Objective:To explore the impact and predictive value of subclinical hypothyroidism on metabolic associated fatty liver disease (MAFLD).Methods:The study adopted a cross-sectional design. A total of 6 597 adults with subclinical hypothyroidism who received health check-ups in the Health Examination Center of Hebei General Hospital from 2018 to 2022 were selected as subjects. The participants were randomly divided into a modeling group (4 617) and a validation group (1 980) in a 7∶3 ratio. The basic information and serological indicators were collected in those subjects. The general clinical characteristics of the MAFLD and non-MAFLD participants were compared in the modeling group. Multivariate logistic regression analysis was used to explore the impact of thyroid hormone on MAFLD. The R language was applied to construct a nomogram prediction model, and internal validation and external validation of the model were performed to evaluate the prediction performance of the model.Results:Of the 6 597 study subjects, the prevalence of MAFLD was 31.14% (2 054/6 597). After preliminary comparison of baseline data from the modeling group, multivariate analysis showed that free triiodothyronine/free thyroxine (FT3/FT4) ( OR=20.945,95% CI: 5.862-74.838), male ( OR=1.394,95% CI: 1.109-1.752), age ( OR=1.017,95% CI: 1.009-1.025), diastolic blood pressure ( OR=1.017,95% CI: 1.004-1.021), waist circumference ( OR=1.049,95% CI:1.031-1.067), body mass index ( OR=1.236,95% CI:1.175-1.299), triglyceride ( OR=1.525,95% CI: 1.379-1.687), low density lipoprotein cholesterol ( OR=1.493,95% CI: 1.313-1.698), fasting glucose ( OR=1.173,95% CI:1.101-1.249) were risk factors for MAFLD, and high density lipoprotein cholesterol ( OR=0.243,95% CI: 0.163-0.363), aspartate aminotransferase/alanine aminotransferase ( OR=0.242,95% CI: 0.183-0.320) were protective factors for MAFLD (all P<0.05). The results showed that the two groups of calibration curves showed the consistency between MAFLD occurrence risk and actual occurrence risk. The AUCs of internal and external validation was 0.897 and 0.907, respectively. This nomogram model had a high predictive value. Conclusion:FT3/FT4 is an independent risk factor for MAFLD in the population with subclinical hypothyroidism. The prediction model incorporating thyroid function and metabolic parameters may provide clinical value for early identification of MAFLD.

Objective:To explore the impact and predictive value of subclinical hypothyroidism on metabolic associated fatty liver disease (MAFLD).Methods:The study adopted a cross-sectional design. A total of 6 597 adults with subclinical hypothyroidism who received health check-ups in the Health Examination Center of Hebei General Hospital from 2018 to 2022 were selected as subjects. The participants were randomly divided into a modeling group (4 617) and a validation group (1 980) in a 7∶3 ratio. The basic information and serological indicators were collected in those subjects. The general clinical characteristics of the MAFLD and non-MAFLD participants were compared in the modeling group. Multivariate logistic regression analysis was used to explore the impact of thyroid hormone on MAFLD. The R language was applied to construct a nomogram prediction model, and internal validation and external validation of the model were performed to evaluate the prediction performance of the model.Results:Of the 6 597 study subjects, the prevalence of MAFLD was 31.14% (2 054/6 597). After preliminary comparison of baseline data from the modeling group, multivariate analysis showed that free triiodothyronine/free thyroxine (FT3/FT4) ( OR=20.945,95% CI: 5.862-74.838), male ( OR=1.394,95% CI: 1.109-1.752), age ( OR=1.017,95% CI: 1.009-1.025), diastolic blood pressure ( OR=1.017,95% CI: 1.004-1.021), waist circumference ( OR=1.049,95% CI:1.031-1.067), body mass index ( OR=1.236,95% CI:1.175-1.299), triglyceride ( OR=1.525,95% CI: 1.379-1.687), low density lipoprotein cholesterol ( OR=1.493,95% CI: 1.313-1.698), fasting glucose ( OR=1.173,95% CI:1.101-1.249) were risk factors for MAFLD, and high density lipoprotein cholesterol ( OR=0.243,95% CI: 0.163-0.363), aspartate aminotransferase/alanine aminotransferase ( OR=0.242,95% CI: 0.183-0.320) were protective factors for MAFLD (all P<0.05). The results showed that the two groups of calibration curves showed the consistency between MAFLD occurrence risk and actual occurrence risk. The AUCs of internal and external validation was 0.897 and 0.907, respectively. This nomogram model had a high predictive value. Conclusion:FT3/FT4 is an independent risk factor for MAFLD in the population with subclinical hypothyroidism. The prediction model incorporating thyroid function and metabolic parameters may provide clinical value for early identification of MAFLD.

Objective:To explore the mediating effect of carotid atherosclerosis on serum triglyceride-glucose (TyG) index and silent lacunar infarction (SLI) in non-diabetic population.Methods:A total of 2 482 patients were selected from the Health Examination Center of Hebei General Hospital from January 2019 to December 2021. The basic demographic information, biochemical parameters, calculated TyG index and carotid plaque score were collected. SLI was diagnosed according to the criteria formulated by Chinese Society of Neurology. Participants were divided into SLI group and non-SLI group according to whether there was SLI in brain magnetic resonance imaging, and the non-SLI group was normal without lacunar infarction. Spearman correlation analysis was used to explore the correlation between TyG index and carotid plaque score. Binary Logistic regression analysis was used to analyze the effects of TyG index and carotid plaque score on SLI. Bootstrap was used to explore whether carotid plaque score mediated the association between TyG index and SLI.Results:There were 471 patients (18.98%) who had SLI, and 2 011 patients (81.02%) did not. Carotid plaque score [2(0, 5) vs 1(0, 3)] and TyG index [7.17(6.81, 7.64) vs 7.12(6.77, 7.54)] were increased in the SLI group compared with the non-SLI group ( Z=-4.213, Z=-2.636, P<0.05). Spearman correlation analysis showed that carotid plaque score was positively correlated with TyG index ( r=0.083, P<0.001). Multivariate Logistic regression analysis indicated that carotid plaque score ( OR=1.047, 95% CI 1.002-1.094) and TyG index ( OR=1.329, 95% CI 1.106-1.598) were independent risk factors for SLI ( P<0.05). Mediated effect analysis showed that TyG index had a direct effect on the incidence of SLI (β=0.265, 95% CI 0.102-0.428). Carotid plaque score partially mediated the effect of TyG index on the incidence of SLI (β=0.024, 95% CI 0.009-0.043), and the mediating effect accounted for 8.30% of the total effects. Conclusion:In non-diabetic population, TyG index and carotid plaque score are predictors of SLI, and the carotid plaque score is a partial mediator in the effect of TyG index on the incidence of SLI.