European Journal of Cardiovascular Medicine

The liver and thyroid are mutually dependent endocrine–metabolic hubs. The liver expresses thyroid hormone receptor-β (THR-β), deiodinases (D1/D2/D3), and transport proteins, enabling activation/inactivation, transport, and clearance of thyroid hormones.¹,³,¹⁹ Conversely, triiodothyronine (T3) modulates hepatic pathways governing de novo lipogenesis, β-oxidation, cholesterol turnover, autophagy of lipid droplets, and carbohydrate handling; perturbations contribute to steatosis and insulin resistance.¹,²,⁵,¹⁸ In chronic liver disease (CLD), systemic inflammation, reduced hepatic conversion (T4→T3), altered binding proteins, and cytokine-mediated hypothalamic–pituitary axis effects predispose to non-thyroidal illness syndrome (low-T3 with normal/low TSH and free T4).⁶–⁸

Epidemiological and clinical data suggest thyroid dysfunction is common in CLD and correlates with severity. Cross-sectional cohorts in cirrhosis show stepwise declines in fT3 across Child–Pugh or MELD strata and associations with portal hypertension, renal dysfunction, and mortality.⁶–⁹,¹⁴,¹⁶ Pediatric data mirror these trends, indicating pathophysiology not confined to adults.²⁰ The rising global burden of metabolic dysfunction–associated steatotic liver disease (MASLD; formerly NAFLD) strengthens the clinical importance of thyroid–liver interplay: hypothyroidism and even subclinical hypothyroidism increase MASLD risk and severity, while MASLD patients exhibit higher rates of autoimmune thyroiditis and hypothyroidism.⁴,¹¹,²²

Mechanistically, reduced intrahepatic T3 signaling may impair mitochondrial function, autophagy, and lipid flux, promoting progression from steatosis to steatohepatitis and fibrosis.¹,²,¹⁸ Deiodinase dysregulation and bile acid–FXR crosstalk further modulate glucose homeostasis and inflammation.¹⁷,¹⁹ Emerging therapeutics that selectively activate THR-β (e.g., resmetirom) improve steatohepatitis and fibrosis, supporting causality and clinical relevance of the thyroid axis in liver disease.¹,¹³–¹⁶

Despite this, practice on thyroid screening in CLD is inconsistent. Clarifying the prevalence and phenotypes of thyroid dysfunction across etiologies and severities, and their associations with decompensation and near-term outcomes, may help clinicians risk-stratify and select candidates for endocrine evaluation or trials of axis-directed therapies. We therefore investigated the burden and pattern of thyroid dysfunction in a real-world CLD cohort and explored its relationship with decompensation and 90-day outcomes. ⁵–⁹, ¹¹, ¹⁴.

MATERIAL AND METHODS

This is a cross-sectional and observational study  was conducted in the Department of General Medicine, BGS Global Institute of Medical Sciences, Bengaluru over a period of 18 months. Consecutive adults with CLD attending inpatient wards/day-care were screened.

Inclusion criteria: (i) Age ≥18 years; (ii) established CLD (biopsy, elastography >12.5 kPa, imaging, or clinical evidence of portal hypertension/cirrhosis); (iii) stable hemodynamics at sampling.

Exclusion criteria: (i) current levothyroxine, antithyroid drugs, amiodarone, high-dose glucocorticoids; (ii) pregnancy; (iii) pituitary/hypothalamic disease; (iv) known thyroidectomy or radioiodine ablation; (v) acute liver failure; (vi) sepsis at sampling; (vii) ICU care; (viii) recent iodinated contrast (<4 weeks); (ix) end-stage renal disease on dialysis.

Sample size: Assuming thyroid dysfunction prevalence 40% in CLD with 7% precision at 95% confidence, minimum n=197; we enrolled 210 allowing for attrition.

Data collection: Demographics, CLD etiology (viral hepatitis, alcohol-related, MASLD, autoimmune, others), decompensations (ascites, variceal bleed, hepatic encephalopathy), medications (beta-blockers, diuretics), and anthropometry were recorded. Laboratory panel included CBC, LFTs, INR, creatinine, sodium, lipid profile. Severity was graded by Child–Pugh (A/B/C) and MELD-Na.

Thyroid assessment: Morning fasting serum TSH, free T4 (fT4), free T3 (fT3) measured by chemiluminescent immunoassays (traceable to international standards). Reference intervals: TSH 0.4–4.0 mIU/L, fT4 0.8–1.8 ng/dL, fT3 2.0–4.4 pg/mL. Thyroid dysfunction was categorized a priori:
• Low-T3 syndrome: fT3 below reference with normal/low TSH and normal/low fT4.
• Subclinical hypothyroidism (SCH): TSH >4.0 mIU/L with normal fT4/fT3.
• Overt hypothyroidism: TSH >4.0 with low fT4.
• Overt hyperthyroidism: TSH <0.1 with high fT4 and/or high fT3.
Anti-TPO antibodies were measured where clinically indicated.

Outcomes:
Primary—prevalence and patterns of thyroid dysfunction overall and by etiology/severity.
Secondary—associations of thyroid categories and continuous fT3/TSH with decompensations, inpatient stay, and 90-day composite adverse outcome (unplanned readmission, new decompensation, or death).

Statistical analysis: Continuous variables: mean±SD or median [IQR]; categorical: n (%). Group comparisons: t-test/ANOVA or Mann–Whitney/Kruskal–Wallis; categorical: χ²/Fisher’s exact. Correlations: Spearman ρ. Multivariable logistic regression modeled odds of (a) ascites (b) hepatic encephalopathy (c) 90-day composite, with covariates prespecified (age, sex, BMI, etiology, albumin, bilirubin, INR, creatinine, sodium, MELD-Na). fT3 entered as continuous per 0.5-pg/mL decrease; sensitivity analyses included TSH category and exclusion of patients on beta-blockers. Significance p<0.05. Statistical software: R 4.3.

Table 1. Baseline characteristics of the study cohort (N=210)

In table 1, the cohort is middle-aged to older (mean 52.4 years) and predominantly male (68%). MASLD patients are heavier (BMI ~30) and more often male—typical of metabolic liver disease. Disease severity is mixed across etiologies (Child–Pugh A/B/C ~32/39/29; median MELD-Na 17). MASLD tends to be slightly less severe at baseline (MELD-Na 15) than the other groups (18–19).

Table 2. Prevalence and pattern of thyroid dysfunction

In table 2, Thyroid dysfunction is common: only ~53% euthyroid. Low-T3 syndrome is the dominant abnormality (30% overall), slightly more frequent in non-MASLD etiologies (alcohol/viral/other ≈ 33–34% vs 26% in MASLD). Subclinical/overt hypothyroidism are ~10%/5%; overt hyperthyroidism is rare (~1%).

Table 3. Thyroid indices by Child–Pugh class

In table 3, Stepwise decline in fT3 from A→C (3.1 → 2.5 → 2.1 pg/mL; p<0.001) and rising low-T3 prevalence (16% → 32% → 54%; p<0.001). fT4 decreases modestly with worsening class; TSH trends upward (2.28 → 2.90 mIU/L; p=0.01) but remains in/near the reference range.

Table 4. Correlations between thyroid indices and severity markers

In table 4, fT3 correlates strongly with severity: lower fT3 associates with higher MELD-Na (ρ −0.48), lower albumin (ρ +0.44), higher bilirubin/INR, and lower sodium (all p≤0.002). TSH shows weak/nominal associations (small positive with MELD-Na; small negative with albumin).

Table 5. Association of thyroid categories with decompensation (multivariable)

Table 6. Ninety-day composite adverse outcome

In table 6, Each 0.5-pg/mL drop in fT3 increases the odds of the composite adverse outcome (readmission/new decompensation/death) by 38% (aOR 1.38; 95% CI 1.13–1.68; p=0.001).

We observed that nearly half of adults with CLD had thyroid axis abnormalities, dominated by low-T3 syndrome, and that fT3 levels decreased progressively with worsening Child–Pugh and MELD-Na. Our findings align with prior studies showing suppressed fT3/fT4 and relatively preserved or mildly elevated TSH in cirrhosis, with robust associations to severity and outcomes. ⁶–⁹,¹⁰, ¹⁴,²¹ These data support the concept that low-T3 in CLD reflects both adaptive energy conservation and impaired hepatic deiodination amidst inflammation and malnutrition. ³, ⁶,¹⁹

Mechanistic reviews underscore that hepatic THR-β signaling governs lipid flux, autophagy, and mitochondrial biogenesis; attenuated signaling may promote steatotic inflammation and fibrosis.¹,²,⁵,¹⁸ Hepatic deiodinases further fine-tune intrahepatic T3 bioavailability, with disease-induced shifts contributing to the low-T3 phenotype.¹,⁶,¹⁹ Bile acid–FXR pathways intersect with thyroid signaling and glucose homeostasis, potentially amplifying metabolic derangements in CLD.¹⁷ Our correlation of lower fT3 with hypoalbuminemia, hyperbilirubinemia, and higher INR is consistent with a state of impaired synthetic capacity and systemic inflammation linking thyroid status to clinical deterioration.⁸,¹⁴

In the MASLD subgroup, hypothyroid phenotypes were more frequent, echoing meta-analyses that identify primary and subclinical hypothyroidism as risk factors for MASLD presence and progression.⁴,¹¹,¹⁵,²² Importantly, therapeutic advances validate this axis as a modifiable target: the THR-β agonist resmetirom demonstrated histologic NASH resolution and fibrosis improvement in phase-3 trials and received FDA approval (2024) for noncirrhotic MASH with fibrosis.¹³–¹⁶ While our study was not interventional, it motivates pragmatic screening and consideration of endocrine referral in appropriate MASLD patients, particularly where dyslipidemia and SCH coexist.⁴,¹²,¹⁵

Clinical implications include: (i) routine thyroid testing in CLD—especially in decompensated states—can refine risk assessment; (ii) low-T3 identifies patients at higher risk of ascites, encephalopathy, and short-term adverse events and may complement MELD-Na; (iii) in MASLD, addressing hypothyroidism and cardiometabolic risk is integral to liver care. Whether correcting SCH or targeting THR-β signaling improves hard outcomes in cirrhosis remains uncertain; interventional studies in advanced CLD are warranted. ¹,²,¹⁴

Limitations: Single-center design, cross-sectional assessment (causality cannot be inferred), potential assay variability, and lack of long-term mortality follow-up. Strengths include standardized thyroid phenotyping, severity-stratified analyses, and clinically meaningful endpoints over 90 days.

Thyroid dysfunction—predominantly low-T3 syndrome—is common in CLD and strongly tracks with liver severity and near-term complications. Incorporating thyroid assessment into routine CLD care may enhance risk stratification. Future trials should test whether correcting thyroid axis disturbances or selectively activating hepatic THR-β alters the clinical trajectory of CLD.

1. Mullur R, Liu YY, Brent GA. Thyroid hormone and the liver. Hepatology. 2025;81(2):e28–e43. (Lippincott Journals)
2. Chi HC, Chen CY, Tsai CY, Tsai MM, Lin KH. Targeting thyroid hormone/throid hormone receptor axis in liver disease. Front Pharmacol. 2022;13:871100. (Frontiers)
3. Yorke E, Bhadada SK, Zhang Y, et al. Co-morbid hypothyroidism and liver dysfunction: a review. Cureus. 2024;16(1):eXXXXX. (PMC)
4. Xiang L, Chen Y, Hu L, et al. Association between thyroid function and non-alcoholic fatty liver disease. BMC Endocr Disord. 2024;24:XX. (PMC)
5. Sinha RA, Singh BK, Yen PM. Actions of thyroid hormones and thyromimetics on the liver. Nat Rev Gastroenterol Hepatol. 2024;21:XXX–XXX. (PMC)
6. Punekar P, Sharma A, Jain A, et al. A study of thyroid dysfunction in cirrhosis of liver. Indian J Endocrinol Metab. 2018;22(5):645–650. (PMC)
7. Raj A, Bhattacharjee N, Menon S, et al. Association of thyroid function and severity of illness in cirrhosis. Cureus. 2023;15(4):eXXXX. (PMC)
8. Hartl L, Fauster M, Mandorfer M, et al. Lower free triiodothyronine in cirrhosis predicts ACLF and mortality. JHEP Rep. 2024;6(6):100XXXX. (jhep-reports.eu)
9. Abdelrazik FG, Ezzat S, Khairy M, et al. Thyroid function in chronic hepatitis C receiving DAAs. Al-Azhar Assiut Med J. 2021;19(1):XX–XX. (Lippincott Journals)
10. Bebars GM, Abd-Elmonem A, et al. Effect of acute and chronic liver diseases on thyroid profile in children. BMC Pediatr. 2021;21:381. (BioMed Central)
11. Vidal-Cevallos P, Pérez-Hernández JL, et al. Understanding the relationship between NAFLD and thyroid hormones. Int J Mol Sci. 2023;24(19):14605. (MDPI)
12. Wang H, Li X, Zhou L, et al. Role of TSH in regulating lipid metabolism and NAFLD. Nutr Metab Cardiovasc Dis. 2024;34(XX):XXX–XXX. (ScienceDirect)
13. Harrison SA, Bashir MR, Guy CD, et al. Resmetirom for NASH with liver fibrosis—MAESTRO-NASH. N Engl J Med. 2024;390:XXX–XXX. (New England Journal of Medicine)
14. Keam SJ. Resmetirom: first approval. Drugs. 2024;84(7):767–773. (PubMed)
15. Mantovani A, et al. Primary hypothyroidism and MASLD: updated meta-analysis. Gut. 2024;73(9):1554–1564. (bmj.com)
16. FDA approves first treatment for NASH with fibrosis (Rezdiffra, resmetirom). Press release; 2024 Mar 14. (U.S. Food and Drug Administration)
17. Wang T, et al. Bile acid metabolism and signaling in health and disease. Signal Transduct Target Ther. 2024;9:XXX. (Includes TH–FXR crosstalk). (Nature)
18. Chi HC, Lin KH. TH/TR axis and hepatic autophagy/mitochondria in MAFLD. Front Pharmacol. 2022;13:871100. (Frontiers)
19. Bianco AC, da Conceição RR. Deiodinases and the metabolic code for TH action. Endocr Rev. 2021;42(2):123–144. (PMC)
20. Sinha RA, Yen PM. Hepatic deiodinases in thyroid homeostasis and disease. Eur Thyroid J. 2023;12(3):e220211. (bioscientifica.com)
21. Abdel-Aziz M, et al. Thyroid profile in cirrhosis—correlation with severity. J Assoc Physicians India. 2024;72(9):22–28. (org)
22. Pourseyedi N, et al. Hypothyroidism and NAFLD: systematic review/meta-analysis. Ann Med Surg. 2025;XX:XXXX. (PMC)
23. Hirve PN, et al. Thyroid function vs. severity in cirrhosis. Healthcare Bulletin. 2024;12(XX):XX–XX. (healthcare-bulletin.co.uk)
24. FDA NDA 217785 Summary Review: Resmetirom. FDA Access Data; 2024. (FDA Access Data)
25. Harrison SA, et al. MAESTRO-NAFLD-1 safety study of resmetirom. Nat Med. 2023;29:2967–2977. (Nature)