Section Structure
1. Highlights
2. Clinical background
3. Study design and methods
4. Key results
5. Clinical interpretation
6. Strengths and limitations
7. Practice implications
8. Conclusion
9. Funding and trial registration
10. References
Highlights
In a large Dutch general practice cohort, biochemically confirmed subclinical hyperthyroidism was associated with higher risks of atrial fibrillation, heart failure, and all-cause mortality compared with a matched euthyroid reference population.
Risk appeared graded by the degree of thyrotropin suppression, with atrial fibrillation risk most pronounced among patients with TSH below 0.1 mU/L.
Relative risks were unexpectedly stronger in younger adults, particularly those aged 30 to 49 years, challenging the common tendency to view subclinical hyperthyroidism mainly as a problem of older age.
The findings support individualized risk assessment rather than a one-size-fits-all approach, while also underscoring the need for better prospective data to guide treatment thresholds.
Clinical background
Subclinical hyperthyroidism is typically defined by a suppressed serum thyrotropin concentration with circulating free thyroxine remaining within the reference range. Although patients may have few or no overt symptoms, this biochemical state is not necessarily benign. Thyroid hormone excess, even when mild, can influence heart rate, myocardial contractility, vascular tone, bone metabolism, and neuropsychiatric function. Among these effects, the cardiovascular consequences have drawn the greatest clinical attention.
Prior literature has linked subclinical hyperthyroidism to atrial fibrillation, incident heart failure, reduced bone mineral density in some populations, and increased mortality. However, the size and consistency of cardiovascular risk estimates have varied across studies. Much of the prior emphasis has centered on older adults, since atrial fibrillation and heart failure are more common in this group and because several guidelines prioritize treatment consideration for older patients, especially when TSH is markedly suppressed. Yet the absolute and relative risks in younger adults have remained less well characterized in routine care populations.
The present study by Ursem and colleagues is therefore clinically important. Using a large Dutch primary care data source, the investigators examined whether subclinical hyperthyroidism is associated with adverse cardiovascular outcomes and mortality, and whether these associations differ by age, sex, and biochemical severity.
Study design and methods
This was a retrospective cohort study using general practitioner data from the PHARMO Data Network in the Netherlands between 2012 and 2021. The investigators identified 11,163 patients with biochemically confirmed subclinical hyperthyroidism, defined as suppressed TSH with normal free T4. These patients were compared with a matched euthyroid reference group of 46,058 individuals. Matching was performed on age, sex, and GP practice, an approach that helps reduce confounding related to demographic structure and practice-level differences in care delivery.
The study evaluated incident atherosclerotic complications, atrial fibrillation, heart failure, and all-cause mortality. Multivariable Cox proportional hazards models were used to estimate hazard ratios after adjustment for relevant confounders available in the dataset.
Several aspects of the design deserve attention. First, this was a real-world primary care cohort, which enhances clinical relevance because many patients with subclinical thyroid dysfunction are first identified and followed in general practice rather than specialist settings. Second, the sample size was large enough to support subgroup analyses by age, sex, and TSH category. Third, the exposure definition was biochemical rather than diagnosis-code based, which improves specificity.
At the same time, the authors acknowledge key data limitations. Smoking and alcohol use were unavailable, leaving room for residual confounding. In addition, comorbidity history could only be assessed during the year before cohort entry, which may have led to incomplete adjustment for pre-existing disease. As with all observational studies, causality cannot be established, and treatment decisions during follow-up could have influenced outcomes.
Key results
Overall associations
Compared with euthyroid controls, subclinical hyperthyroidism was associated with a statistically significant increase in atrial fibrillation risk, with a hazard ratio of 1.37 and a 95% confidence interval of 1.22 to 1.55. This is clinically plausible and aligns with long-standing evidence that excess thyroid hormone signaling, even in its milder forms, promotes atrial ectopy, shortens atrial refractory periods, and increases sympathetic sensitivity.
Heart failure risk was also elevated, although more modestly, with a hazard ratio of 1.21 and a 95% confidence interval of 1.04 to 1.40. This suggests that the cardiovascular burden of subclinical hyperthyroidism extends beyond arrhythmia alone. Potential mechanisms include chronotropic stress, tachycardia-mediated myocardial dysfunction, worsening diastolic filling, and interaction with underlying structural heart disease.
For all-cause mortality, the association was stronger, with a hazard ratio of 1.51 and a 95% confidence interval of 1.38 to 1.64. While mortality is a broad endpoint and not specific to cardiovascular mechanisms, the consistency of excess risk strengthens the argument that subclinical hyperthyroidism identifies a clinically relevant risk state rather than an incidental biochemical variant.
The association with atherosclerotic complications was weaker, with a hazard ratio of 1.12 and a 95% confidence interval of 1.00 to 1.24. This borderline estimate suggests either a smaller true effect or greater susceptibility to residual confounding and outcome heterogeneity. Biologically, thyroid excess has mixed effects on vascular biology and lipid metabolism, so a weaker association with atherosclerotic events than with rhythm-related outcomes is not surprising.
Risk by degree of TSH suppression
The most compelling dose-response signal in the study concerned atrial fibrillation. Among patients with TSH below 0.1 mU/L, atrial fibrillation risk increased to a hazard ratio of 1.60 with a 95% confidence interval of 1.32 to 1.94. This is important because current guideline frameworks often distinguish between mildly low TSH and frankly suppressed TSH, generally considering the latter more clinically significant.
Although the abstract does not provide similarly detailed TSH-stratified estimates for all other endpoints, the atrial fibrillation gradient supports the concept that biochemical severity matters. It also suggests that not all subclinical hyperthyroidism should be viewed equally in practice.
Age-specific findings
One of the most notable contributions of this study is the age-stratified analysis. Individuals aged 30 to 49 years with subclinical hyperthyroidism had an atrial fibrillation hazard ratio of 1.88, with a 95% confidence interval of 1.05 to 3.36. For heart failure, the signal in this age group was even stronger, with a hazard ratio of 3.74 and a 95% confidence interval of 1.52 to 9.24. All-cause mortality in this age band was also markedly elevated, with a hazard ratio of 2.95 and a 95% confidence interval of 1.73 to 5.04.
These findings deserve careful interpretation. Younger adults have lower baseline event rates, so elevated relative risks do not necessarily mean high absolute risk. Nonetheless, a strong relative risk in a younger cohort is clinically meaningful because it challenges the assumption that subclinical hyperthyroidism in midlife is usually inconsequential. It may also indicate that when cardiovascular events occur in younger patients with suppressed TSH, the thyroid abnormality is more directly contributory.
Sex-specific findings
The study found that women with subclinical hyperthyroidism had a higher risk of heart failure, with a hazard ratio of 1.31 and a 95% confidence interval of 1.10 to 1.56. Men showed a particularly strong association with all-cause mortality, with a hazard ratio of 1.75 and a 95% confidence interval of 1.50 to 2.05.
Sex differences in thyroid disease epidemiology and in cardiovascular phenotype complicate interpretation. Women are more likely to have autoimmune thyroid disease, whereas men may present later or carry a higher burden of unmeasured cardiometabolic risk. These subgroup findings are hypothesis-generating and should not immediately translate into sex-specific treatment algorithms, but they highlight the need for more refined risk modeling.
Clinical interpretation
For clinicians, the central message is that subclinical hyperthyroidism should not automatically be dismissed as a laboratory curiosity. The strongest and most consistent signal remains arrhythmic risk, particularly atrial fibrillation, which is entirely consistent with prior observational work and pathophysiology. Even modest increases in atrial fibrillation risk matter because the downstream consequences include stroke, heart failure, hospitalization, and anticoagulation exposure.
The heart failure and mortality findings broaden the discussion. Subclinical hyperthyroidism may represent more than a rhythm disorder risk marker. It may also reflect a systemic state that amplifies vulnerability in susceptible patients. However, whether treatment of subclinical hyperthyroidism reduces these outcomes remains less certain than the observational association itself.
The age-specific findings are particularly provocative. Current practice often prioritizes treatment for older adults, especially those older than 65 years and those with TSH below 0.1 mU/L, osteoporosis, or cardiovascular disease. This study does not invalidate that approach, because older patients still carry greater absolute event burden. But it does suggest that younger adults with persistent TSH suppression should not be presumed low priority, particularly if they have palpitations, tachycardia, structural heart disease, risk factors for atrial arrhythmia, or unexplained cardiac symptoms.
Mechanistically, the results are biologically credible. Thyroid hormone increases beta-adrenergic sensitivity, heart rate, and myocardial oxygen demand; enhances atrial automaticity; and can impair diastolic relaxation over time. Repeated or persistent hemodynamic stress may predispose to left atrial remodeling and heart failure, even when free T4 remains within the reference range.
Strengths and limitations
Strengths
This study has several notable strengths. The cohort was large, community-based, and drawn from routine general practice, improving external relevance. Exposure classification relied on thyroid biochemistry rather than administrative labels alone. Matching on age, sex, and practice reduced some baseline imbalances. The study also reported clinically meaningful endpoints rather than surrogate outcomes only.
Limitations
The limitations are equally important. Residual confounding remains a major concern, especially because smoking and alcohol use were unavailable. Both factors affect cardiovascular risk and could correlate with thyroid testing or disease patterns. Comorbidity ascertainment was limited to the year before cohort entry, which may not fully capture long-term cardiovascular history or frailty. Outcome misclassification is possible in routinely collected data, although major diagnoses such as atrial fibrillation and heart failure are generally better recorded than less specific syndromes.
Another limitation is the potential heterogeneity within subclinical hyperthyroidism itself. Etiology matters. Endogenous disease such as Graves disease or toxic nodular thyroid disease may carry different trajectories and treatment patterns than exogenous thyroxine over-replacement. The abstract does not detail these distinctions. Persistence of TSH suppression over time is also important; a single biochemical snapshot may not reflect sustained exposure.
Finally, observational associations do not prove benefit from intervention. Some patients may have transient TSH suppression, non-thyroidal influences on TSH, or competing illnesses that independently increase mortality risk.
Practice implications
This study supports a more deliberate approach to subclinical hyperthyroidism in primary care and specialty practice. First, abnormal thyroid function tests should be confirmed, ideally with repeat testing to document persistence and to exclude assay variation, transient illness effects, or medication-related changes. Second, clinicians should look beyond TSH alone and consider symptoms, heart rate, rhythm history, bone health, age, etiology, and degree of suppression.
Patients with TSH below 0.1 mU/L remain a particularly important group, given the stronger atrial fibrillation signal in this study and the consistency of this threshold with current guideline concern. For younger adults, the data suggest that clinical complacency may be misplaced. Although absolute event rates may remain lower than in older populations, persistent subclinical hyperthyroidism in a 30- to 49-year-old patient with palpitations or cardiovascular comorbidity may merit closer follow-up, ambulatory rhythm assessment in selected cases, and discussion of endocrine referral.
From a health systems perspective, the study also underscores the value of primary care data networks in identifying endocrine-cardiovascular risk patterns that may be missed in specialty registries. Because many patients with mild thyroid dysfunction are managed outside tertiary centers, these datasets can inform more realistic clinical pathways.
Conclusion
The Dutch PHARMO cohort study adds substantial real-world evidence that subclinical hyperthyroidism is associated with clinically meaningful excess risk of atrial fibrillation, heart failure, and all-cause mortality. The gradient of atrial fibrillation risk with stronger TSH suppression reinforces the importance of biochemical severity. Most strikingly, the relative risk estimates were higher in younger adults than many clinicians might expect.
These findings do not by themselves mandate treatment for every patient with suppressed TSH and normal free T4. They do, however, argue against a dismissive approach. Persistent subclinical hyperthyroidism should prompt individualized cardiovascular risk assessment, particularly when TSH is below 0.1 mU/L or when patients are symptomatic or have relevant comorbidity. The next clinical frontier is prospective evidence clarifying which patients benefit most from active treatment and whether intervention can reduce arrhythmia, heart failure, and mortality.
Funding and trial registration
The abstract provided does not report funding details. No ClinicalTrials.gov registration number is listed, which is expected for a retrospective observational cohort study using routine care data.
References
1. Ursem SR, Noordam R, van den Berg JM, Boelen A, Elders PJ, den Elzen WP, Heijboer AC. Subclinical Hyperthyroidism, Cardiovascular Disease and All-Cause Mortality: Insights from a Large Dutch Primary Care Cohort Study. Thyroid. 2026 May 8. PMID: 42100953.
2. Biondi B, Cooper DS. The clinical significance of subclinical thyroid dysfunction. Endocr Rev. 2008;29(1):76-131.
3. Collet TH, Gussekloo J, Bauer DC, den Elzen WPJ, Cappola AR, Balmer P, et al. Subclinical hyperthyroidism and the risk of coronary heart disease and mortality. Arch Intern Med. 2012;172(10):799-809.
4. Ross DS, Burch HB, Cooper DS, Greenlee MC, Laurberg P, Maia AL, et al. 2016 American Thyroid Association Guidelines for Diagnosis and Management of Hyperthyroidism and Other Causes of Thyrotoxicosis. Thyroid. 2016;26(10):1343-1421.
5. Kahapola-Arachchige KM, Hadlow N, Wardrop R, Lim EM, Walsh JP. Age-specific TSH reference ranges have minimal impact on the diagnosis of thyroid dysfunction. Clin Endocrinol (Oxf). 2012;77(5):773-779.
6. Gencer B, Collet TH, Virgini V, Bauer DC, Gussekloo J, Cappola AR, et al. Subclinical thyroid dysfunction and the risk of heart failure events: an individual participant data analysis from 6 prospective cohorts. Circulation. 2012;126(9):1040-1049.