (VCTE). A positive test was defined as an LSM of 8 kPa and greater or an alanine aminotransferase (I Graupera PhD, A Sòria MD, concentration of at least 1·5 the upper limit of normal, or a combination of both. Individuals with at least one criterion M Pérez-Guasch RN, A Juanola PhD, E Pose PhD, were referred for hepatology consultation to confirm chronic liver disease with fibrosis. The primary outcome was the Prof P Ginès PhD); Fundació de prevalence of LSM of 8 kPa and greater. Recerca Clínic Barcelona- Institut d’Investigacions Findings We enrolled 30 199 individuals (mean age 58 years; 57% [17 203 of 30 199] women, 89% [24 440 of 27 481] Biomèdiques August Pi i Sunyer (FCRB-IDIBAPS), White) from nine European countries. Metabolic factors were present in 70% (21 084 of 30 024) of participants, Barcelona, Spain (I Graupera, and 59% (15 107 of 25 488) of participants reported alcohol use, with 6·1% (1771 of 29 081) of participants reporting A Sòria, Prof N Fabrellas PhD, harmful consumption. The positive screening rate was 6·9% with a prevalence of 4·6% with an LSM of 8 kPa and A Diaz PhD, greater. Elevated LSM was strongly associated with obesity, type 2 diabetes, and harmful alcohol use. Of the participants A Arslanow PhD; J Pich BSc, E Bonfill BSc, M Pérez-Guasch, referred, 61% (1491 of 2457) completed a hepatology evaluation, and chronic liver disease with fibrosis was confirmed A Thu Ma PhD, A Juanola, E Pose, in 32% (477 of 1491) of participants, yielding an overall estimated prevalence of 1·6% (477 of 30 199). Steatotic liver Prof P Ginès); Centro de disease accounted for 93% (443 of 477) of cases. Investigación Biomédica en Red Enfermedades Hepáticas y Digestivas (CIBEREHD), Madrid, Interpretation Undiagnosed liver fibrosis is common in the general population in Europe and is primarily driven by Spain (I Graupera, A Sòria, metabolic factors and alcohol consumption. Early detection is pivotal as it could allow personalised interventions that Prof N Fabrellas, can prevent progression to cirrhosis and complications. A Jiménez-Masip MD, R M Morillas PhD, A Diaz, M Pérez-Guasch, A Juanola, Funding The European Commission, Fondo de Investigación Sanitaria de Salud, the Spanish Ministry of Economy, E Pose, L Ibáñez-Samaniego PhD, Industry, and Competitiveness, and the European Regional Development Fund. Prof R Bañares PhD, J M Pericàs PhD, Prof P Ginès); Centre for Liver Research, Copyright © 2026 Elsevier Ltd. All rights reserved, including those for text and data mining, AI training, and similar Department of technologies. Gastroenterology and Hepatology, Odense University Introduction identification of fibrosis is pivotal, offering an opportunity Hospital (OUH), Odense, Denmark. (Prof M Thiele PhD, Chronic liver disease is the 12th leading cause of death to intervene before complications arise and, thereby, K T Bech MD, H L Schnefeld MSc, worldwide, responsible for 2·3 million deaths in 2021.1 potentially improve patient outcomes.3 P Andersen MSc, I Villesen PhD, Liver fibrosis—the main hallmark of chronic liver Early detection has gained importance with the J K Hansen MD, Prof A Krag PhD); disease—progresses silently over decades and often development of novel therapeutics targeting moderate Department of Clinical Research, Faculty of Health remains undetected until cirrhosis with complications and severe fibrosis in patients with metabolic dysfunction- Sciences, University of develops later in life.2 Given that morbidity and mortality associated steatotic liver disease (MASLD).4,5 Despite the Southern Denmark (SDU), increase substantially when cirrhosis develops, early development of accurate non-invasive fibrosis tests6 such 1448 Articles Odense, Denmark Research in context (Prof M Thiele, K T Bech, Prof S Detlefsen PhD, I Villesen, Evidence before this study evaluation. Metabolic risk factors and alcohol use were present J K Hansen, Prof A Krag); Chronic liver disease is a leading cause of premature mortality in more than 70% of the population. Type 2 diabetes and Department of Hepatology, worldwide. Liver fibrosis—the principal determinant of disease harmful alcohol use were the strongest predictors of positive Hôpital Beaujon, Assistance progression—is often undetected until advanced stages when screening and confirmed chronic liver disease, with a clear Publique-Hôpitaux de Paris, Clichy, Université de Paris, there is little hope for cure. Although non-invasive tests and dose–response to elevated liver stiffness (≥8 kPa), which rose Paris, France (Prof L Castera PhD, guideline recommendations for earlier detection are from 1·3% (no risk factors) to 21·0% (four metabolic factors) B Boutouria MD); Unitat de increasingly available, population-level evidence to support and 37·0% when combined with harmful alcohol use. Notably, Suport a la Recerca Metropolitana Nord, Fundació liver fibrosis screening in the general population remains to be nearly half of referred participants had lower liver stiffness Institut Universitari per a la poor. We searched PubMed on Oct 20, 2025, using the search measurements at hepatology assessment, underscoring the Recerca a l’Atenció Primària de terms “liver fibrosis screening”, “assessment of liver fibrosis”, importance of confirmatory testing for accurately establishing a Salut Jordi Gol i Gurina and “liver fibrosis tests”, together with “general population”. diagnosis of chronic liver disease. Overall, the prevalence of (IDIAPJGol), Mataró, Barcelona, Spain (G Pera BSc, P Toran PhD, We searched for prospective studies published between chronic liver disease with fibrosis in the screened population C Chacon BSc, I Arteaga MD, Jan 1, 2010, to Oct 20, 2025, for English-language studies was 1·6%. L Caballeria PhD); Unit of assessing liver fibrosis screening in unselected populations Internal Medicine and Implications of all the available evidence using invasive or non-invasive methods. Of the 605 records Hepatology, Department of Early detection of liver fibrosis is essential to enable timely, Medicine (DIMED), University identified, 66 contained relevant information and were personalised interventions that can prevent disease and Hospital of Padova, Italy assessed in detail. Most studies were excluded because they progression. However, confirmatory testing is crucial to (S Piano PhD, M Tonon MD, were restricted to selected high-risk populations, had small S Incicco MD, V Calvino MD, accurately diagnose chronic liver disease. Future screening sample sizes, did not report relevant prevalence estimates, R Gagliardi MD, strategies should be adapted to local epidemiological patterns Prof P Angeli PhD); Faculty of or lacked protocolised confirmatory diagnostic assessment of and health-care system capacities to maximise effectiveness Medicine and Health Sciences, chronic liver disease following screening. and equity in liver disease prevention. Given the high University of Barcelona, Barcelona, Spain (A Sòria, Added value of this study prevalence of metabolic and alcohol-related risk factors in the A Diaz, J Pich, This is the largest prospective population-based liver fibrosis general population, effective screening approaches should R Hernández-Ibáñez MD, screening study that has included more than 30 000 individuals extend beyond narrowly defined high-risk groups and include a Prof P Ginès); Faculty of Nursing, University of from nine European countries. Unlike previous studies, screen- broad segment of the population. Barcelona, Barcelona, Spain positive participants underwent complete hepatology (Prof N Fabrellas); Centre d’examens de santé, CPAM, Bobigny, France (J Moussy MD); Département de recherche as liver stiffness measurement (LSM) by vibration- moderately high alcohol consumption (MetALD); and 3) clinique, Hôpital Avicenne, controlled transient elastography (VCTE), liver fibrosis alcohol-related liver disease secondary to very high alcohol APHP, Bobigny, France remains vastly underdiagnosed and most patients are use regardless of concomitant metabolic dysfunction.12 (Prof V Lévy PhD); Department of Gastroenterology, diagnosed late at a symptomatic, decompensated stage.7 MASLD affects approximately one-third of adults globally Hepatology and Clinical Increased LSM values are associated with advanced liver and harmful alcohol use remains to be a large contributor Nutrition, University Hospital fibrosis and a high probability of liver-related outcomes to liver-related mortality.13 Although interactions between Dubrava, Croatia (A Madir MD, such as complications of portal hypertension, hepato- alcohol and cardiometabolic risk factors have been studied F Pastrovic MD, P D Kujundzic MD, cellular carcinoma, and liver-related death.8 Despite in high-risk cohorts, population-based data clarifying their Prof Ivica Grgurevic PhD); having a similar age-standardised mortality rate as lung relative contribut ions to hepatic fibrosis are still low. University of Zagreb School of cancer and a higher prevalence than colorectal cancer, Therefore, we aimed to assess the prevalence of medicine, Zagreb, Croatia there are no population-wide screening programmes for undiagnosed liver fibrosis and its associated risk factors in (S Kukic MD); Division of Hepatology, Gastroenterology the early detection of cirrhosis.9 Previous studies have the LiverScreen study, incorporating a comprehensive and Liver Transplantation, estimated liver fibrosis using non-invasive tests but they confirmatory assessment for individuals across a large Department of Internal are all single-country studies and do not incorporate European adult population with positive screening results. Medicine II, Slovak Medical University Faculty of Medicine, structured hepatology evaluation or liver biopsy for F D Roosevelt University diagnostic confirmation of people with increased LSM, Methods Hospital, Banska Bystrica, limiting its ability to accurately define chronic liver Study design Slovakia (D Jan Havaj MD, disease with fibrosis or differentiate aetiologies.10,11 Thus, This prospective population-based multinational cohort S Adamcova-Selcanova MD, D Zilincanova MD, robust data on the estimated prevalence of undiagnosed study was carried out across 35 sites, including primary K K Sulejova MD, liver fibrosis across different countries and health-care health-care centres and screening units and 16 associated L Skladany PhD); Department of settings are required to justify and inform any future tertiary hospitals across nine European countries: Spain, Gastroenterology and systematic screening initiatives.3,9 Denmark, Italy, Slovakia, Croatia, UK, France, The Hepatology, Erasmus MC- University Medical Centre, The nomenclature of steatotic liver disease is based on Netherlands, and Germany (appendix pp 8–10). Details Rotterdam, Netherlands the presence of steatosis and metabolic or alcohol risk on site selection have been provided in the appendix (p 3). (J Pustjens MD, L A van Kleef MD, factors and recognises three subtypes: (1) MASLD usually The primary objectives were to assess the prevalence of R J de Knegt MD, related to type 2 diabetes and obesity; (2) MASLD with liver fibrosis, as estimated by an elevated LSM of 8 kPa or W P Brouwer PhD); Liver Unit, Articles Hospital Universitari Vall above, and evaluate the relationship between LSM considered by the investigator to preclude adequate d´Hebron, Vall d’Hebron (≥8 kPa) and metabolic risk factors, alcohol consumption, conduct of the study, and major extrahepatic diseases Institut de Recerca (VHIR), Vall and the interaction between both metabolic risk factors that could impair short-term prognosis, such as d’Hebron Barcelona Hospital Campus, Barcelona, Spain and alcohol in a random population aged 40 years or malignancy, congestive heart failure (New York Heart (A Jiménez-Masip, L Pagès MD, older across Europe. Association Grade 4), chronic obstructive pulmonary D Rojo MD, J M Pericàs); UCL The study was approved by the individual ethics disease (GOLD >3), and advanced chronic kidney disease Institute for Liver and committees (appendix p 7) and is reported in accordance (serum creatinine >3 mg/dL or undergoing renal Digestive Health, Royal Free Hospital, University College of with the STROBE guidelines for observational studies replacement therapy). Participants with a positive London (UCL), London, UK (appendix pp 30–31). screening test were subsequently referred to a second (M Zoncapè MD, M D Melo MD, visit with an hepatologist to confirm the presence of Prof E Tsochatzis PhD); Participants chronic liver disease. All participants gave written Department of Medicine II, Saarland University Medical We invited individuals from the general population to informed consent to participate in the study. Center, Saarland University, participate via three different methods: (1) a random Homburg, Germany selection of people attending primary health-care centres Procedures (S N Weber PhD, for a regular health check-up or follow-up (Spain, Croatia, At the screening visit, trained nurses carried out the Prof J M Schattenberg PhD); Department of Internal France, and Germany); (2) a random selection of social examinations, following standard operating procedures. Medicine I, University Medical security numbers or ZIP codes followed by online letter Visits included a medical history and measurement of Center of the Johannes invitations or phone calls (Denmark, Spain, Italy, UK, weight, height, waist and hip circumferences, blood Gutenberg-University, Mainz, The Netherlands); and (3) individuals who participated pressure, and blood tests for liver biochemistry, in addition Germany (Prof P R Galle PhD, Prof J M Schattenberg); King’s in existing population screening programmes or to blood glucose, glycated haemoglobin, lipid profile, and Mill Hospital, Mansfield, UK occupational health checks (Spain, Germany, Slovakia, serum creatinine. Alcohol drinking pattern was based on (R Harris MD); Gastroenterology and France). For details on recruitment procedures by self-reported alcohol consumption using standardised and Hepatology Service, site and country, see the appendix (pp 8–10). alcohol questionnaires, including weekly alcohol intake Hospital General Universitario, Gregorio Marañon, Gregorio The inclusion criteria were participants aged 40 years and the Alcohol Use Disorders Identification Test-Concise Marañon Research Institute, and older, and who were able to give informed consent. (AUDIT-C) questionnaire. Harmful alcohol use was Madrid, Spain The exclusion criteria were known chronic liver disease defined as current consumption of standard drink units of (L Ibáñez-Samaniego, (except liver steatosis), mental incapacity, language 14 units or more per week for women and 21 units or Prof R Bañares PhD); School of Medicine, Universidad barrier, insufficient social support or any other reason more per week for men. Diagnostic criteria for metabolic dysfunction (arterial hypertension, dyslipidaemia, type 2 diabetes, and people with overweight or people with A B obesity) and alcohol consumption patterns definition are reported in the appendix (p 4). 67 074 participants invited 2457 referred for a hepatology visit We estimated liver fibrosis using LSM by VCTE with the FibroScan device (Echosens, Paris, France) according 30 909 accepted to the instructions by the manufacturer (appendix p 5). 256 visit incomplete VCTE was done following a minimum fasting period of 3 h. Measurements were considered reliable if at least 368 excluded 187 were younger than 710 not visited at tertiary ten valid readings were obtained and met one of the 40 years hospital following criteria: an IQR range-to-median ratio (IQR/ 64 VCTE not 473 cause not reported performed 127 loss to follow-up P50) of 30% or lower for median liver stiffness (P50 44 previous chronic 64 not referred ≥7·1 kPa), or a P50 of less than 7·1 kPa regardless of the liver disease 43 patient withdrawals IQR. Measurements that did not meet these criteria 37 current malignancy 3 died or had severe 36 other medical condition were classified as unreliable. VCTE failure was defined as the inability to obtain any valid LSM. We recorded all study data in an electronic database called REDCap 30 541 at the screening visit 1491 hepatology visit complete (version 13.7.1). Vanderbilt University, TN, USA) on servers hosted by and located in the Region of Southern VCTE Denmark, using pseudonymised data. Gender U Fa n il r u e r li e a † b le* 477 CLD with 1014 no evidence information was not assessed and sex and ethnicity fibrosis of CLD with variables were self-reported by participants at study entry fibrosis using a prespecified question in the REDCap electronic 30 199 completed the visit case report form (appendix p 5). Ethnicity data were not collected for participants in the Slovakian cohort because Figure 1: Trial profile this site did not use the REDCap questionnaire and, The invitation and enrolment process for the screening visit (A), and referral for hepatology consultation and final therefore, did not request or collect ethnicity data. evaluation (B). CLD=chronic liver disease. VCTE=vibration-controlled transient elastography. *273 (0·9%) of We defined a positive screening test as an LSM of at 30 541 participants had an interquartile range-to-median ratio (IQR/P50) of at least 30% for median liver stiffness (P50 ≥7·1 kPa). †For 69 (0·2%) of 30 541 participants, we could not obtain any valid liver stiffness measurement. least 8·0 kPa or an alanine aminotransferase (ALT) of at 1450 Articles least 1·5 upper limit of normal (ULN). Participants Complutense, Madrid, Spain meeting at least one of the two criteria were referred for (L Ibáñez-Samaniego, hepatology consultation to assess the presence, staging, Prof R Bañares); Liver Unit, Hospital Germans Trias i Pujol, and aetiology of chronic liver disease. We selected a Instituto de Investigación threshold of 8·0 kPa to define elevated LSM because an Overall cohort Germans Trias i Pujol (IGTP), LSM of less than 8 kPa is associated with a very low (N=30 199) Badalona, Spain (R M Morillas); probability of fibrosis14,15 and liver-related events.8,16 We Age (years) 58 (10) Department of Pathology, Hospital Clinic, Barcelona, also included an ALT-based criterion to identify Sex Spain (A Diaz); Department of participants with possible liver inflammation that might Female 17 203 (57%) Pathology, Odense University be missed with VCTE alone.17,18 Individuals with Male 12 996 (43%) Hospital (OUH), Odense, Denmark (Prof S Detlefsen); unreliable or failed LSM were also referred for further Ethnicity Epidemiology, Statistics, and evaluation. White 24 440 (89%) Prevention Institute, The hepatology consultation to confirm the presence of African 1171 (4·2%) University of Zurich, Zurich, chronic liver disease with fibrosis consisted of a Latin-American 985 (3·6%) Switzerland (M Serra-Burriel PhD); comprehensive evaluation including biochemical testing Asian 537 (2·0%) Department of Clinical for viral, metabolic, and immune diseases; abdominal Mixed 94 (0·3%) Pharmacology, Clinical Trial ultrasound; repeat VCTE to confirm elevated LSM; and, if Other 140 (0·5%) Unit, Hospital Clínic, Barcelona, participants consented, liver biopsy as part of standard Smoking 4273 (15%) Spain (J Pich, E Bonfill); European Liver Patients’ clinical care. Definitions of chronic liver disease aetiologies BMI (kg/m²)* 27 (5) Association, Brussels, Belgium have been shown in the appendix (p 5). Pathology was Normal 10 259 (34%) (Marko Korenjak); Echosens, analysed blindly using digitalised central reading by Overweight 11 857 (40%) Paris, France two pathologists (AD and SD; appendix pp 5–6). (C Fournier-Poizat MD, Obese 8065 (26%) A Llorca MSc, We defined chronic liver disease with fibrosis as Abdominal obesity† 14 071 (47%) M-C Gourmelon MSc); presence of at least one of the following criteria: (1) any Metabolic risk factors Department of Public Health, stage of fibrosis (F1=mild fibrosis, ≥F2=significant Erasmus MC, University Arterial hypertension 10 488 (35%) fibrosis, ≥F3=advanced fibrosis, and ≥F4=cirrhosis) Medical Center Rotterdam, Type 2 diabetes 3140 (10%) Rotterdam, Netherlands confirmed by biopsy; (2) abdominal ultrasound with at Dyslipidaemia 16 031 (53%) (Prof H J de Koning PhD); Health, least two of the following findings typical of chronic liver At least one metabolic risk factor 21 084 (70%) Safety and Emergencies of disease with fibrosis: heterogeneous parenchyma, SEAT, CUPRA and the Harmful alcohol use‡ 1771 (6·1%) enlarged right liver lobe, enlarged left liver lobe including Volkswagen Group Companies Glucose (mg/dL) 99 (27) in Spain, Martorell, Spain caudate lobe, irregular liver surface, rounded edges, Glycated haemoglobin (%) 5·6 (0·8) (A Torrejón MSc); Institut Catala extrahepatic venous collaterals, spleen length greater de la Salut (ICS), Ambit Total cholesterol (mg/dL) 205 (43) than 13 cm, or ascites; and (3) an LSM of at least 10 kPa d’Atenció Primària, Barcelona, on VCTE performed at the hepatology visit.19 We selected HDL cholesterol (mg/dL) 59 (16) Spain (R Hernández-Ibáñez, LDL cholesterol (mg/dL) 124 (37) J Hoyo MD); Instituto de a threshold of 10 kPa to diagnose chronic liver disease Triglycerides (mg/dL) 106 (75) Investigación Germans Trias i with fibrosis in the absence of liver biopsy as this Pujol (IGTP), Badalona, Spain AST (IU/L) 24 (9) threshold is associated with high risk of liver-related (L Muñoz MD); Pathological outcomes across aetiologies and is recommended in ALT (IU/L) 22 (13) Anatomy Department, GGT (IU/L) 22 (19) Hospital Germans Trias i Pujol, clinical guidelines for the diagnosis of possible Badalona, Spain AP (IU/L) 72 (24) compensated advanced chronic liver disease.16 Adverse (R López-Martos MD); events were systematically assessed and recorded at both Platelet count (10⁹/L) 247 (61) Department of Public Health and Primary Care visits. Continuous variables with normal distribution are expressed as mean (SD) and (S J Griffin PhD) and MRC non-normal variables are expressed as P50 (IQR). ALT=alanine aminotransferase. AP=alkaline phosphatase. AST=aspartate aminotransferase. GGT=gamma- Epidemiology Unit (S J Griffin), Statistical analysis glutamyl-transpeptidase. *BMI categories used: normal (BMI <25·0 kg/m²); University of Cambridge School Quantitative variables were described using either mean overweight (BMI ≥25·0 kg/m² and <29·9 kg/m²; obese (BMI ≥30·0 kg/m²; Asian of Clinical Medicine, and SD (normal distributed variables) or median (P50) and origin BMI ≥25·0 kg/m²). †Abdominal obesity was defined as a waist Cambridge, UK; Hannover circumference of at least 88 cm for women and 102 cm for men. ‡Harmful Medical School (MHH), IQR (skewed variables). Categorical variables were alcohol use was defined as individuals who reported current alcohol consumption Hannover, Germany described using frequencies and percentages. equal to or greater than 14 units per week for women and 21 units per week for (Prof M Manns PhD); Clinic of Categorisation of continuous variables was done using men. For the participants in whom the standard drinking units questionnaire was Surgery and Specialized well established cutoff points (eg, BMI 25 kg/m² and not available, the Alcohol Use Disorders Identification Test-Concise (AUDIT-C) Medicine, Oslo University responses were changed to units. Missing data: ethnicity (9% missing); smoking Hospital and University of 30 kg/m²) or close to a quartile or median (eg, aged (6% missing); alcohol consumption (4% missing); glucose (11% missing), Oslo, Oslo, Norway ≥60 years). We applied univariable and multivariable glycated haemoglobin (39% missing); total cholesterol (10% missing); HDL (Prof T H Karlsen PhD§); Roger logistic regression models to assess the association cholesterol (12% missing); LDL cholesterol (14% missing), triglycerides (10% Williams Institute of Liver missing); AST (13% missing); ALT (9% missing); GGT (10% missing); AP (16% Studies, Faculty of Life Sciences between selected variables and LSM using different cutoff missing); and platelet count (16% missing). Ethnicity data were not collected for and Medicine, King’s College points (8 kPa, 10 kPa, and 15 kPa), adjusted by age, sex, and participants from the Slovakian cohort. London and King’s College country. We selected the following predictor variables Hospital, London, UK based on existing evidence: obesity, type 2 diabetes, arterial Table 1: Baseline characteristics (Prof P N Newsome PhD); Articles hypertension, dyslipidaemia, harmful alcohol use, and Results standard liver tests. A similar approach was used with Between May 2, 2018, and Dec 19, 2024, 67 074 individuals chronic liver disease with fibrosis as the dependent were invited and 30 909 participants from nine European variable. In this case, all the explanatory variables countries prospectively accepted (figure 1). The corresponded to the baseline visit. All the logistic participation rate according to the recruitment method regression models provided adjusted odds ratios (aORs) and country has been shown in the appendix (pp 8–10). A and 95% CI on tables or figures (forest plots). We total of 368 participants were excluded due to the determined the optimal LSM cutoff at the screening visit predefined exclusion criteria and, in addition, to detect chronic liver disease with fibrosis by maximising 342 participants had unreliable or failed LSM at the the product of sensitivity and specificity.20 No imputation screening visit. Therefore, the primary outcome was was done for missing data, as most of the missing values assessed in 30 199 participants (figure 1) with a mean age were due to centre-specific data collection policies of 58 years; 57% (17 203 of 30 199) were women, and 89% (eg, analyse only ALT and no AST, or fasting glucose and (24 440 of 30 199) were White. The characteristics of the no glycated haemoglobin). All tests were two-sided and overall study population and those of the participants statistical significance was defined as a p value of less categorised by country, sex, age, and recruitment method than 0·05. The statistical analysis was performed with have been shown in table 1 and the appendix (pp 11–15). Stata (version 19). Of the 30 199 participants, 2075 (6·9%) participants The sample size was calculated to assess the usefulness screened positive: 1376 (4·6%) participants with an LSM of VCTE as a screening method for detecting substantial of 8 kPa and greater, of whom 171 (0·6%) participants liver fibrosis. Because liver biopsy remains the gold also exhibited increased ALT, and 699 (2·3%) participants standard for fibrosis staging, the calculation considered with ALT levels of at least 1·5 ULN but normal LSM the number of biopsies required to assess concordance values. The median LSM in the whole series was 4·6 kPa between biopsy and VCTE. We first estimated the (range 1·5–75·0 kPa; IQR 1·8 kPa), whereas the median population prevalence of LSM at least 8 kPa to be 6%.16,17 value for the participants with an LSM of 8 kPa and Assuming that half of the participants with elevated LSM greater was 10·2 kPa (IQR 4·0 kPa). The prevalence of would agree to have a liver biopsy performed, with a LSM values of 10 kPa and greater was 2·5% and 5% of bilateral alpha error and a statistical power of 95%, 0·8% for 15 kPa and greater. The prevalence of increased we would need to screen a minimum of 28 887 participants LSM and the mean LSM values by country have been to obtain histological scores from at least 450 biopsies. shown in the appendix (p 17). The LiverScreen study is registered on ClinicalTrials. Elevated LSM correlated with several demographic and gov (NCT03789825) and is completed, and the protocol clinical factors including age and male sex, and metabolic was published in 2022.21 risk factors such as dyslipidaemia, arterial hypertension, people who are overweight or people with obesity, and Role of the funding source type 2 diabetes, and harmful alcohol use (figure 2). The The funders of the study had no role in study design, number of metabolic risk factors also correlated with data collection, data analysis, data interpretation, or elevated LSM. Although we observed a prevalence of an writing of the report. LSM of 8 kPa and greater in 1·3% of participants without 20 12 4 <60 ≥60 Female Male No Yes No Yes No Yes Normal Over- Obesity No Yes years years weight Age Sex Dyslipidaemia Arterial Harmful Overweight or Type 2 hypertension alcohol use obesity diabetes Figure 2: Prevalence of liver stiffness measurement (LSM; ≥8 kPa) according to age, sex, and presence or absence of metabolic conditions and harmful alcohol use Data show the prevalence of LSM of 8 Kpa and greater and 95% CI in brackets for each condition. 1452 )%( ecnelaverP Division of Gastroenterology and Hepatology, Mayo Clinic College of Medicine and Science, Rochester, MN, USA (Prof P S Kamath PhD); NIHR Nottingham Biomedical Research Centre, Nottingham University Hospitals NHS Trust and the University of Nottingham, Nottingham, UK (Prof I N Guha PhD); Center for Health Economics Research Hannover (CHERH), Hannover Medical School (MHH), Hannover, Germany (Prof F Lammert PhD); Universitat Autònoma de Barcelona, Barcelona, Spain (Prof J M Pericàs); Unité d’Hépatologie, Hôpital Avicenne, Assistance Publique- Hôpitaux de Paris, Université Sorbonne Paris Nord, Bobigny, France (Prof D Roulot PhD) §Prof Manns died in August, 2025 Correspondence to: Dr Isabel Graupera, Liver Unit, Hospital Clínic of Barcelona, 08036 Barcelona, Spain igraupe@clinic.cat See Online for appendix 14% (13·2–15·7) 10% (9·3–10·6) 8% 8% 6% 6% 6% (7·2–8·3) (6·7–9·3) (3·5 4 – % 4·1) (5·2–6·0) (3·1 3 – % 3·6) (5·8–6-6) (2·5 3 – % 3·1) (5·7–6·4) (2·6 3 – % 3·1) (3·8 4 – % 4·3) 2% (3·0 3 – % 3·6) (3·2 3 – % 3·6) (1·6–2·1) Risk factors for chronic liver disease Articles metabolic risk factors and without evidence of harmful alcohol consumption, the prevalence increased stepwise to 2·2%, 5·2%, 9·4%, and 20·7% among participants with one, two, three, or four metabolic risk factors 35 (figure 3). The presence of harmful alcohol use was 30 associated with an additional increase in the prevalence 25 of elevated LSM, beyond that conferred by metabolic risk factors alone (1·9%, 6·5%, 9·3%, 15·3%, and 37·1% in the case of zero, one, two, three, and four metabolic risk 15 factors, respectively; figure 3). Although current harmful alcohol use was associated with increased LSM, former 5 alcohol use, either low or harmful, showed no association (appendix pp 18, 26). Similar associations were observed 0 when analysing LSM 10 kPa and 15 kPa cutoffs (appendix p 18). Type 2 diabetes and harmful alcohol consumption in combination showed the strongest association with Figure 3: Prevalence of liver stiffness measurement (LSM; ≥8 kPa) according to the number of metabolic increased LSM (table 2). In addition, we observed a conditions and its association with harmful alcohol use progressive increase in the prevalence of LSM of at least The numbers equate to the following: 0 is no metabolic condition; 1 is one metabolic condition; 2 is two 8 kPa across increasing BMI categories (appendix p 18). metabolic conditions; 3 is three metabolic conditions; 4 is four metabolic conditions. The blue bars indicate no Similarly, median LSM values increased progressively harmful alcohol use and pink bars indicate harmful alcohol use. Data show the prevalence of LSM of 8Kpa and greater and 95% CI in brackets for each condition. Odds ratios for the association between harmful alcohol use and with the number of metabolic conditions present and LSM of 8 kPa and greater were calculated within each stratum of metabolic conditions, comparing participants showed a further elevation in individuals who also with harmful alcohol use to those without harmful alcohol use but with the same number of metabolic conditions. reported harmful alcohol use (appendix p 19). The odds ratio was 1·41 in participants with no metabolic conditions, 2·97 in those with one metabolic condition, 1·79 in those with two metabolic conditions, 1·62 in those with three metabolic conditions, and 1·79 in those with The laboratory variables associated with increased LSM four metabolic conditions. The interaction between harmful alcohol use and the number of metabolic conditions were elevated concentrations of ALT, aspartate amino- was not statistically significant, with a p value of 0·072. transferase (AST), gamma-glutamyl-trans peptidase (GGT), alkaline phosphatase, glycated haemog lobin, with normal liver enzymes, fewer risk factors, and for total cholesterol, and triglycerides, in addition to a lower individuals whose LSM values were only slightly platelet count and HDL cholesterol (appendix p 20). On increased over the threshold of 8 kPa (appendix p 23). Of multivariable analysis that included both clinical and the 1491 participants who completed the second visit, a laboratory variables, the following binary predictors were liver biopsy was done in 331 (22%) participants. The inde pendently associated with an increased LSM main reasons for not doing a liver biopsy were decrease (≥8 kPa): age, male sex, harmful alcohol use, people who in LSM to 8 kPa or lower at the second visit in were overweight or people with obesity, type 2 diabetes, 715 (48%) of 1491 participants, and lack of consent for arterial hypertension, and increased ALT and GGT values biopsy in 234 (16%) participants. A diagnosis of (figure 4). Analyses using the same variables modelled as chronic liver disease with fibrosis was made in continuous predictors yielded similar results (appendix 477 of 1491 participants, almost one-third (32%) of p 21). Country-specific analyses yielded similar findings individuals who completed the evaluation for liver (appendix p 22). Findings remained unchanged after disease, which represents an overall estimated prevalence excluding individuals in whom VCTE measurements of chronic liver disease with fibrosis of at least 1·6% could be less reliable (BMI ≥36 kg/m² or an ALT (477 of 30 199 participants), considering that more than ≥100 IU/L; appendix p 23). Factors associated with an one-third of individuals who screened positive were not increased LSM of 10 kPa and greater or 15 kPa and evaluated further (figure 1). Most of the participants with greater on multivariable analyses were similar (appendix a liver biopsy had fibrosis (273 [82%] of 331 participants), p 27). and 93 (28%) participants had advanced liver fibrosis We referred 2457 participants for a hepatology (58 participants with F3 liver fibrosis and 35 participants consultation: 2075 (84%) participants because of with F4 liver fibrosis). In participants who met the screening positive, 342 (14%) participants because of chronic liver disease criteria but did not have biopsy unreliable or failed VCTE, and 40 (2%) due to confirmation, 102 (50%) of 204 participants had an LSM investigator’s oversight. Of the 2457 participants, of 10 kPa and greater but no ultrasound evidence of 1491 (61%) completed the follow-up evaluation (figure 1). chronic liver disease, 70 (34%) participants fulfilled the Excluding the three countries with the lowest follow-up ultrasound criteria of chronic liver disease only, and evaluation rates (Germany [40%], Slovakia [25%], and 32 (16%) participants exhibited both an LSM of 10 kPa The Netherlands [15%]), the hepatology visit completion and greater and ultrasound criteria indicative of chronic rate was 76% (1276 of 1674). Overall, there was a reduction liver disease. Presence and severity of liver fibrosis on in LSM values at the hepatology visit compared with liver biopsy with respect to LSM values at screening and those at the screening visit, particularly in participants at the hepatology consultation has been shown in the )%( ecnelaverP 37·1% (25·2–50·3) 20·7% (18·1–23-5) 15·3% (10·9–20·6) 9·3% 9·4% 6·5% (6·6–12·6)(8·5–10·5) 5·2% (4·6–8·9) 1·3% 1·9% 2·2% (4-6–5·8) (1·1–1·6) (0·9–3·4) (1·9–2·5) 0 1 2 3 4 Number of metabolic conditions Articles referred for hepatology evaluation, individuals in whom Obesity chronic liver disease with fibrosis was confirmed had a GGT ≥1·5 × ULN higher baseline LSM compared with participants in whom it was not confirmed (11·0 vs 8·5 kPa). The optimal Type 2 diabetes LSM cutoff value at screening for predicting histological ALT ≥1·5 × ULN liver fibrosis was 10·3 kPa (Se 60% and Sp 73%; appendix Harmful alcohol use p 28) whereas the optimal cutoff value for predicting Male chronic liver disease with fibrosis was 10·0 kPa (Se 63% Arterial hypertension and Sp 75%; appendix p 28). LSM values at screening had Overweight a higher accuracy to diagnose chronic liver disease with fibrosis than liver enzymes alone or FIB-4 (appendix Age ≥60 (years) p 29). The prevalence of chronic liver disease with Dyslipidaemia fibrosis in participants with LSM at screening between 0 1 2 3 4 5 8 kPa and 9·9 kPa was 27% (115 of 429) versus 56% (284 of 504) in participants with LSM at screening of at Figure 4: Multivariable analysis of factors related to increased liver stiffness measurement (LSM; ≥8 kPa) least 10 kPa. Values shown are adjusted odds ratio and 95% CI, adjusted by country, sex, and all variables included in the model. ALT=alanine aminotransferase. GGT=gamma-glutamyl-transpeptidase. ULN=upper limit of normal. Discussion appendix (p 24). The decrease in LSM values at the In this large multinational cohort study of hepatology visit compared with LSM values at screening 30 199 individuals, who were randomly selected from the was observed in all fibrosis stages. We found chronic general population from nine European countries, we liver disease with fibrosis in 13% (47 of 439) of participants found that 4·6% of participants had an increased LSM who at screening had an ALT of at least 1·5 ULN without using VCTE, and 2·9% had an increased ALT, indicative increased LSM, 39% (305 of 773) participants with an of possible undiagnosed chronic liver disease. We LSM of 8 kPa and greater alone, and 67% (80 of 119) subsequently confirmed that at least 1·6% of the whole participants meeting both screening criteria. Most population had chronic liver disease with fibrosis. The individuals with chronic liver disease had steatotic liver presence and number of metabolic conditions, including disease (93% of 477 participants), of whom 73% had obesity, type 2 diabetes, dyslipidaemia, and arterial MASLD, 15% with MetALD, and 5% with alcohol-related hypertension were strongly associated with increased liver disease, while the remaining cases were due to LSM, and coexistence of harmful alcohol use further miscellaneous causes and only four individuals had viral increased this association. hepatitis (appendix p 24). Of the 342 participants with This study shows a high overall prevalence of metabolic unreliable or failed VCTE, 153 (45%) were evaluated, and conditions, including being overweight or people with 21 (14%) of 153 participants had chronic liver disease obesity, type 2 diabetes, arterial hypertension, and with fibrosis. dyslipidaemia, in addition to current alcohol consump- Among the participants who completed the liver tion of any amount (70% of participants had at least disease evaluation, the prevalence of chronic liver disease one metabolic risk factor, and 59% reported any alcohol with fibrosis was higher among those with metabolic consumption, respectively), closely matching national conditions and harmful alcohol use (appendix p 27). estimates and supporting the representativeness of the Type 2 diabetes and harmful alcohol use were the cohort. The prevalence of harmful alcohol use (6·1%) two conditions associated with the highest prevalence of was slightly lower than WHO estimates, possibly chronic liver disease with fibrosis. Among participants reflecting both non-standardised alcohol reporting across 1454 selbairaV LSM≥8 kPa LSM≥10 kPa LSM≥15 kPa n (aOR) 95% CI n (aOR) 95% CI n (aOR) 95% CI No risk factors 340 (1 [ref]) ·· 167 (1 [ref]) ·· 54 (1 [ref]) ·· Type 2 diabetes 106 (3·4) 2·7–4·3 67 (4·0) 3·0–5·4 28 (4·8) 3·0–7·8 Harmful alcohol use 52 (2·6) 1·9–3·5 29 (2·8) 1·9–4·2 15 (4·8) 2·7–8·7 Obesity 386 (3·8) 3·3–4·5 186 (3·6) 2·9–4·5 45 (2·6) 1·8–3·9 Type 2 diabetes and harmful alcohol use 18 (12·5) 7·2–21·6 12 (14·2) 7·4–27·0 6 (20·9) 8·6–51·2 Type 2 diabetes and obesity 273 (11·7) 9·8–14·0 168 (12·7) 10·1–16·0 57 (11·5) 7·7–17·0 Harmful alcohol use and obesity 40 (5·6) 4·0–8·0 26 (7·0) 4·5–10·8 9 (7·9) 3·8–16·3 Type 2 diabetes, harmful alcohol use and obesity 30 (25·9) 16·3–41·1 17 (22·5) 12·8–39·4 5 (18·7) 7·2–48·7 Data are n (aOR) and 95% CI, adjusted by age, sex, and country. aOR=adjusted odds ratio. LSM=liver stiffness measurement. Table 2: Logistic regression analysis of individual and combined risk factors with increased LSM (≥8, ≥10, and ≥15 kPa) Odds ratio For more on the national estimates see https://ec.europa. eu/eurostat Articles studies and under-reporting of alcohol consumption.22 cancer but requires proof in randomised controlled Among individual factors, obesity and type 2 diabetes studies. showed the strongest correlation with elevated LSM, Our findings provide relevant information for any whereas the combination of alcohol consumption and future screening strategies for liver fibrosis. Screening type 2 diabetes showed the strongest association across efforts should primarily target individuals with established risk factor combinations. risk factors, particularly type 2 diabetes, obesity, and MASLD was by far the most common aetiology of harmful alcohol use, although these conditions are chronic liver disease with fibrosis, followed by MetALD prevalent in a large proportion of the adult population. In and alcohol-related liver disease. Together, these such populations, an initial screening test should be three aetiologies accounted for 93% of all chronic liver followed by confirmatory evaluation supported by well- disease with fibrosis, underscoring the importance of structured organisational and logistical systems to ensure metabolic conditions and alcohol consumption as adherence to repeated assessment.3 In this study, LSM primary drivers of chronic liver disease with fibrosis in was applied as the initial screening test; however, in nearly European countries. These findings are consistent with a half of participants with an initial LSM of 8 kPa and modelling study suggesting that MASLD will become the greater, the values fell below this threshold at the follow-up leading cause of decompensated cirrhosis and liver assessment consistent with previous reports showing cancer in the near future.23 Nevertheless, the importance declines in 33–45% of cases on repeat assessment.28 In our of alcohol as a major aetiological factor of chronic liver cohort, individuals with a healthier liver profile— disease cannot be overstated, given that alcohol characterised by low liver enzyme concentrations, few risk consumption remains to be the highest in Europe among factors, and slight or moderate increase in baseline LSM all continents, and is increasing in many countries.24 values—were most likely to show such decreases, Moreover, under-reporting of alcohol consumption is a suggesting that test—retest variability of VCTE and well recognised limitation that can bias the estimated regression to the mean are important contributors, prevalence of SLD subcategories, likely leading to an particularly in low-prevalence populations.28 underestimation of the true contribution of alcohol.25 Nevertheless, lifestyle changes between visits that were The low prevalence of hepatitis B as an aetiological factor not captured could also have contributed. Importantly, of chronic liver disease in this study could be explained rather than implying that LSM is ineffective as a by the effect of the universal hepatitis B vaccination in all screening tool, these findings highlight the need for the included countries. structured confirmatory pathways to minimise false- Abnormalities in liver tests, particularly ALT and AST positive diagnoses in population-based screening and have for long been associated with the presence of ensure accurate identification of individuals with chronic liver disease of various aetiologies.26 Our previously unrecognised liver fibrosis. An alternative findings show that, although most liver tests were approach to a single LSM measurement is a stepwise associated with LSM elevation, ALT and GGT had the strategy using blood-based risk scores, such as guideline- strongest association. Nevertheless, it is important to recommended FIB-4, followed by confirmatory LSM emphasise that increased ALT in the absence of before specialist referral. In our cohort, FIB-4 increased LSM, was associated with chronic liver disease underperformed compared with LSM for detecting with fibrosis in only 13% of cases, compared with 67% chronic liver disease, suggesting that this is not a good when increased ALT occurred together with increased method for fibrosis screening in a low-prevalence LSM. In some countries, including the USA, GGT is not population. Whether blood-based tests specifically systematically used as a liver test. However, recent developed for general population screening (eg, LiverRisk, studies from 2021 have shown that GGT is an excellent LiverPRO, SAFE, MAF-5, or CLivD) could be more biomarker of liver fibrosis.27 Our results confirm these suitable as first-line tools and deserve further study.29–33 findings and support the use of GGT in the assessment Future studies should define optimal referral pathways of fibrosis. combining blood-based and imaging-based non-invasive Our study underscores the asymptomatic nature of tests and evaluate their cost-effectiveness and health- chronic liver disease, and the importance of imaging or system impact. Ultimately, randomised controlled trials blood-based testing for case finding patients before are needed to determine whether population-based decompensation. There are effective treatments to fibrosis screening improves clinical outcomes. Finally, reverse type 2 diabetes, obesity, and harmful alcohol use. our findings suggest that when using LSM with VCTE Furthermore, the two approved pharmacological agents for fibrosis screening in the general population, a to treat metabolic dysfunction-associated steatohepatitis, threshold of 10 kPa identifies substantial fibrosis (≥F2) resmetirom and semaglutide, have been approved for more accurately than 8 kPa, despite the reduction in moderate or advanced hepatic fibrosis but not cirrhosis.4,5 sensitivity. This finding has implications for the potential Therefore, early diagnosis followed by a therapy targeted use of LSM by VCTE as surrogate of fibrosis for at reversing disease aetiology could potentially prevent pharmacological treatment indication or guidance in the the progression to decompensated cirrhosis or liver future, indicating that confirmatory testing is likely to be Articles required before treatment initiation when fibrosis is more than 90% of confirmed chronic liver disease cases, identified through general population screening. which is a timely finding given the approval of therapies The main strengths of this study include the evaluation for MASLD and the emergence of pharmaceutical trials in of a large cohort of adults without known liver disease, alcohol-related liver disease and MetALD. These findings randomly assigned from nine European countries, the also highlight the importance and urgency of primary standardised use of VCTE across all participating centres prevention measures to reduce the widespread prevalence supported by structured training and identical devices, of metabolic risk factors and harmful alcohol consumption and the rigorous assessment of chronic liver disease across Europe. among participants who screened positive. The rigorous LiverScreen Consortium Investigators assessment of chronic liver disease individuals who Loes Alferink, Marifé Alvarez, Peter Andersen, Paolo Angeli, screened positive has been largely absent from previous Alba Ardèvol, Anita Arslanow, Ingrid Arteaga, Rafael Bañares, Katrine T Bech, Luca Beggiato, Zahia Ben Abdesselam, Lucy Bennett, population-based studies, limiting accurate estimation of Eva Bonfill, Bahija Boutouria, Alessandra Brocca, M Teresa Broquetas, the true prevalence of undiagnosed chronic liver disease Willem P Brouwer, Llorenç Caballeria, Valeria Calvino, Laurent Castera, with fibrosis. Almost two-thirds of participants with Marta Cervera, Carla Chacón, Mar Coll, Carmen Corps, either increased LSM or ALT values underwent a Marika Crescenzi, Fernando Cucchietti, Robert J de Knegt, Sarwa Darwish Murad, Ana de la Arada, Sönke Detlefsen, Alba Diaz, comprehensive hepatology assessment to confirm or José Diéguez Bande, Galadriel Diez-Fadrique, Petra Dinjar Kujundzic, rule out the presence of chronic liver disease with Peter Eddowes, Vanessa Esnault, Carme Expósito-Martínez, fibrosis. Although this proportion was lower than that Núria Fabrellas, Josep Lluis Falcó, Rosa Fernández, Céline Fournier, reported in established screening programmes, such as Matilde Fuentes, Desiree Fyle, Roberta Gagliardi, Peter R Galle, Edgar García, Zaira Garcia, Montserrat García-Retortillo, Esther Garrido, colorectal cancer screening (76–84%),34 several structural Marina Gigante, Pere Ginès, Rosa Gordillo Medina, and contextual factors likely contributed, including little Marie-Caroline Gourmelon, Jordi Gratacós-Ginès, Isabel Graupera, integration with hepatology services at some centres, Ivica Grgurevic, Simon J Griffin, Indra Neil Guha, Eva Guix, COVID-19-related disruptions and lower public Isaías Gundín, Rebecca Harris, Daniel Jan Havaj, Elena Hernández Boluda, Rosario Hernández-Ibáñez, awareness of chronic liver disease compared with cancer. María Teresa Hernando, Jordi Hoyo, Luis Ibáñez-Samaniego, This study has some limitations that should be Arfan MA Ikram, Simone Incicco, Mads Israelsen, Harry LA Janssen, acknowledged. First, the generalisability of our findings is Alba Jiménez-Masip, Adrià Juanola, Ralf Kaiser, Patrick S Kamath, restricted by the demographic characteristics of the cohort, Tom H Karlsen, Maria Kjærgaard, Harry J de Koning, Marko Korenjak, Aleksander Krag, Johanne Kragh Hansen, Marcin Krawczyk, which was predominantly White and included only adults Sandro Kukic, Christian Labenz, Philippe Laboulaye, Irina Lambert, aged 40 years and older. Liver fibrosis patterns can differ Frank Lammert, Simon Langkjær Sørensen, Sonia Lazaro Pi, in younger individuals and in ethnically diverse Elsa Ledain, Vincent Lévy, Katrine Prier Lindvig, Anne Llorca, populations due to variations in genetic susceptibility, Raquel López-Martos, Guirec Loyer, Ann T. Ma, Anita Madir, Michael Manns, Denise Marshall, M. Lluïsa Martí, metabolic phenotypes, and lifestyle factors. Accordingly, Alba Martínez-Escudé, Ricard Martínez Sala, Andrea Martini, our prevalence estimates should be applied with caution Roser Masa-Font, Maykon Diego Melo, Maurice Michel, in other demographic settings, and further studies in Jane Møller Jensen, Rosa M Morillas, Joël Moussy, Laura Muñoz, more diverse populations are warranted. Second, Ruth Nadal, JM Navarrete, Phillip N Newsome, Vibeke Nielsen, Gloria Osorio, Alba Pachón-Camacho, Laura Pagès, Frane Pastrovic, serological markers of hepatitis C and B were not tested in Guillem Pera, Martina Pérez-Guasch, Juan Manuel Pericàs, all participants. The uncommon cases with hepatitis C or B Salvatore Piano, Judit Pich, Elisa Pose, Judit Presas Escobet, and normal ALT and LSM could have been missed. Third, Jesse Pustjens, Francesc Ramos-Roure, Matthias Reichert, Diego Rojo, direct biomarkers of alcohol consumption, such as Dominique Roulot, Gemma Saez, Susana Sánchez Berdial, Maria Sánchez-Morata, Raul Sancho, Jörn M Schattenberg, phosphatidylethanol, were not measured. Fourth, fewer Helle Lindholm Schnefeld, Svetlana Adamcova-Selcanova, liver biopsies were performed than anticipated, which was Feliu Serra-Burriel, Miquel Serra-Burriel, Lubomir Skladany, likely to be due to a combination of factors, including a Louise Skovborg Just, Anna Sòria, Christiane Stern, Patricia Such, reduction in LSM values at the time of hepatology visits in Karolina Kristina Sulejova, Maja Thiele, Marta Tonon, Pere Toran, Antoni Torrejón, Emmanuel A Tsochatzis, Laurens A van Kleef, almost half of participants who were screened positive Ida Villesen, Paulien van Wijngaarden, Ana Viu, Zhanhui Wang Jiang, and the challenges in patient follow-up and referral Susanne Nicole Weber, Tracey Wildsmith, Nicola Zeni, imposed by the COVID-19 pandemic. Nevertheless, this Daniela Zilincanova, and Mirko Zoncapè. study represents the largest general population cohort to Contributors date in which the greatest number of liver biopsies has All authors reviewed the manuscript, had access to and reviewed the results of the data analyses, and take responsibility for the fidelity of the been performed. report to the protocol. The study coordinator and statistician team In summary, we found that almost 5% of the middle- (IGra, GP, PG, AS, JPi, EB, and AA) had full access to the complete aged general population across Europe exhibit evidence of dataset from all participating centres and verified the integrity of the data possible liver fibrosis. As elevated LSM strongly associates and the accuracy of the analyses. Each local principal investigator had access to, and takes responsibility for the accuracy and completeness of with type 2 diabetes, obesity, and harmful alcohol use, risk the data contributed by their centre (MTh, AK, PAng, SP, PT, WPB, factors that can be effectively treated by pharmaceutical DRou, LCas, IGrg, LS, JMP, FL, JS, RB, ET, ING, and RM). All authors and non-pharmaceutical interventions, there is a need for approved the decision to submit the manuscript for publication. IG and systematic strategies to ensure early case finding. MASLD, MT designed the study, acquired data, developed the database, designed the statistical analyses and interpretation, and drafted and revised the MetALD, and alcohol-related liver disease accounted for 1456 Articles manuscript. LC designed the study, acquired data, designed the consulting fees from GSK. DRoj reports support from Boehringer statistical analyses and interpretation, and drafted and revised the Ingelheim for travel expenses to attend the EU for the Study of Liver manuscript. GP designed the study, undertook statistical analyses and (EASL) congress 2025. SG has received honoraria from AstraZeneca and interpretation, and drafted and revised the manuscript. SP acquired data, Eli Lilly for contributing to postgraduate educational meetings; and has designed the statistical analyses and interpretation, and drafted and served as an unpaid trustee for the Novo Nordisk UK Research revised the manuscript. AS acquired data, interpreted data, and Foundation. TK has consulted or attended advisory boards for Gilead, contributed to manuscript drafting and revision. NF designed the study, Merck Sharp & Dohme, Boehringer Ingelheim, Falk Pharma, and interpreted data, and contributed to manuscript drafting and revision. Rectify Pharma. PSK has received consulting fees from GSK. PT, CC, KTB, HLS, MTo, SI, JM, VL, AM, SK, DJH, SA-S, JP, LAVK, JMS has received consulting fees from Akero, Alexion, Altimmune, AJ-M, LP, MZ, SNW, PRG, RH, and LI-S acquired data, interpreted data, Astra Zeneca, 89Bio, Boehringer Ingelheim, Gilead Sciences, GSK, and contributed to manuscript drafting and revision. AD and SD blindly Ipsen, Inventiva Pharma, Madrigal Pharmaceuticals, Merck Kríya read liver biopsies acquiring data, interpreted data, and contributed to Therapeutics, Echosens, Eli Lilly, E-Therapeutics, Merck Sharp & manuscript drafting and revision. MS-B designed the study, designed Dohme, Novo Nordisk, Roche, and Vantage Biosciences; reports the statistical analyses and interpretation, and drafted and revised the honoraria as a speaker for AbbVie, Boehringer Ingelheim, IPSEN, manuscript. AA, PAng, JPi, EB, MK, CF-P, AL, M-CG, HJK, and MP-G Echosens, Gilead Sciences, MedScape, Novo Nordisk, and Madrigal acquired data, interpreted data, and contributed to manuscript drafting Pharmaceuticals; reports stock or stock options from Hepta Bio; and also and revision. ATM acquired data, interpreted data, participated in the stock or stock options from HepaPharm. ET reports grants from formal analysis and methodology, and contributed to manuscript National Institute for Health and Care Research, European Association drafting and revision. AJ, EP, IA, LM, IV, JKH, VC, RG, BB, FP, PDK, for the Study of the Liver, Medical Research Council, and EU Horizon DZ, KKS, DRoj, RK, MDM, AT, RH-I, JH, RL-M, and RMM acquired 2020; has received consulting fees from Siemens, Boehringer, Merck data, interpreted data, and contributed to manuscript drafting and Sharp & Dohme, NovoNordisk, Orphalan, Univar, and Madrigal; and has revision. SG, MM, TK, PN, and PSK interpreted data and contributed to reported honoraria as a speaker for Boehringer, Echosens, AstraZeneca, manuscript drafting and revision. RB acquired data, interpreted data, AbbVie, Univar, Orphalan, Gilead, and NovoNordisk. WPB has and contributed to manuscript drafting and revision. ING designed the participated in advisory boards of Novo Nordisk; has participated in trials study, acquired data, interpreted data, and contributed to manuscript for Inventiva Pharma, 89BIO, and Boehringer Ingelheim; and has drafting and revision. JMS, FL, ET, WPB, JMP, LS, IGr, DR, PA acquired served as speaker for Novo Nordisk, Boehringer Ingelheim, and Eli Lilly. data, interpreted data and contributed to manuscript drafting and JMP reports grants from the European Commission, Vall d’Hebron revision. AK and LlC designed the study, interpreted the data, and University Hospital, La Caixa Foundation, and ISCIII and PERIS- contributed to manuscript drafting, and revision. PG designed the study, Generalitat de Catalunya; has received consulting fees from Boehringer interpreted the data, and drafted and revised the manuscript. Ingelheim, Novo Nordisk, and Madrigal; has received speaking fees from Boehringer Ingelheim, Novo Nordisk, and Madrigal; and has Declaration of interests received support from Boehringer-Ingelheim, Novo Nordisk, and IG reports receiving grants (grant number: PI22/00776) funded by Madrigal for attending meetings. DR has reported support for meetings Instituto de Salut Carlos III (ISC III) and co-funded by the EU and and travels from Gilead. PA has reported honoraria as a speaker for Pfizer NASH-Aspire programme; honoraria as a speaker for Boehringer Grifols and CSL Behring; and has reported a patent from Biovie. AK has Ingelheim and Mayoly; and participation on advisory boards of reported an EU Horizon 2020 grant award for this specific project; has Boehringer Ingelheim. MT reports receiving grants from GSK; has reported support for this Article from the Region of Southern Denmark; consulted or attended advisory boards for Boehringer Ingelheim, Astra and reports grants from Novo Nordisk Foundation, EU Horizon 2020, Zeneca, Novo Nordisk, and GSK; has been an unpaid board member at Innovation fund Denmark, Danish National Research Foundation, in Evido Health until 2025; is a board member at Alkohol & Samfund Region of Southern Denmark, AstraZeneca, and Institute for Healthcare until 2025; received speaking fees from Echosens, Madrigal, Takeda, and Improvement; has received royalties from Gyldendal; and has reported Novo Nordisk; and is co-founder at Evido Health. LC has consulted or consulting fees from Madrigal; honoraria as a speaker for Norgine, attended advisory boards for Boehringer Ingelheim, Boston Mescape, and NovoNordisk; patents from Region of Southern Denmark pharmaceutical, Echosens, Gilead, GSK, Madrigal, Merck Sharp & and University of Southern Denmark; has reported participation on Dohme, Novo Nordisk, Pfizer, Sagimet, and Siemens Healthineers, and advisory boards of Madrigal, GSK, Siemens, Novo Nordisk, Boehringer has received speaking fees from AstraZeneca, Boehringer Ingelheim, Ingelheim, and W4Cure; is Secretary of the General European Echosens, Gilead, Madrigal, and Novo Nordisk. SP reports receiving Association for the Study of The Liver 2023–25; and has received the grants from the EU and the Italian Ministry of Health; has received drug rifaximin from Norgine for an investigator-initiated study, free consulting fees from Boehringer Ingelheim and Plasma Protein Enhanced Liver Fibrosis test measurements from Siemens for an Therapeutics Association; reports honoraria as a speaker for Grifols and investigator-initiated study provision of FibroScan devices from Ferring; and has participated in an advisory board of the University of Echosens to conduct investigator-initiated research, including as part of Barcelona. AS was granted a Río Hortega grant by the EU and Instituto the LiverScreen project and extracellular matrix biomarker assays de Salud Carlos III, Acción Estratégica en Salud (grant number: supplied by Nordic Bioscience for investigator-initiated studies; CM23/00133) and by a mobility grant by Instituto de Salud Carlos III, and reports other financial or non-financial interests as a board member Acción Estratégica en Salud 2021–23 (grant number: MV24/00039). and co-founder of Evido. PG has received research funding from Gilead MTo has reported support for attending meetings from Gilead and & Grifols; has consulted or attended advisory boards for Gilead, RallyBio, Grifols. MSB has reported support for this Article from European SeaBeLife, Merck, Sharp and Dohme, Ocelot Bio, Boehringer, Roche Commission under the program H20/20 (grant number: 847989). LP has Diagnostics International, Boehringer Ingelheim, and Astra Zeneca; and received research funding from Río Hortega (grant number: has received speaking fees from Pfizer. All other authors have declared CM25/00085). PRG has consulted or attended advisory boards for Bayer, no competing interests. Boston Scientific, AstraZeneca, Adaptimmune, Bristol Myers Squibb, Eisai, Merck Sharp & Dohme, Sirtex, Lilly, Roche, and Guerbet; has Data sharing received speaking fees from Bayer, Boston Scientific, AstraZeneca, Data from this manuscript can be requested by qualified researchers. Adaptimmune, Bristol Myers Squibb, Eisai, Merck Sharp & Dohme, Before the use of the data, proposals need to be approved by all partners Sirtex, Lilly, Roche, and Guerbet; and has received support from Roche of the LiverScreen Consortium, and a data sharing agreement will need for attending meetings. RMM reports honoraria as a speaker for Gilead to be signed. Approval will depend on the scientific value of the proposal, and support for attending meetings and travel for Gilead and AbbVie. compatibility with the original patient consent, and data protection CFP reports other financial or non-financial interests as a full-time legislation. employee of Echosens; and has stock or stock options in Echosens. Acknowledgments MCG reports other financial or non-financial interests as a full-time The provider of the VCTE devices (Echosens, Paris, France) did not employee of Echosens. AL reports other financial or non-financial participate in the data analysis but reviewed the final manuscript. This interests as a full-time employee of Echosens. EP has received Articles study was funded by the European Commission under the program 16 de Franchis R, Bosch J, Garcia-Tsao G, et al. Baveno VII—renewing H20/20 (grant number 847989) to investigate screening for liver fibrosis consensus in portal hypertension. J Hepatol 2022; 76: 959–74. and its applicability in European countries and by a grant from the 17 Kwo PY, Cohen SM, Lim JK. ACG clinical guideline: evaluation of Italian Ministry of Health (grant number: RF-2018-12366098). This work abnormal liver chemistries. Am J Gastroenterol 2017; 112: 18–35. was supported by European Institute of Innovation and Technology 18 Newsome PN, Cramb R, Davison SM, et al. Guidelines on the (EIT) Health, a Knowledge and Innovation Community of EIT, which is management of abnormal liver blood tests. Gut 2018; 67: 6–19. a body of the EU (project no: 20308). This work was also funded by 19 Schwope RB, Katz M, Russell T, Reiter MJ, Lisanti CJ. The many Fondo de Investigación Sanitaria de Salud (grant number: PI18/01330 faces of cirrhosis. Abdom Radiol 2020; 45: 3065–80. funded by Instituto de Salud Carlos III, Spanish Ministry of Economy, 20 Liu X. Classification accuracy and cut point selection. Stat Med Industry, and Competitiveness and by the European Regional 2012; 31: 2676–86. Development Fund to Núria Fabrellas. We acknowledge Beatriz Márquez 21 Graupera I, Thiele M, Ma AT, et al, and the LiverScreen Consortium and Roser Poblet for their assistance in preparing the manuscript, and investigators. LiverScreen project: study protocol for screening for Nicki van Berckel for her support with project management. We also liver fibrosis in the general population in European countries. thank Laia Asso Ministral for her support in the development of the BMC Public Health 2022; 22: 1385. study. 22 WHO. Alcohol, health and policy response in the European Union in 2019. https://www.who.int/europe/publications/m/item/alcohol- References -health-and-policy-response-in-the-european-union-in-2019 1 Naghavi M, Ong KL, Aali A, et al, and the GBD 2021 Causes of (accessed Dec 15, 2025). Death Collaborators. Global burden of 288 causes of death and life 23 Le P, Tatar M, Dasarathy S, et al. Estimated burden of metabolic expectancy decomposition in 204 countries and territories and dysfunction-associated steatotic liver disease in US adults, 2020 to 811 subnational locations, 1990–2021: a systematic analysis for the 2050. JAMA Netw Open 2025; 8: e2454707–2454707. Global Burden of Disease Study 2021. Lancet 2024; 403: 2100–32. 24 Karlsen TH, Sheron N, Zelber-Sagi S, et al. The EASL-Lancet Liver 2 Ginès P, Krag A, Abraldes JG, Solà E, Fabrellas N, Kamath PS. Liver Commission: protecting the next generation of Europeans against cirrhosis. Lancet 2021; 398: 1359–76. liver disease complications and premature mortality. Lancet 2022; 3 Johansen S, Åberg F, Tsochatzis EA, Krag A. Screening for 399: 61–116. advanced steatotic liver disease. Lancet Gastroenterol Hepatol 2025; 25 Krag A, Torp N, Younossi ZM, Israelsen M. Reporting discrepancy 10: 842–54. of alcohol intake affecting estimated prevalence of MetALD and 4 Harrison SA, Bedossa P, Guy CD, et al, and the MAESTRO-NASH ALD. Lancet Gastroenterol Hepatol 2025; 10: 282–84. Investigators. A Phase 3, Randomized, Controlled Trial of 26 Radcke S, Dillon JF, Murray AL. A systematic review of the Resmetirom in NASH with Liver Fibrosis. N Engl J Med 2024; prevalence of mildly abnormal liver function tests and associated 390: 497–509. health outcomes. Eur J Gastroenterol Hepatol 2015; 27: 1–7. 5 Sanyal AJ, Newsome PN, Kliers I, et al, and the ESSENCE Study 27 Chen LW, Huang MS, Shyu YC, Chien RN. Gamma-glutamyl Group. Phase 3 trial of semaglutide in metabolic dysfunction- transpeptidase elevation is associated with metabolic syndrome, associated steatohepatitis. N Engl J Med 2025; 392: 2089–99. hepatic steatosis, and fibrosis in patients with nonalcoholic fatty 6 Castera L, Rinella ME, Tsochatzis EA. Noninvasive assessment of liver disease: a community-based cross-sectional study. liver fibrosis. N Engl J Med 2025; 393: 1715–29. Kaohsiung J Med Sci 2021; 37: 819–27. 7 Nilsson E, Anderson H, Sargenti K, Lindgren S, Prytz H. Clinical 28 Lindfors A, Strandberg R, Hagström H. Screening for advanced course and mortality by etiology of liver cirrhosis in Sweden: liver fibrosis due to metabolic dysfunction-associated steatotic liver a population based, long-term follow-up study of 1317 patients. disease alongside retina scanning in people with type 2 diabetes: Aliment Pharmacol Ther 2019; 49: 1421–30. a cross-sectional study. Lancet Gastroenterol Hepatol 2025; 10: 125–37. 8 Lin H, Lee HW, Yip TCF, et al, and the VCTE-Prognosis Study 29 Lindvig KP, Thorhauge KH, Hansen JK, et al. Development, Group. Vibration-controlled transient elastography scores to predict validation, and prognostic evaluation of LiverPRO for the prediction liver-related events in steatotic liver disease. JAMA 2024; of significant liver fibrosis in primary care: a prospective cohort 331: 1287–97. study. Lancet Gastroenterol Hepatol 2025; 10: 55–67. 9 Thiele M, Kamath PS, Graupera I, et al. Screening for liver fibrosis: 30 Sripongpun P, Kim WR, Mannalithara A, et al. The steatosis- lessons from colorectal and lung cancer screening. associated fibrosis estimator (SAFE) score: a tool to detect low-risk Nat Rev Gastroenterol Hepatol 2024; 21: 517–27. NAFLD in primary care. Hepatology 2023; 77: 256–67. 10 Luo N, Zhang X, Huang J, Chen H, Tang H. Prevalence of steatotic 31 Åberg F, Luukkonen PK, But A, et al. Development and validation liver disease and associated fibrosis in the United States: results of a model to predict incident chronic liver disease in the general from NHANES 2017-March 2020. J Hepatol 2024; 80: e70–71. population: the CLivD score. J Hepatol 2022; 77: 302–11. 11 Man S, Deng Y, Ma Y, et al. Prevalence of liver steatosis and fibrosis 32 Serra-Burriel M, Juanola A, Serra-Burriel F, et al, and the in the general population and various high-risk populations: LiverScreen Consortium Investigators. Development, validation, a nationwide study with 5·7 million adults in China. Gastroenterology and prognostic evaluation of a risk score for long-term liver-related 2023; 165: 1025–40. outcomes in the general population: a multicohort study. Lancet 12 Rinella ME, Lazarus JV, Ratziu V, et al, and the NAFLD 2023; 402: 988–96. Nomenclature consensus group. A multisociety Delphi consensus 33 van Kleef LA, Francque SM, Prieto-Ortiz JE, et al. Metabolic statement on new fatty liver disease nomenclature. J Hepatol 2023; Dysfunction-Associated Fibrosis 5 (MAF-5) score predicts liver 79: 1542–56. fibrosis risk and outcome in the general population with metabolic 13 Israelsen M, Francque S, Tsochatzis EA, Krag A. Steatotic liver dysfunction. Gastroenterology 2024; 167: 357–67. disease. Lancet 2024; 404: 1761–78. 34 Jørgensen SF, Larsen PT, Erichsen R, Andersen B, Rebolj M, 14 Roulot D, Costes JL, Buyck JF, et al. Transient elastography as a Njor S. Adherence to follow-up and resource use after abnormal screening tool for liver fibrosis and cirrhosis in a community-based FIT-screening: evaluation of the Danish colorectal cancer screening population aged over 45 years. Gut 2011; 60: 977–84. program. Endosc Int Open 2024; 12: e649–58. 15 Caballería L, Pera G, Arteaga I, et al. High prevalence of liver fibrosis among European adults with unknown liver disease: a population-based study. Clin Gastroenterol Hepatol 2018; 16: 1138–45. 1458 --- [PDF原文](https://sci-net.xyz/storage/7751365/09f4b1a5d2c5155f682304178b515c32c2769f7a34c8febab2cbbe70cc96d3a0/Prevalence-of-liver-fibrosis-in-the-general-population-the-LiverScreen-project-a-multinational.pdf) DOI: 10.1016/S0140-6736(26)00354-5