Evidence on the link between hypothyroidism and non-alcoholic fatty liver disease: an updated systematic review

Abstract

Background

Numerous studies have investigated the relationship between hypothyroidism and non-alcoholic fatty liver disease (NAFLD), while the underlying mechanisms are not well understood. This systematic review evaluated the mentioned relationship among non-diabetic adults.

Methods

This systematic review was written in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A comprehensive literature search was conducted using PubMed, Scopus, Web of Science, and Google Scholar until January 2024. After a critical analysis, 29 observational studies were included in the present systematic review.

Results

The available observational evidence suggests a potential association between hypothyroidism and NAFLD. Regarding this complex relationship, patients with hypothyroidism may be more likely to develop NAFLD compared to those with normal thyroid function. Several factors are involved in the development of hypothyroidism-induced NAFLD.

Conclusions

Thyroid hormones regulate energy and metabolism, suggesting their relevance in NAFLD. Increased awareness and optimized strategies are needed for mutual screening and managing thyroid disease and NAFLD coexistence.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12902-025-01977-2.

Introduction

Non-alcoholic fatty liver disease (NAFLD), the leading cause of chronic liver disorders, is the hepatic demonstration of metabolic syndrome (MetS). NAFLD is a condition manifested by the accumulation of lipids in hepatocytes without alcohol consumption. The disease encompasses a range of pathologies ranging from simple hepatic steatosis to non-alcoholic steatohepatitis (NASH), which can extend to liver fibrosis and cirrhosis [1–3]. The global prevalence of NAFLD is rapidly increasing, affecting more than 32% of the population worldwide [4]. Based on the multiple-hit theory, the pathogenesis of NAFLD provides clarification by a sequence or simultaneous combination of genetic factors and metabolic disorders, including insulin resistance (IR), changes in the gut microbiome, activation of the immuno-inflammatory systems, and oxidative stress. The mentioned pathologies lead to the activation of hepatic stellate cells and Kupffer cells, which contribute to the development of fibrosis [5]. In recent years, many specialists have advocated changing the terminology, switching from NAFLD to metabolic-associated fatty liver disease (MAFLD). Consequently, MAFLD could be detected by the existence of hepatic steatosis and at least one of the approved metabolic criteria: (a) overweight/obesity, (b) type 2 diabetes (T2D), and (c) metabolic disorders [6]. Accordingly, NAFLD contributes to both liver-related and extra-hepatic complications, resulting in a sharp trend in morbidity and mortality [7, 8]. Endocrinopathies such as T2D, MetS, cardiovascular diseases, and chronic kidney diseases, as well as thyroid dysfunction, are the main extra-hepatic disorders of NAFLD [9, 10].

Hypothyroidism is characterized by the reduction or absence of thyroid hormones. It can be manifested at birth as “congenital” or develop later “acquired.” The most common cause of hypothyroidism is primary hypothyroidism due to the dysfunction of the thyroid gland. On the other hand, secondary or central hypothyroidism is caused by problems in the pituitary gland or hypothalamus [11]. Hypothyroidism includes overt hypothyroidism with evident clinical symptoms, including elevated thyroid-stimulating hormone (TSH) levels (> 10 mIU/l) and reduced free T4 (FT4) levels. In comparison, subclinical hypothyroidism is indicated by a moderate rise in TSH levels (TSH > 4 mIU/l), along with normal thyroid hormone levels. Approximately 3% of the general population is affected by overt hypothyroidism. However, subclinical hypothyroidism is found in 5–10% of the global population, as well as 8–10% of individuals aged over 65 years [12]. It should be noted that thyroid hormones have a crucial impact on multiple organs, especially the liver. Impairments in the function of thyroid hormones could lead to hypercholesterolemia, which is a key factor in the pathophysiology of hypothyroidism-induced NAFLD [13]. To date, numerous studies have investigated the relationship between thyroid disorders (especially hypothyroidism) and the onset and progression of NAFLD; however, the exact association and related mechanisms seem to remain unclear [14–17]. Considering the global burden of NAFLD and associated complications and the sharp trend in the prevalence of thyroid dysfunction, investigating this epiphenomenon seems necessary [4, 18].

Although previously published reviews have collected the results of original studies on the association between hypothyroidism and NAFLD, they missed numerous observational studies [19–22]. Furthermore, several observational studies have since been published [23–34]. Therefore, the current systematic review aimed to assess the association between NAFLD and hypothyroidism in non-diabetic adults. To our knowledge, the present study is the first comprehensive systematic review of observational studies in this area conducted on non-diabetic adults.

Methods

Search strategy

This systematic review was written in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [35]. We searched PubMed, Scopus, Web of Science, and Google Scholar until January 2024, followed by a hand search of the reference lists of related articles and reviews. We restricted our search to articles published in English. The full search strategy for each database is provided in Supplementary Table 1.

Eligibility criteria

Two authors (NP and SA) reviewed the titles, abstracts, and full text of all articles obtained. Any disagreement between the two authors was resolved by discussion. Observational studies (cross-sectional, case-control, and cohort studies) that investigated the association between hypothyroidism and NAFLD in non-diabetic adults (aged ≥ 18 years) were eligible to be included in this systematic review. We excluded animal studies and in vitro models and studies on children or patients suffering from chronic liver diseases such as viral hepatitis, autoimmune hepatitis, liver fibrosis and cirrhosis, NASH, alcohol-related liver disease (ALD), Hashimoto’s thyroiditis, cancer, cardiovascular diseases, gastrointestinal and biliary disorders, diabetes mellitus, and polycystic ovary syndrome. Furthermore, letters, comments, and review articles were not eligible for inclusion in the current study.

Screening, data extraction, and quality assessment

Two researchers (NP and SA) independently extracted the required data using prespecified data extraction forms. Consensus was achieved through discussing and resolving disagreements. The information extracted from eligible studies was the first author’s name, date of publication, study design, study population and characteristics, method for diagnosis of NAFLD, definition of thyroid dysfunction, and outcomes. The quality of studies included in the present systematic review was assessed by two independent reviewers (NP and SA) using a 9-point scoring system for case-control and cohort studies and a 10-point scoring system for cross-sectional studies according to the Newcastle-Ottawa Scale (NOS). However, all observational studies meeting the inclusion criteria were included in the review, regardless of their methodological quality [36]. Studies with ≥ 7 points were considered high quality. Any discrepancy between reviewers was resolved by the third reviewer (HT).

Results

We initially identified 4008 eligible studies through database searching. Excluding duplicates, 2949 titles and abstracts remained for further screening. Of these, 2904 articles were excluded because they did not fulfill the inclusion criteria. Eventually, 45 studies were obtained according to the research topic. After a critical analysis, 29 observational studies were included in our systematic review (Fig. 1). The included studies were comprised of 19 cross-sectional studies [19, 20, 22, 24, 29–34, 37–45], five case-control studies [21, 23, 26, 27, 46], and five cohorts [25, 28, 47–49]. Details of the included studies are presented in Table 1 (cross-sectional studies) and Table 2 (case-control and cohort studies).

Fig. 1.

Flow diagram of the literature search and study selection process

Table 1.

Main characteristics and findings from cross-sectional studies

Study	Year	Study design	Study population	Method for diagnosis of NAFLD	Definition of thyroid dysfunction	Main findings
Xu et al. [37]	2011	Cross-sectional	878 elderly subjects with euthyroidism (Male:555, Female:323)	Ultrasonography	TSH > 4.5 mIU/l or < 0.5 mIU/l FT4 > 14.41 pmol/l or 7.85 pmol/l	Significantly associated with the change in thyroid function and NAFLD. [OR = 0.847, 95%CI: 0.743–0.966]
Chung et al. [38]	2012	Cross-sectional	2324 subjects with hypothyroidism	Ultrasonography	Subclinical hypothyroidism: TSH ≥ 4.1 mIU/l and normal FT4 Overt hypothyroidism: FT4 < 0.7 ng/dL	Significant associations between subclinical hypothyroidism, hypothyroidism, and NAFLD. [OR = 1.38, 95%CI: 1.17–1.62]
Zhang et al. [39]	2012	Cross-sectional	1322 subjects (Male:545, Female:777)	Ultrasonography	TSH > 2.5 mIU/mL TSH ≤ 2.5 mIU/mL	Lack of effect of TSH level change on the prevalence of NAFLD
Eshraghian et al.[40]	2014	Cross-sectional	832 participants (127 individuals with NAFLD)	Ultrasonography	Subclinical hypothyroidism: TSH > 5.2 mIU/l Overt hypothyroidism: TSH > 5.2 mIU/l FT4 < 11.5 Euthyroidism: TSH (0.2–5.2) MIU/L FT4 (11.5–23) pmol/L	No association between thyroid dysfunction and NAFLD
Tao et al. [42]	2014	Cross-sectional	739 subjects with euthyroidism	Ultrasonography	Euthyroidism: TSH: (3.1–6.8) pmol/L FT4: (0.3–4.5) mIU/L FT3: (12–22) mIU/L	Association of serum FT4 and TSH levels with NAFLD in the general population [for FT4 levels; OR = 2.21, 95%CI: 1.21–4.02 and for TSH levels; OR = 0.39, 95%CI: 0.17–0.87]
Liu et al. [41]	2015	Cross-sectional	2576 subjects with euthyroidism	Ultrasonography	Euthyroidism: TSH: (0.55–4.78) mIU/L FT4: (11.5–22.7) pmol/L FT3: (3.5–6.5) pmol/L	Negative correlations between serum FT4 level and NAFLD in pre-menopausal women. [OR = 0.77, 95%CI: 0.61–0.97]
Gokmen et al. [43]	2016	Cross-sectional	115 subjects with euthyroidism/ hypothyroidism (Male:75, Female:40)	Ultrasonography	Euthyroidism: TSH (0.5-4) mIU/L Hypothyroidism: TSH ≥ 4.1 mIU/L	Absence of direct association of hypothyroidism with NAFLD
Van den Berg et al. [44]	2016	Cross-sectional	20,289 subjects with euthyroidism	FLI ≥ 60	Euthyroidism: TSH: (0.5-4) mIU/L FT4: (11 -19.5) pmol/L FT3: (4.4–6.7) pmol/L	Higher FT3 and FT3/ FT4 ratio and lower FT4 in euthyroid subjects with suspected NAFLD [OR = 1.44, 95%CI: 1.39–1.49]
Janovsky et al. [19]	2018	Cross-sectional	10,539 participants	Ultrasonography	Quartiles of TSH levels within the reference range: (0.46–1.44, 1.45–1.97, 1.98–2.68, 2.69–4.68)	Overall associations between TSH quartiles and NAFLD; lack of significant associations after adjusting for metabolic syndrome factors [P < 0.001]
Lee et al. [20]	2018	Cross-sectional	3452 subjects (Euthyroidism: 3324, subclinical hypothyroidism: 128)	HSI ≥ 36	Euthyroidism: TSH: (0.62–6.68) mIU/L FT4: (0.82–1.76) ng/mL Subclinical hypothyroidism: TSH > 6.68 FT4: (0.82–1.76) ng/mL	Association between subclinical hypothyroidism and increased risk of NAFLD, only in males [OR = 2.37, 95%CI: 1.09–5.12]
Liu et al. [45]	2018	Cross-sectional	1773 subjects with euthyroidism	Ultrasonography and FLI ≥ 60	Euthyroidism: TSH: (0.35–4.94) mIU/L FT4: (9.01–19.05) pmol/L FT3: (2.63–5.7) pmol/L	Positive association between FT3 and TSH levels and the risk of NAFLD in euthyroid subjects; For TSH levels and ultrasound-proven NAFLD [OR = 1.10, 95%CI: 1.05–1.39, P = 0.024] For FT3 levels and ultrasound-proven NAFLD [OR = 1.25, 95%CI: 1.12–1.40]
Grewal et al. [22]	2020	Prospective cross-sectional	100 patients with hypothyroidism + 100 euthyroid individuals	Ultrasonography and liver enzymes	-	Association between the presence of hypothyroidism and the risk of NAFLD [OR = 2.07, 95%CI: 1.02–4.19]
Lai et al. [24]	2021	Retrospective cross-sectional study	4610 participants	Ultrasonography	FT3: 2.63–5.7 pmol/L FT4: 9.01–19.05 pmol/L TSH: 0.35–4.94 mU/L	Independent associations between TFQIFT3 and FT3/FT4 with the risk of dyslipidemia and NAFLD[P < 0.05]
Shao et al. [29]	2022	Retrospective cross-sectional study	81 patients with NAFLD and normal thyroid function + 34 healthy individuals	Abdominal CT	TSH, FT4, FT3, TT3 and TT4	Serum FT4 level as an independent risk factor of NAFLD in euthyroid subjects [OR = 0.73, 95%CI: 0.54–1.001]
Xiao et al. [30]	2022	Retrospective cross-sectional study	178 patients with hypopituitarism	Ultrasonography	Low serum FT4 levels with a low, normal, or mildly elevated TSH.	Involvement of thyroid dysfunction in NAFLD by regulating whole-body metabolism and lipid utilization
Vázquez et al. [31]	2022	Retrospective cross-sectional study	3697 patients with NAFLD	Transient elastography	LNTF was defined as: TSH (2.5–4.5 mIU/L), divided into three different cut-off points (> 4.5 to 5.0, > 3.1, and > 2.5 mIU/L).	Lack of any associations between LNTF and NAFLD or MAFLD
Augustine et al. [32]	2023	Cross-sectional	100 patients with newly diagnosed overt hypothyroidism	Ultrasonography and LFT	Thyroid profile (total T3, total T4, and TSH)	A significant association between TSH levels and NAFLD findings[P < 0.001]
Elshinshawy et al. [33]	2023	Cross-sectional	60 participants with newly diagnosed hypothyroidism + 30 healthy participants with TSH level < 4.5mIU/L	Transient elastography	Overt hypothyroidism: TSH > 4.5 mIU/L and FT4 < 0.9 ng/dl Subclinical hypothyroidism: TSH > 4.5 mIU/L and normal FT4 (0.9–1.7 ng/dl)	Significantly higher risk of NAFLD in subclinical or overt hypothyroidism individuals compared with those with normal thyroid function [P < 0.001] (Free T4 decreases the risk as OR < 1)
Patel et al. [34]	2023	Cross-sectional Comparative Study	142 subjects	Ultrasonography	Biochemical parameters for thyroid profile	Significantly increased severity of NAFLD in patients with overt hypothyroidism[P < 0.001]

NAFLD, Non-alcoholic fatty liver disease; TSH, Thyroid stimulating hormone; OR, Odds ratio; CI, Confidence interval; FT4, Free thyroxine; FT3, Free triiodothyronine; FLI, Fatty liver index; HIS, Hepatic steatosis index; LFT, Liver function tests; TFQI, Thyroid feedback quantile-based index; LNTF, Low-normal thyroid function; CT, Computed tomography; TT3, Total triiodothyronine; TT4, Total thyroxine

Table 2.

Main characteristics and findings from case-control studies and cohort studies

Study	Year	Study design	Study population	Method for diagnosis of NAFLD	Definition of thyroid dysfunction	Main findings
Pagadala et al. [46]	2012	Case-control	233 Cases with NAFLD and 430 controls	Liver biopsy	Clinical diagnosis of hypothyroidism and undergoing thyroid replacement therapy	Significantly higher prevalence of hypothyroidism in patients with NAFLD, compared with controls [P < 0.01] (Higher BMI (OR, 1.04; P = 0.03) was associated with the presence of hypothyroidism in NAFLD)
Lee et al. [47]	2015	Retrospective cohort study	18,544 subjects: Euthyroid control (n = 17,052) Subclinical hypothyroidism (n = 1,303) Overt hypothyroidism (n = 189)	Ultrasonography	Subclinical hypothyroidism: TSH > 4.2 mIU/l FT4 (0.97–1.68 ng/Dl) Overt hypothyroidism: TSH > 4.2 mIU/l FT4 < 0.97 ng/dL	Lack of association between subclinical and overt hypothyroidism and an increased incidence of NAFLD
Bano et al. [48]	2016	Prospective cohort study	9419 participants	Ultrasonography	Subjects with normal TSH and FT4 levels	Increased risk of NAFLD in individuals with hypothyroidism, compared with euthyroid subjects [OR = 1.24, 95%CI: 1.01–1.53]
Kim et al. [49]	2020	Cohort study	10,144 adults	Ultrasonography	“Low-normal” thyroid function (TSH > 2.5–4.5) and “Strict-normal” thyroid function (TSH= (0.39–2.5) mIU/l and normal T4 level (4.5–13.2) ug/dL). Subclinical hypothyroidism (TSH > 4.5 mIU/l and normal T4 level)	Association between low thyroid function and NAFLD as well as being a predictor for all-cause and cardiovascular mortality [HR = 1.24, 95%CI: 1.02–1.50]
Popescu et al. [21]	2020	Case-control	124 patients with subclinical or manifested hypothyroidism	Ultrasonography	Subclinical hypothyroidism: TSH > 4.2 mIU/l and normal FT4	Significant associations between hypothyroidism (subclinical or overt) and NAFLD [P < 0.001]
Labenz et al. [23]	2021	Case-control	57,483 patients with NAFLD + 57,383 patients without liver disease	Unclear	Unclear	Associations between hypothyroidism [OR = 1.17, 95%CI: 1.10–1.24, P < 0.001] and autoimmune thyroiditis [OR = 1.53, 95%CI: 1.35–1.73] with a moderately higher risk of NAFLD
Loosen et al. [25]	2021	Retrospective cohort study	40,582 patients without NAFLD + 40,582 patients with NAFLD	Unclear	Unclear	Association between NAFLD and an increased incidence of hypothyroidism [HR = 1.53,95%CI: 1.43–1.64, P < 0.001] and autoimmune thyroiditis [HR = 1.55, 95%CI: 1.34–1.79]
Shatla et al. [26]	2021	Case-control	325 subjects with a recently diagnosed overt hypothyroidism + 325 matched subjects	Ultrasonography	Overt hypothyroidism: (TSH > 4.2 mIU/mL and FT4 < 0.94 ng/dL) Eu-thyroid state: normal TSH, FT4, and FT3 levels	Significant associations between overt hypothyroidism and NAFLD [P < 0.001]
Fan et al. [27]	2022	Case-control	14,797 adults	Ultrasonography	Genetically predicted NAFLD (based on Genome-wide association)	Significant associations between higher TSH levels and an increased risk of NAFLD [OR = 1.29, 95%CI: 1.10–1.52]
Gu et al. [28]	2022	Cohort study	6462 subjects without baseline NAFLD	Ultrasonography	3.5 < FT3 < 6.5 pmol/L 11.5 < FT4 < 22.7 pmol/L 0.55 < TSH < 4.78 mIU/L	Associations between high-normal FT3 [HR = 1.30, 95%CI: 1.12–1.51] and low-normal TSH [HR = 0.82, 95%CI: 0.71–0.95] with higher incidence of NAFLD in middle-aged and older euthyroid subjects

NAFLD, Non-alcoholic fatty liver disease; TSH, Thyroid stimulating hormone; OR, Odds ratio; CI, Confidence interval; FT4, Free thyroxine; FT3, Free triiodothyronine; T4, Thyroxine; ICD, International classification of disease; HR, Hazard ratio

The quality of studies assessed via the NOS is provided in Table 3 (cross-sectional studies), Table 4 (case-control studies), and Table 5 (cohort studies). According to the NOS, 20 studies had high quality (NOS ≥ 7) [19–21, 23, 24, 26–28, 30, 31, 37–42, 44, 45, 47, 48], six studies obtained six stars [29, 32–34, 46, 49], and three studies achieved a score of 5 [22, 25, 43].

Table 3.

Quality assessment of cross-sectional studies

Cross-sectional studies	Representative	Sample size	Nonrespondent	Ascertainment	Comparability	Outcome assessment	Statistical test	Sum up	Quality
Xu et al., 2011 [37]	*	0	0	**	*	**	*	7	Good
Chung et al., 2012 [38]	*	*	*	**	**	**	*	10	Very good
Zhang et al., 2012 [39]	*	*	*	**	*	**	0	8	Good
Eshraghian et al., 2014 [40]	*	*	*	**	0	**	*	8	Good
Tao et al., 2014 [42]	*	0	*	**	*	**	*	8	Good
Liu et al., 2015 [41]	*	*	*	**	*	**	*	9	Very good
Gokmen et al., 2016 [43]	*	0	0	*	0	**	*	5	Satisfactory
Van den Berg et al., 2016 [44]	*	*	*	**	*	**	*	9	Very good
Janovsky et al., 2018 [19]	*	*	0	**	*	**	*	8	Good
Lee et al., 2018 [20]	*	*	*	**	*	**	*	9	Very good
Liu et al., 2018 [45]	*	*	0	**	*	**	*	8	Good
Grewal et al., 2020 [22]	*	0	0	0	*	**	*	5	Satisfactory
Lai et al., 2021 [24]	*	*	*	**	**	**	*	10	Very good
Shao et al., 2022 [29]	*	0	0	**	0	**	*	6	Satisfactory
Xiao et al., 2022 [30]	*	0	0	*	**	**	*	7	Good
Vazquez et al., 2022 [31]	*	*	0	**	**	**	*	9	Very good
Augustine et al., 2023 [32]	*	0	0	**	0	**	0	6	Satisfactory
Elshinshawy et al., 2023 [33]	*	0	0	**	*	**	0	6	Satisfactory
Patel et al., 2023 [34]	*	0	0	**	0	**	*	6	Satisfactory

Table 4.

Quality assessment of case-control studies

Case-control studies	Case definition	Representative	Control selection	Control definition	Comparability	Ascertainment	Same method	Nonresponse	Sum up	Quality
Pagadala et al., 2012 [46]	*	*	*	*	**	0	0	0	6	Medium
Popescu et al., 2020 [21]	*	*	0	*	*	*	*	*	7	High
Labenz et al., 2021 [23]	*	*	*	*	**	0	*	0	7	High
Shatla et al., 2021 [26]	*	*	*	*	**	*	*	0	8	High
Fan et al., 2022 [27]	*	*	*	*	*	*	*	0	7	High

Table 5.

Quality assessment of cohort studies

Cohort studies	Representative	Selection	Ascertainment	Start of study	Comparability	Assessment	Length of follow-up	Adequacy of follow-up	Sum up	Quality
Lee et al., 2015 [47]	*	*	*	*	*	*	*	0	7	High
Bano et al., 2016 [48]	*	*	*	*	**	0	*	*	8	High
Kim et al., 2020 [49]	*	0	*	*	**	0	*	0	6	Medium
Loosen et al., 2021 [25]	*	0	*	*	0	*	*	0	5	Medium
Gu et al., 2022 [28]	*	*	*	*	*	*	*	*	8	High

Characteristics of included studies

The main characteristics of the twenty-nine qualified publications are illustrated in Tables 1 and 2. The studies were conducted between 2011 and 2023, with sample sizes ranging from 90 to 114,866 adult participants. Among the included studies, nine were conducted in China [24, 28–30, 37, 39, 41, 42, 45], three in South Korea [20, 38, 47], three in the United States of America [27, 46, 49], three in India [22, 32, 34], two in the Netherlands [44, 48], two in Germany [23, 25], one in Iran [14], one in Turkey [43], one in Romania [21], one in Mexico [31], one in Saudi Arabia [26], one in Egypt [33], and one in Brazil [19].

Evidence from cross-sectional studies

As mentioned, NAFLD—the barometer of metabolic health—is closely associated with metabolic abnormalities. In this context, some cross-sectional studies have suggested a relationship between hypothyroidism and NAFLD [19, 20, 22, 24, 29, 30, 32–34, 37, 38, 41, 42, 44, 45]. However, some studies have contradicted this relationship [31, 39, 40, 43]. Patel et al. [34] reported that NAFLD was more severe in patients with overt hypothyroidism and more prevalent in untreated hypothyroid subjects. They also documented a significant association between thyroid status and NAFLD (p < 0.001). Likewise, a recently published study indicated that individuals with subclinical hypothyroidism and overt hypothyroidism had a higher risk of developing NAFLD compared with individuals with normal thyroid function (p < 0.001) [32]. These studies shed light on the bi-directional association between thyroid status and NAFLD, which has been more precisely investigated by Xiao et al. [30], who assessed 134 patients with hypopituitarism, divided into NAFLD and non-NAFLD groups based on ultrasonography findings. After an adjustment for age and gender, they documented that thyroid disorders might be involved in NAFLD by controlling whole-body metabolism and mainly lipid utilization. Chung et al. [38] conducted a study on 4648 subjects (2324 cases with hypothyroidism and age and sex-matched controls) by categorizing hypothyroidism into subclinical hypothyroidism or overt hypothyroidism. The study found a significant association between NAFLD and hypothyroidism (30.2% in patients and 19.5% in controls, respectively, with a p-value of 0.001). Moreover, the prevalence of NAFLD and abnormal liver enzymes escalated steadily with the increase in the severity of hypothyroidism (for NAFLD, subclinical: 29.9% and overt: 36.3%; for abnormal ALT, 20.1% and 25,9%, p < 0.001 respectively).

Far from general thyroid dysfunction, changes in thyroid hormones (specifically FT4 levels) are also involved. Hence, changes in thyroid function (even within the reference range) are associated with NAFLD. Xu et al. [37] observed significantly lower levels of serum FT4 in patients with NAFLD by performing a cross-sectional study on 878 euthyroid elderly Chinese subjects. Statistical analysis revealed that serum FT4 levels were markedly related to the risk of NAFLD (Odds ratio (OR) = 0.847, p = 0.013). Another cross-sectional study conducted in 2015 illustrated a negative correlation between serum FT4 levels and NAFLD in premenopausal women [41]. Collectively, these findings provide a possible association between FT4 and TSH levels and the presence of NAFLD in the general population [32, 42]. A recent retrospective cross-sectional study by Shao et al. [29] manifested the independent relation between FT4 and NAFLD in normal thyroid function (OR = 0.73; 95% Confidence interval (CI): 0.54–1.001), while lower levels of FT4 (within the normal range) were associated with an increased risk of NAFLD in euthyroid individuals. They found that thyroid function was not only associated with aminotransferases and ultrasound findings but also was related to liver-related indices such as the fatty liver index (FLI). Van den Berg et al. [44] gathered euthyroid subjects from the lifeline cohort study and assessed NAFLD by using validated FLI (FLI ≥ 60 was defined as NAFLD). Findings showed that euthyroid participants with suspected NAFLD were characterized by higher free T3 (FT3) levels, lower FT4 levels, and higher FT3/FT4 ratio (p < 0.0001), probably resulting from central obesity. However, the studies conducted by Zhang et al. [39], Eshraghian et al. [40], and Gokmen et al. [43] showed different results. Firstly, Zhang et al. [39] illustrated significant correlations between serum TSH levels, body mass index (BMI), and body fat accumulation in females, while changes in TSH levels did not affect the prevalence of NAFLD. In this line, it was reported that thyroid disorders were not correlated with NAFLD, and the changes in thyroid hormones were attributed to sick euthyroid syndrome [40]. Another recently published study found no association by defining a condition known as low-normal thyroid function (LNTF), which was not associated with NAFLD or even MAFLD. LNTF was characterized by TSH levels between 2.5 mIU/L and 4.5 mIU/L with normal FT4 [31].

Evidence from case-control studies

In the milieu of current knowledge, all reviewed case-control studies demonstrated an association between hypothyroidism and NAFLD [21, 23, 26, 27, 50]. Fan et al. [27] included 14,797 adult participants and observed a dose-response relationship between TSH levels and NAFLD (confirmed by ultrasonography) (OR = 1.29; 95% CI:1.10–1.52). As a result, circulating serum levels of TSH were associated with NAFLD. Another case-control study was conducted on 57,483 patients with NAFLD and 57,483 controls, illustrating that hypothyroidism (OR = 1.17; 95% CI: 1.10–1.24) and autoimmune thyroiditis (OR = 1.53; 95% CI: 1.35–1.73) were associated with a higher risk of NAFLD. Moreover, hypothyroidism significantly affects the prevalence of NAFLD in both men (OR = 1.31; 95% CI: 1.15–1.48) and women (OR = 1.12; 95% CI: 1.05–1.21) [23]. These findings were also approved by Shatla et al. [26], who manifested a positive association between overt hypothyroidism and the prevalence of NAFLD (p < 0.001). Indeed, it is mentioned that this link may serve as a risk factor for NAFLD [21]. The diagnosis of NAFLD has strengthened evidence from case-control studies through biopsy as the gold standard assessment model. In this regard, Pagadala et al. [50] surveyed 246 biopsy-proven NAFLD subjects and 430 age, sex, and body mass-matched controls. In line with previous reports, they documented a higher prevalence of hypothyroidism in biopsy-proven patients with NAFLD in comparison with the controls (21% in cases vs. 9.5% in controls, respectively, p < 0.01). Taken together, it should be noted that case-control studies suggest the inter-relationship between thyroid function and NAFLD.

Evidence from cohort studies

Adding strength to current knowledge, most of the cohort studies showed an association between hypothyroidism and NAFLD [25, 28, 48, 49]. Among the five reviewed original studies, only one study showed contradictory results [47]. Bano et al. [48] conducted a prospective cohort study, and 9419 participants were included. After 10 years of follow-up, results indicated that hypothyroidism was associated with a 1.24-fold higher risk of NAFLD (CI, 1.01–1.53) compared to euthyroiditis. Other studies have also indicated similar results after following 50,727 participants [25, 49]. In addition, findings by Gu et al. [28] illustrated that high-normal FT3 and low-normal TSH levels were independently associated with a higher incidence of NAFLD (p < 0.001) in middle-aged and older euthyroid adults. These results were addressed following the enrollment of 6462 individuals without baseline NAFLD in the cohort study. Meanwhile, a study in 2015 presented different results on 18,544 subjects without baseline NAFLD. These findings proposed that the incidence of NAFLD did not differ significantly based on the baseline thyroid hormonal status, even after multivariable adjustment (subclinical hypothyroidism group: hazard ratio (HR) = 0.965, 95% CI = 0.814–1.143; overt hypothyroidism group: HR = 1.25, 95% CI = 0.830–1.899). Based on their investigation, subclinical and overt hypothyroidism was not associated with an increased risk of NAFLD incidence [47].

Discussion

Although not entirely understood, NAFLD is believed to arise from a combination of genetic, epigenetic, and metabolic etiologies [51]. Based on our extensive search, we elucidated a potent association between hypothyroidism and NAFLD by reviewing the included studies. In this context, most of the mentioned studies with different methodologies highlighted a positive inter-relationship between the disease by evaluating different subtypes of thyroid dysfunction and NAFLD assessment through invasive and non-invasive methods. Several factors have been implicated in the development of hypothyroidism-induced NAFLD [52]. Impaired lipid metabolism, IR, oxidative stress, and the involvement of inflammatory cytokines and hormones are predominant underlying mechanisms [53, 54]. Indeed, an imbalance in energy metabolism, particularly the excess delivery of carbohydrates and fat into the liver, is a key parameter in the development of NAFLD [55, 56].

Hypothalamic-pituitary axis

T4 (3,3′,5,5′-tetraiodo-l-thyronine) and T3 (3,5,3′-triiodo-l-thyronine) are thyroid hormones [57]. The thyroid gland predominantly produces T4, but most of the systemic T3 is created through deiodination of T4 in peripheral tissues, making T3 the most potent thyroid hormone [58]. The secretion of thyroid hormones from the thyroid gland is regulated by the hypothalamic-pituitary axis [59]. Thyrotropin-releasing hormone (TRH) is a hormone released by the hypothalamus. It acts on the pituitary gland by binding to G-coupled TRH receptors on the thyrotrope. This binding causes an increase in intercellular cyclic adenosine monophosphate and the production of TSH. TSH stimulates the production and release of thyroid hormones by binding to a TSH receptor on the thyroid follicular cell, which is a G protein-coupled receptor [60]. The thyroid gland controls numerous biological functions in the liver, adipose tissue, central nervous system, and cardiovascular system by producing and releasing two main thyroid hormones—T3 and T4— into the bloodstream [61–63].

Obesity and insulin resistance

Adipocyte expansion, which presents as overweight and obesity, plays a central role in the development of cardiometabolic disorders, particularly NAFLD. Obesity is widely recognized as a major contributor to hepatic steatosis, and its prevalence is directly associated with higher BMI [64–66]. Epidemiological studies have confirmed the increased prevalence of hypothyroidism among overweight individuals compared to the general population [67–69]. Hypothyroidism leads to a decreased metabolic rate, reduced resting energy expenditure, and impaired gluconeogenesis and lipolysis, which may result in weight gain. These changes are closely associated with IR, another key mechanism in NAFLD pathogenesis [70]. Hypothyroidism is associated with reduced insulin secretion and an elevated homeostatic model assessment for IR (HOMA-IR) index, which indicates systemic IR. This state increases lipolysis in adipose tissue, raising circulating free fatty acids (FFAs). Excess FFAs are delivered to the liver, promoting lipid accumulation and triggering hepatic IR, which aggravates NAFLD [71, 72]. The metabolic interplay among hypothyroidism, obesity, and IR creates a pathologic cycle that perpetuates hepatic lipid accumulation and steatosis [60]. These features overlap with metabolic syndrome (MetS) components and align with the newer concept of metabolic-associated fatty liver disease (MAFLD) [73]. Moreover, overt and subclinical hypothyroidism are the independent risk factors for NAFLD. At the hepatic level, IR impairs the suppression of endogenous glucose production and promotes de novo lipogenesis [74]. Accumulation of lipid mediators (lipoxins) within hepatocytes further disrupts insulin signaling, contributing to hepatic dysfunction [75]. This lipid overload initiates lipotoxicity, hepatocellular stress, and inflammation, forming a key pathophysiological link between IR and the development of NAFLD [76].

Lipid homeostasis

Thyroid hormones play a significant role in hepatic lipid metabolism by acting through thyroid hormone receptors (THRs), which function as nuclear transcription factors upon ligand binding [69]. These receptors regulate lipid synthesis, mobilization, degradation, and aspects of glucose metabolism via modulation of gene expression and interaction with other nuclear receptors [77, 78]. Among THR isoforms, THRβ —particularly THR-β1 expressed in the liver—mediates most of the metabolic actions of T3, while THRα predominates in tissues such as the heart and brown adipose tissue. THR-β2 is mainly expressed in the hypothalamus and pituitary, contributing to feedback regulation of the hypothalamic-pituitary-thyroid axis [77]. In the liver, activation of the THRβ pathway influences key metabolic processes, including de novo lipogenesis, fatty acid β-oxidation, mitophagy, and cholesterol synthesis, thereby affecting circulating levels of low-density lipoprotein (LDL) cholesterol, apolipoprotein B, and lipoprotein(a) [12]. THRs regulate gene transcription by binding to thyroid hormone response elements, directly inducing lipogenic genes, and modulating transcription factors such as sterol regulatory element-binding protein-1 C (SREBP1C) and carbohydrate-responsive element binding protein (ChREBP), central to hepatic lipid synthesis [79, 80]. Additionally, T3 indirectly influences lipid metabolism by activating peroxisome proliferator-activated receptor gamma (PPAR-γ) and its coactivator PPAR-γ coactivator 1-alpha (PGC-1α), along with targets such as acyl-CoA oxidase and fibroblast growth factor-21 (FGF-21) [81]. T3 also enhances LDL receptor gene expression, increasing LDL receptor density and lipoprotein lipase activity. Consequently, hypothyroidism, characterized by low T3 levels, is often associated with dyslipidemia, particularly elevated LDL cholesterol. Both T3 and T4 further regulate the synthesis of apolipoprotein B, promoting very low density lipoprotein (VLDL) secretion [82, 83]. Figure 2 represents the effect of hypothyroidism on lipid homeostasis as the mutual link with NAFLD. Evidence has suggested the role of increased TSH levels in reducing hydroxymethylglutaryl-CoA reductase phosphorylation, thereby increasing cholesterol biosynthesis and resulting in hypercholesterolemia [84]. Moreover, elevated expression of regulatory SREBP-2 regulates cholesterol metabolism [73]. Hypothyroidism and increased TSH levels are also associated with reduced hepatic lipoprotein lipase activity, which eliminates hepatic triglyceride clearance and progressions in liver steatosis [84].

Fig. 2.

Possible pathogenic effects of hypothyroidism on lipid homeostasis as a link between hypothyroidism and non-alcoholic fatty liver disease. SREBP-1c, Sterol regulatory element-binding protein-c; SREBP-2, Sterol regulatory element-binding protein-2; ChREBP, Carbohydrate-responsive element-binding protein; PPAR, Peroxisome proliferator-activated receptor; LDLR, Low-density lipoprotein receptor; AMPK, AMP-activated protein; HMG-CoA reductase, 3-hydroxy 3-methylglutaryl-coenzyme A; LDL-c, Low-density lipoprotein cholesterol; LPL, Lipoprotein lipase; NAFLD, Non-alcoholic fatty liver disease; TG, Triglycerides; TSH, Thyroid stimulating hormone; T4, Thyroxine; T3, Triiodothyronine

Oxidative stress and chronic inflammation

Oxidative-inflammatory pathways have been introduced as novel potential game players in the link between NAFLD and hypothyroidism. Hypothyroidism has been associated with elevated oxidative stress markers, besides its subsequent effects on lipid and glucose metabolism [84, 85]. Mitochondria, the cell’s powerhouse, is crucial for fatty acid oxidation, oxidative phosphorylation, electron transfer, and ATP generation. An increase in FFA and mitochondrial dysfunction results in free radical generation and lipid peroxidation, damaging hepatocytes and activating pro-inflammatory cytokines. Cytokines such as tumor necrosis factor-alpha (TNF-α) and transforming growth factor-beta can enhance the progression of NAFLD to fibrosis by activating hepatic stellate cells [86]. Pioneered studies have also proposed the involvement of leptin and adiponectin in NAFLD-hypothyroidism comorbidity by their roles as adipokines as well as cytokines [13].

Hormones

Thyroid hormones also modify hepatic lipid deposition via triggering leptin and adiponectin, cytokines that are also significant in the pathogenesis of hepatic steatosis [87, 88]. Leptin and adiponectin have been implicated in the multi-hit progression of NAFLD to fibrosis. Leptin stimulates beta-oxidation and suppresses lipogenesis, while many studies have found an inverse correlation between adiponectin, triglyceride, and LDL cholesterol [69]. Patients with hypothyroidism experience elevated leptin levels [84]. On the other hand, leptin plays a role in regulating appetite and, thus, in the metabolic consequences of NAFLD [89]. Leptin is also involved in the aggravation of hepatic IR by triggering the dephosphorylation of insulin receptor substrate-1 [90]. Conversely, adiponectin, a key insulin-sensitizing and anti-inflammatory adipokine, is typically reduced in hypothyroid states, contributing to increased hepatic fat accumulation and metabolic stress [91]. In addition, fibroblast growth factor-21 (FGF-21), another metabolic hormone elevated in hypothyroidism, has been implicated in NAFLD pathogenesis through its effects on lipid metabolism and hepatic stress responses [84]. The imbalance of these hormones reflects a broader endocrine disturbance that exacerbates NAFLD progression [92].

Liver homeostasis

Thyroid hormones, particularly T3, maintain hepatic homeostasis by regulating mitochondrial biogenesis, respiration, and mitophagy. T3 influences nuclear and mitochondrial gene expression by activating key transcription factors such as PGC-1α and nuclear respiratory factor 1 (NRF-1) transcription factors, which enhance oxidative phosphorylation and fatty acid oxidation [92, 93]. Furthermore, T3 supports mitophagy—the selective degradation of dysfunctional mitochondria—through activating proteins such as PTEN-induced kinase 1 (PINK1), death-associated protein kinase 2 (DAPK2), and sequestosome-1, helping preserve mitochondrial quality and energy efficiency [94–96]. T3 also induces hepatic lipophagy, where intracellular lipid droplets are targeted for lysosomal degradation, reducing hepatic triglyceride load [97, 98]. Figure 3 summarizes the pathological consequences of hypothyroidism linked to the progression of NAFLD. Finally, thyroid hormones indirectly affect hepatic homeostasis by regulating adipocyte gene expression, impacting lipid storage and mobilization, and adipose tissue thermogenesis [63].

Fig. 3.

Proposed pathological consequences of hypothyroidism, linked to the progression of non-alcoholic fatty liver disease. REE, Resting energy expenditure; BMI, Body mass index; FFA, Free fatty acid; HOMA-IR, Homeostatic model assessment for insulin resistance; TSH, Thyroid stimulating hormone; T4, Thyroxine; T3, Triiodothyronine; IRS-1, Insulin receptor substrate-1; FGF-21, Fibroblast growth factor-21; NAFLD, Non-alcoholic fatty liver disease; PGC-1α, Peroxisome proliferator-activated receptor coactivator 1-alpha; NRF-1, Nuclear respiratory factor-1; CPT-1, Carnitine palmitoyltransferase-1; PINK1, PTEN-induced kinase-1; ERRα, Estrogen-related receptor-α; DAPK-2, Death-associated protein kinase-2; MED-1, Mediator complex subunit-1

Limitations

The present systematic review has some strengths and limitations. A significant strength of the present study was that we tried to review the related observational studies with various methodologies to achieve a better point of view. Moreover, this study is an updated systematic review that includes all observational studies conducted from the beginning to now on the relationship between NAFLD and hypothyroidism. This study has some limitations. First, the methods of NAFLD diagnosis varied across the included studies, with most using ultrasonography and a minority employing liver biopsy, transient elastography, or indices such as FLI. This diagnostic heterogeneity may have introduced variability in sensitivity and specificity, potentially contributing to bias in assessing the true prevalence and association with hypothyroidism. Second, as the present systematic review only included observational studies, the results could not represent an association that reflects cause and effect. Included studies with lower methodological quality (NOS score < 7) may have introduced a risk of bias, affecting the strength of the evidence presented. Therefore, the findings should be interpreted with caution. Moreover, further research is necessary to assess the impact of genetic factors, molecular pathways, and gene variations on the link between NAFLD and hypothyroidism and related hormonal and metabolic response mechanisms.

Conclusion

The available observational evidence indicates a possible link between hypothyroidism and NAFLD. Regarding this intricate association, it seems that most of the patients with hypothyroidism are prone to have NAFLD in comparison to those with normal thyroid function. Increased awareness and optimized strategies are needed for mutual screening and managing thyroid disease and NAFLD coexistence. Given the inclusion of observational studies with varying methodological quality, the findings and overall conclusions of this review should be interpreted with caution.

Clinical relevance

Although causality cannot be established due to the observational nature of the included studies, the available evidence suggests an association between hypothyroidism and NAFLD. Therefore, regular evaluation of thyroid function in patients with NAFLD may contribute to early identification and management of underlying thyroid dysfunction, potentially improving related clinical outcomes.

From a clinical standpoint, it is important to assess serum TSH and free T4 levels in patients with NAFLD, particularly those with obesity, IR, or other metabolic abnormalities. Conversely, individuals diagnosed with overt or subclinical hypothyroidism may benefit from liver steatosis screening using noninvasive methods such as ultrasonography or hepatic indices. Early detection of these coexisting conditions enables timely lifestyle changes and therapeutic interventions, which can positively impact both liver and thyroid health.

Further research directions

Given the observational nature of the included studies, future research should prioritize well-designed prospective cohort studies and randomized controlled trials to better elucidate the causal relationship between hypothyroidism and NAFLD. Additionally, studies exploring molecular and genetic pathways linking thyroid dysfunction with hepatic steatosis are warranted. Research investigating the impact of thyroid hormone replacement therapy on NAFLD progression and outcomes would also provide valuable clinical insights.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

The authors’ responsibilities were as follows: N.P. and S.A. wrote the original paper; H.T. and M.E.M. contributed to the conception of the article and provided advice and consultation; M.E.M. and H.T. contributed to the final revision of the manuscript. All authors read and approved the final version of the manuscript.

Funding

This study was funded by the ‘Research Vice-Chancellor’ of Tabriz University of Medical Sciences, Tabriz, Iran (Ethics code: IR.TBZMED.REC.1402.617) (grant number: 72261).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

All procedures performed in this study were in accordance with the ethical standards of the Ethics Committee of Tabriz University of Medical Science (IR.TBZMED.REC.1402.617).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.