Bidirectional relationship between nonalcoholic fatty liver disease and serum creatinine-to-body weight ratio as a proxy for skeletal muscle mass index

Effects of the triglyceride-glucose index on non-alcoholic fatty liver disease: Causal evidence from longitudinal cohort studies

BACKGROUND

Insulin resistance (IR) is strongly related to non-alcoholic fatty liver disease (NAFLD). Triglyceride-glucose (TyG) index serves as a novel substitute indicator for IR. However, research on the effect of TyG index on NAFLD remains sparse. This study aims to investigate the causal association between TyG index and incident NAFLD.

METHODS

The primary cohort consisted of 27,052 participants from the Beijing Health Management Cohort, while the external validation cohort included 75,023 participants from the Taiwan MJ Cohort. Entropy balancing for continuous treatments (EBCT) combined with logistic regression and targeted maximum likelihood estimation (TMLE) were used to evaluate the causal association between the TyG index and incident NAFLD.

RESULTS

During a median follow-up of 2.49 years in the primary cohort, 6,168 participants (median age: 36.0 years) developed incident NAFLD. EBCT combined with logistic regression revealed the odds ratio (95 % CI) of NAFLD risk was 1.742 (1.478-2.054) for each 1-unit increase in the baseline TyG index. In the TMLE model, the risk ratio (95 % CI) for NAFLD was 1.540 (1.406-1.687) in the Q4 (quartile 4) group compared with the Q1 group. These findings were consistent with those from the external validation cohort, reinforcing the robustness of the causal relationship between the TyG index and NAFLD incidence.

CONCLUSIONS

The advanced double-robust estimation method suggests that a higher baseline TyG index may be causally associated with an increased NAFLD risk, providing more reliable evidence for its role as a simple biomarker and demonstrating the utility of double-robust estimation causal inference models in epidemiology.

Establishment of a machine learning predictive model for non-alcoholic fatty liver disease: A longitudinal cohort study

BACKGROUND AND AIMS

Non-alcoholic fatty liver disease (NAFLD) is a common chronic liver disease, which lacks effective drug treatments. This study aimed to construct an eXtreme Gradient Boosting (XGBoost) prediction model to identify or evaluate potential NAFLD patients.

METHODS AND RESULTS

We conducted a longitudinal study of 22,140 individuals from the Beijing Health Management Cohort. Variable filtering was performed using the least absolute shrinkage and selection operator. Random Over Sampling Examples was used to address imbalanced data. Next, the XGBoost model and the other three machine learning (ML) models were built using balanced data. Finally, the variable importance of the XGBoost model was ranked. Among four ML algorithms, we got that the XGBoost model outperformed the other models with the following results: accuracy of 0.835, sensitivity of 0.835, specificity of 0.834, Youden index of 0.669, precision of 0.831, recall of 0.835, F-1 score of 0.833, and an area under the curve of 0.914. The top five variables with the greatest impact on the onset of NAFLD were aspartate aminotransferase, cardiometabolic index, body mass index, alanine aminotransferase, and triglyceride-glucose index.

CONCLUSION

The predictive model based on the XGBoost algorithm enables early prediction of the onset of NAFLD. Additionally, assessing variable importance provides valuable insights into the prevention and treatment of NAFLD.