Dear Editor,

Recently, a multi-society expert panel updated the terminology for steatotic liver disease (SLD), including metabolic dysfunction-associated SLD (MASLD), metabolic dysfunction and alcohol-related SLD (MetALD) [1,2]. This classification focuses on the roles of cardiometabolic risk factors and alcohol consumption. A recent study reported that the age-adjusted prevalence for MASLD was 31.9% and Met-ALD 2.2%, along with a significant rise in advanced fibrosis among individuals with SLD in the United States (US) [3]. The prevalence of MASLD is notably higher in individuals with cardiometabolic risk factors, especially in those with type 2 diabetes who have an increased risk of advanced fibrosis and cirrhosis [4]. We examined the prevalence of MASLD, Met-ALD, and hepatic fibrosis using transient elastography among adults with normal glucose tolerance, prediabetes, and diabetes in the US from 2017 to 2023.

We analyzed data from the 2017–2023 US National Health and Nutrition Examination Survey (NHANES). Of the 15,300 adults (≥18 years) examined at the mobile examination center (n=15,300), we excluded 3,101 individuals with incomplete alcohol questionnaires, missing clinical variables, or ineligible transient elastography data. Our final study cohort consisted of 12,199 adults with complete data. We evaluated the controlled attenuation parameter (CAP) score and liver stiffness measurements performed with Fibroscan model 502 V2 Touch (Echosens, Waltham, MA, USA). We applied two CAP score cut-offs: 1) ≥263 dB/m (sensitivity fixed at 90%) and 2) ≥285 dB/m (optimizing sensitivity and specificity) to categorize SLD [3,5]. In individuals with SLD, 1) MASLD (SLD with one or more cardiometabolic criteria) and 2) MetALD (both SLD and at least one cardiometabolic criteria, along with alcohol consumption of 140– 350 g/week for women and 210–420 g/week for men) were categorized. Transient elastography values of ≥8 kPa (≥F2) [6,7], ≥11.6 kPa (≥F3) [8], and ≥13.1 kPa (≥F4)7 indicated significant fibrosis, advanced fibrosis, and cirrhosis, respectively, in individuals with SLD. Diabetes was defined as either diagnosis of diabetes or HbA1c ≥6.5%, while prediabetes as HbA1c 5.7–6.4% in individuals without diabetes [9]. We applied sample weights to adjust for the complex NHANES design, re-constituting representative US population-level data. Agestandardized prevalence estimates were calculated using direct standardization with age-based proportions from the 2020 US Census, enabling comparisons stratified by diabetes status independent of age.

The mean age of this cohort was 47.2 years, with 50.0% men. The age-adjusted prevalence of prediabetes was 19.8% (95% confidence interval [CI] 18.6–21.0%), and diabetes was 13.1% (95% CI 12.3–14.0%). Figure 1A shows that the age-adjusted prevalence of MASLD (≥263 dB/m) was highest in individuals with diabetes (77.7%, 95% CI 73.4–81.9%), followed by prediabetes (56.5%, 95% CI 53.0–60.1%), and normal glucose tolerance (33.8%, 95% CI 31.8– 35.8%). When we applied another cut-off (≥285 dB/m) [5], the prevalence of MASLD was 23% in normal glucose tolerance, 42.4% in prediabetes, and 68.6% in diabetes (Fig. 1B). In individuals with MASLD (Fig. 1A), the age-adjusted prevalence of significant fibrosis (≥F2) was highest in diabetes (29.9%), followed by prediabetes (15.8%), and normal glucose tolerance (11.5%). For advanced fibrosis and cirrhosis, prevalences were 12.7%, 10.3% for diabetes, 6.1%, 4.8% for prediabetes, and 4.3%, 3.1% for normal glucose tolerance, respectively. Using a cut-off value of ≥285 dB/m to define MASLD, the prevalence of fibrosis and cirrhosis was nearly identical (Fig. 1B). Regarding MetALD, the ageadjusted prevalence was similar among individuals with normal glucose tolerance, prediabetes, and diabetes (Fig. 1C, 1D). While a slightly higher prevalence of advanced fibrosis and cirrhosis was observed in diabetes, there was no significant difference across the three glucose intolerance sub-groups.

In this study utilizing a nationally representative sample of US adults, we noted that (1) the prevalence of the MASLD was alarmingly high among individuals with diabetes (range 68.6–77.7%) and prediabetes (42.4–56.5%); (2) significant fibrosis and cirrhosis were observed in nearly one-third and one-tenth of individuals with MASLD and diabetes, respectively; (3) the prevalence of MetALD and associated hepatic fibrosis were comparable across different glucose status categories. The global prevalence of nonalcoholic fatty liver disease (NAFLD) in individuals with diabetes is estimated to be 55.5% (95% CI 47.3–63.7%), while in the United States, the prevalence is approximately 51.8% based on three studies [10]. Our study is the first to provide a population-based estimate of MASLD, MetALD, and fibrosis stages in individuals with diabetes and prediabetes. Our findings are comparable to the NHANES 2017-2018 analysis of individuals with previous nomenclature, NAFLD [9]. However, the prevalence of significant fibrosis was slightly higher in diabetes in the current study (24.6% vs. 29.9%) and notably higher in prediabetes (9.3% vs. 15.8%) [9]. Including cardiometabolic criteria in MASLD versus NAFLD enhances the accuracy of prediction of hepatic fibrosis. The limitations of our study include the absence of universally accepted cut-off guidelines for CAP score and liver stiffness. Nevertheless, we utilized cut-off points that are well-established in previous landmark studies [5-7]. Our findings reiterate that individuals with prediabetes and diabetes may benefit by screening for a pre-emptive diagnosis of MASLD and MASLD-related fibrosis.