An artificial intelligence-generated model predicts 90-day survival in alcohol-associated hepatitis: A global cohort study.

BACKGROUND AND AIMS: Alcohol-associated hepatitis (AH) poses significant short-term mortality. Existing prognostic models lack precision for 90-day mortality. Utilizing artificial intelligence in a global cohort, we sought to derive and validate an enhanced prognostic model. APPROACH AND RESULTS: The Global AlcHep initiative, a retrospective study across 23 centers in 12 countries, enrolled patients with AH per National Institute for Alcohol Abuse and Alcoholism criteria. Centers were partitioned into derivation (11 centers, 860 patients) and validation cohorts (12 centers, 859 patients). Focusing on 30 and 90-day postadmission mortality, 3 artificial intelligence algorithms (Random Forest, Gradient Boosting Machines, and eXtreme Gradient Boosting) informed an ensemble model, subsequently refined through Bayesian updating, integrating the derivation cohort's average 90-day mortality with each center's approximate mortality rate to produce posttest probabilities. The ALCoholic Hepatitis Artificial INtelligence Ensemble score integrated age, gender, cirrhosis, and 9 laboratory values, with center-specific mortality rates. Mortality was 18.7% (30 d) and 27.9% (90 d) in the derivation cohort versus 21.7% and 32.5% in the validation cohort. Validation cohort 30 and 90-day AUCs were 0.811 (0.779-0.844) and 0.799 (0.769-0.830), significantly surpassing legacy models like Maddrey's Discriminant Function, Model for End-Stage Liver Disease variations, age-serum bilirubin-international normalized ratio-serum Creatinine score, Glasgow, and modified Glasgow Scores ( p < 0.001). ALCoholic Hepatitis Artificial INtelligence Ensemble score also showcased superior calibration against MELD and its variants. Steroid use improved 30-day survival for those with an ALCoholic Hepatitis Artificial INtelligence Ensemble score > 0.20 in both derivation and validation cohorts. CONCLUSIONS: Harnessing artificial intelligence within a global consortium, we pioneered a scoring system excelling over traditional models for 30 and 90-day AH mortality predictions. Beneficial for clinical trials, steroid therapy, and transplant indications, it's accessible at: https://aihepatology.shinyapps.io/ALCHAIN/ .

Auditory and visual P300 event-related potentials to detect minimal hepatic encephalopathy

Introduction: the diagnosis of minimal hepatic encephalopathy (MHE) requires psychometric tests, although new methods are needed since their sensitivity, specificity, and accuracy are low. The P300 event-related potential (ERP) is obtained by auditory and visual stimuli, although only the auditory P300 has been used to detect MHE. This study aimed to compare the diagnostic features of auditory and visual P300 to detect MHE.Materials and methods: sixty patients with liver cirrhosis and thirty-five healthy controls completed the Psychometric Hepatic Encephalopathy Score (PHES), the critical flicker frequency (CFF), and auditory and visual P300 tests. MHE was diagnosed if PHES and CFF scores were abnormal.Results: fifty-three cirrhotic patients (aged 54.5 ± 8.6 years) completed all tests. Abnormal scores were obtained for PHES (49.1 %) and CFF (67.9 %). The proportion of MHE was 21.4 %. The area under the receiver operating (ROC) curves (AUROC) for auditory P300 was better than for visual P300 for distinguishing MHE from controls (AUROC: 0.792 vs 0.725; p < 0.005 for both; accuracy: 73.8 % vs 70.2 %; sensitivity: 72.2 % for both; specificity: 74.2 vs 69.7, respectively). Among cirrhotic patients, only auditory P300 was useful to detect MHE (AUROC: 0.723; p < 0.05; 77.4 % accuracy; 61.1 % sensitivity, and 81.8 % specificity). Conclusions: auditory P300 sensitivity, specificity, and accuracy were similar to those of CFF. Our results showed that only auditory P300 is useful to differentiate patients with MHE, although both modalities, auditory and visual, differentiated patients with cirrhosis from controls. Thus, we consider that visual P300 is not suitable for detecting MHE. (AU)

Malnutrition is a key prognostic factor related to high mortality-rate in patients with severe alcoholic hepatitis / La desnutrición es un factor pronóstico clave relacionado con elevada mortalidad en pacientes con hepatitis alcohólica severa

Background and aim: comparatively with European or North-American populations, severe alcoholic hepatitis has a high mortality rate in Mexican population, becoming as high as 50 to 81% in those classified as ABIC B or C; this is true even when they receive specific therapy with steroids or pentoxifylline. The aim of this study was to know which clinical factors are related to early mortality (first 30 days) in Mexican patients with severe alcoholic hepatitis. Subjects and methods: this was a retrospective cohort study that included patients with severe alcoholic hepatitis, defined by a Maddrey’s discriminant function ≥ 32, treated at a tertiary care center in a period of five years (2010 to 2015).Results: seventy-six patients were included, 72 (94.7%) were males, mean age was 43 ± 9.1 year-old, and 58 (76.3%) had also cirrhosis. According to the subjective global assessment (SGA), 38 (50%) had severe malnutrition, 22 (28.9%) were at risk of malnutrition, and 16 (21.1%) were well-nourished. At 30 days, 46 patients (60.5%) died. In the multivariate analysis, only the presence of severe malnutrition was associated with 30-day mortality: OR = 6.4; 95% CI: 1.9-22.1; p = 0.003.Conclusions: the nutritional status seems to be a cardinal prognostic factor associated with early mortality (first 30 days). Malnutrition can explain the high mortality rate observed in Mexican patients with severe alcoholic hepatitis

Antecedentes: en comparación con otras poblaciones europeas o norteamericanas, la hepatitis alcohólica severa tiene una mortalidad muy elevada en población mexicana atendida en hospitales federales (generalmente, pacientes de medio socioeconómico bajo) y llega a alcanzar entre el 50% y el 81% en aquellos catalogados como ABIC B o C; esto es cierto a pesar del tratamiento específico con esteroide o pentoxifilina. Objetivo: conocer qué factores clínicos están asociados a una mortalidad temprana (30 días) en pacientes mexicanos con hepatitis alcohólica severa. Material y métodos: se realizó un estudio de cohorte histórica que incluyó a pacientes con hepatitis alcohólica severa, definida por una función discriminante de Maddrey mayor a 32, atendidos en el Hospital General de México durante un periodo de cinco años (2010-2015). Resultados: se incluyeron 76 pacientes, de los cuales 72 (94,7%) fueron hombres, la media de edad fue de 43 ± 9,1 años; 58 (76,3%) de ellos tenían alteraciones sugestivas de cirrosis en el ultrasonido. De acuerdo con la valoración global subjetiva (VGS), 38 (50%) presentaban desnutrición grave, 22 (28,9%) se encontraban en riesgo de desnutrición y 16 (21,1%) estaban bien nutridos. La mortalidad global a 30 días en esta cohorte fue de 46 pacientes (60,5%). En un modelo multivariado, solo la desnutrición severa se asoció con mortalidad a 30 días: OR = 6,4; IC 95%: 1,9-22,1; p = 0,003. Conclusión: el estado nutricional es el determinante más importante asociado a mortalidad temprana (30 días). La desnutrición severa explica claramente la elevada mortalidad que se observa en pacientes con hepatitis alcohólica severa, en población mexicana