Low-normal thyroid function, and subclinical hypothyroidism were present in 23.3% and 7.1% of participants, respectively. Prevalence of ultrasonographic-defined and CAP-defined MASLD in the overall population was 39.2%, and 52.8%, respectively.

As shown in Table 1, age, systolic/diastolic BP, FPG, HbA1c, TC, ALT, AST, LSM, and TSH showed a progressive increase from the group with strict-normal thyroid function to the group with overt hypothyroidism (all P for trend < 0.05). This was also the case for the prevalence of cardiometabolic disorders (all P for trend < 0.05).

FT3 and fT4 showed a progressive decrease (P for trend < 0.001).

Since accumulating evidence has shown sex disparities in the epidemiology, progression, and outcomes of hypothyroidism and MASLD [15, 16], we stratified the analyses by sex. In men, the MASLD prevalence increased from 50.3% in individuals with strict-normal thyroid function to 62.9% in patients with overt hypothyroidism using ultrasonographic definition (Fig. 2A), and from 63.9% to 72.8% in the same categories using CAP definition (P trend < 0.001, Fig. 2B). In women, the corresponding figures were 16.5% and 33.4%, respectively, using ultrasonographic definition (P trend < 0.001, Fig. 2A) and 28.9% and 49%, respectively, using CAP definition, respectively (P trend < 0.001, Fig. 2B).

Metabolic dysfunction-associated steatotic liver disease (MASLD) and fibrotic burden across the spectrum of hypothyroidism in men and women. A Prevalence of ultrasonographic-defined MASLD across the spectrum of hypothyroidism in men and women; B Prevalence of controlled attenuation parameter-defined MASLD across the spectrum of hypothyroidism in men and women; C Prevalence of fibrotic stages across the spectrum of hypothyroidism in men; D Prevalence of fibrotic stages across the spectrum of hypothyroidism in women; E Prevalence of fibrotic burden in each hypothyroidism status in men. F Prevalence of fibrotic burden in each hypothyroidism status in women

Liver fibrosis stage across the spectrum of hypothyroidism is presented in Fig. 2C–F. Among both men and women, low-normal thyroid function group, subclinical hypothyroidism group, and overt hypothyroidism group all have more fibrosis present, including mild fibrosis, significant fibrosis, advanced fibrosis, and cirrhosis, than strict-normal thyroid function group (P trend < 0.001, Fig. 2C–F). The low-normal thyroid function group have the similar liver fibrotic burden to subclinical hypothyroidism group. The highest liver fibrotic burden was noted in the overt hypothyroidism group. The significant fibrosis and advanced fibrosis reached 16.0% and 9.6%, respectively, in patients with overt hypothyroidism in men (Fig. 2C). In women, the corresponding figures were 8.4% and 3.9%, respectively (P trend < 0.001, Fig. 2D).

Since it is common for an individual to have multiple cardiometabolic disorders and each additional cardiometabolic disorder increases the risk of fibrosis progression [11], we further investigated the effect of the number of cardiometabolic disorders on fibrotic burden across the spectrum of hypothyroidism. As shown in Fig. 3A–H, among both genders, for a given thyroid function, for both the MASLD definition, the proportions of significant and advanced fibrosis increased progressively from individuals having one cardiometabolic disorder to individuals having two, three, four, and five cardiometabolic disorders (all P trend < 0.001). For example, significant liver fibrosis in male patients with subclinical hypothyroidism was present in 8.4%, 10.7%, 13.2%, 20.6%, and 28.1% of MASLD (CAP definition) patients having one, two, three, four, and five cardiometabolic disorders (Fig. 3C). The corresponding figures in women were 5.6%, 9.1%, 11%, 18.6%, and 19.4% (Fig. 3D), respectively. Both the proportions of significant and advanced fibrosis were the highest in patients with overt hypothyroidism and five cardiometabolic disorders among both genders.

Fibrotic burden across the spectrum of hypothyroidism by metabolic dysfunction-associated steatotic liver disease (MASLD) status and the number of cardiometabolic disorders in men and women. The term “5 risk factors” indicates individuals with five cardiometabolic risk factors that used to define MASLD. A Prevalence of significant liver fibrosis in male patients with ultrasonographic-defined MASLD across the spectrum of hypothyroidism by the number of cardiometabolic disorders; B Prevalence of significant liver fibrosis in female patients with ultrasonographic-defined MASLD across the spectrum of hypothyroidism by the number of cardiometabolic disorders; C Prevalence of significant liver fibrosis in male patients with CAP-defined MASLD across the spectrum of hypothyroidism by the number of cardiometabolic disorders; D Prevalence of significant liver fibrosis in female patients with CAP-defined MASLD across the spectrum of hypothyroidism by the number of cardiometabolic disorders; E Prevalence of advanced liver fibrosis in male patients with ultrasonographic-defined MASLD across the spectrum of hypothyroidism by the number of cardiometabolic disorders; F Prevalence of advanced liver fibrosis in female patients with ultrasonographic-defined MASLD across the spectrum of hypothyroidism by the number of cardiometabolic disorders; G Prevalence of advanced liver fibrosis in male patients with CAP-defined MASLD across the spectrum of hypothyroidism by the number of cardiometabolic disorders; H Prevalence of advanced liver fibrosis in female patients with CAP-defined MASLD across the spectrum of hypothyroidism by the number of cardiometabolic disorders

Because hypothyroidism often coexisted with diabetes and obesity and these two factors outperformed other cardiometabolic disorders in driving fibrosis progression [11], among both men and women, we additionally divided MASLD individuals into three subgroups in the following manner. First, we determined MASLD + diabetes subgroup based on the presence of diabetes (Group 1). Next, in the absence of diabetes, we determined MASLD + BMI ≥ 23 kg/m2 subgroup based on the presence of BMI ≥ 23 kg/m2 (Group 2). Last, the rest of MASLD patients were categorized as MASLD + other cardiometabolic disorders (Group 3). Among both genders, both significant fibrosis (Fig. 4A–D) and advanced fibrosis (Fig. 4E–H) increased progressively with increasing hypothyroidism grade in patients with each MASLD subgroup, regardless of the definition (P trend < 0.001). For example, in men, advanced fibrosis in group 1 (MASLD was defined by CAP value) was present in 12.1% of participants with strict-optimal thyroid function and increased in prevalence in those with low-normal thyroid function (12.3%), subclinical hypothyroidism (13%), and overt hypothyroidism (18.3%) (P trend < 0.001) (Fig. 4G). Moreover, for a given thyroid function, for both the MASLD definition, the proportions of significant and advanced fibrosis were the highest in group 1, followed by group 2, and group 3 (P trend < 0.001) (Fig. 4A–H). For example, in men, advanced fibrosis in patients with subclinical hypothyroidism was present in 13%, 7%, and 6.5% of participants in group 1, group 2, and group 3 (MASLD was defined by CAP value), respectively (P trend < 0.001) (Fig. 4G).

Fibrotic burden across the spectrum of hypothyroidism by metabolic dysfunction-associated steatotic liver disease (MASLD) status and cardiometabolic traits in men and women. The term “other risk factors” indicates MASLD individuals with cardiometabolic risk factors that used to define MASLD in the absence of diabetes and overweight/obesity. A Prevalence of significant liver fibrosis in ultrasonographic-defined MASLD across the spectrum of hypothyroidism by cardiometabolic traits in men; B Prevalence of significant liver fibrosis in ultrasonographic-defined MASLD across the spectrum of hypothyroidism by cardiometabolic traits in men in women; C Prevalence of significant liver fibrosis in CAP-defined MASLD across the spectrum of hypothyroidism by cardiometabolic traits in men; D Prevalence of significant liver fibrosis in CAP-defined MASLD across the spectrum of hypothyroidism by cardiometabolic traits in women; E Prevalence of advanced liver fibrosis in ultrasonographic-defined MASLD across the spectrum of hypothyroidism by cardiometabolic traits in men; F Prevalence of advanced liver fibrosis in ultrasonographic-defined MASLD across the spectrum of hypothyroidism by cardiometabolic traits in women; G Prevalence of advanced liver fibrosis in CAP-defined MASLD across the spectrum of hypothyroidism by cardiometabolic traits in men; H Prevalence of advanced liver fibrosis in CAP-defined MASLD across the spectrum of hypothyroidism by cardiometabolic traits in women

Because hypothyroidism correlates strongly with dyslipidemia and it can promote fibrosis progression [11, 17], we also stratified MASLD individuals by dyslipidemia status. Both significant and advanced liver fibrosis were more present in low-normal thyroid function group, subclinical hypothyroidism group, and overt hypothyroidism group than in the strict-normal thyroid function group (all P for trend < 0.001), regardless of dyslipidemia status (Supplementary Table 2). Individuals with hypertriglyceridemia or with decreased HDL-C levels have a greater fibrotic burden than their counterparts across the spectrum of hypothyroidism.

We found that ALT levels were strongly associated with hypothyroidism status (Supplementary Fig. 1). We hence stratified our analysis by ALT levels. Both significant and advanced liver fibrosis increased across the spectrum of hypothyroidism (all P for trend < 0.001), regardless of ALT levels (Supplementary Table 3). Individuals with elevated ALT levels have a greater fibrotic burden than their counterparts across the spectrum of hypothyroidism.

Cholestasis might have effects on LSM values. To avoid this effect, we repeated our analysis after excluding patients with abnormal bilirubin (including total and direct bilirubin), results were essentially the same (Supplementary Table 4). To dispel the impact of cholestasis and/or liver congestion, or the potential impact of autoimmune liver disease on the diagnostic accuracy of LSM for fibrosis, we also did sensitive analyses after excluding patients with abnormal bilirubin as well as transaminase levels (including ALT, AST, and GGT). We found that even in subjects with normal bilirubin and transaminase levels, the prevalence of significant and advanced liver fibrosis increased progressively from low-normal thyroid function through subclinical hypothyroidism, to overt hypothyroidism (Supplementary Fig. 2–4).

We then performed a multivariate logistic regression analysis to determine whether thyroid function status was associated with the risk of liver fibrosis. As shown in Table 2, in multivariable analyses, which included demographic and cardiometabolic risk factors, low-normal thyroid function, subclinical hypothyroidism, and overt hypothyroidism in men all increased the risk of significant fibrosis (Table 2), with overt hypothyroidism incurred the highest risk. This was also the case for risk of advanced fibrosis. Women showed a similar pattern. The extent of contributions of the confounding factors were shown in Table 2.

When participants who taking thyroid medications were included in the analysis, results were essentially the same (data were not shown).