Performance of AI-Enabled Electrocardiogram in the Prediction of Metabolic Dysfunction-Associated Steatotic Liver Disease.

BACKGROUND AND AIMS: Accessible noninvasive screening tools for metabolic dysfunction-associated steatotic liver disease (MASLD) are needed. We aim to explore the performance of a deep learning-based artificial intelligence (AI) model in distinguishing the presence of MASLD using 12-lead electrocardiogram (ECG). METHODS: This is a retrospective study of adults diagnosed with MASLD in Olmsted County, Minnesota, between 1996 and 2019. Both cases and controls had ECGs performed within 6 years before and 1 year after study entry. An AI-based ECG model using a convolutional neural network was trained, validated, and tested in 70%, 10%, and 20% of the cohort, respectively. External validation was performed in an independent cohort from Mayo Clinic Enterprise. The primary outcome was the performance of ECG to identify MASLD, alone or when added to clinical parameters. RESULTS: A total of 3468 MASLD cases and 25,407 controls were identified. The AI-ECG model predicted the presence of MASLD with an area under the curve (AUC) of 0.69 (original cohort) and 0.62 (validation cohort). The performance was similar or superior to age- and sex-adjusted models using body mass index (AUC, 0.71), presence of diabetes, hypertension or hyperlipidemia (AUC, 0.68), or diabetes alone (AUC, 0.66). The model combining ECG, age, sex, body mass index, diabetes, and alanine aminotransferase had the highest AUC: 0.76 (original) and 0.72 (validation). CONCLUSIONS: This is a proof-of-concept study that an AI-based ECG model can detect MASLD with a comparable or superior performance as compared with the models using a single clinical parameter but not superior to the combination of clinical parameters. ECG can serve as another screening tool for MASLD in the nonhepatology space.

Performance of American Gastroenterological Association Clinical Care Pathway for the risk stratification of patients with nonalcoholic fatty liver disease in the US population.

BACKGROUND AND AIMS: The American Gastroenterological Association (AGA) recently launched the Clinical Care Pathway for the Risk Stratification and Management of Patients with NAFLD to identify adults with significant fibrosis. We aimed to examine this pathway's performance in the US population. APPROACH AND RESULTS: Using the 2017-2018 National Health and Nutrition Examination Survey data, we identified participants aged ≥18 with available Fibrosis-4 (FIB-4) score and liver stiffness measurement (LSM) in the absence of other liver diseases. Based on the AGA clinical pathway, FIB-4 < 1.3 and LSM < 8 kilopascals (kPa) by vibration-controlled transient elastography (VCTE) are associated with low risk of significant fibrosis. Using these cutoffs, we examined the pathway performance using negative predictive value (NPV) and positive predictive value (PPV) and explored alternative risk-stratification strategies. There were 2322 participants with available data (projected to 94.2 million US adults). The NPV of LSM ≥ 8 kPa among those with FIB-4 < 1.3 was 90%, whereas the PPV among those with FIB-4 1.3-2.67 was 13%. As diabetes was a strong predictor of fibrosis, we propose a simple, alternative strategy to eliminate the indeterminate FIB-4 range and perform VCTE in those with FIB-4 ≥ 1.3 and diabetes. This strategy would decrease the number of VCTEs from 14.5 to 4.9 million and increase PPV from 13% to 33% without compromising the NPV among those who did not undergo VCTE. CONCLUSION: The implementation of the current AGA clinical pathway would lead to overutilization of VCTE. An alternative strategy using FIB-4 ≥ 1.3 and diabetes to select adults undergoing second-line testing will improve this pathway's performance and minimize unnecessary VCTEs.

Clinical course of non-alcoholic fatty liver disease and the implications for clinical trial design.

BACKGROUND & AIMS: The predicted risk and timeline to progression to liver-related outcomes in the population with NAFLD are not well-characterized. We aimed to examine the risk and time to progression to cirrhosis, hepatic decompensation and death in a contemporary population over a long follow-up period, to obtain information to guide endpoint selection and sample size calculations for clinical trials on NAFLD-related cirrhosis. METHODS: This is a retrospective study of prospectively collected data in a medical record linkage system, including all adults diagnosed with NAFLD between 1996-2016 by clinical, biochemical and radiological criteria in Olmsted County, Minnesota and followed until 2019. Liver-related outcomes and death were ascertained and validated by individual medical record review. Time and risk of progression from NAFLD to cirrhosis to decompensation and death were assessed using multistate modeling. RESULTS: A total of 5,123 individuals with NAFLD (median age 52 years, 53% women) were followed for a median of 6.4 (range 1-23) years. The risk of progression was as follows: from NAFLD to cirrhosis: 3% in 15 years; compensated cirrhosis to first decompensation: 33% in 4 years (8%/year); first decompensation to ≥2 decompensations: 48% in 2 years. Albumin, bilirubin, non-bleeding esophageal varices and diabetes were independent predictors of decompensation. Among the 575 deaths, 6% were liver related. Therapeutic trials in compensated cirrhosis would require enrolment of a minimum of 2,886 individuals followed for >2 years to detect at least a 15% relative decrease in liver-related endpoints. CONCLUSION: In this population-based cohort with 23 years of longitudinal follow-up, NAFLD was slowly progressive, with liver-related outcomes affecting only a small proportion of people. Large sample sizes and long follow-up are required to detect reductions in liver-related endpoints in clinical trials. LAY SUMMARY: For patients with compensated non-alcoholic steatohepatitis-related cirrhosis, the time spent in this state and the risk of progression to decompensation are not well-known in the population. We examined the clinical course of a large population-based cohort over 23 years of follow-up. We identified that adults with compensated cirrhosis spend a mean time of 4 years in this state and have a 10% per year risk of progression to decompensation or death. The risk of further progression is 3-fold higher in adults with cirrhosis and one decompensating event. These results are reflective of placebo arm risks in drug clinical trials and are essential in the estimation of adequate sample sizes.

Impact of Program Changes Including Telemedicine and Telephonic Care During the COVID-19 Pandemic in Preventing 30-Day Hospital Readmission for Patients in a Care Transitions Program.

INTRODUCTION/OBJECTIVES: To describe health outcomes of older adults enrolled in the Mayo Clinic Care Transitions (MCCT) program before and during the COVID-19 pandemic compared to unenrolled patients. METHODS: We conducted a retrospective cohort study of adults (age >60 years) in the MCCT program compared to a usual care control group from January 1, 2019, to September 20, 2022. The MCCT program involved a home, telephonic, or telemedicine visit by an advanced care provider. Outcomes were 30- and 180-day hospital readmissions, emergency department (ED) visit, and mortality. We performed a subgroup analysis after March 1, 2020 (during the pandemic). We analyzed data with Cox proportional hazards regression models and hazard ratios (HRs) with 95% CIs. RESULTS: Of the 1,012 patients total, 354 were in the MCCT program and 658 were in the usual care group with a mean (SD) age of 81.1 (9.1) years overall. Thirty-day readmission was 16.9% (60 of 354) for MCCT patients and 14.7% (97 of 658) for usual care patients (HR, 1.24; 95% CI, 0.88-1.75). During the pandemic, the 30-day readmission rate was 15.1% (28 of 186) for MCCT patients and 14.9% (68 of 455) for usual care patients (HR, 1.20; 95% CI, 0.75-1.91). There was no difference between groups for 180-day hospitalization, 30- or 180-day ED visit, and 30- or 180-day mortality. CONCLUSIONS: Numerous factors involving patients, providers, and health care delivery systems during the pandemic most likely contributed to these findings.