-
Views
-
Cite
Cite
Wei Fu, Junlong Zhao, Guobin Chen, Linya Lv, Letter to the Editor From Fu et al: “Machine Learning Reveals the Contribution of Lipoproteins to Liver Triglyceride Content and Inflammation”, The Journal of Clinical Endocrinology & Metabolism, Volume 110, Issue 2, February 2025, Pages e548–e549, https://doi.org/10.1210/clinem/dgae579
- Share Icon Share
Extract
We read with interest the recent study titled “Machine Learning Reveals the Contribution of Lipoproteins to Liver Triglyceride Content and Inflammation” by Tavaglione et al (1), published in The Journal of Clinical Endocrinology & Metabolism. This study, which explored the critical role of lipoprotein control in the management of metabolic-associated steatotic liver disease (MASLD) and metabolic-associated steatohepatitis (MASH), stands out for its comprehensive analysis and valuable insights. While expressing our deep appreciation for the meticulous work and notable contributions to this study, some questions and constructive suggestions are offered for further refinement.
First, to enhance the robustness of conclusions, it is advisable to employ more analytical methods. Although several key covariates, such as age, sex, body mass index, type 2 diabetes, and alcohol consumption, were adjusted, it is recommended to adjust for additional important covariates. These include socioeconomic factors (income, education level, occupation) (2), lifestyle factors (detailed dietary habits, physical activity, and smoking), and psychological factors (stigma, anxiety, and depression) (3). Utilizing hierarchical models to simultaneously adjust for individual- and group-level covariates as well as propensity score matching to balance the control and experimental groups can further improve the analysis. Sensitivity analyses should also be performed by excluding special populations (eg, excessive alcohol users, specific medication users, or individuals with unique genetic backgrounds) and by using varying disease definition standards (4).