Abstract
Context
Thyroid hormones play an important role in metabolic homeostasis, and higher levels have been associated with cardiometabolic risk.
Objective
To examine the association of cardiometabolic risk factors with TSH levels in US youth.
Methods
Cross-sectional study of youth aged 12 to 18 years without known thyroid abnormalities from 5 National Health and Nutrition Examination Survey cycles (n = 2818) representing 15.4 million US children. Subclinical hypothyroidism (SH) was defined as thyrotropin (TSH) levels of 4.5 to 10 mIU/L. Assessed cardiometabolic risk factors include abdominal obesity (waist circumference >90th percentile), hypertriglyceridemia (triglyceride ≥130 mg/dL), low high-density lipoprotein cholesterol (<40 mg/dL), elevated blood pressure (systolic and diastolic blood pressure ≥90th percentile), hyperglycemia (fasting blood glucose ≥100 mg/dL, or known diabetes), insulin resistance (homeostatic model for insulin resistance > 3.16), and elevated alanine transferase (≥ 50 for boys and ≥44 U/L for girls). Age and sex- specific percentiles for thyroid parameters were calculated.
Results
In this cohort of youth (51.3% male), 31.2% had overweight/obesity. The prevalence of SH was 2.0% (95% CI 1.2-3.1). The median TSH levels were higher in youth with overweight/obesity (P < 0.001). Adjusting for age, sex, race/ethnicity, and obesity, youth with TSH in the fourth quantile had higher odds of abdominal obesity (OR 2.53 [1.43-4.46], P = .002), insulin resistance (OR 2.82 [1.42-5.57], P = .003), and ≥2 cardiometabolic risk factors (CMRF) (OR 2.20 [1.23-3.95], P = .009).
Conclusion
The prevalence of SH is low in US youth. The higher odds of insulin resistance and cardiometabolic risk factors in youth with TSH levels >75th percentile requires further study.
Keywords: thyroid function, youth, cardiometabolic risk
Thyroid hormones play an important role in normal growth and development as well as regulating metabolism, thermogenesis, and cardiovascular function (1). Many of the metabolic activities regulated by thyroid hormones are related to the anabolism and/or catabolism of macromolecules that affect energy homeostasis, and thyroid hormones, triiodothyronine (T3) and thyroxine (T4) have direct effects on both cholesterol and fatty acid synthesis and metabolism (2, 3). The feedback loop of thyroid function is regulated by the thyrotropin (TSH) secreted from the anterior pituitary in response to the circulating thyroid hormone(s). TSH levels in the blood are the most sensitive markers of optimal thyroid function and mild elevation of TSH (4.5-10 mIU/L), called subclinical hypothyroidism (SH), has been considered the earliest manifestation of thyroid dysfunction. In adults, SH has been associated with adverse cardiovascular outcomes, such as carotid intima thickness (4, 5), coronary heart disease (6), heart failure (7), stroke (8), nonalcoholic fatty liver disease (9), and the role of thyroid replacement therapy is being debated (10, 11).
Cohort studies in children and adolescents have found an association between elevated TSH levels or SH and obesity (12-17). An association has also been demonstrated between SH and cardiometabolic risk factors, such as hypercholesterolemia (18, 19), carotid intima thickness (19), and glucose dysmetabolism (20-22). Whether these associations are a result of the insulin resistance seen with obesity or are independently associated with TSH remains to be established. There is sparsity of such studies from population-based cohorts.
The objective of this study was to assess the relationship between TSH levels and cardiometabolic risk factors namely abdominal obesity, dyslipidemia, glucose intolerance, elevated blood pressure, insulin resistance, and elevated alanine transferase (23) in a nationally representative sample of children and adolescents from the US from the National Health and Nutrition Examination Survey (NHANES). Additionally, we aimed to identify the reference percentiles for measures of thyroid function by age and gender in children with normal weight.
Material and Methods
Study population
The data for this study were obtained from combining 5 2-year cycles of NHANES where thyroid function tests were measured (1999-2000; 2001-2002, 2007-2008, 2009-2010, 2011-2012). NHANES is a cross-sectional, multistage complex nationally representative survey that uses interview, physical examination, and laboratory data to assess the health and nutritional status of the noninstitutionalized civilian population in the United States (24).
Study cohort
Subjects for the study are youth between the ages of 12 and 18 years at the time of examination. Data pertinent to the thyroid function, lipid levels, liver enzymes, measures of glucose homeostasis in fasting samples, anthropometric and demographic measures were obtained. Subjects with history of thyroid disease or those on thyroid supplements or drugs that influence thyroid function, such as steroids, lithium, amiodarone, or beta blockers, were excluded. Additionally, subjects with clinical hypothyroidism, hyperthyroidism, or those with abnormal creatinine were excluded. As antithyroid antibodies (antithyroidperoxidase [anti-TPO] or antithyroglobulin [anti-TG]) may cause SH, those with levels greater than the reference ranges were excluded. The interview data contains information on race/ethnicity that was classified as non-Hispanic white, non-Hispanic black, Hispanic, or other. The average response rate for the study cohort was 80% (range 75-85). The Columbia University Irving Medical Center Institutional Review Board considered the study exempt as data are publicly available without obligations.
Anthropometric measurement
NHANES survey instruments and protocols have been established and ongoing in 2-year cycles since 1999 (24). Briefly, anthropometric measures include weight measured to the nearest 0.1 kg using a digital weight scale with participants wearing the standard mobile examination center gown. Standing height was measured to the nearest 0.1 cm using a stadiometer with a fixed vertical backboard and an adjustable head piece. Body mass index (BMI) was calculated as weight (kg)/height (m) squared. Subjects with BMI between the 5th and 85th percentile were considered normal weight, 85th to 94th percentile overweight, and ≥95th percentile obese (25). Waist circumference was measured in the horizontal plane to the nearest 0.1 cm at the junction between the uppermost lateral border of the right ilium and the midaxillary line, which extends from the armpit down the side of the torso. Blood pressure was measured 3 times and the average was used.
Biochemistry assessment
TSH was measured with a 2-site immunoenzymatic assay and the total T4 was measured with a competitive binding immunoenzymatic assay summarized in eTable 1 (26). The total T4 levels for the 2011 to 2012 cycle were adjusted to align with the measurements from previous years for instrument change (27). TG antibody and TPO antibody were measured using a sequential 2-step immunoenzymatic “sandwich” assay (27-30). Triglycerides were measured enzymatically in serum or plasma using a series of coupled reactions in which triglycerides are hydrolyzed to produce glycerol. High-density cholesterol (HDL-C) was measured by heparin manganese precipitation or a direct immunoassay method. The fasting plasma glucose and insulin levels were forward calibrated to the methodology used in 2011 to 2012 (31-35). Alanine aminotransferase (ALT) was measured by the catalytic action that causes the interconversion of amino acids and α-ketoacids by transfer of amino groups.
Cardiometabolic risk factors
We used the organizational framework offered by metabolic syndrome (MetS) to assess the cardiometabolic risk factors in children and adolescents (23). For this study, the criteria defined by the International Diabetes Federation (36) were updated for hypertriglyceridemia and elevated blood pressure according to current guidelines as follows: (1) abdominal obesity (waist circumference > 90th percentile for 10-16 years and/or ≥94 cm in males and ≥80 cm for females); (2) hypertriglyceridemia (triglyceride ≥130 mg/dL) (37); (3) low high-density cholesterol (HDL-C <40 mg/dL); (4) elevated blood pressure (systolic and diastolic blood pressure ≥90th percentile) (38); (5) hyperglycemia (fasting blood glucose ≥100 mg/dL, or reported diagnosis of diabetes). Two additional measures of metabolic risk used were insulin resistance measured by the homeostatic model for insulin resistance (HOMA-IR = fasting insulin (µU/L) × fasting glucose (mmol/L)/22.5) (39) with >3.16 considered high (40) and ALT levels as a measure for hepatic steatosis (≥50 U/L for boys and ≥44 U/L for girls) (41).
Data analysis
BMI and waist circumference were transformed into age- and sex-appropriate percentiles and z-score respectively using the CDC 2000 growth charts (42). The thyroid related variables were examined by probability density estimation to assess the shape of distribution. The summary of the distribution is presented as median with standard deviation for continuous variables or percentage for categorical variables. The characteristics of participants were compared using complex survey design adjusted differences in the median.
TSH quantiles were generated using quantile regression adjusting for age and sex to account for the absence of the pubertal status in the NHANES dataset. The fourth quantile was considered to be relevant for cardiometabolic risk. After ensuring absence of temporal trends between the survey cycles, the 5-survey cycles were combined for analysis and 10-year survey weights were constructed by rescaling the NHANES survey weights so that the sum of the weights matches the survey population at the midpoint of that period (43). Univariate and multivariable analysis using complex survey-based generalized linear models with a logarithmic link was used to assess the odds of the cardiometabolic risk factors with TSH quantiles while adjusting for the age, gender, race/ethnicity and level of obesity. A binary classification of TSH using the definition of SH (TSH >4.5 mIU/L with normal free T4 levels) was used to assess the association of the risks with SH (10, 44). Survey based goodness of fit tests were used to test the model fit. Interaction between TSH quantiles and other predictors were tested using design-adjusted Wald test and likelihood ratio tests.
We imputed the missing data for thyroid function tests by multivariate imputation by chained equations using predictive mean matching to ascertain the percentiles. The imputation was verified by random verification of the complete dataset with similar methods. Sensitivity analysis for the results was performed using the complete cases. The terms with P < .05 were considered statistically significant and 95% CI was calculated. Age- and sex-specific percentiles of the thyroid parameters among subjects with normal weight (BMI 5-85th percentiles) were calculated using the imputed dataset by Generalized Additive Models for Location Scale and Shape (GAMLSS) package using different Box-Cox transformation distribution family and the best fitted model was determined by the Akaike Information Criterion.
Statistical analyses were performed in R v3.5.1 using tidyverse, survey, quantreg, mice, gamlss packages and Stata v 16.1 (StataCorp LLC, College Station, TX: svy suite of commands).
Results
Characteristics of study subjects
Of the 3093 youth between the ages of 12 and 18 years with thyroid function data in 5 NHANES cycles, 278 were excluded (TSH ≥10 mIU/L = 5, total T4 >13.2 µg/dL = 10, antithyroid antibodies = 153, thyroid-related meds = 20, no survey weights = 87). The weighted final study cohort of 2818 represents 15.4 million US youth (51.7% boys, 95% CI 49.7-53.8%); of which 32.0% (95% CI 29.5-34.6%) had BMI ≥85th percentile. The characteristics of the study subjects are shown in Table 1 grouped by BMI percentile. The distribution of age and gender was similar between the 2 groups. The prevalence of obesity was higher in Hispanic and non-Hispanic Black youth. The cardiometabolic risk factors were higher in the group with overweight/obesity by BMI percentile. The TSH levels were significantly different between the 2 groups, while the total and free T4 and total T3 levels were not.
Table 1.
Demographic distribution of the cohort
| BMI < 85th percentile Median (95% CI) N = 1732 (10.3 million) | BMI ≥ 85th percentile Estimate (95% CI) N = 1046 (4.9 million) | P-value | |
|---|---|---|---|
| Proportion of population, % | 68.0 (65.3-70.5) | 32.0 (29.5-34.6) | |
| Male, % | 51.4 (48.9-53.9) | 53.5 (49.2-57.7) | .44 |
| Age, mean (years) | 15.5 (15.3-15.6) | 15.4 (15.2-15.6) | .41 |
| Race/ethnicity, % | |||
| Hispanic | 15.8 (12.8-19.3) | 22.7 (17.9-28.3) | .001 |
| Non-Hispanic Black | 12.7 (10.6-15.2) | 19.6 (15.9-24.0) | <.001 |
| Non-Hispanic White | 62.8 (58.8-66.6) | 50.0 (44.0-56.0) | <.001 |
| BMI z-score | 0.08 (0.03-0.13) | 1.76 (1.72-1.80) | <.001 |
| Waist circumference z-score | -0.05 (-0.02-0.08) | 1.42 (1.38-1.46) | <.001 |
| Abdominal obesity, % | 6.4 (4.9-8.4) | 93.2 (91.0-94.8) | <.001 |
| Waist-height ratio > 0.5, % | 12.3 (9.2-16.3) | 93.0 (90.8-94.7) | <.001 |
| Systolic BP, mm Hg | 107.0 (106.5-107.5) | 112.0 (111.0-113.0) | <.001 |
| TSH (uIU/mL) | 1.33 (1.28-1.39) | 1.49 (1.43-1.54) | <.001 |
| Total T4 (ug/dL) | 7.50 (7.45-7.55) | 7.60 (7.50-7.70) | .07 |
| Total T3 (ng/dL) | 129.0 (127.0-131.0) | 131.5 (129.0-133.9) | .12 |
| Free T4 (ng/dL) | 0.80 (0.78-0.81) | 0.80 (0.77-0.81) | .31 |
| Free T3 (pg/mL) | 3.63 (3.60-3.65) | 3.70 (3.65-3.75) | .009 |
| Triglyceride (mg/dL) | 66.0 (63.1-68.9) | 83.0 (77.6-88.4) | <.001 |
| HDL-cholesterol (mg/dL) | 52.0 (51.0-53.0) | 45.0 (44.0-46.0) | <.001 |
| HOMA-IR | 1.74 (1.66-1.81) | 3.24 (3.01-3.46) | <.001 |
| ALT (U/L) | 15.0 (14.5-15.5) | 19.0 (18.5-19.5) | <.001 |
SI unit conversion: total T4 (nmol/L) = multiply by 12.9; Total T3 (nmol/L) = multiply by 0.0154; fT4 (pmol/L) = multiply by 12.9.
fT3 (pmol/L) = multiply by 1.5362; TG (mmol/L) = multiply by 0.0113; HDL-C (mmol/L) = multiply by 0.0259.
Abbreviations: ALT, alanine aminotransferase; BMI, body mass index; BP, blood pressure; HDL, high-density lipoprotein; HOMA-IR, homeostatic model for insulin resistance; TSH, thyroid stimulating hormone; T4, thyrotropin; T3, thyroxine.
BMI percentile adjusted for age and sex.
In the subgroup analysis by gender, the TSH levels were similar in both, but other thyroid parameters were statistically different. In the cardiometabolic risk factors, waist circumference z-scores were lower in girls with obesity. HDL-C levels were higher and ALT levels were lower in the girls, and HOMA-IR was similar in either gender with obesity (eTable 2 (26)). We found no differences in the temporal trends in TSH, Total T4 and Total T3 by NHANES cycles (eFig. 1 (26)). The distribution of the cardiometabolic risk factors by the TSH quantiles and gender is shown in eTable 3 (26).
Cardiometabolic risk factors by TSH levels
The cardiometabolic risk factors were significantly higher in the group with overweight/obesity by BMI percentile, as were the TSH levels (Table 1). There was a positive correlation between TSH levels and BMI-z score (r = 0.1, P < .001), waist circumference (r = 0.2, P < .001), HOMA-IR (r = 0.1, P < .001), and triglycerides (r = 0.1, P < .001). A greater number of subjects with overweight and obesity were seen in the higher age- and sex-adjusted quantiles (Fig. 1). In the univariate regression analyses weighted by the 10-year constructed weights, the fourth quantile of TSH was associated with increased odds of abdominal obesity, hypertriglyceridemia, elevated blood pressure, abnormal HOMA-IR, abnormal ALT levels, and multiple risk factors compared with the first quantile (Table 2, Model 1). This association continued to have statistical significance while controlling for age, gender and race/ethnicity (Model 2). When controlling for the level of obesity in addition to age, gender and race/ethnicity, the fourth quantile for TSH was significantly associated with abdominal obesity, elevated HOMA-IR and ≥2 cardiometabolic risk factors (Model 3). Two percent (95% CI 1.2-3.1) subjects had SH by the conventional definition, with no differences between the 2 BMI categories. No association was seen between the presence of SH and any of the cardiometabolic risk factors. No significant interactions were identified between the TSH quantiles and other predictors. In this cohort, similar results were obtained in repeat analyses including the subjects with antithyroid antibodies.
Figure 1.
Proportion of individuals based on BMI percentile in each quantile of TSH.
Table 2.
Association between fourth quantile of TSH and cardiometabolic risk factors
| Regression models | Subpopulation size (population represented) | Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|---|---|
| OR* (95% CI) | P value | OR* (95% CI) | P value | OR* (95% CI) | P value | ||
| Abdominal obesity | 2524 (13.7 million) | 3.56 (2.39-5.32) | <.001 | 3.63 (2.43-5.42) | <.001 | 2.36 (1.22-4.58) | .01 |
| Hypertriglyceridemia | 1118 (12.1 million) | 2.13 (1.04-4.34) | .04 | 2.14 (1.02-4.51) | .04 | 1.83 (0.85-3.92) | .12 |
| Low HDL cholesterol | 2551 (13.9 million) | 1.04 (0.65-1.65) | .87 | 1.00 (0.62-1.61) | .99 | 0.72 (0.44-1.19) | .20 |
| Elevated blood pressure | 2205 (12.0 million) | 1.85 (1.10-3.08) | .02 | 2.07 (1.26-3.43) | .005 | 1.55 (0.92-2.59) | .09 |
| Hyperglycemia | 1118 (12.1 million) | 1.45 (0.80-2.60) | .21 | 1.25 (0.70-2.22) | .43 | 1.36 (0.78-2.35) | .27 |
| HOMA-IR | 1109 (12.0 million) | 3.37 (1.91-5.94) | <.001 | 3.57 (2.02-6.29) | <.001 | 2.85 (1.46-5.59) | .003 |
| Elevated ALT level | 2549 (13.9 million) | 3.91 (1.13-13.52) | .03 | 4.22 (1.16-15.29) | .03 | 2.17 (0.67-7.01) | .19 |
| Risk score | 1118 (12.1 million) | 2.36 (1.39-4.00) | .002 | 2.26 (1.30-3.89) | .004 | 2.16 (1.18-3.96) | .01 |
TSH quantile defined by quantile regression model adjusted for age and sex. 1st quantile used as reference. fourth quantile of TSH => 2.05 mIU/L. The values in bold are statistically significant.
Model 1, unadjusted; Model 2, adjusted for age, sex and race/ethnicity; Model 3, adjusted for age, sex, race/ethnicity and BMI.
Abbreviations: OR, odds ratio; HDL, high-density lipoprotein; HOMA-IR, homeostatic model for insulin resistance; ALT, alanine aminotransferase.
Age- and sex-specific percentile of TSH levels among normal weight subjects
Generalized Additive Models were used to derive the age- and sex-specific percentiles of thyroid parameters based on the values for the subjects with BMI between the 5th and 85th percentile (Table 3). Of the 1791 subjects with BMI between the 5th and 85th percentile, TSH levels were imputed for 318 (17.7%), total T4 for 328 (18.3%), total T3, free T4, and free T3 levels for 923 (51.5%) individuals. TSH levels decreased with increasing age and were consistently lower in girls than in the boys in each age category. On the other hand, the total T4 levels increased with age and were higher in girls compared to the boys. The total T3 levels decreased with increasing age and were higher in girls. The levels of the free T4 as well as free T3 remained remarkably consistent across ages. The percentile curves can be seen in eFig. 2 (26).
Table 3.
Age- and sex-specific percentiles for thyroid parameters
| Percentiles for boys | Percentiles for girls | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sample | Sample | |||||||||||||||
| Age (years) | size | 2.5 | 5 | 25 | 50 | 75 | 95 | 97.5 | size | 2.5 | 5 | 25 | 50 | 75 | 95 | 97.5 |
| 12 | 115 | 0.5 | 0.6 | 1.1 | 1.6 | 2.2 | 3.7 | 4.4 | 105 | 0.4 | 0.5 | 0.9 | 1.4 | 1.9 | 3.4 | 4.2 |
| 13 | 140 | 0.5 | 0.6 | 1.1 | 1.5 | 2.2 | 3.6 | 4.3 | 144 | 0.4 | 0.5 | 0.9 | 1.3 | 1.9 | 3.3 | 4.1 |
| 14 | 133 | 0.5 | 0.6 | 1 | 1.5 | 2.1 | 3.5 | 4.1 | 145 | 0.4 | 0.5 | 0.9 | 1.3 | 1.8 | 3.3 | 4 |
| 15 | 134 | 0.5 | 0.6 | 1 | 1.4 | 2 | 3.3 | 4 | 114 | 0.4 | 0.5 | 0.9 | 1.2 | 1.8 | 3.2 | 3.9 |
| 16 | 144 | 0.5 | 0.6 | 1 | 1.4 | 1.9 | 3.2 | 3.8 | 129 | 0.4 | 0.5 | 0.8 | 1.2 | 1.7 | 3.1 | 3.8 |
| 17 | 136 | 0.4 | 0.5 | 0.9 | 1.3 | 1.9 | 3.1 | 3.7 | 125 | 0.4 | 0.5 | 0.8 | 1.2 | 1.7 | 3 | 3.7 |
| 18 | 115 | 0.4 | 0.5 | 0.9 | 1.3 | 1.8 | 3 | 3.5 | 112 | 0.4 | 0.5 | 0.8 | 1.1 | 1.6 | 2.9 | 3.6 |
| Total T4 level (µg/dL) | ||||||||||||||||
| 12 | 115 | 5.2 | 5.6 | 6.6 | 7.5 | 8.4 | 9.9 | 10.5 | 105 | 5.3 | 5.7 | 6.9 | 7.7 | 8.5 | 10.4 | 11.2 |
| 13 | 140 | 5.2 | 5.5 | 6.5 | 7.3 | 8.2 | 9.7 | 10.3 | 144 | 5.3 | 5.6 | 6.8 | 7.6 | 8.5 | 10.3 | 11.1 |
| 14 | 133 | 5.1 | 5.4 | 6.4 | 7.2 | 8.1 | 9.6 | 10.2 | 145 | 5.3 | 5.6 | 6.8 | 7.6 | 8.5 | 10.3 | 11.1 |
| 15 | 134 | 5.1 | 5.4 | 6.4 | 7.2 | 8.1 | 9.6 | 10.1 | 114 | 5.3 | 5.7 | 6.9 | 7.7 | 8.5 | 10.4 | 11.2 |
| 16 | 144 | 5.1 | 5.4 | 6.5 | 7.3 | 8.2 | 9.7 | 10.2 | 129 | 5.4 | 5.8 | 6.9 | 7.8 | 8.7 | 10.5 | 11.3 |
| 17 | 136 | 5.2 | 5.5 | 6.6 | 7.4 | 8.3 | 9.8 | 10.4 | 125 | 5.5 | 5.9 | 7.1 | 7.9 | 8.8 | 10.7 | 11.5 |
| 18 | 115 | 5.3 | 5.6 | 6.7 | 7.5 | 8.4 | 9.9 | 10.5 | 112 | 5.6 | 6 | 7.2 | 8.1 | 9 | 11 | 11.8 |
| Total T3 level (ng/dL) | ||||||||||||||||
| 12 | 115 | 110 | 116 | 137 | 151 | 167 | 192 | 201 | 105 | 97.2 | 103 | 123 | 137 | 152 | 183 | 196 |
| 13 | 140 | 105 | 112 | 131 | 145 | 160 | 184 | 193 | 144 | 91.6 | 97.2 | 116 | 129 | 144 | 173 | 185 |
| 14 | 133 | 101 | 107 | 125 | 139 | 153 | 176 | 185 | 145 | 88.2 | 93.6 | 111 | 124 | 138 | 166 | 178 |
| 15 | 134 | 96.5 | 102 | 120 | 133 | 146 | 169 | 177 | 114 | 86 | 91.3 | 108 | 121 | 135 | 162 | 173 |
| 16 | 144 | 93.1 | 98.4 | 116 | 128 | 141 | 163 | 171 | 129 | 84.7 | 89.9 | 107 | 119 | 133 | 160 | 171 |
| 17 | 136 | 90.6 | 95.8 | 113 | 125 | 138 | 159 | 166 | 125 | 84.7 | 89.9 | 107 | 119 | 133 | 160 | 171 |
| 18 | 115 | 88.6 | 93.7 | 110 | 122 | 135 | 155 | 162 | 112 | 86.1 | 91.3 | 109 | 121 | 135 | 162 | 173 |
| Free T4 level (ng/dL) | ||||||||||||||||
| 12 | 115 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 | 105 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 |
| 13 | 140 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 | 144 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 |
| 14 | 133 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 | 145 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 |
| 15 | 134 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 | 114 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 |
| 16 | 144 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 | 129 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 |
| 17 | 136 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 | 125 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 |
| 18 | 115 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 | 112 | 0.6 | 0.6 | 0.7 | 0.8 | 0.9 | 1 | 1.1 |
| Free T3 level (pg/mL) | ||||||||||||||||
| 12 | 115 | 3.3 | 3.4 | 3.8 | 4 | 4.3 | 4.8 | 5 | 105 | 3 | 3.1 | 3.5 | 3.7 | 4 | 4.5 | 4.7 |
| 13 | 140 | 3.2 | 3.3 | 3.7 | 3.9 | 4.2 | 4.7 | 4.9 | 144 | 2.9 | 3 | 3.4 | 3.6 | 3.8 | 4.3 | 4.5 |
| 14 | 133 | 3.1 | 3.3 | 3.6 | 3.8 | 4 | 4.5 | 4.8 | 145 | 2.9 | 3 | 3.3 | 3.5 | 3.8 | 4.2 | 4.4 |
| 15 | 134 | 3.1 | 3.2 | 3.5 | 3.7 | 3.9 | 4.4 | 4.6 | 114 | 2.8 | 2.9 | 3.2 | 3.4 | 3.7 | 4.1 | 4.3 |
| 16 | 144 | 3 | 3.1 | 3.4 | 3.6 | 3.9 | 4.3 | 4.5 | 129 | 2.7 | 2.8 | 3.1 | 3.4 | 3.6 | 4 | 4.2 |
| 17 | 136 | 2.9 | 3 | 3.4 | 3.5 | 3.8 | 4.2 | 4.4 | 125 | 2.7 | 2.8 | 3.1 | 3.3 | 3.5 | 4 | 4.2 |
| 18 | 115 | 2.9 | 3 | 3.3 | 3.5 | 3.7 | 4.1 | 4.3 | 112 | 2.7 | 2.8 | 3.1 | 3.3 | 3.5 | 4 | 4.2 |
SI units conversion: Total T4 (nmol/L) = multiply by 12.9; Total T3 (nmol/L) = multiply by 0.0154; fT4 (pmol/L) = multiply by 12.9.
fT3 (pmol/L) = multiply by 1.5362.
Abbreviations: TSH, thyroid stimulating hormone; T4, thyrotropin; T3, thyroxine.
Discussion
This is the first study to comprehensively examine the association of TSH levels with cardiometabolic risk factors in children from a nationally representative noninstitutionalized population in the United States comprising of 5 NHANES cycles over 10 years. The prevalence of SH was low (2.0%, 95% CI 1.2-3.1) with no differences in children with normal BMI or those with overweight and obesity. This is remarkably different from the report of SH in 14.2% children in the Korean National Health and Nutrition Examination Surveys reported by Lee et al. with statistically significant higher proportion in children with obesity (45). The higher prevalence in Korean youth may represent the higher cardiometabolic risks seen in Asians for age- and sex-adjusted BMI (46-49). Further, this study did not report on subjects with anti-thyroid antibodies that may confound the results.
We noted an association of higher TSH levels with abdominal obesity, insulin resistance and multiple cardiometabolic risk factors while adjusting for BMI. Higher TSH levels have been previously reported in children with obesity (12-15). Animal and human studies have identified a complex relationship of thyrotropin, thyroid hormones and glucose metabolism (50). Whether such relationships are mediated by the direct effects of TSH and thyroid hormones on skeletal muscle and adipose tissue or via hepatic metabolism or mitochondrial oxidation remains to be identified (50, 51). Euglycemic–hyperinsulinemic clamp studies in humans have demonstrated decreased insulin-mediated glucose disposal in individuals with hypothyroidism (52) that reverted with treatment (52, 53). Postulated mechanisms for this phenomenon include reduction in the blood flow to the skeletal muscle and adipose tissue as demonstrated in a study that measured the differences in the arteriovenous sampling in hypothyroid individuals compared with normal controls (54) or a reduction in glucose uptake by skeletal muscle and adipose tissue by reduction in expression of glucose transporter, GLUT4 (55). In a longitudinal study of health check-up program, Jun et al. have shown that the risk of incident Type 2 diabetes was significantly increased with each 1 mIU/mL increment in TSH after adjusting for confounding factors and that individuals in the highest tertile of TSH rise were at greater risk (56, 57). Insulin resistance has been known to modify the long-term cardiovascular complications of childhood obesity (58). The findings from our study suggest that the association between the higher TSH levels and insulin resistance begins in adolescence and may be the harbinger of the cardiometabolic risk with SH seen in adult life.
Circulating TSH is an important homeostatic hormone required for maintaining the body milieu and linear association of risk factors with TSH levels is unlikely. The association of cardiometabolic risks with the higher quantile may suggest that perhaps, the currently accepted upper range of normal levels of TSH requires modification. This issue has previously been debated in adult population with the report of mean TSH of 1.50 mIU/L (95% CI 1.46-1.54) in 13 344 people evaluated in NHANES III (59). In a subsequent study, it was reported that the prevalence of TPO antibody was higher in the individuals with TSH levels >2 mIU/L, especially in those of older age. Therefore, the tail of the distribution with the higher TSH levels may represent occult thyroid dysfunction and an accurate population reference range remains an open question (60, 61). The American Thyroid Association currently maintains that in children levothyroxine replacement therapy may be initiated with TSH >10 mIU/L with signs/symptoms consistent with thyroid disease. It is also noted that if treatment is initiated, the goal of therapy is to maintain the TSH levels in the lower half of reference range, optimally between 0.5 and 2.0 mIU/L (62). These guidelines leave the range between 2 and 10 mIU/L without specific guidance, often resulting in referral to pediatric endocrinology providers, with disproportionately higher referrals of children with obesity. The debate on whether or not children with SH with or without obesity should be treated continues (63). Whether the higher odds of cardiometabolic risk in the fourth quantile translates into therapeutic relevance can only be determined by longitudinal studies in future.
To provide guidance to the clinicians caring for children, we have estimated the percentiles for the thyroid hormones in youth between 12 and 18 years of age with BMI between the fifth and 85th percentiles without TPO or TG antibodies. The TSH levels decreased with increasing age from 12 to 18 years, while the total T4 levels increased and the free T4 levels remained stable across the ages. In a study of Danish children, similar trends were seen, although the absolute levels were higher perhaps reflecting differences in assay or differences between populations (64). The fourth quantile of TSH most closely aligns with the 75th percentile of TSH. Future studies should address the optimal cut-off for TSH levels.
There are several limitations of this study. First, we have combined the results from thyroid function assays over 5 NHANES cycles over 10 noncontiguous years. While the distribution of the measured hormonal levels overlaps in these 5 cycles, the possibility of differences across platforms remains. The NHANES survey does not include assessment for puberty. We used quantiles adjusted for age and sex to account for this lack of information, that may or may not capture the influence of puberty on thyroid function. The cardiometabolic risk factors were ascertained at the same time as the measurement of TSH, hence no inference can be drawn about causation. Finally, the study cohort was ascertained nearly a decade ago. NHANES has not undertaken thyroid function assessment after the 2011 to 2012 survey cycle. We observed no temporal differences in the measurements from 1999 to 2012, and hence any subsequent drifts are expected to be small.
Notwithstanding these limitations, we believe that this study provides an important reference for clinicians in practice, both for the reference ranges provided as well as the association of cardiometabolic risk factors with the highest quantile of TSH levels. Future longitudinal studies are needed to estimate if these cardiometabolic risk factors translate into adverse cardiometabolic outcomes and whether intervention with thyroid replacement therapy to maintain the TSH levels within a certain percentile is warranted.
Acknowledgments
The authors gratefully acknowledge the critical review of the manuscript by Bat-Sheva Levine, MD, MPH from Boston Children’s Hospital, Ilene Fennoy, MD, from Columbia University and Stavroula Osganian, MD, ScD, from National Institutes of Health (NIH). We are grateful to Qixuan Chen, PhD, from the Mailman School of Public Health at Columbia University for biostatistical consultation and the faculty and fellows from the Division of Pediatric Endocrinology at Columbia University Medical Center for their valuable input. We thank Morten Asp Vonsild Lund, MD, PhD, from the Department of Biomedical Sciences at University of Copenhagen for sharing their R code. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Financial Support: This publication was supported in part by the National Institute for Diabetes and Digestive and Kidney Diseases, National Institutes of Health (NIH), through Grant Number K23 DK110539 (V.V.T.) and the National Center for Advancing Translational Sciences, NIH, through Grant Number UL1TR001873.
Glossary
Abbreviations
- ALT
alanine aminotransferase
- BMI
body mass index
- HDL-C
high-density lipoprotein cholesterol
- HOMA-IR
homeostatic model for insulin resistance
- MetS
metabolic syndrome
- NHANES
National Health and Nutrition Examination Survey
- SH
subclinical hypothyroidism
- T3
triiodothyronine
- T4
thyroxine
- TG
thyroglobulin
- TSH
thyrotropin
- TPO
thyroidperoxidase
Additional Information
Disclosure Summary: The authors have nothing to disclose.
Data Availability
All data analyzed during this study are publicly available in NHANES data repository, the links to which are listed in References.
References
- 1. Brent GA. Mechanisms of thyroid hormone action. J Clin Invest. 2012;122(9):3035-3043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Duntas LH, Brenta G. The effect of thyroid disorders on lipid levels and metabolism. Med Clin North Am. 2012;96(2):269-281. [DOI] [PubMed] [Google Scholar]
- 3. Duntas LH. Thyroid disease and lipids. Thyroid. 2002;12(4):287-293. [DOI] [PubMed] [Google Scholar]
- 4. Gao N, Zhang W, Zhang YZ, Yang Q, Chen SH. Carotid intima-media thickness in patients with subclinical hypothyroidism: a meta-analysis. Atherosclerosis. 2013;227(1):18-25. [DOI] [PubMed] [Google Scholar]
- 5. Aziz M, Kandimalla Y, Machavarapu A, et al. Effect of thyroxin treatment on Carotid intima-media thickness (CIMT) reduction in patients with Subclinical Hypothyroidism (SCH): a meta-analysis of clinical trials. J Atheroscler Thromb. 2017;24(7):643-659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Collet TH, Bauer DC, Cappola AR, et al. ; Thyroid Studies Collaboration . Thyroid antibody status, subclinical hypothyroidism, and the risk of coronary heart disease: an individual participant data analysis. J Clin Endocrinol Metab. 2014;99(9):3353-3362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Gencer B, Collet TH, Virgini V, et al. ; Thyroid Studies Collaboration . 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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Chaker L, Baumgartner C, den Elzen WP, et al. ; Thyroid Studies Collaboration . Subclinical hypothyroidism and the risk of stroke events and fatal stroke: an individual participant data analysis. J Clin Endocrinol Metab. 2015;100(6):2181-2191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Mantovani A, Nascimbeni F, Lonardo A, et al. Association between primary hypothyroidism and nonalcoholic fatty liver disease: a systematic review and meta-analysis. Thyroid. 2018;28(10):1270-1284. [DOI] [PubMed] [Google Scholar]
- 10. Biondi B, Cappola AR, Cooper DS. Subclinical hypothyroidism: a review. JAMA. 2019;322(2):153-160. [DOI] [PubMed] [Google Scholar]
- 11. Taylor P, Bianco AC. Urgent need for further research in subclinical hypothyroidism. Nat Rev Endocrinol. 2019;15(9):503-504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Krause AJ, Cines B, Pogrebniak E, et al. Associations between adiposity and indicators of thyroid status in children and adolescents. Pediatr Obes. 2016;11(6):551-558. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Reinehr T. Thyroid function in the nutritionally obese child and adolescent. Curr Opin Pediatr. 2011;23(4):415-420. [DOI] [PubMed] [Google Scholar]
- 14. Giannakopoulos A, Lazopoulou N, Pervanidou P, Kanaka-Gantenbein C. The impact of adiposity and puberty on thyroid function in children and adolescents. Child Obes. 2019;15(6):411-415. [DOI] [PubMed] [Google Scholar]
- 15. Aypak C, Türedi O, Yüce A, Görpelioğlu S. Thyroid-stimulating hormone (TSH) level in nutritionally obese children and metabolic co-morbidity. J Pediatr Endocrinol Metab. 2013;26(7-8):703-708. [DOI] [PubMed] [Google Scholar]
- 16. Ayala-Moreno MR, Guerrero-Hernández J, Vergara-Castañeda A, Salazar-Aceves G, Cruz-Mercado DE. Thyroid function in pediatric population with different nutritional status. Bol Med Hosp Infant Mex. 2018;75(5):279-286. [DOI] [PubMed] [Google Scholar]
- 17. Dahl M, Ohrt JD, Fonvig CE, et al. Subclinical hypothyroidism in danish lean and obese children and adolescents. J Clin Res Pediatr Endocrinol. 2017;9(1):8-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Iqbal AM, Lteif AN, Kumar S. Association between mild hyperthyrotropinemia and hypercholesterolemia in children with severe obesity. J Pediatr Endocrinol Metab. 2019;32(6):561-568. [DOI] [PubMed] [Google Scholar]
- 19. Rumińska M, Witkowska-Sędek E, Majcher A, Brzewski M, Krawczyk M, Pyrżak B. Serum TSH level in obese children and its correlations with atherogenic lipid indicators and carotid intima media thickness. J Ultrason. 2018;18(75):296-301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Lundbäck V, Ekbom K, Hagman E, Dahlman I, Marcus C. Thyroid-stimulating hormone, degree of obesity, and metabolic risk markers in a cohort of Swedish children with obesity. Horm Res Paediatr. 2017;88(2):140-146. [DOI] [PubMed] [Google Scholar]
- 21. Ma CG, Shim YS. Association of thyroid-stimulating hormone and thyroid hormones with cardiometabolic risk factors in euthyroid children and adolescents aged 10-18 years: a population-based study. Sci Rep. 2019;9(1):15476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Özer S, Bütün İ, Sönmezgöz E, Yılmaz R, Demir O. Relationships among thyroid hormones and obesity severity, metabolic syndrome and its components in Turkish children with obesity. Nutr Hosp. 2015;32(2):645-651. [DOI] [PubMed] [Google Scholar]
- 23. Magge SN, Goodman E, Armstrong SC. The metabolic syndrome in children and adolescents: shifting the focus to cardiometabolic risk factor clustering. Pediatrics. 2017;140(2):e20171603. [DOI] [PubMed] [Google Scholar]
- 24. Centers for Disease Control and Prevention (CDC). National Health and Nutrition Examination Survey Data, U.S. Department of Health and Human Services, 2007–2011. Hyattsville, MD: National Center for Health Statistics. https://www.cdc.gov/nchs/nhanes/index.htm. Accessed January 30, 2020. [Google Scholar]
- 25. Barlow SE; Expert Committee . Expert committee recommendations regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: summary report. Pediatrics. 2007;120 Suppl 4:S164-S192. [DOI] [PubMed] [Google Scholar]
- 26. Chen X, Deng S, Sena C, Zhou C, Thaker VV. Data from: Relationship of TSH levels with cardiometabolic risk factors in US youth aged 12-18 years and population-based reference percentiles for thyroid function tests. Supplemental data.Columbia University Libraries 2020. ProMED-mail website. 10.7916/d8-qrm2-py07. Deposited September 21, 2020. [DOI] [Google Scholar]
- 27. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 2011–2012 data documentation, codebook and frequencies. THYROD_G. https://wwwn.cdc.gov/nchs/nhanes/2011-2012/THYROD_G.htm [Google Scholar]
- 28. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 2001–2002 data documentation, codebook and frequencies. SSNH4THY. https://wwwn.cdc.gov/Nchs/Nhanes/2001-2002/SSNH4THY.htm [Google Scholar]
- 29. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 2007–2008 data documentation, codebook and frequencies. THYROD_E. https://wwwn.cdc.gov/nchs/nhanes/2007-2008/THYROD_E.htm [Google Scholar]
- 30. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 2009–2010 data documentation, codebook and frequencies. THYROD_F. https://wwwn.cdc.gov/nchs/nhanes/2009-2010/THYROD_F.htm [Google Scholar]
- 31. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 2011–2012 data documentation, codebook and frequencies. GLU_G. https://wwwn.cdc.gov/nchs/nhanes/2011-2012/GLU_G.htm [Google Scholar]
- 32. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 2009–2010 data documentation, codebook and frequencies. GLU_F. https://wwwn.cdc.gov/nchs/nhanes/2009-2010/GLU_F.htm [Google Scholar]
- 33. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 2007–2008 data documentation, codebook and frequencies. GLU_E. https://wwwn.cdc.gov/nchs/nhanes/2007-2008/GLU_E.htm [Google Scholar]
- 34. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 2001–2002 data documentation, codebook and frequencies. L10_2_B. https://wwwn.cdc.gov/Nchs/Nhanes/2001-2002/L10_2_B.htm [Google Scholar]
- 35. Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 1999–2000 data documentation, codebook and frequencies. LAB10AM. https://wwwn.cdc.gov/nchs/nhanes/1999-2000/LAB10AM.htm [Google Scholar]
- 36. Zimmet P, Alberti KG, Kaufman F, et al. ; IDF Consensus Group . The metabolic syndrome in children and adolescents–an IDF consensus report. Pediatr Diabetes. 2007;8(5):299-306. [DOI] [PubMed] [Google Scholar]
- 37. Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents; Lung and Blood Institute National Heart. Expert panel on integrated guidelines for cardiovascular health and risk reduction in children and adolescents: summary report. Pediatrics. 2011;128(Suppl 5):S213-S256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Flynn JT, Kaelber DC, Baker-Smith CM, et al. Clinical practice guideline for screening and management of high blood pressure in children and adolescents. Pediatrics. 2017;140(3):e20171904. [DOI] [PubMed] [Google Scholar]
- 39. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412-419. [DOI] [PubMed] [Google Scholar]
- 40. Keskin M, Kurtoglu S, Kendirci M, Atabek ME, Yazici C. Homeostasis model assessment is more reliable than the fasting glucose/insulin ratio and quantitative insulin sensitivity check index for assessing insulin resistance among obese children and adolescents. Pediatrics. 2005;115(4):e500-e503. [DOI] [PubMed] [Google Scholar]
- 41. Vos MB, Abrams SH, Barlow SE, et al. NASPGHAN clinical practice guideline for the diagnosis and treatment of nonalcoholic fatty liver disease in children: recommendations from the expert committee on NAFLD (ECON) and the North American Society of Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN). J Pediatr Gastroenterol Nutr. 2017;64(2):319-334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Centers for Disease Control and Prevention (CDC). A SAS Program for the 2000 CDC Growth Charts (ages 0 to < 20 years). https://www.cdc.gov/nccdphp/dnpao/growthcharts/resources/sas.htm [Google Scholar]
- 43. Centers for Disease Control (CDC). National Health and Nutrition Examination Survey. Tutorials Module 3. Weighting. https://wwwn.cdc.gov/nchs/nhanes/tutorials/module3.aspx [Google Scholar]
- 44. Surks MI, Ortiz E, Daniels GH, et al. Subclinical thyroid disease: scientific review and guidelines for diagnosis and management. JAMA. 2004;291(2):228-238. [DOI] [PubMed] [Google Scholar]
- 45. Lee MK, Kim YM, Sohn SY, Lee JH, Won YJ, Kim SH. Evaluation of the relationship of subclinical hypothyroidism with metabolic syndrome and its components in adolescents: a population-based study. Endocrine. 2019;65(3):608-615. [DOI] [PubMed] [Google Scholar]
- 46. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet (London, England). 2004;363(9403):157-163. [DOI] [PubMed] [Google Scholar]
- 47. Pan WH, Flegal KM, Chang HY, Yeh WT, Yeh CJ, Lee WC. Body mass index and obesity-related metabolic disorders in Taiwanese and US whites and blacks: implications for definitions of overweight and obesity for Asians. Am J Clin Nutr. 2004;79(1):31-39. [DOI] [PubMed] [Google Scholar]
- 48. Razak F, Anand SS, Shannon H, et al. Defining obesity cut points in a multiethnic population. Circulation. 2007;115(16):2111-2118. [DOI] [PubMed] [Google Scholar]
- 49. Shai I, Jiang R, Manson JE, et al. Ethnicity, obesity, and risk of type 2 diabetes in women: a 20-year follow-up study. Diabetes Care. 2006;29(7):1585-1590. [DOI] [PubMed] [Google Scholar]
- 50. Brenta G. Why can insulin resistance be a natural consequence of thyroid dysfunction? J Thyroid Res. 2011;2011:152850. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Crunkhorn S, Patti ME. Links between thyroid hormone action, oxidative metabolism, and diabetes risk? Thyroid. 2008;18(2):227-237. [DOI] [PubMed] [Google Scholar]
- 52. Rochon C, Tauveron I, Dejax C, et al. Response of glucose disposal to hyperinsulinaemia in human hypothyroidism and hyperthyroidism. Clin Sci (Lond). 2003;104(1):7-15. [DOI] [PubMed] [Google Scholar]
- 53. Stanická S, Vondra K, Pelikánová T, Vlcek P, Hill M, Zamrazil V. Insulin sensitivity and counter-regulatory hormones in hypothyroidism and during thyroid hormone replacement therapy. Clin Chem Lab Med. 2005;43(7):715-720. [DOI] [PubMed] [Google Scholar]
- 54. Dimitriadis G, Mitrou P, Lambadiari V, et al. Insulin action in adipose tissue and muscle in hypothyroidism. J Clin Endocrinol Metab. 2006;91(12):4930-4937. [DOI] [PubMed] [Google Scholar]
- 55. Salvatore D, Simonides WS, Dentice M, Zavacki AM, Larsen PR. Thyroid hormones and skeletal muscle – new insights and potential implications. Nat Rev Endocrinol. 2014;10(4):206-214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Jun JE, Jee JH, Bae JC, et al. Association between changes in thyroid hormones and incident Type 2 diabetes: a seven-year longitudinal study. Thyroid. 2017;27(1):29-38. [DOI] [PubMed] [Google Scholar]
- 57. Jun JE, Jin SM, Jee JH, et al. TSH increment and the risk of incident type 2 diabetes mellitus in euthyroid subjects. Endocrine. 2017;55(3):944-953. [DOI] [PubMed] [Google Scholar]
- 58. Zhang H, Zhang T, Li S, et al. Long-term impact of childhood adiposity on adult metabolic syndrome is modified by insulin resistance: the Bogalusa Heart Study. Sci Rep. 2015;5:17885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. Hollowell JG, Staehling NW, Flanders WD, et al. Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab. 2002;87(2):489-499. [DOI] [PubMed] [Google Scholar]
- 60. Waise A, Price HC. The upper limit of the reference range for thyroid-stimulating hormone should not be confused with a cut-off to define subclinical hypothyroidism. Ann Clin Biochem. 2009;46(Pt 2):93-98. [DOI] [PubMed] [Google Scholar]
- 61. Wartofsky L, Dickey RA. The evidence for a narrower thyrotropin reference range is compelling. J Clin Endocrinol Metab. 2005;90(9):5483-5488. [DOI] [PubMed] [Google Scholar]
- 62. Jonklaas J, Bianco AC, Bauer AJ, et al. ; American Thyroid Association Task Force on Thyroid Hormone Replacement . Guidelines for the treatment of hypothyroidism: prepared by the American thyroid association task force on thyroid hormone replacement. Thyroid. 2014;24(12):1670-1751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63. Niranjan U, Wright NP. Should we treat subclinical hypothyroidism in obese children? BMJ. 2016;352:i941. [DOI] [PubMed] [Google Scholar]
- 64. Gunapalasingham G, Frithioff-Bøjsøe C, Lund MAV, et al. Reference values for fasting serum concentrations of thyroid-stimulating hormone and thyroid hormones in healthy Danish/North-European white children and adolescents. Scand J Clin Lab Invest. 2019;79(1-2):129-135. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data analyzed during this study are publicly available in NHANES data repository, the links to which are listed in References.