Relationship between impaired sensitivity to thyroid hormones and MAFLD with elevated liver enzymes in the euthyroid population

Aims To investigate the association between thyroid hormone sensitivity and metabolic dysfunction-associated fatty liver disease (MAFLD) with elevated liver enzymes in the euthyroid populations. Methods A total of 3929 euthyroid adults from the Second A�liated Hospital of Nanchang University were included in this study. Thyroid hormone sensitivity indices were calculated by thyroid feedback quantile-based index (TFQI), TSH index (TSHI) and thyrotropin thyroxine resistance index (TT4RI). Associations between thyroid hormones sensitivities and risk of MAFLD, MAFLD with elevated liver enzymes, MAFLD with elevated free fatty acid (FFAs) were assessed with logistic regression. Results After adjustment for multiple risk factors, odd ratios (ORs


INTRODUCTION
The overall prevalence of Non-alcoholic fatty liver disease (NAFLD) worldwide was estimated to be 32.4%.Prevalence increased signi cantly over time, from 25.5% in or before 2005 to 37.8% in 2016 or later. 1 In 2020, an international panel of hepatologists proposed a novel terminology, metabolic dysfunction-associated fatty liver disease (MAFLD). 2 Metabolic comorbidities associated with MAFLD included obesity, type 2 diabetes (T2DM), hyperlipidemia, hypertension, and metabolic syndrome. 3The pathogenesis of NAFLD is complex, involving insulin resistance, oxidative stress, lipid peroxidation, and mitochondrial dysfunction. 4High levels of free fatty acids (FFAs) in hepatocytes result in an imbalanced fatty acid metabolism and can cause mitochondrial dysfunction increasing oxidative stress and steatosis. 5Thyroid hormones stimulate lipolysis from fat stores in white adipose tissue and from dietary fat sources to generate circulating FFAs, which are the major source of lipids for the liver.hyperthyroidism increased triglyceride-derived fatty acid uptake in oxidative tissues such as liver and muscle, whereas hypothyroidism increased triglyceride-derived fatty acid uptake in white adipose tissue and decreased its uptake in liver. 6ert hypothyroidism is signi cantly associated with NAFLD and is a risk factor that is independent from other known metabolic risk factors. 7Of note, two studies from 2014 and 2016 demonstrate that serum levels of FT3 and FT4 are inversely associated and that TSH levels are associated with NAFLD in the general population, even among those within the reference range for euthyroid participants. 7,8However, the results from a large cohort study in euthyroid patients were inconsistent, higher serum FT3 and lower serum TSH levels were independently related to a higher incidence of MAFLD. 9These metabolic mechanisms described in clinical thyroid diseases may not be enough to explain the associations found within the normothyroid range.
Laclaustra et al. found that thyroid feedback quantile-based index (TFQI), a new index of central sensitivity to thyroid hormones, was linked with many metabolic disorders in the general population. 10So far, there has been some research on the association between sensitivity to thyroid hormone indices with the risk of T2DM, hyperlipidemia, hypertension, and metabolic syndrome.Nevertheless, the associations of thyroid hormones sensitivity with FFAs and liver enzymes levels remains unclear.Therefore, in this study, we aimed to investigate the relationship between thyroid hormones sensitivity and MAFLD with elevated liver enzymes in the euthyroid population.

Study population
This retrospective cohort study was approved by the Information Management Organization of the Second A liated Hospital of Nanchang University.Given that the study does not involve direct patient data or identi able information, as well as enrolment of participants, the Second A liated Hospital of Nanchang University Research Ethics Committee deemed the research to not require informed consent.
The participants consisted of 3,929 adults (age ≥ 18 years old) who were hospitalised between January 2017 and October 2022.The exclusion criteria were age 14 years old, missing data, type 1 diabetes mellitus, cancer, acute infection, autoimmune liver diseases, viral hepatitis, cirrhosis of liver, treatment with drugs associated with hepatic steatosis (including glucocorticoids, tamoxifen, amiodarone), thyroid dysfunction, replacement of thyroid hormones and anti-thyroid therapies.
A value of 0 indicates normal thyroid hormone sensitivity.Negative values indicate good thyroid hormone sensitivity.The higher the TSHI and TT4RI values, the lower the central sensitivity to thyroid hormone.

Other Variables
Height and weight were measured by standard methods, and body mass index (BMI) was calculated by dividing weight (kg) by the square of height (m).Blood pressure was measured using OMRON HEM-7130 blood pressure monitor with an appropriately sized cuff.Uric acid (UA), liver enzymes [alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma glutamyl transpeptidase (GGT)], and lipid pro le measurements [free fatty acid (FFA), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides (TG)] were measured using Automatic biochemical analyser (Cobas c800, Roche, Switzerland).Glycated haemoglobin (HbA1c) was measured using high-performance liquid chromatography (4500MD, AB SCIEX, USA).Elevated FFAs was de ned as FFAs ≥ 0.4mmol/L.Hyperuricemia was de ned as a uric acid (UA) level ≥ 420 µmol/L in males and ≥ 360 µmol/L (6.0 mg/dL) in females. 13Elevated liver enzymes were de ned as one or more measurement of ALT or AST > 50 U/L in males and > 35 U/L in females or GGT > 60 U/L in males and > 40 U/L in females. 14

MAFLD De nition
Patients were diagnosed as MAFLD based on abdominal ultrasonography diagnosed hepatic steatosis and the presence of any one of the following three conditions: 1) overweight or obesity (BMI ≥ 23kg/m 2 in Asian populations); 2) presence of type 2 diabetes mellitus (T2DM); 3) presence of at least two metabolic abnormalities, including increased waist circumference, arterial hypertension, hypertriglyceridemia, low HDL-C, prediabetes, subclinical in ammation, and insulin resistance. 2 Incident T2DM was de ned as a diagnosis of diabetes after discharge and HbA1c ≥ 6.5% or fasting blood glucose ≥ 7.0 mmol/L or random blood glucose ≥ 11.1 mmol/L, or have received insulin or oral hypoglycaemic drug treatment.Notably, waist circumference and insulin resistance were not assessed to identify metabolic dysfunction because of the scarcity of these data in the database of Second A liated Hospital of Nanchang University.
Normality was examined by the Kolmogorov-Smirnov test.Normally distributed variables were shown as mean and standard deviation, while skewed-distribution variables were described as median and interquartile range.The difference between two groups was compared by the independent samples T test and the Mann-Whitney U test.Correlations between two variables were assessed using Spearman correlations.Binary logistic regression analysis, adjusted for age, sex, BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), and smoking, were used to examine the association of thyroid hormone sensitivity index with MAFLD, MAFLD with elevated liver enzymes.Odds ratios (ORs) for MAFLD, MAFLD with elevated liver enzymes criteria were calculated for 1-SD increase of sensitivity to thyroid hormones indices.We created population-based quartiles for TFQI FT4 , interquartile comparisons were performed for MAFLD, elevated FFAs, hyperuricemia, elevated AST, elevated ALT, elevated GGT, MAFLD with elevated FFAs, and MAFLD with elevated liver enzymes were modeled with logistic regression.Model 1adjusted for sex, age, BMI, SBP, DBP, and smoking.Model 2 further adjusted for HbA1c, and T2DM.Model 3 further adjusted for FFAs ≥ 0.4mmol/L.To evaluate the performance of the thyroid hormone sensitivity indices, we examined the receiver operating characteristics curves (ROC), which plots sensitivity against 1speci city, and calculated the cut-points from ROC results.All calculated p values were two-sided, and a p value < 0.05 was considered statistically signi cant.

DISCUSSION
As far as we are aware, we are the rst to adopt MAFLD diagnostic criteria established by international expert consensus to evaluate the association between impaired sensitivity to thyroid hormones with the risk of MAFLD and MAFLD with elevated liver enzymes.We provide evidence of the association between indices measuring impaired sensitivity to thyroid hormones and prevalence of MAFLD, T2DM, hyperuricemia, elevated liver enzymes, elevated FFAs as well as between these indices and the risk of MAFLD with elevated liver enzymes in the euthyroid population, and these correlations remained signi cant even after adjusting for multiple risk factors.
The liver plays an essential physiological role in the regulation of thyroid hormones (THs), and vice versa, THs are essential for the metabolism of cholesterol, glucose, and hepatic lipids. 15,16THs act to increase cholesterol metabolism and simultaneously stimulates lipogenesis and fatty acid oxidation, with predominant action on oxidation. 17THs can stimulate hydroxymethyl glutaryl coenzyme A (HMG CoA) reductase, the key enzyme involved in cholesterol synthesis. 18Moreover, triiodothyronine (T3) can stimulate hepatic de novo lipogenesis, and this process is also seen in hyperthyroid patients. 19In addition, THs can stimulate FFAs delivery to the liver for re-esteri cation to TG by directly stimulating the fatty acids synthase and enzymes acetyl-CoA carboxylase, subsequently affecting fat accumulation in the liver. 20,21In hypothyroidism, the development and progression of MAFLD is enhanced by the decreased activity of hepatic lipase, which reduce the clearance of hepatic TG and promote hepatic TG accumulation. 6,22Furthermore, biologically, thyroxine (T4) must be converted into triiodothyronine (T3) in the body to exert its effects, multiple interdependent pathways involved in lipid metabolism are affected by higher FT3 levels, including increased the release of FFAs for TG synthesis in the liver, enhanced the delivery of nonesteri ed fatty acids to the liver. 9,19,23However, hepatic fatty acids in MAFLD may impair nuclear thyroid hormone receptor activity. 24Therefore, liver dysfunction might account for a major variation in the bioavailability of THs and clinical evidence also supports the different effects of various liver diseases on thyroid hormone metabolism. 25,26om a clinical point of view, the correlation between THs and MAFLD also remains disputable.Previous attempts to relate MAFLD with the thyroid axis either showed an association with TSH but not with fT4 or yielded inconsistent results. 27One cohort study revealed a negatively linear association between FT4 levels and incident MAFLD, but no signi cant association between TSH and incident MAFLD was observed. 8On the other side, one study demonstrated that FT3 levels were positively associated with incidence of MAFLD, and that TSH levels were negatively associated with incident MAFLD, but there was no association between FT4 and incident MAFLD. 9Previous studies have reported that the serum concentration of liver enzymes is frequently abnormal in patients with hypothyroidism, and hypothyroidism is associated with an increased risk of MAFLD. 6,15And, it is worth noting that even in the euthyroid subjects, elevated TSH levels were reported to associate with MAFLD and advanced brosis. 7,28,29Thus, we speculate that the contradictory results may re ect the close association between sensitivity to thyroid hormone with MAFLD.Our results offer an explanation for thyroid pro les commonly found in patients with MAFLD.That is, at the normal level, measurements of impaired sensitivity to thyroid hormones are cross-sectionally associated with MAFLD, independently of the degree of BMI, HbA1c, and FFAs, suggesting that there might be other underlying mechanisms linking MAFLD and impaired sensitivity to thyroid hormones.
The current study has some limitations.First, the cross-sectional design is insu cient to establish causality between sensitivity to thyroid hormone indices with risk of MAFLD and MAFLD with elevated liver enzymes.Second, liver biopsy was not used to accurately detect MAFLD, there was limited accuracy for detecting mild hepatic lipid accumulation.Finally, we did not evaluate thyroid-related antibodies, which might be potential confounding factors.Given that many factors may be associated with metabolic syndrome, it is not possible to adjust for unavailable variables and unknown factors, and the potential residual confounding must be considered when interpreting the study results.
In conclusion, higher values in indices of sensitivity to thyroid hormone are associated with MAFLD, MAFLD with elevated liver enzymes, T2DM, hyperuricemia, elevated liver enzymes, and MAFLD with elevated FFAs in the euthyroid population.This study provides evidence for the signi cance of thyroid hormones in their interactions with liver enzymes and fatty acid metabolism, while future studies are warranted to con rm these ndings and underlying mechanism.These indices may facilitate the monitoring and evaluation of new therapies that focus on energy expenditure.

Supplementary Files
This is a list of supplementary les associated with this preprint.Click to download.

Table 1
Baseline characteristics of participants with Non-MAFLD and MAFLD Cr creatinine, AST aspartate aminotransferase, ALT alanine aminotransferase, GGT gamma glutamyl transpeptidase, ALP alkaline phosphatase, T-BIL total bilirubin, D-BIL direct bilirubin, I-BIL indirect bilirubin, FT3 free triiodothyronine, FT4 free thyroxine, TSH thyroid stimulating hormone, TFQIFT3 the thyroid feedback quantile-based index calculated by FT3, TFQIFT4 the thyroid feedback quantile-based index calculated by FT4, TSHI TSH index, TT4RI thyrotropin thyroxine resistance index Correlation of thyroid parameters and serum lipid pro les and liver enzymes levels Table GGT (U/L)25.43 (17.30, 42.01) 22.60 (15.89, 37.11) 31.72 (21.10, 50.Data were expressed as the mean ± SD or median (upper and lower quartiles) or number (%).P values are calculated by t-test and Mann-Whitney tests for continuous variables, Chi-square test for categorical variables.Bold indicates p value < 0.01 BMI body mass index, SBP systolic blood pressure, DBP diastolic blood pressure, HbA1c glycated hemoglobin A1c, FFAs free fatty acids, TG triglyceride, TC total cholesterol, HDL-C high-density lipoprotein-cholesterol, LDL-C low-density lipoprotein-cholesterol, UA uric acid,

Table 2
Relationship of thyroid parameters with lipid pro les and liver enzymes .069(95% CI 1.613, 2.655) and 1.699 (1.287, 2.242), respectively, and the association was independent of T2DM and HbA1c, given that after adjustment, OR estimations were only partially reduced to 1.778 (1.378, 2.293) and 1.466 (1.105, 1.945), respectively (all p 0.001).For participants with MAFLD with elevated liver enzymes, the association with TFQI FT4 remained statistically signi cant even after adjustment for FFAs≥0.4mmol/L(p 0.01) (Table MAFLD prevalence, T2DM prevalence, MAFLD with elevated FFAs prevalence, MAFLD with elevated liver enzymes prevalence, elevated FFAs, hyperuricemia, elevated AST, elevated ALT, and elevated GGT were progressively higher across TFQI FT4 quartiles after adjustment for sex, age, BMI, SBP, DBP, and smoking (all p 0.001).ORs of the fourth versus the rst TFQI FT4 quartile for MAFLD and MAFLD with elevated liver enzymes were 2