Background <p>Metabolic dysfunction-associated steatotic liver disease (MASLD) and type 2 diabetes mellitus (T2DM) are increasingly prevalent metabolic disorders worldwide, with a complex bidirectional relationship between them. Currently, there is a lack of research on the trajectories of blood glucose and insulin concentration over time after eating in patients with MASLD and T2DM.</p> Methods <p>This clinical cohort included diagnosed T2DM patients in a large hospital over the past five years, was divided into an observation group (with MASLD) and a control group (without MASLD). The postprandial time trends of blood glucose and insulin concentration were analysed within two hours after eating. A strategy of backward iterative feature elimination combined with propensity score matching (PSM) was employed to screen for potential associated factors that might influence these trends.</p> Results <p>In total, there were 521 in the observation group and 373 in the control group. In terms of blood glucose, the postprandial time-concentration trajectories for both groups shown a significant time main effect (<i>F</i> = 1145.567, <i>P</i> &lt; 0.001), a significant group main effect (<i>F</i> = 15.340, <i>P</i> &lt; 0.001), and a significant time*group interaction effect (<i>F</i> = 2.873, <i>P</i> = 0.035); After matching all the factors, the time*group interaction effect of blood glucose was not significant, but the differences from group main effect still existed. In terms of insulin, the postprandial time-concentration trajectories for both groups shown a significant time main effect (<i>F</i> = 309.429, <i>P</i> &lt; 0.001), a significant group main effect (<i>F</i> = 6.319, <i>P</i> &lt; 0.012), and a significant time*group interaction effect (<i>F</i> = 20.057, <i>P</i> &lt; 0.001), but the trajectories crossed; After matching 4 factors such as Smoking, Essential Hypertension (EH), Body Mass Index (BMI), Triglyceride (TG) and Ca<sup>2+</sup>, neither the group main effect nor the time*group interaction effect on insulin was significant any more.</p> Conclusion <p>The postprandial trends of blood glucose and insulin concentration over time shown significant differences between T2DM patients with and without MASLD. IL-6 might be associated with the insulin resistance, while EH and Ca<sup>2+</sup> might be related to the islet β-cell function. Smoking and TG might participate in both of the above processes. The strategy of backward iterative with PSM had demonstrated a relatively satisfactory effect in feature screening.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Analysis of postprandial time trends and influencing factors of blood glucose and insulin in type 2 diabetes mellitus (T2DM) with metabolic dysfunction-associated steatotic liver disease (MASLD): a retrospective study based on propensity score matching (PSM)

  • Ge-yao Qi,
  • Fei Wang,
  • Yuan-bo Shi,
  • Juan Feng,
  • Jin Xu

摘要

Background

Metabolic dysfunction-associated steatotic liver disease (MASLD) and type 2 diabetes mellitus (T2DM) are increasingly prevalent metabolic disorders worldwide, with a complex bidirectional relationship between them. Currently, there is a lack of research on the trajectories of blood glucose and insulin concentration over time after eating in patients with MASLD and T2DM.

Methods

This clinical cohort included diagnosed T2DM patients in a large hospital over the past five years, was divided into an observation group (with MASLD) and a control group (without MASLD). The postprandial time trends of blood glucose and insulin concentration were analysed within two hours after eating. A strategy of backward iterative feature elimination combined with propensity score matching (PSM) was employed to screen for potential associated factors that might influence these trends.

Results

In total, there were 521 in the observation group and 373 in the control group. In terms of blood glucose, the postprandial time-concentration trajectories for both groups shown a significant time main effect (F = 1145.567, P < 0.001), a significant group main effect (F = 15.340, P < 0.001), and a significant time*group interaction effect (F = 2.873, P = 0.035); After matching all the factors, the time*group interaction effect of blood glucose was not significant, but the differences from group main effect still existed. In terms of insulin, the postprandial time-concentration trajectories for both groups shown a significant time main effect (F = 309.429, P < 0.001), a significant group main effect (F = 6.319, P < 0.012), and a significant time*group interaction effect (F = 20.057, P < 0.001), but the trajectories crossed; After matching 4 factors such as Smoking, Essential Hypertension (EH), Body Mass Index (BMI), Triglyceride (TG) and Ca2+, neither the group main effect nor the time*group interaction effect on insulin was significant any more.

Conclusion

The postprandial trends of blood glucose and insulin concentration over time shown significant differences between T2DM patients with and without MASLD. IL-6 might be associated with the insulin resistance, while EH and Ca2+ might be related to the islet β-cell function. Smoking and TG might participate in both of the above processes. The strategy of backward iterative with PSM had demonstrated a relatively satisfactory effect in feature screening.