Thyroid Sensitivity Indices Lack Validation for Cardiovascular Risk in Euthyroid Adults

Quantitative indices designed to measure tissue sensitivity to thyroid hormones in individuals with normal circulating blood levels reflect regulatory set points within the pituitary gland rather than verified physiological responsiveness in cardiovascular organs, according to a comprehensive narrative review published in Frontiers in Endocrinology. Investigators led by Peiwen Wang from Peking Union Medical College Hospital and Guojie Ye from the Aerospace Center Hospital evaluated the clinical and epidemiological evidence linking these surrogate indices to heart conditions. Across decades of research, clinical endocrinology focused primarily on overt hyperthyroidism and hypothyroidism. However, epidemiologists increasingly apply mathematical transformations to routine blood tests—specifically free thyroxine and thyroid-stimulating hormone—to calculate central sensitivity metrics, including the thyroid feedback quantile-based index, its parametric counterpart, the thyroid-stimulating hormone index, and the thyrotroph thyroxine resistance index, alongside the peripheral free triiodothyronine to free thyroxine ratio. The review concludes that while these screening formulas show cross-sectional associations with certain cardiometabolic profiles, they lack essential validation against physiological reference standards and must not guide clinical decision-making.

The fundamental biological regulation governing these derived metrics rests on the hypothalamic-pituitary-thyroid axis, where thyrotropin-releasing hormone from the hypothalamus prompts the pituitary gland to secrete thyroid-stimulating hormone, stimulating follicular cells in the thyroid to synthesize and release thyroxine and smaller quantities of triiodothyronine. Circulating thyroxine operates predominantly as a long-lived prohormone that peripheral deiodinase enzymes convert into triiodothyronine, the active ligand exhibiting roughly tenfold greater affinity for nuclear thyroid hormone receptor isoforms alpha and beta. Over ninety-nine percent of circulating thyroid hormones bind tightly to carrier proteins such as thyroxine-binding globulin, transthyretin, and albumin, leaving only tiny unbound free fractions capable of entering target cells through specialized membrane transporters. Within healthy feedback architecture, circulating thyroxine exerts potent negative feedback on both the hypothalamus and pituitary, amplifying minute shifts in thyroxine into substantial adjustments in thyroid-stimulating hormone. Central indices capitalize on this tightly coupled circuit, yet the review stresses that an equilibrium set point established between the pituitary and thyroid does not demonstrate how responsive peripheral tissues, such as the myocardium or vascular endothelium, remain to active hormone exposure.

A critical methodological limitation identified by the review is that the four widely cited central indices are not independent biological markers, but rather mathematical reformulations derived from the exact same laboratory pair of free thyroxine and thyroid-stimulating hormone. When researchers analyze observational datasets and report identical statistical correlations across multiple central indices, they are presenting a single underlying association expressed through four distinct formulas rather than multiple converging lines of biological evidence. Furthermore, none of these mathematical constructions has ever been evaluated against a physiological reference standard, such as observing true organ-level responses to administered exogenous thyroid hormone. The reliability of these measurements is further undermined by substantial within-person biological variation captured in routine testing. In healthy adults sampled over time, indices of individuality remain remarkably low, measuring roughly 0.49 for thyroid-stimulating hormone, between 0.54 and 0.58 for triiodothyronine, and between 0.59 and 0.60 for free thyroxine. Such low individuality figures indicate that individual set points cannot be reliably characterized from a single blood draw, rendering static index values vulnerable to random temporal fluctuations.

Among cardiovascular domains, atherogenic dyslipidemia demonstrates the most consistent epidemiological association and possesses the clearest mechanistic foundation. Thyroid hormone receptor beta represents the dominant receptor isoform expressed in human hepatocytes, where triiodothyronine stimulates low-density lipoprotein receptor expression, suppresses proprotein convertase subtilisin/kexin type 9 through a sterol regulatory element-binding protein-2 dependent pathway, and accelerates cholesterol clearance alongside bile acid synthesis. Multiple cross-sectional studies in euthyroid populations demonstrate that higher central index scores and reduced free triiodothyronine to free thyroxine ratios track higher circulating total cholesterol, low-density lipoprotein cholesterol, triglycerides, and remnant cholesterol. This biochemical pathway mirrors observations in patients harboring inherited monogenic mutations in the thyroid hormone receptor beta gene, who exhibit resistance to thyroid hormone accompanied by blunted lipid-lowering responses and impaired regulation of proprotein convertase subtilisin/kexin type 9 despite elevated endogenous thyroid hormone. Nevertheless, the authors emphasize that while hepatic receptor biology explains lipid clearance, calculating a pituitary feedback formula remains an indirect surrogate that cannot directly measure intrahepatic enzymatic signaling.

In sharp contrast to the consistent lipid findings, epidemiological data linking thyroid hormone sensitivity indices to cardiac arrhythmias, arterial hypertension, and vascular atherosclerosis show contradictory directions and pronounced vulnerability to methodological bias. Investigations into atrial fibrillation highlight these discrepancies: an observational analysis of Chinese patients diagnosed with coronary heart disease linked higher central resistance indices to an inverse, protective relationship with atrial fibrillation, showing an odds ratio of 0.783, whereas a smaller Spanish cohort reported an odds ratio of 8.13 with an exceptionally wide 95 percent confidence interval spanning from 1.33 to 158.20. Similarly, studies evaluating vascular atherosclerosis reached opposing conclusions depending on patient sampling. Chinese angiography investigations associated reduced sensitivity with increased carotid intima-media thickness and greater coronary stenosis severity, reporting an odds ratio of 1.50, whereas a large Korean screening population associated higher parametric resistance scores with lower odds of atherosclerotic cardiovascular disease, yielding a negative log-odds of minus 0.31. The authors explain that restricting studies to hospitalized patients or those undergoing catheterization introduces severe collider bias and reverse causation, distorting baseline epidemiological relationships.

The evaluation of cardiac rhythm and blood pressure across cohort studies similarly yields conflicting and inconclusive signals that fail to demonstrate consistent diagnostic utility. Although thyroid hormone enhances sinus node automaticity and governs the transcription of electrophysiological ion channels responsible for cardiac depolarization and repolarization, population studies have paradoxically linked elevated central indices and lower peripheral conversion ratios to both tachyarrhythmias and bradycardia, the latter defined as a resting heart rate equal to or below sixty beats per minute. Monogenic resistance cohorts also show that up to twenty percent of individuals develop atrial fibrillation or atrial flutter despite high circulating hormone concentrations, underscoring that tissue response cannot be deduced from systemic levels alone. Regarding systemic hemodynamics, where thyroid hormone normally reduces peripheral vascular resistance and enhances renal blood flow, several community and outpatient studies observed modest correlations between elevated central indices and higher hypertension prevalence. However, blood pressure associations remain the weakest and least specific in the published literature, with no prospective studies confirming that sensitivity indices predict new-onset hypertension independently of established clinical determinants.

The interaction between thyroid signaling and heart failure represents an equally complex clinical landscape where peripheral alterations often reflect illness severity rather than primary etiology. The low-triiodothyronine state, characterized by depressed free triiodothyronine alongside normal free thyroxine and thyroid-stimulating hormone, occurs in approximately fifteen to thirty percent of individuals diagnosed with chronic heart failure and closely mirrors hemodynamic deterioration. At the cellular level, active thyroid hormone normally promotes the faster alpha isoform of myosin heavy chain, supports sarcoplasmic/endoplasmic reticulum calcium ATPase-mediated calcium reuptake, preserves myocardial contractility, and lowers systemic vascular resistance. While small randomized trials and meta-analyses demonstrate that short-term triiodothyronine infusions or analog administration like diiodothyropropionic acid can increase stroke volume, improve left ventricular ejection fraction by a weighted mean difference of 5.61 percentage points, and reduce circulating norepinephrine and brain natriuretic peptide, these investigations examined only short-term surrogate markers. The review stresses that reduced peripheral hormone conversion in decompensated, malnourished heart failure patients represents an adaptive consequence of systemic catabolism and inflammation rather than proof that baseline tissue resistance causes heart failure.

Longitudinal and genetic investigations have provided critical insight into the limits of causal interpretation, demonstrating that elevated resistance indices correlate with general cardiometabolic vulnerability without establishing direct etiology. Prospective cohort data remain scarce across the international literature, with two prominent exceptions tracking broad health outcomes rather than distinct cardiac events. In the German Study of Health in Pomerania, elevated parametric feedback indices predicted incident type 2 diabetes over a five-year follow-up period, recording an incidence rate ratio of 1.64 in the highest exposure group, while the Spanish Di@bet.es community study documented that individuals in the top five percent of the central feedback index distribution faced an elevated risk of all-cause mortality over 7.3 years, with a relative risk of 2.61. Complementing these cohort findings, a Mendelian randomization analysis involving over 105,000 participants from the Copenhagen General Population Study demonstrated that genetically predicted lower thyroid-stimulating hormone increased the risk of atrial fibrillation, myocardial infarction, and aortic valve stenosis, but exerted no measurable causal effect on stroke or heart failure. Because this genetic instrument isolates circulating hormone regulation rather than tissue sensitivity per se, it does not confirm that altered tissue responsiveness drives cardiovascular pathology.

A profound translational barrier identified by the authors is the structural mismatch between the physiological axes measured by risk indices and the molecular targets engaged by therapeutic agents. Observational studies demonstrate that individuals with reduced central thyroid sensitivity display an elevated prevalence of metabolic dysfunction-associated steatohepatitis, prompting interest in selective thyromimetics. The primary clinical example is resmetirom, a liver-directed thyroid hormone receptor beta agonist that demonstrated significant resolution of nonalcoholic steatohepatitis, improved hepatic fibrosis, and lowered low-density lipoprotein cholesterol by fourteen to sixteen percent in the phase 3 MAESTRO-NASH clinical trial, securing accelerated approval from the United States Food and Drug Administration in 2024. However, the authors emphasize that the therapeutic success of a liver-targeted receptor beta agonist does not validate central screening indices. A patient with normal or abnormal central feedback metrics can still benefit from hepatic receptor engagement, and confusing pituitary set-point markers with hepatic receptor responsiveness represents a fundamental category error that fails to identify which patients require therapy.

Acute clinical confounders routinely encountered in cardiology practice further impair the interpretability of derived sensitivity scores. Patients undergoing cardiovascular evaluation frequently receive diagnostic and therapeutic interventions that directly perturb thyroid hormone metabolism, including iodinated contrast media administered during coronary angiography, amiodarone therapy that blocks peripheral monodeiodination, beta-blockers, systemic glucocorticoids, and the metabolic stress of critical illness. When clinical investigators collect single blood samples in acute care units or shortly after invasive catheterization, calculated indices reflect these acute pharmacology and systemic stress perturbations rather than an enduring biological trait. To advance the field from unverified associations toward practical clinical utility, future research must establish intra-individual stability across repeated samplings, validate formulas against standardized physiological hormone challenges, rigorously control for contrast and medication exposures, and demonstrate true incremental predictive value using formal metrics such as the concordance statistic, net reclassification improvement, and integrated discrimination improvement within prospective registries.

Subject of Research: Cardiovascular validity and translational gaps of thyroid hormone sensitivity indices in euthyroid populations

Article Title: Thyroid hormone sensitivity and cardiovascular disease in the euthyroid population: indices, mechanisms, and translational gaps

Article References: Ye, G., Zhang, Y., Bai, Y., & Wang, P. (2026). Thyroid hormone sensitivity and cardiovascular disease in the euthyroid population: indices, mechanisms, and translational gaps. Frontiers in Endocrinology, 17, Article 2019099. https://doi.org/10.3389/fendo.2026.2019099

Image Credits: AI Generated

DOI: 10.3389/fendo.2026.2019099

Keywords: thyroid hormone sensitivity, cardiovascular disease, euthyroid population, thyroid feedback quantile-based index, TSH index, thyrotroph T4 resistance index, FT3 to FT4 ratio, atherogenic dyslipidemia, pituitary set point, atrial fibrillation, atherosclerosis, heart failure, resmetirom, thyroid hormone receptor beta, Mendelian randomization, incremental predictive value, cardiology confounders