The oxidative balance score impacts serum FT4 levels and all-cause mortality in Euthyroid Participants

doi:10.21203/rs.3.rs-4973287/v1

Abnormal fluctuations in thyroid function within the reference range were strongly associated with increased all-cause mortality. This study aimed to analyze the association between oxidative balance score (OBS) and biomarkers of thyroid function in euthyroid adults and to further investigate whether oxidative balance status affects all-cause mortality by regulating these biomarkers. 5727 euthyroid adults were selected from the National Health and Nutrition Examination Survey (NHANES). Weighted linear regression investigated the potential association of OBS with free thyroxine (FT4) and thyrotropin (TSH). In addition, COX proportional hazard models and restricted cubic spline (RCS) were used to investigate the association between OBS, FT4, TSH, and all-cause mortality. Mediating analyses were performed to determine whether FT4 and TSH modulated the relationship between OBS and all-cause mortality. The results showed that OBS was negatively correlated with FT4 concentration in euthyroid adults. In addition, OBS and FT4 were significantly associated with all-cause mortality. Mediation analysis showed that serum FT4 level mediated the relationship between OBS and FT4 to a certain extent. Among euthyroid individuals, antioxidant OBS significantly correlates with lower FT4 levels. Adherence to an antioxidant diet and lifestyle may help maintain healthier serum FT4 levels, significantly reducing the risk of all-cause mortality.

Health sciences/Health care

Health sciences/Medical research

Oxidative Balance Score

Thyrotropin

Free Thyroxine

All-cause mortality

Diet

Lifestyle

Thyroid hormones (TH) play a pivotal role in mammals in various fields of growth and development, neural differentiation, and energy metabolism in the cardiovascular system¹. The hypothalamic-pituitary-thyroid axis maintains normal thyroid function by regulating the concentration of thyroxine (T4) and thyroid-stimulating hormone (TSH) in the bloodstream^1,2. The current clinical definition of normal thyroid function is primarily based on assessing thyroid function biomarkers within the normal reference range, typically the 2.5th to 97.5th percentile^3,4. However, this approach to defining the range has certain limitations in practice, as it fails to consider the potential risks associated with clinical outcomes entirely. Consequently, the existing reference intervals may not effectively identify individuals at higher risk. Several studies have demonstrated that variations in circulating thyroid hormones may increase the risk of atrial fibrillation⁵, cardiovascular and cerebrovascular disease^6,7, and mortality⁸, even when within the reference range for thyroid function. Indeed, various thyroid diseases and their complications are closely associated with oxidative stress^9,10.

Oxidative stress is characterized by an imbalance between the production of reactive oxygen species and the antioxidant defense system. This results in the disturbance of REDOX signaling and could potentially lead to harm at the molecular level, ultimately contributing to the onset of different diseases^11-13. The Oxidative Balance Score (OBS) is a tool created for evaluating the oxidative stress conditions caused by diet and lifestyle, it is calculated based on a variety of dietary (pro-oxidant and antioxidant nutrients) and lifestyle exposures (smoking, alcohol consumption, obesity, and physical activity), a higher OBS indicating antioxidants are superior to pro-oxidants¹⁴. Several studies have shown that a reduction in OBS is linked to a higher likelihood of developing inflammatory and chronic diseases, as well as an increased risk of mortality from all causes¹⁵. However, more comprehensive studies are needed to investigate the association between OBS and thyroid function, particularly in euthyroid individuals.

To sum up, both oxidative stress and thyroid function biomarkers are associated with all-cause mortality. Therefore, the aim of this study was to analyze the relationship between oxidative balance score (OBS) and thyroid function biomarkers among euthyroid adults and to further investigate whether oxidative balance status influences all-cause mortality by modulating these biomarkers.

2.1 Study design

This research utilized information from the 2007-2012 National Health and Nutrition Examination Survey (NHANES), a cross-sectional study conducted by the National Centers for Disease Prevention¹⁶. Participants in the NHANES provided written informed consent and were approved by the National Center for Health Statistics Ethics Review Board¹⁷. Among the 30,442 subjects from 2007-2012, we excluded those who (1) were under the age of 20 (n = 12729), (2) had dietary and lifestyle OBS elemental deficiencies (n = 2033), (3) did not have thyroid function measurements (n = 7648), (4) reported a history of a previous diagnosis of thyroid disease (n = 2033), (5) TSH not in the range of 0.4 to 4.5 mUI/L and FT4 concentrations not in the range of 9 to 25 pmol/L (n = 1291), (6) pregnant at the time the survey was conducted (n = 64), (7) missing education level (n = 7), (8) missing iodine status (n = 186), and (9) missing mortality status (n = 7). 5727 participants remain in the current analysis (Participant Flowchart, Figure 1).

2.2 Oxidative balance score

The OBS was determined using a combination of 16 dietary nutrients and 4 lifestyle factors, encompassing 15 antioxidant elements and 5 pro-oxidant components^14,18,19. Physical activity was assessed using the NHANES-recommended metabolic equivalent of task (MET) assignment, which includes one week of “(1) vigorous work-related activity, (2) moderate work-related activity, (3) walking or biking for transportation, (4) intense leisure-time physical activity, and (5) moderate leisure-time physical activity”. The resulting score was determined based on the activity level, with 0 points for low-intensity activity, 1 for moderate-intensity activity, and 2 for high-intensity activity. Participants were categorized into heavy drinkers (≥15 g/d for women and ≥30 g/d for men), non-heavy drinkers (0-15 g/d for women and 0-30 g/d for men), and non-drinkers, receiving 0, 1, and 2 points respectively²⁰. The body mass index (BMI) was categorized based on the World Health Organization's (WHO) classification for obesity (BMI ≥ 30 kg/m2), overweight (BMI = 25-30 kg/m2), and normal weight (BMI < 25 kg/m2), with corresponding scores of 0-2. The remaining OBS antioxidant factors included dietary fiber, carotenoids, riboflavin, niacin, vitamin B6, total folate, vitamin B12, vitamin C, vitamin E, calcium, magnesium, zinc, copper, and selenium. The factors were divided into three categories using a three-quartile approach, and the scores for each category (1-3) ranged from 0 to 2. Higher scores indicated an increase in antioxidant levels. The remaining pro-oxidant factors included total iron and fat. Unlike antioxidants, the top tertile of pro-oxidants received a score of 0, while the bottom tertile received a score of 2. The overall OBS was calculated by adding all the components together. For more information on how each element was scored, please see (Table S1).

2.3 Measurement of serum FT4, TSH, and urinary iodine

FT4 and TSH were measured using immunoenzymatic assays²¹. Concerning the clinical practice guidelines of the American Association of Clinical Endocrinologists, we defined a participant with normal thyroid function as having a serum TSH concentration of 0.4-4.5 mUI/L and an FT4 concentration of 9-25 pmol/L^22,23. The concentration of iodine in the urine was measured using mass spectrometry with inductively coupled plasma and dynamic reaction cells ²¹.

2.4 Ascertainment of Outcomes

The mortality status and time of death for NHANES participants were obtained through the National Death Index (NDI). The NCHS website provides corresponding methods for accessing this information (https://www.cdc.gov/nchs/data-linkage/mortality-public.htm). The International Classification of Diseases, 10th edition (ICD-10) was used to establish the cause of death. This study examined all-cause mortality as the primary outcome.

2.5 Covariates

We incorporated variables that have been demonstrated to impact thyroid function and the risk of mortality in prior research, such as information on gender, age, race/ethnicity, level of education, status of iodine intake, and health condition. Participants' iodine intake status was defined based on urinary iodine concentration (UIC), with UIC <100 μg/L as iodine deficiency, 100-299 μg/L as usual, and ≥300 μg/L as iodine overdose²⁴; the definition of Diabetes Mellitus (DM) included a physician's report, glycosylated hemoglobin A1c (HbA1c) ≥ 6.5%, fasting blood glucose ≥ 7.0 μmol/L, or an oral glucose tolerance test (OGTT) ≥ 11.1 mmol/L. Hyperlipidemia is defined as total cholesterol ≥ 200 mg/dL, triglycerides ≥ 150 mg/dL, LDL ≥ 130 mg/dL, or HDL < 40 mg/dL²⁵. The definition of hypertension includes self-reported diagnosis by a medical professional or an average of three consecutive measurements indicating systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg. Cardiovascular disease was characterized by physician diagnoses obtained through personal interviews and included conditions such as coronary heart disease, angina, heart failure, heart disease, and stroke.

2.6 Statistical Analyses

In performing the statistical analyses, we followed the CDC recommendations and chose the appropriate weights to apply to the data analyses¹⁷. Descriptive analysis was conducted to examine the characteristics of participants, as well as their serum FT4, TSH levels, and overall health status. Continuous variables were presented as median values, while categorical variables were expressed as percentages. Multivariate linear regression models examined the association between OBS and FT4, TSH. Due to the skewed data distribution, natural logarithm conversion is performed for OBS and FT4, TSH. The results are then expressed as percent differences²¹. We selected the covariates for adjustment based on the findings from the previous NHANES study. Variance Inflation Factor (VIF) less than 10 was considered free of multicollinearity. We used Model 1 (unadjusted), Model 2 (adjusted for age, gender, and race), and Model 3 (adjusted for age, gender, race, education, and iodine status) to explore potential associations between OBS and normal thyroid function. The continuous OBS was converted into categorical variables using quartile methods, and the P value for trends was computed. Sensitivity analyses were performed to exclude participants with serum TPOAb (>9 IU / mL) and TgAb (>4 IU/mL) to minimize the effect of pre-existing immune disorders of thyroid tissue. In addition, we performed a stratified analysis. A multivariate COX proportional hazard model was employed to investigate the association between OBS, FT4, TSH, and all-cause mortality. Restricted cubic spline was employed to investigate the potential for a nonlinear relationship between these three variables and all-cause mortality. When examining the relationship between OBS and all-cause mortality, we also made adjustments for diabetes, hypertension, hyperlipidemia, and cardiovascular disease. In analyzing the association between FT4, TSH, and all-cause mortality, we further refined our model by including adjustments for smoking, alcohol consumption, and BMI. To assess whether the impact of OBS and all-cause mortality were influenced by FT4 or TSH, mediation effect analyses were conducted while controlling for the covariates including age, sex, race, education, diabetes mellitus, hypertension, hyperlipidemia, and Cardiovascular disease²⁶. The R software (version 4.2.2) and EmpowerStats were utilized for all statistical analyses. Statistical significance was determined based on a two-sided P-value below 0.05.

3.1 Participant characteristics

The median age of the participants was 44 years (95%CI: 31, 57). Approximately 51.7% of the participants were male, and non-Hispanic whites accounted for a larger proportion of the participants at about 67.8%. Regarding iodine intake, 17.3% of participants had excessive intake while 34.3% had insufficient intake. Additionally, it was found that 46.7% of the participants were smokers. Furthermore, among the participants, 11.7%, 64.9%, 30.4%, and 5.92% had diabetes, hyperlipidemia, hypertension, and cardiovascular disease respectively. Over a median follow-up period of 133 months, it was observed that approximately 9.5% of the participants died. For more detailed characteristics of the participants please refer to (Table 1).

3.2 Association of OBS with FT4 and TSH in euthyroid participants.

We explored the correlation between OBS and FT4, TSH by multivariate-adjusted linear regression analysis. The results indicated a notable inverse relationship between OBS and FT4 after adjusting all confounding variables. Further, we divided OBS according to the quartile method. Compared to the lowest quantile of OBS, the serum FT4 in the third quantile of OBS decreased by 2.70% (-4.91, -0.41%), and that in the fourth quantile of OBS decreased by 2.59% (-4.81, -0.30%) (Table 2). The observed trend of decreasing FT4 with increasing OBS was statistically significant (p < 0.05). To validate the robustness of these results, we also performed sensitivity analyses by excluding individuals with serum TPOAb levels exceeding 9 IU/mL and TgAb levels surpassing 4 IU/mL. This was done to minimize the influence of pre-existing immune disorders that may impact thyroid tissue. These analyses' findings aligned with our previous results (Table S2).

(Table S3) displays the association between OBS and FT4 in various gender subcategories. In the model adjusting for all confounding, we observed that in male participants, OBS showed a more significant negative correlation with FT4, whereas in female participants, although the two showed the same tendency to be negatively correlated, the statistical significance of this association was relatively weak. When analyzing the interaction of gender on OBS, there was no interaction of gender on OBS (P for interaction > 0.05).

3.3 Association between OBS and all-cause mortality

(Table 3) Our results indicated that, even after controlling for age, gender, race, education, diabetes, hyperlipidemia, hypertension, and cardiovascular disease, there was an inverse relationship between OBS and mortality. Participants with OBS in the highest percentile (75th-100th percentile) had a significant 31% lower risk of death compared to participants with OBS in the lowest percentile (0th-25th percentile). Restricted cubic spline plots provided a more visual representation (Figure 2A).

3.4 Correlation between indicators of thyroid function and all-cause mortality

(Table 4) In a COX proportional hazards model adjusting for age, gender, race, education, body mass index, smoking, alcohol consumption, diabetes, hyperlipidemia, hypertension, and cardiovascular disease, we observed a significant positive association between FT4 and mortality. In addition, we stratified FT4 according to quartiles, and participants with FT4 levels in the 75th-100th percentile had a substantial 40% increased risk of death compared with participants with FT4 levels in the 0th-25th percentile (P<0.05). The Restricted Cubic Spline (RCS) further reinforces this finding, providing a more visual depiction of the trend (Figure 2 B and C)

3.5 Association between OBS, FT4, TSH, and all-cause mortality.

(Figure 3A and B) We performed a mediation effect analysis to examine whether there was a potential association between OBS and mortality mediated by thyroid function. After comprehensive adjustment for various confounders, it was observed that FT4 partially mediated between OBS and the risk of all-cause mortality, with a mediation effect share of 5.5% (p < 0.01).

Considering the effect of gender, we conducted a subgroup analysis of participants (Table S4). In men, there was a 5.2% mediation effect of serum FT4 between OBS and all-cause mortality. However, we did not find a significant role for FT4 between OBS and all-cause mortality in the female population.

This study evaluated the association of OBS and reference range biomarkers of thyroid function (FT4, TSH) with all-cause mortality using data from 5727 participants in NHANES. Our results showed significant cross-sectional associations between antioxidant OBS and lower FT4 levels. Further analysis showed that FT4 partially mediated the association between OBS and all-cause mortality, suggesting that oxidative stress may also affect all-cause mortality by influencing FT4 levels in euthyroid individuals.

4.1 Negative correlation between OBS and FT4

This is the first cross-sectional study conducted in euthyroid participants to examine the association between diet and lifestyle related oxidative stress states and thyroid function. Our findings indicated a negative association between OBS and circulating FT4 levels, which remained consistent even after excluding participants with abnormal TPOAb or TgAb. Stratified analyses by gender showed that the relationship between OBS and FT4 was more statistically significant in male participants. This may be related to differences in men's and women's endocrine, metabolic, and immune systems^27,28.

A cross-sectional study, encompassing individuals with normal and abnormal thyroid function, revealed a significant negative correlation between OBS and FT4²⁹, a finding consistent with our findings. However, given that abnormal fluctuations in the normal range of thyroid function-related biomarkers are also strongly associated with mortality, and that oxidative stress is not only a key factor in hypothyroidism but also constitutes one of the underlying pathologic processes in hyperthyroidism ^30-33, the present study focused on a group of subjects whose thyroid function was within the reference range. We performed a sensitivity analysis (excluding patients with positive thyroid peroxidase antibodies and positive thyroglobulin antibodies) to verify more precisely the association between OBS and FT4 and TSH levels in a population with normal thyroid function. In addition, a long-term cohort study of 324 euthyroid people showed that adherence to the Mediterranean dietary pattern reduced FT3 and FT4 levels in normal thyroid functioning people³⁴. Meanwhile, the Dietary Antioxidant Index (CDAI), A comprehensive assessment of six dietary antioxidants, including vitamins A, C, and E, as well as manganese (Mn), selenium (Se), and zinc (Zn), was strongly associated with a significant reduction in FT4 levels in another cross-sectional study²⁴. However, when the research focus turned to the complementary effects of nutrients, the results were different from what was expected. A randomized controlled trial showed that participants who received zinc (Zn) and selenium (Se) supplements for 8 weeks did not experience significant changes in their thyroid hormone levels compared with placebo³⁵. This highlights that nutrients do not exist in isolation, but are absorbed and utilized by the body as the building blocks of the overall diet. Therefore, compared with the supplementation of a single nutrient, a comprehensive review, and optimization of individual dietary patterns may be more effective in maintaining the stability and health of thyroid function. After all, the normal operation of thyroid function is inseparable from the fine regulation and dynamic balance of a variety of trace elements, which together promote the synthesis and metabolism of thyroid hormones. In addition, the link between lifestyle and thyroid hormone levels has also been established. A cross-sectional study of 5,766 participants clearly showed that smoking habits were significantly associated with an increase in FT4 levels³⁶. At the same time, most studies report a negative association between BMI and FT4 levels. ^37,38. However, regarding the specific effects of alcohol consumption and physical activity on thyroid function, there are still diverging results and no unified conclusion has been reached^39,40.

4.2 Antioxidant OBS reduces all-cause mortality.

Prior studies in epidemiology have consistently demonstrated a negative correlation between OBS and the risk of all-cause mortality, cardiovascular disease mortality, and cancer mortality⁴¹. Our study also found a notable link between OBS levels and all-cause mortality risk. Specifically, we observed that participants with higher OBS levels (antioxidant OBS) had a 31% lower risk of all-cause mortality than those with lower OBS levels (pro-oxidant OBS).

4.3 Lower FT4 levels in the reference range reduce all-cause mortality

In our study, we included only participants whose thyroid function was within the normal reference range. Higher circulating FT4 is strongly associated with all-cause mortality, even within the reference range for thyroid function.

A research study conducted on the adult population of the United States investigated the potential associations between thyroid hormones (FT3, FT4, and TSH) and death rates among elderly individuals. The research revealed a significant link between reduced FT4 levels and lower mortality in older adults with normal thyroid function⁴². Furthermore, an analysis using data from the National Health and Nutrition Examination Survey (NHANES) found that higher levels of FT4 were linked to an elevated risk of all-cause mortality⁴³. The reason for this association may be that high levels of FT4 are commonly associated with an increased risk of conditions such as hip fractures, cardiovascular disease, and atrial fibrillation in adults with normal thyroid function.

4.4 FT4 might play a role in the relationship between OBS and mortality

After thoroughly examining the relationship between OBS, thyroid-related biomarkers, and all-cause mortality, our findings revealed that FT4 mediated 5.5% of the impact of OBS on all-cause mortality. Moreover, this mediating effect is more significant in men. Although this percentage appears relatively modest, its clinical implications cannot be overlooked. This finding suggests that in euthyroid participants, higher serum FT4 levels are strongly associated with increased all-cause mortality, and OBS modulates FT4 levels to some extent, thereby affecting mortality. Additionally, it implies that antioxidant diets and lifestyles may indirectly impact mortality by influencing thyroid function. However, it is important to note that the influence of OBS on all-cause mortality through FT4 was relatively modest; there may be several additional unidentified factors that could have a more substantial impact on the relationship between OBS and overall mortality.

4.5 Strengths and limitations

Our study demonstrates significant strengths on several fronts. Firstly, our study is the first in-depth exploration of the association between OBS and normal thyroid functioning subjects specifically. To evaluate the strength of the findings, we performed sensitivity and stratification analyses to confirm the rigor and dependability of the study. Secondly, this study is the first to explore the mediating role of thyroid function biomarkers between OBS and all-cause mortality risk. This finding provides new perspectives for understanding the complex relationship between the two.

In addition, it should be noted that our study has certain limitations. Firstly, our study population was primarily comprised of subjects with normal thyroid function, which resulted in our inability to explore relevant conditions in the hyperthyroid and hypothyroid populations. Secondly, although we have adjusted for multiple covariates and performed sensitivity analyses to assess the robustness of our results, there may still be other unaccounted-for confounding factors that could impact our results. Conclusion

Overall, a notable association was noted between antioxidant OBS and decreased FT4 levels in participants with normal thyroid function, and a strong association between OBS and FT4 with all-cause mortality. Following an antioxidant-rich diet and healthy lifestyle choices may help maintain optimal levels of circulating FT4, thereby significantly reducing the risk of all-cause mortality. This offers fresh perspectives on our comprehension of the connection between oxidative stress, thyroid function, and mortality.

5 Data Availability

This study analyzed publicly available data sets. These data can be found at: https://www.cdc.gov/nchs/nhanes/ index.htm.

6 Ethics Approval and Consent to Participate

Participants in the NHANES provided written informed consent and were approved by the National Center for Health Statistics Ethics Review Board.

7 Author Contributions

Conceptualization, XM, ZZ, LM, and QX. Methodology, XM, ZZ, LM, and QX. Software, QX, and ML. Validation, XM, and LM. Formal analysis, QX, BZ, SJ. Data curation, LM, HG, and CX. Writing—original draft preparation, QX, SJ, ML, BZ, QX, CX, and HG. Writing—review and editing, XM, ZZ, QX, and LM. Investigation, XM. Supervision, LM. Funding acquisition, XM. All authors have read and agreed to the published version of the manuscript.

8 Funding

This study was supported by the Clinical Research Award of the First Affiliated Hospital of Xi'an Jiaotong University, China (No. XJTU1AF-CRF-2023-020).

9 Acknowledgments

None.

10 Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Stathatos, N. Thyroid physiology. Med Clin North Am 96, 165-173, doi:10.1016/j.mcna.2012.01.007 (2012).
Choi, S. et al. Thyroxine-binding globulin, peripheral deiodinase activity, and thyroid autoantibody status in association of phthalates and phenolic compounds with thyroid hormones in adult population. Environ Int 140, 105783, doi:10.1016/j.envint.2020.105783 (2020).
Ozarda, Y. et al. Distinguishing reference intervals and clinical decision limits - A review by the IFCC Committee on Reference Intervals and Decision Limits. Crit Rev Clin Lab Sci 55, 420-431, doi:10.1080/10408363.2018.1482256 (2018).
Xu, Y. N. et al. The optimal healthy ranges of thyroid function defined by the risk of cardiovascular disease and mortality: systematic review and individual participant data meta-analysis. Lancet Diabetes Endo 11, 743-754, doi:10.1016/S2213-8587(23)00227-9 (2023).
Baumgartner, C. et al. Thyroid Function Within the Normal Range, Subclinical Hypothyroidism, and the Risk of Atrial Fibrillation. Circulation 136, 2100-2116, doi:10.1161/CIRCULATIONAHA.117.028753 (2017).
van Tienhoven-Wind, L. J. & Dullaart, R. P. Low-normal thyroid function and novel cardiometabolic biomarkers. Nutrients 7, 1352-1377, doi:10.3390/nu7021352 (2015).
Chaker, L. et al. Thyroid Function Within the Reference Range and the Risk of Stroke: An Individual Participant Data Analysis. J Clin Endocrinol Metab 101, 4270-4282, doi:10.1210/jc.2016-2255 (2016).
Xu, Y. et al. The optimal healthy ranges of thyroid function defined by the risk of cardiovascular disease and mortality: systematic review and individual participant data meta-analysis. Lancet Diabetes Endocrinol 11, 743-754, doi:10.1016/S2213-8587(23)00227-9 (2023).
Cheng, W. J. et al. Acupuncture Relieves Stress-Induced Depressive Behavior by Reducing Oxidative Stress and Neuroapoptosis in Rats. Front Behav Neurosci 15, 783056, doi:10.3389/fnbeh.2021.783056 (2021).
Marcocci, C., Leo, M. & Altea, M. A. Oxidative stress in graves' disease. Eur Thyroid J 1, 80-87, doi:10.1159/000337976 (2012).
Marrocco, I., Altieri, F. & Peluso, I. Measurement and Clinical Significance of Biomarkers of Oxidative Stress in Humans. Oxid Med Cell Longev 2017, 6501046, doi:10.1155/2017/6501046 (2017).
Luo, J., Mills, K., le Cessie, S., Noordam, R. & van Heemst, D. Ageing, age-related diseases and oxidative stress: What to do next? Ageing Res Rev 57, 100982, doi:10.1016/j.arr.2019.100982 (2020).
Liguori, I. et al. Oxidative stress, aging, and diseases. Clin Interv Aging 13, 757-772, doi:10.2147/CIA.S158513 (2018).
Zhang, W. et al. Association between the Oxidative Balance Score and Telomere Length from the National Health and Nutrition Examination Survey 1999-2002. Oxid Med Cell Longev 2022, 1345071, doi:10.1155/2022/1345071 (2022).
Talavera-Rodriguez, I. et al. Association between an oxidative balance score and mortality: a prospective analysis in the SUN cohort. Eur J Nutr 62, 1667-1680, doi:10.1007/s00394-023-03099-8 (2023).
Xie, Z. et al. Mediation of 10-Year Cardiovascular Disease Risk between Inflammatory Diet and Handgrip Strength: Base on NHANES 2011-2014. Nutrients 15, doi:10.3390/nu15040918 (2023).
Johnson, C. L. et al. National health and nutrition examination survey: analytic guidelines, 1999-2010. Vital Health Stat 2, 1-24 (2013).
Lee, J. H., Son, D. H. & Kwon, Y. J. Association between oxidative balance score and new-onset hypertension in adults: A community-based prospective cohort study. Front Nutr 9, 1066159, doi:10.3389/fnut.2022.1066159 (2022).
Chen, X. et al. Interplay of sleep patterns and oxidative balance score on total cardiovascular disease risk: Insights from the National Health and Nutrition Examination Survey 2005-2018. J Glob Health 13, 04170, doi:10.7189/jogh.14.04170 (2023).
Xu, Z. et al. Oxidative balance score was negatively associated with the risk of metabolic syndrome, metabolic syndrome severity, and all-cause mortality of patients with metabolic syndrome. Front Endocrinol (Lausanne) 14, 1233145, doi:10.3389/fendo.2023.1233145 (2023).
Sun, Y. et al. Relationship between Blood Trihalomethane Concentrations and Serum Thyroid Function Measures in U.S. Adults. Environ Sci Technol 55, 14087-14094, doi:10.1021/acs.est.1c04008 (2021).
Garber. Clinical Practice Guidelines for Hypothyroidism in Adults: Cosponsored by the American Association of Clinical Endocrinologists and the American Thyroid Association (vol 22, pg 1200, 2012). Thyroid 23, 251-251, doi:10.1089/thy.2012.0205.cxn2 (2013).
Airaksinen, J. et al. Subclinical hypothyroidism and symptoms of depression: Evidence from the National Health and Nutrition Examination Surveys (NHANES). Compr Psychiatry 109, 152253, doi:10.1016/j.comppsych.2021.152253 (2021).
Liu, N., Ma, F., Feng, Y. & Ma, X. The Association between the Dietary Inflammatory Index and Thyroid Function in U.S. Adult Males. Nutrients 13, doi:10.3390/nu13103330 (2021).
Chen, X. et al. Interplay of sleep patterns and oxidative balance score on total cardiovascular disease risk: Insights from the National Health and Nutrition Examination Survey 2005-2018. Journal of Global Health 13, doi:ARTN 0417010.7189/jogh.13.04170 (2023).
Imai, K., Keele, L. & Tingley, D. A general approach to causal mediation analysis. Psychol Methods 15, 309-334, doi:10.1037/a0020761 (2010).
Mammen, J. S. R. & Cappola, A. R. Autoimmune Thyroid Disease in Women. JAMA 325, 2392-2393, doi:10.1001/jama.2020.22196 (2021).
Bauer, M., Glenn, T., Pilhatsch, M., Pfennig, A. & Whybrow, P. C. Gender differences in thyroid system function: relevance to bipolar disorder and its treatment. Bipolar Disord 16, 58-71, doi:10.1111/bdi.12150 (2014).
Song, L. et al. Association between the oxidative balance score and thyroid function: Results from the NHANES 2007-2012 and Mendelian randomization study. PLoS One 19, e0298860, doi:10.1371/journal.pone.0298860 (2024).
Rostami, R., Aghasi, M. R., Mohammadi, A. & Nourooz-Zadeh, J. Enhanced oxidative stress in Hashimoto's thyroiditis: inter-relationships to biomarkers of thyroid function. Clin Biochem 46, 308-312, doi:10.1016/j.clinbiochem.2012.11.021 (2013).
Ates, I. et al. The effect of oxidative stress on the progression of Hashimoto's thyroiditis. Arch Physiol Biochem 124, 351-356, doi:10.1080/13813455.2017.1408660 (2018).
Rostami, R. et al. Serum Selenium Status and Its Interrelationship with Serum Biomarkers of Thyroid Function and Antioxidant Defense in Hashimoto's Thyroiditis. Antioxidants (Basel) 9, doi:10.3390/antiox9111070 (2020).
Diana, T. et al. Stimulatory TSH-Receptor Antibodies and Oxidative Stress in Graves Disease. J Clin Endocrinol Metab 103, 3668-3677, doi:10.1210/jc.2018-00509 (2018).
Zupo, R. et al. Adherence to a Mediterranean Diet and Thyroid Function in Obesity: A Cross-Sectional Apulian Survey. Nutrients 12, doi:10.3390/nu12103173 (2020).
Zavros, A. et al. The Effects of Zinc and Selenium Co-Supplementation on Resting Metabolic Rate, Thyroid Function, Physical Fitness, and Functional Capacity in Overweight and Obese People under a Hypocaloric Diet: A Randomized, Double-Blind, and Placebo-Controlled Trial. Nutrients 15, doi:10.3390/nu15143133 (2023).
Gruppen, E. G. et al. Cigarette smoking is associated with higher thyroid hormone and lower TSH levels: the PREVEND study. Endocrine 67, 613-622, doi:10.1007/s12020-019-02125-2 (2020).
Wolffenbuttel, B. H. R. et al. Thyroid function and metabolic syndrome in the population-based LifeLines cohort study. BMC Endocr Disord 17, 65, doi:10.1186/s12902-017-0215-1 (2017).
Knudsen, N. et al. Small differences in thyroid function may be important for body mass index and the occurrence of obesity in the population. J Clin Endocrinol Metab 90, 4019-4024, doi:10.1210/jc.2004-2225 (2005).
Roa Duenas, O. H. et al. Thyroid Function and Physical Activity: A Population-Based Cohort Study. Thyroid 31, 870-875, doi:10.1089/thy.2020.0517 (2021).
Tian, L., Lu, C. & Teng, W. Association between physical activity and thyroid function in American adults: a survey from the NHANES database. BMC Public Health 24, 1277, doi:10.1186/s12889-024-18768-4 (2024).
Hernandez-Ruiz, A. et al. A Review of A Priori Defined Oxidative Balance Scores Relative to Their Components and Impact on Health Outcomes. Nutrients 11, doi:10.3390/nu11040774 (2019).
Cappola, A. R. et al. Thyroid function in the euthyroid range and adverse outcomes in older adults. J Clin Endocrinol Metab 100, 1088-1096, doi:10.1210/jc.2014-3586 (2015).
Lawton, R. I., Sabatini, B. L. & Hochbaum, D. R. Longevity, demographic characteristics, and socio-economic status are linked to triiodothyronine levels in the general population. Proc Natl Acad Sci U S A121, e2308652121, doi:10.1073/pnas.2308652121 (2024).

Table 1. Characteristics of Study Participants in NHANES 2007–2012.
Characteristic	Weighted N = 76,849,583 Unweighted n = 5,727
Age	44 (31, 57)
FT₄ (pmol/L)	10.30 (9.00, 11.60)
TSH (mUI/L)	1.59 (1.11, 2.25)
OBS	24 (18, 28)
Gender
Male	51.7%
Female	48.3%
Race
Mexican American	8.9%
Other Hispanic	5.6%
Non-Hispanic White	67.8%
Non-Hispanic Black	11.1%
Other Race	6.6%
Education
Less Than 9th Grade	6.3%
9-11th Grade	13.3%
High School Grad/GED or equivalent	80.4%
UIC
Iodine deficient (<100 ug/L)	34.3%
Normal (100-299 ug/L)	48.4%
Excessive iodine intake(≥300 ug/L)	17.3%
Smoking
Yes	46.7%
No	53.3%
Drinking
Yes	22.6%
No	72.3%
BMI
Obesity	33.2%
Overweight	33.5%
Normal	33.3%
Diabetes
Yes	11.7%
No	88.3%
Hyperlipidemia
Yes	64.9%
No	35.1%
Hypertension
Yes	30.4%
No	69.6%
CVD
Yes	94.08(%)
No	5.92(%)
Death
Yes	90.5%
No	9.5%

Median (IQR) or n (%). Abbreviations: TSH, thyroid-stimulating hormone; FT4, free thyroxine; OBS, Oxidative Balance Score; BMI, body mass index; CVD, Cardiovascular disease; UIC stands for urine iodine concentration to evaluate iodine intake.

Table 2. Adjusted Percent Difference (%) and 95% CI in Serum Thyroid Function Measures in Relation to OBS among euthyroid Participants 2007–2012.
OBS	FT4 % (95% CI)	TSH % (95% CI)
Model 1
Continuous	-1.60（-3.62,0.46）*	9.14(0.46,18.58) *
Q1	Ref	Ref
Q2	1.13(-1.12,3.47)	-2.28(-10.67,7.15)
Q3	-1.32(-3.53,0.93)	7.89(-1.37,18.03)
Q4	-1.01(-3.17,1.18)	5.68(-3.17,15.35)
P for trend	0.089	0.053
Model 2
Continuous	-2.50(-4.72, -0.46) *	5.68(-2.95,15.08)
Q1	Ref	Ref
Q2	0.42(-1.83,2.80)	-3.84(-12.10,5.20)
Q3	-2.37(-4.57, -0.12) *	3.51(-5.38,13.24)
Q4	-2.16(-4.35,0.09) *	2.80(-6.03,12.46)
P for trend	0.008	0.230
Model 3
Continuous	-2.95(-5.16, -0.92) *	4.23(-4.50,13.76)
Q1	Ref	Ref
Q2	0.28(-2.05,2.56)	-4.28(-12.50,4.71)
Q3	-2.70(-4.91, -0.41) *	2.33(-6.46,12.20)
Q4	-2.59(-4.81, -0.30) *	1.62(-7.32,11.43)
P for trend	0.003	0.361

Abbreviations: FT₄, free thyroxine; TSH, thyroid-stimulating hormone; Q, quartile; OBS, Oxidative Balance Score. Estimates are derived from a complex survey design, with three models utilized: Model 1 (unadjusted), Model 2 (adjusted for age, gender, and race), and Model 3 (adjusted for age, gender, race, education, and UIC). In the model, natural logarithm conversion was applied to adjust OBS and thyroid function indexes. The results were presented as the percentage difference in serum thyroid function per 10-unit increase in OBS. Percent differences = [e^(ln10×β)– 1] × 100. *P < 0.05.

Table 3. Association between OBS and all-cause mortality.
	HR (95%CI) p-value
OBS
Continuous	0.98(0.97,1.00)0.008
0-25h	Ref
25-50th	0.88(0.69,1.13)0.308
50-75th	0.87(0.67,1.13)0.300
75-100th	0.69(0.52,0.92)0.012

Abbreviations: HR: Hazard Ratio, CI: Confidence Interval; OBS, Oxidative Balance Score. Adjusted for Age, Gender, Race, Education, Diabetes mellitus, Hypertension, Hyperlipidemia, and Cardiovascular disease.

Table 4. Association between FT₄, TSH, and all-cause mortality.
	HR (95%CI) p-value
FT4
Continuous	1.13(1.07,1.19)<0.001
0-25h	Ref
25-50th	0.88(0.68,1.14)0.340
50-75th	1.19(0.91,1.55)0.208
75-100th	1.40(1.07,1.85)0.016
P for trend	<0.001
TSH
Continuous	1.04(0.93,1.16)0.491
0-25h	Ref
25-50th	0.98(0.73,1.31)0.877
50-75th	1.02(0.76,1.35)0.916
75-100th	0.99(0.75,1.30)0.921
P for trend	0.822

Abbreviations: HR: Hazard Ratio, CI: Confidence Interval; FT₄, free thyroxine; TSH, thyroid-stimulating hormone; Adjusted for Age, Gender, Race, Education, Smoking status, Drinking status, BMI, Diabetes mellitus, Hypertension, Hyperlipidemia, and Cardiovascular disease.

No competing interests reported.

SupplementaryMaterial.docx
Supplementary Materials The following are available, Table S1: Oxidative balance score (OBS) items and score assignment. Table S2: Sensitivity analyses. Table S3: Association between oxidative balance scores and thyroid function in euthyroid participants by gender group stratification. Table S4: Mediation analysis between OBS and all-cause mortality by gender group stratification.

The oxidative balance score impacts serum FT4 levels and all-cause mortality in Euthyroid Participants

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods