Utility of Fatty Liver Index to predict reversion to normoglycemia in people with prediabetes

by Carla Busquets-Cortés, Miquel Bennasar-Veny, Ángel Arturo López-González, Sergio Fresneda, Manuela Abbate, Aina M. Yáñez

Background

Fatty Liver Index (FLI) is strongly associated with changes in glycemic status and incident Type 2 Diabetes (T2D). The probability of reverting to normoglycemia from a state prediabetes could be determined by FLI, however such relationship remains poorly understood.

Aim

To determine the clinical interest of using FLI to estimate prediabetes reversion at 5 years in patients with impaired fasting plasma glucose at baseline, and identify those factors associated with changes in FLI, that could contribute to the reversion of prediabetes.

Methods

This 5-year cohort study included 16,648 Spanish working adults with prediabetes. Prediabetes was defined as fasting plasma glucose (FPG) between 100 and 125 mg/dl according to the ADA criteria, while prediabetes reversion was defined as a FPG <100 mg/dL. The population was classified as: FLI <30 (no hepatic steatosis), FLI 30–59 (intermediate status), and FLI ≥60 (hepatic steatosis).

Results

At 5 years follow-up, 33.7% of subjects reverted to normoglycemia (annual rate of 6.7%). The adjusted binomial logistic regression model showed that scoring FLI <30 (OR 1.544; 95% CI 1.355–1.759), performing at least 150 min/week of physical activity (OR 4.600; 95% CI 4.088–5.177) and consuming fruits and vegetables daily (OR 1.682; 95% CI 1.526–1.855) were associated with the probability of reverting form prediabetes to normoglycemia. The ROC curve for prediction of reversion showed that FLI (AUC 0.774;95% CI 0.767–0.781) was a better predictor than FPG (AUC 0.656; 95% CI 0.648–0.664).

Conclusions

Regular physical activity, healthy dietary habits and absence of hepatic steatosis are independently associated with the probability of reversion to normoglycemia in adult workers with prediabetes at baseline. Low FLI values (especially FLI< 30) may be useful to predict the probability of prediabetes reversion, especially in active subjects with healthy eating habits, and thus identify those who might benefit from early lifestyle intervention.