MASLD Fibrosis and ASCVD Risk in Prediabetes vs Diabetes

Key Takeaways

• In 6519 adults with MASLD at a Seoul health-screening center, mean predicted 10-year ASCVD risk rose stepwise across normoglycemia, prediabetes, and diabetes at 3.4%, 6.2%, and 15.2%.
• Higher KASL/EASL fibrosis severity also became more common as glycemia worsened, with the merged intermediate-high/high tier present in 1.3%, 2.3%, and 5.8% of those groups.
• Within prediabetes, greater fibrosis severity was associated with significantly higher 5-year clinical ASCVD incidence.
• In established diabetes, 5-year clinical ASCVD incidence remained high across fibrosis tiers without a significant gradient by fibrosis severity.
• Age-stratified analyses suggested that the fibrosis-ASCVD association was more apparent in younger individuals than in older adults.

Metabolic dysfunction-associated steatotic liver disease (MASLD) is a cardiometabolic condition in which atherosclerotic cardiovascular disease (ASCVD) is a major source of illness and death. In Seoul, South Korea, an unresolved question was whether a fibrosis-4-plus-vibration-controlled transient elastography pathway still separates cardiovascular risk once glycemia has advanced from normoglycemia to prediabetes or diabetes. Investigators therefore examined that sequential noninvasive staging approach in adults undergoing health screening at a single South Korean center.

In a single-center retrospective health-screening cohort at Severance Hospital in Seoul, South Korea, from 2012 to 2020, Jeong et al. cohort study of MASLD fibrosis stage and incident ASCVD in Hepatology Communications included adults with MASLD who underwent concurrent fibrosis-4 (FIB-4) testing and vibration-controlled transient elastography (VCTE).

The baseline cohort included 6519 participants, with 3243 classified as normoglycemia, 2369 as prediabetes, and 907 as diabetes; 3221 had follow-up records for survival analysis. Glycemic status was defined by prior diabetes diagnosis or antidiabetic medication use, or by fasting glucose and HbA1c thresholds; prediabetes was defined by fasting glucose/HbA1c criteria. The Korean Association for the Study of the Liver/European Association for the Study of the Liver (KASL/EASL) pathway used FIB-4 cutoffs of <1.3, 1.3–2.67, and >2.67, with a low-risk cutoff of <2.0 for adults 65 years or older, followed by liver stiffness cutoffs of <8.0 kPa, 8.0–12.0 kPa, and ≥12.0 kPa; the American Gastroenterological Association (AGA) pathway was also assessed. Incident clinical ASCVD included myocardial infarction, fatal coronary heart disease, stable or unstable angina, ischemic stroke, transient ischemic attack, and peripheral artery disease, while hemorrhagic stroke and nonischemic heart failure were excluded. Cumulative incidence was estimated with Aalen-Johansen methods, and associations were modeled with Fine-Gray competing-risk analyses using death as the competing event.

At baseline, mean predicted 10-year ASCVD risk rose from 3.4% in normoglycemia to 6.2% in prediabetes and 15.2% in diabetes.

The highest KASL/EASL fibrosis tier likewise became more common across those strata at 1.3%, 2.3%, and 5.8%. Predicted cardiovascular risk also increased within fibrosis tiers across all glycemic groups.

During follow-up, 269 incident clinical ASCVD events occurred among the 3221 participants in the survival cohort.

In prediabetes, 5-year clinical ASCVD incidence was 22% in the merged upper KASL/EASL category (intermediate-high/high) versus 9.5% in the low tier, with an adjusted subdistribution hazard ratio of 2.62 (95% CI 1.14–6.01; p=0.024). In diabetes, 5-year incidence remained high across tiers at 18.5%, 15.8%, and 15.2%, and adjusted hazards were not significantly different. Normoglycemia showed an upward but non-significant trend, and AGA-stratified hazard ratios could not be estimated reliably in some strata because no diabetic participants occupied the very low tier and some higher-risk groups had no events.

These findings come from a retrospective, single-center health-screening cohort, with fibrosis staged noninvasively rather than histologically and outcomes captured through one institutional electronic medical record system. Age and glycemic-control subgroup analyses were exploratory, and some AGA estimates were constrained by sparse events and by the pathway’s structure in diabetes. For North American readers, the cardiovascular comparison was framed with the American College of Cardiology/American Heart Association Pooled Cohort Equations, but the cohort itself came from a single South Korean center.

In this MASLD cohort, the association between fibrosis severity and incident ASCVD varied by glycemic status, with the clearest main-analysis signal in prediabetes and a more apparent pattern in younger individuals. Established diabetes was accompanied by high 5-year clinical ASCVD incidence regardless of fibrosis tier. FIB-4/VCTE pathways were not validated here as standalone ASCVD prediction tools or replacements for established risk calculators.

Clinician Questions

How did investigators define clinical ASCVD in adults with MASLD?

In this Seoul cohort, incident clinical ASCVD was the longitudinal composite outcome and included nonfatal myocardial infarction, fatal coronary heart disease, stable or unstable angina, ischemic stroke, transient ischemic attack, and peripheral artery disease; hemorrhagic stroke and nonischemic heart failure were excluded.

Why couldn’t the AGA fibrosis pathway be fully compared in MASLD patients with diabetes?

Within adults with MASLD and diabetes, the AGA framework could not provide a full reference-to-higher-tier comparison because its very low-risk category requires the absence of diabetes. The authors also reported that some intermediate- and higher-risk AGA strata had no incident ASCVD events, which prevented reliable subdistribution hazard ratio estimation.

Did age change how fibrosis severity related to ASCVD risk in MASLD?

Age-stratified analyses suggested that older adults had substantially higher baseline predicted ASCVD risk despite similar liver stiffness, while fibrosis-related separation of ASCVD risk appeared more evident in younger individuals. After multivariable adjustment, however, the age-stratified KASL/EASL associations did not remain statistically significant, so that subgroup pattern should be viewed as exploratory.

Can FIB-4 and VCTE be used as standalone ASCVD prediction tools in MASLD?

In adults with MASLD, FIB-4 and VCTE were not validated here as independent replacements for established ASCVD risk calculators. The analysis did not include time-dependent area-under-the-curve comparisons or direct head-to-head model validation, so fibrosis assessment is best interpreted as potentially additive information rather than a standalone cardiovascular prediction approach.