Dear Editor,
We have read with great interest the recent paper by Iwaki et al. [1], which investigated the prognosis of patients with biopsy-confirmed metabolic dysfunction-associated steatotic liver disease (MASLD). The study concluded that the prognosis of MASLD is similar to that previously reported for non-alcoholic fatty liver disease (NAFLD).
As the most prevalent chronic liver disease worldwide, steatotic liver disease (SLD) substantially impacts the health outcomes of a growing population [2,3]. The transition from the term “NAFLD” to “metabolic dysfunction‐associated fatty liver disease (MAFLD)” and “MASLD” marks a critical juncture in comprehending and communicating this prevalent condition [4,5]. The divergent definitions carry significant implications for clinicians, researchers, and policymakers, given the distinctive epidemiological profiles and outcomes linked to each definition. Beyond semantic discrepancies, it is crucial to consider how well each definition can predict health outcomes when deciding which definition to use.
We obtained data from the United States National Health and Nutrition Examination Survey (NHANES) III database, which enrolled participants from 1988 to 1994, with followup through December 31, 2019. According to the National Death Index (NDI), we defined individuals with death from heart disease as those with cardiovascular-related mortality (Supplementary Methods), which is the most common cause-specific mortality for NAFLD [6]. Those with poor health status or comorbid conditions at the start of the study were excluded to prevent bias.
Baseline characteristics were presented as medians for continuous variables and counts with percentages for categorical variables. We performed Cox regression models to determine the hazard ratio for mortality. The Akaike Information Criterion (AIC), which considers both the goodness of fit and the complexity of the model, was used to compare the relative quality of the models. A lower AIC value indicates a better balance between model fit and complexity.
Of 13,856 participants enrolled in this study, 4,708 demonstrated the absence of hepatic steatosis as proven by ultrasound. Stratified by the NAFLD, MAFLD, MASLD and metabolic dysfunction and alcohol-associated liver disease (MetALD) definitions, 3,896 participants were categorized in the MAFLD group, 2,190 in the NAFLD group, 2,205 in the MASLD group, and 235 in the MetALD group. The baseline characteristics of the respective cohorts are elucidated in Supplementary Table 1.
Compared to non-SLD individuals, those with steatosis liver disease—under various classification criteria—exhibited statistically significant correlations with mortality after adjusting for demographic variables, metabolic risk factors, and fibrotic risk factors (Supplementary Table 2). Among the three models, NAFLD and MASLD demonstrated comparable AIC values in predicting the prognostic efficacy for all-cause and cardiovascular-related mortality. Conversely, the AIC value for the MAFLD model was higher than that of its counterparts (Fig. 1A). After adjusting for all possible confounding factors, consistent outcomes were observed across all subgroup analyses, stratified by age, sex, and obesity status (Fig. 1B).
The findings suggest that the definition of MAFLD may be more complex and less effective at predicting all-cause and cardiovascular-related mortality. Indeed, the diagnostic criteria for MAFLD include non-specific serologic characteristics in addition to cardiometabolic traits, and they do not account for alcohol consumption, which complicates mortality risk assessment [5]. Conversely, the greater predictive value of MASLD for mortality renders comprehensive management of metabolic syndromes more crucial, including regular screening for cardiovascular risk factors and more aggressive lifestyle interventions.
Our study reveals that NAFLD and MASLD share similar hazard ratios and predictive power, aligning with the findings of Iwaki. Consequently, building on the findings of Iwaki, we further investigated the prognostic value of different definitions of steatotic liver disease. The primary distinctions between NAFLD and MASLD lie in the inclusion of other chronic liver diseases and cardiometabolic characteristics within the diagnostic criteria for MASLD. However, given the high prevalence of metabolic risk factors among NAFLD patients, prior research has shown minimal differences between MASLD and NAFLD [7]. Therefore, it is reasonable to assume that the results from NAFLD studies remain applicable under the new MASLD definition.
Nonetheless, we acknowledged the limitations of this study. Firstly, we focused only on mortality rates due to the lack of disease incidence data. However, mortality remains a standardized approach to disease burden. Secondly, ultrasound's lower sensitivity may miss mild hepatic steatosis, causing misclassification bias. However, despite lower sensitivity and lower hazard ratio (HR) values, our findings remained significant, thus proving the robustness of the conclusions instead. Thirdly, using self-reported data and lacking data on lifetime drinking patterns may introduce biases. Fourthly, the study cohort, enrolled over three decades ago, may not represent the current population. Therefore, prospective studies should track the long-term outcomes of MASLD patients using more precise diagnostic tools and novel longitudinal cohort studies.
In conclusion, our study suggested that NAFLD and MASLD have similar predictive power, whereas MAFLD is less effective in predicting all-cause and cardiovascular-related mortality. Our results favored using MASLD as the standard terminology for risk stratification in clinical practice.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
Print
