Association Between Long-Term Exposure to Environmental Fine Particulate Matter and the Prevalence of Thyroid Disorders: A National Cross-Sectional Study in China

Abstract

Background:

Exposure to particles with an aerodynamic diameter of ≤2.5 μm (PM_2.5) is associated with the occurrence of thyroid dysfunction among pregnant women and neonates, but it is not known if this association occurs in the general population. We aimed to determine the association of prolonged exposure to PM_2.5 with the prevalence of thyroid disorders among adults in China.

Methods:

A nationally representative cross-sectional study of thyroid disorders, iodine status, and diabetes status was carried out in all 31 provinces across China from 2015 to 2017. In total, 73,900 adults aged 18 years and older were included. Serum concentrations of thyroid hormones, thyrotropin, and thyroid antibodies and the urine iodine concentration were measured. The environmental concentration of PM_2.5 for each participant’s residential address at a spatial resolution of 1 × 1 km was estimated.

Results:

The average long-term exposure to PM_2.5 at residential addresses was 66.41 μg/m³, ranging from 17.58 μg/m³ to 120.40 μg/m³. Compared with that of individuals with lower exposure levels, the prevalence of thyroid diseases such as autoimmune thyroiditis and subclinical hypothyroidism was greater in those with PM_2.5 concentrations within the third quartile range (60.18 to 73.78 μg/m³). Compared with those in the first quartile (17.58 to 46.38 μg/m³), participants in the highest PM_2.5 quartile (73.78 to 120.40 μg/m³) presented an increased risk of overt hypothyroidism (OR 1.23 [CI 0.94–1.61]), subclinical hypothyroidism (1.10 [1.01–1.21]), autoimmune thyroiditis (1.09 [1.00–1.18]), and thyroglobulin antibody positivity (1.17 [1.07–1.29]). However, there was no association between PM_2.5 exposure and overt hyperthyroidism, subclinical hyperthyroidism, Graves’ disease, or thyroid peroxidase antibody positivity (p > 0.05). Each 10 μg/m³ increase in the PM_2.5 concentration was associated with an increased risk of overt hypothyroidism (OR 1.05 [1.00–1.11]), subclinical hypothyroidism (1.02 [1.00–1.03]), and thyroglobulin antibody positivity (1.02 [1.00–1.04]). Furthermore, a nearly linear exposure–response relationship was observed between long-term PM_2.5 exposure and thyroglobulin antibody positivity.

Conclusions:

PM_2.5 exposure was associated with thyroid disorders among Chinese adults. A dose–response relationship between PM_2.5 exposure and autoimmune thyroiditis, as well as thyroglobulin antibody positivity, was also observed.

Introduction

Environmental air pollution, specifically particulate matter (PM_2.5, particles with an aerodynamic diameter ≤2.5 μm), has been acknowledged as a pervasive global environmental and public health concern.¹ The World Health Organization (WHO) data show that almost all of the global population (99%) breathes air that exceeds the WHO’s annual mean PM_2.5 standard, which is set at 10 μg/m³, with low- and middle-income countries having the highest exposure.² Pollution is the most common environmental cause of disease and premature death worldwide.³ A Global Burden of Disease study estimated that air pollution contributed to 213.3 million healthy life years lost due to premature death and disability in 2019.⁴

Exposure to air pollution has been reported to be associated with a range of adverse health outcomes, including cardiovascular diseases and respiratory conditions.^3
–5 Thyroid disease, specifically subclinical hypothyroidism, is among the most prevalent endocrine disorders.⁶ Subclinical hypothyroidism has been reported to be associated with an increased risk of coronary events and mortality, as well as increased cancer-related mortality, diabetes, and other related conditions.⁷

While the relationship between iodine intake and thyroid disorders remains significant, previous surveys have indicated that even with adequate iodine intake, the prevalence of subclinical hypothyroidism among the Chinese population is 12.9%.⁸ Several studies have documented the detrimental impact of exposure to environmental air pollution on thyroid function.^9,10 Given the potential link between air pollution and maternal thyroid hormone deficiency, as well as impaired neurodevelopment in children, most related studies have focused on mothers or newborns, with less attention given to the general population.^11
–13 Previous research has been limited to specific geographic areas and lacks representativeness.¹² Furthermore, some studies have analyzed city-wide levels of pollution exposure rather than individual-level pollution exposure, which can lead to inaccurate estimates.¹⁴ Before 2013, the absence of a conventional air quality monitoring network in China hindered epidemiological studies on the chronic health effects of PM_2.5 exposure. This study utilized the Space-Time Extra-Trees (STET) model to obtain high-quality, high-resolution (1 × 1 km) data on PM_2.5 exposure levels.¹⁵

This study integrated pollutant data with a nationally representative cross-sectional study conducted across 31 provinces of China from 2015 to 2017. The aim of this study was to evaluate the relationship between long-term PM_2.5 exposure and the prevalence of thyroid diseases in the general population of China.

Methods

Sampling and study population

The Thyroid disorders, Iodine status, and Diabetes Epidemiological (TIDE) study, which was conducted from 2015 to 2017, was a population-based cross-sectional investigation. A comprehensive description of the study design has been previously published elsewhere.^8,16 A detailed flowchart outlining the study’s methodology is presented in Supplementary Figure S1. Briefly, 31 provincial regions in mainland China were considered. A nationally representative sample was obtained through a multistage stratified random sampling method based on demographic data, including age, sex distribution, urban–rural proportions, and 2010 national census data.¹⁷ In total, 80,937 Chinese adults (≥18 years old) were invited to participate in the survey, and 73,900 individuals were included in the final analysis. Because this study focused on the general adult population in China, the inclusion criteria for the study participants were as follows: (1) aged 18 years and older, (2) residing in the selected community for a minimum of 5 years, (3) not taking iodine-containing medications or contrast agents within the past three months, and (4) not pregnant. Individual information about each participant and the results of thyroid ultrasonography were obtained by trained interviewers who were proficient in data collection and qualified physicians who were trained by the same instructor, respectively. The questionnaire included demographic variables, behavioral factors, family history of chronic diseases, and personal medical history. Ultrasonography was performed using the same ultrasonic frequency instrument (LOGIQ 100 PRO, 7.5 MHz; GE Healthcare). Fasting blood and spot urine samples were collected from each participant. All samples were transported through a cold chain system to the central laboratory in Shenyang, China, for centralized testing. The levels of serum thyrotropin (TSH), thyroid peroxidase antibodies (TPOAbs), and thyroglobulin antibodies (TgAbs) were measured using an electrochemiluminescence immunoassay on a Cobas e601 analyzer (Roche Diagnostics) in a central laboratory in Shenyang. Serum-free thyroxine (fT4) and free triiodothyronine (fT3) levels were measured only if the level of TSH was outside the reference range. The urinary iodine concentration (UIC) was determined through inductively coupled plasma mass spectrometry (Agilent 7700x; Agilent Technologies).

The research protocol was approved by the Ethics Committee of the First Hospital of China Medical University. Following a comprehensive explanation of the research procedures, written informed consent was provided by all participants.

Air pollution exposure assessment

This study used the STET model to integrate remote sensing, meteorological, and land use data for estimating PM_2.5 environmental concentrations at a spatial resolution of 1 × 1 km. The detailed methodologies have been previously published elsewhere.¹⁸

On the basis of this model, we determined the monthly PM_2.5 exposure concentration for each participant’s residential address from 2010 to 2017 using longitudinal and latitudinal data. We subsequently derived the average monthly PM_2.5 concentrations for the past five years at the time of questionnaire completion.

Diagnostic criteria

The diagnostic criteria for thyroid disorders are listed in Table 1.

Table 1.

Diagnostic Criteria for Thyroid Disorders

Thyroid disorders	Diagnostic criteria
Overt hyperthyroidism	TSH <0.27 mIU/L; fT4 > 22 pmol/L; or fT3 > 6.8 pmol/L
Subclinical hyperthyroidism	TSH <0.27 mIU/L; fT4 and fT3 within the normal range (fT4 12.0–22.0 pmol/L; fT3 3.1–6.8 pmol/L)
Overt hypothyroidism	TSH >4.2 mIU/L; fT4 < 12 pmol/L
Subclinical hypothyroidism	TSH >4.2 mIU/L; fT4 within 12–22 pmol/L
Graves’ disease	Overt hyperthyroidism or subclinical hyperthyroidism; TRAb >1.75 IU/L or a diffuse goiter on B-mode ultrasonography
Autoimmune thyroiditis	TPOAb >34 IU/ml or TgAb >115 IU/ml
TPOAb positivity	TPOAb >34 IU/ml
TgAb positivity	TgAb >115 IU/ml

fT3, free triiodothyronine; fT4, free thyroxine; TgAb, thyroglobulin antibody; TPOAb, thyroid peroxidase antibody; TRAb, TSH receptor antibodies; TSH, thyrotropin.

Statistical analyses

We present participant characteristics as the means with corresponding confidence intervals (CIs) for continuous variables and percentages with CIs for categorical variables. The Kruskal–Wallis test was used to compare groups for continuous variables, whereas categorical variables were analyzed using χ² tests or Fisher’s exact tests. To estimate the overall population characteristics of Chinese adults in the general population aged 18 years and older in this study, computations were weighted using the 2010 Chinese Population Census data along with the sampling scheme of the present survey. The weighting coefficient represented the inverse of the adjusted probability for respondents to acquire the data; each case in the analysis was assigned a specific coefficient (individual weight) derived by multiplying the coefficient by the actual population sharing the same sex, age, province, and location characteristics.

We categorized the participants into four groups according to quartiles of exposure to PM_2.5 (17.58 to 46.38 μg/m³, 46.38 to 60.18 μg/m³, 60.18 to 73.78 μg/m³, and 73.78 to 120.40 μg/m³). We utilized generalized linear models to compute the adjusted odds ratios (ORs) and CIs for examining variations in the prevalence of thyroid disorders at different PM_2.5 exposure levels. Adjustments were made in the models for several potential confounding factors. In Model 1, we adjusted for age and sex. Model 2 was further adjusted for race/ethnicity, body mass index (BMI), UIC, education level, average annual household income, smoking history, family history of thyroid disorders, and urbanization. Based on Model 2, interaction terms were incorporated to estimate the potential effect of PM_2.5 exposure on thyroid disorders. The statistical significance of any interaction was assessed, prompting subsequent stratification of risk factors if deemed significant (Supplementary Tables S1–S5). The associations between PM_2.5 concentrations and thyroid disorders were evaluated on a continuous scale with restricted cubic spline curves on the basis of weighted generalized linear model analysis. This approach allowed for the exploration of potential nonlinear effects and facilitated the construction of an exposure–response curve. Multiple comparisons were also performed (Supplementary Table S7). In addition, the ORs for the risk of thyroid disorders per 10 μg/m³ increase in PM_2.5 exposure level were estimated.

A bilateral p value of <0.05 was considered statistically significant. All the statistical analyses were performed using R software, version 4.2.3.

Results

In the TIDE project, 87,900 individuals were invited to participate, and a total of 80,937 participants completed the survey, yielding an overall response rate of 92.1%. Information from 73,900 participants was utilized for the present analysis after excluding those lacking details on age and sex (n = 167), thyroid function testing (n = 2300), and specific residential addresses (n = 4570) (Fig. 1).

FIG. 1.

Flowchart of study participants included and excluded in analyses.

Characteristics of participants

Table 2 presents the background characteristics of the study population categorized by overall levels and quartiles of PM_2.5 exposure. Overall, 73,900 individuals were included, 52.4% of whom were male (n = 37,560). Overall, 46.8%, 66.0%, and 94.6% of the participants were aged younger than 40 years, had less than a high school education, and had no family history of thyroid disorders, respectively. The median UIC in the general adult population was 251.83 µg/L. Approximately 40.22% of the general adults had UICs ranging between 100 and 200 μg/L. From 2000 to 2017, participants were exposed to an average PM_2.5 concentration of 66.41 μg/m³ in their residential areas, ranging from 17.58 μg/m³ to 120.40 μg/m³. Compared with the group exposed to lower levels of PM_2.5, the group exposed to higher levels had a smaller proportion of smokers but a higher BMI and UIC.

Table 2.

General Characteristics of Adults Living in Mainland China and According to Quartiles of PM_2.5 ^a Concentrations

Characteristics	Total	Quartiles of PM_2.5 levels				P for difference
		First	Second	Third	Fourth
		[17.58, 46.38] µg/m³	[46.38, 60.18] µg/m³	[60.18, 73.78] µg/m³	[73.78, 120.40] µg/m³
No of participants	73,900	19,136	17,810	18,213	18,741
Mean PM2.5 level (µg/m3)	66.41 (66.20 to 66.61)	39.56 (39.45 to 39.67)	51.43 (51.37 to 51.50)	70.03 (69.95 to 70.10)	86.67 (86.49 to 86.85)	< 0.001
Male	52.40 (51.84 to 52.96)	50.97 (50.03 to 51.92)	53.87 (52.86 to 54.88)	52.74 (51.75 to 53.72)	52.14 (51.23 to 53.05)	0.01
Age at baseline (years)						< 0.001
<40	46.82 (46.08 to 47.57)	47.41 (46.37 to 48.46)	50.56 (49.49 to 51.63)	41.81 (40.72 to 42.89)	47.78 (46.80 to 48.77)
40–60	16.14 (15.80 to 16.49)	16.49 (15.84 to 17.14)	13.67 (13.07 to 14.28)	18.87 (18.09 to 19.65)	15.50 (14.89 to 16.12)
≥60	37.03 (36.54 to 37.52)	36.09 (35.19 to 37.00)	35.77 (34.80 to 36.74)	39.33 (38.33 to 40.32)	36.71 (35.82 to 37.61)
Race/Ethnicity						< 0.001
Han	95.14 (94.35 to 95.92)	87.79 (87.17 to 88.40)	92.73 (92.27 to 93.18)	96.06 (95.83 to 96.28)	99.81 (99.78 to 99.85)
Others	4.86 (4.70 to 5.02)	12.21 (11.60 to 12.83)	7.27 (6.82 to 7.73)	3.94 (3.72 to 4.17)	0.19 (0.15 to 0.22)
Body mass index (kg/m²)						< 0.001
<24	52.88 (52.19 to 53.58)	58.80 (57.77 to 59.82)	55.47 (54.42 to 56.52)	50.97 (49.91 to 52.02)	49.55 (48.58 to 50.52)
24–28	33.05 (32.54 to 33.57)	30.24 (29.37 to 31.12)	31.45 (30.49 to 32.41)	34.98 (33.99 to 35.97)	34.17 (33.28 to 35.06)
≥28	14.06 (13.67 to 14.45)	10.96 (10.08 to 11.85)	13.08 (12.41 to 13.75)	14.05 (13.34 to 14.76)	16.28 (15.60 to 16.96)
Urine iodine concentration (μg/L)						< 0.001
<100	17.76 (17.33 to 18.19)	22.64 (21.80 to 23.49)	20.66 (19.77 to 21.56)	13.42 (12.66 to 14.18)	16.41 (15.68 to 17.14)
100–200	40.22 (39.60 to 40.85)	41.61 (40.58 to 42.65)	45.42 (44.36 to 46.49)	37.55 (36.51 to 38.59)	38.41 (37.45 to 39.37)
200–300	23.21 (22.77 to 23.65)	22.33 (21.51 to 23.14)	21.72 (20.86 to 22.57)	24.67 (23.82 to 25.51)	23.54 (22.75 to 24.34)
≥300	18.80 (18.37 to 19.23)	13.42 (12.78 to 14.05)	12.20 (11.56 to 12.84)	24.36 (23.45 to 25.27)	21.64 (20.84 to 22.43)
Education						< 0.001
Less than high school	65.95 (65.26 to 66.64)	74.06 (73.18 to 74.94)	58.93 (57.85 to 60.02)	69.69 (68.75 to 70.63)	62.91 (61.98 to 63.85)
High school and above	34.05 (33.48 to 34.62)	25.94 (25.06 to 26.82)	41.07 (39.98 to 42.15)	30.31 (29.37 to 31.25)	37.09 (36.15 to 38.02)
Income per year (¥):						< 0.001
≤30,000	44.81 (44.17 to 45.46)	47.53 (46.50 to 48.55)	39.38 (38.35 to 40.41)	46.82 (45.77 to 47.87)	44.97 (44.01 to 45.93)
>30,000	55.19 (54.56 to 55.82)	52.47 (51.45 to 53.50)	60.62 (59.59 to 61.65)	53.18 (52.13 to 54.23)	55.03 (54.07 to 55.99)
Smoking history	27.30 (26.85 to 27.75)	28.09 (27.23 to 28.95)	27.50 (26.59 to 28.41)	27.83 (26.91 to 28.76)	26.41 (25.59 to 27.23)	0.02
Family history of thyroid disorders	5.43 (5.21 to 5.65)	5.67 (5.13 to 6.21)	7.85 (7.27 to 8.43)	4.72 (4.35 to 5.09)	4.45 (4.13 to 4.78)	< 0.001
Urban residence	51.34 (50.81 to 51.88)	34.82 (33.94 to 35.70)	63.20 (62.32 to 64.08)	45.05 (44.08 to 46.02)	57.99 (57.07 to 58.91)	< 0.001

Values are percentage (confidence interval) unless stated otherwise.

Particles of diameter ≤2.5 μm.

Associations between PM_2.5 and thyroid disorders

Table 3 presents the prevalence of thyroid disorders in individuals with different levels of PM_2.5 exposure. Overall, the weighted prevalence of autoimmune thyroiditis was 13.9% (CI 13.5–14.2) and that of subclinical hypothyroidism was 12.8% (12.5–13.2). Notably, the prevalence of TgAb positivity increased with increasing PM_2.5 exposure. Moreover, we observed that participants with a higher prevalence of thyroid disorders were primarily exposed to PM_2.5 concentrations in the third quartile.

Table 3.

Weighted Prevalence of Thyroid Disorders according to Quartiles of PM_2.5 ^a Concentrations

Thyroid disorders	Total	Quartiles of PM_2.5 levels				P for trend
		First	Second	Third	Fourth
		[17.58, 46.38] µg/m³	[46.38, 60.18] µg/m³	[60.18, 73.78] µg/m³	[73.78, 120.40] µg/m³
Overt hyperthyroidism	0.78 (0.69 to 0.87)	0.78 (0.61 to 0.94)	0.81 (0.62 to 1.01)	0.87 (0.67 to 1.07)	0.70 (0.54 to 0.85)	0.61
Subclinical hyperthyroidism	0.43 (0.36 to 0.50)	0.44 (0.32 to 0.55)	0.46 (0.32 to 0.60)	0.48 (0.30 to 0.66)	0.38 (0.28 to 0.48)	0.43
Overt hypothyroidism	0.97 (0.87 to 1.07)	0.91 (0.76 to 1.07)	0.76 (0.60 to 0.91)	1.21 (1.00 to 1.42)	0.96 (0.76 to 1.17)	0.17
Subclinical hypothyroidism	12.81 (12.46 to 13.17)	12.32 (11.65 to 13.00)	10.44 (9.85 to 11.04)	13.81 (13.12 to 14.50)	13.71 (13.03 to 14.39)	0.01
Graves’ disease	0.54 (0.46 to 0.61)	0.52 (0.39 to 0.64)	0.60 (0.43 to 0.77)	0.59 (0.43 to 0.75)	0.48 (0.35 to 0.61)	0.62
Autoimmune thyroiditis	13.87 (13.51 to 14.23)	13.31 (12.72 to 13.91)	13.56 (12.84 to 14.29)	14.38 (13.66 to 15.10)	14.00 (13.31 to 14.68)	0.73
Thyroid peroxidase antibodies positivity	9.99 (9.68 to 10.30)	9.91 (9.39 to 10.43)	10.04 (9.41 to 10.68)	10.38 (9.74 to 11.02)	9.74 (9.16 to 10.33)	0.23
Thyroglobulin antibodies positivity	9.44 (9.14 to 9.73)	8.62 (8.14 to 9.10)	9.02 (8.43 to 9.61)	9.64 (9.08 to 10.20)	9.98 (9.40 to 10.56)	0.04

Values are percentage (confidence interval) unless stated otherwise.

Particles of diameter ≤2.5 μm.

Table 4 presents the crude and adjusted ORs and CIs for the association of thyroid disorder risk with long-term exposure to PM2.5. We observed an association between higher levels of PM_2.5 (compared with first-quartile PM_2.5 levels) and an increased risk of subclinical hypothyroidism (in the third (OR 1.10 [CI 1.01–1.20]) and fourth (1.10 [1.01–1.21]) quartiles). The risk of overt hypothyroidism and autoimmune thyroiditis was found to be highest under the PM_2.5 concentrations in the third quartile, with a 40% (OR 1.40 [CI 1.10–1.78]) increase in the risk of overt hypothyroidism and a 12% (1.12 [1.03–1.21]) increase in the risk of autoimmune thyroiditis, as determined by Model 2 in the multivariate analysis. In addition, after Model 2 was implemented for variable adjustments, an increase of 10 μg/m³ in the PM_2.5 concentration was associated with an increased risk of overt hypothyroidism (1.05 [CI 1.00–1.11]) and subclinical hypothyroidism (1.01 [1.00–1.03]). We observed a significant association between higher levels of PM_2.5 exposure and an increased risk of TgAb positivity. Even after adjusting for multiple variables, this association persisted. The ORs of participants exposed to PM_2.5 concentrations in the second, third, and fourth quartiles were 1.05 (CI 0.95–1.15), 1.14 (1.04–1.25), and 1.17 (1.07–1.29), respectively. An increased odds of TgAb positivity was associated with a 10 mg/m³ increase in the PM_2.5 concentration (OR 1.02 [CI 1.00–1.04]). However, no associations were found between PM_2.5 exposure and the risk of overt hyperthyroidism, subclinical hyperthyroidism, Graves’ disease, or TPOAb positivity. Further analysis of the associations between autoimmune thyroiditis-related hypothyroidism and an increased risk of autoimmune overt hypothyroidism and exposure to PM_2.5 was conducted for the third (OR 1.82 [CI 1.32–2.49]) and fourth (1.71 [1.20–2.42]) quartiles. A 10 μg/m³ increase in the PM_2.5 concentration was significantly associated with an 11% increase in the prevalence of autoimmune overt hypothyroidism (1.11 [1.05–1.18]) after multivariate adjustment in Model 2 (Supplementary Tables 6).

Table 4.

Thyroid Disorders Weighted Prevalence Associated with Long Term Exposure to Fine Particulate Matter (PM_2.5)^*. Values are Odds Ratios (95% Confidence Intervals) Unless Stated Otherwise

Characteristics	Quartiles of PM_2.5 levels							P for trend	PM_2.5 per 10 µg/m³ increment	P value
	First	Second	P value	Third	P value	Fourth	P value
	[17.58, 46.38] µg/m³	(46.38, 60.18] µg/m³	P value	(60.18, 73.78] µg/m³	P value	(73.78, 120.40] µg/m³	P value
Overt hyperthyroidism
Sample size	19,136	17,810		18,213		18,741
No of cases	146	139		146		126
Crude model	1.00	41.05 (0.76 to 1.44)	0.79	1.12 (0.82 to 1.54)	0.47	0.89 (0.66 to 1.21)	0.47	0.45	0.99 (0.93 to 1.05)	0.70
Model 1	1.00	1.05 (0.76 to 1.45)	0.75	1.15 (0.84 to 1.57)	0.39	0.90 (0.66 to 1.22)	0.49	0.46	0.99 (0.93 to 1.05)	0.74
Model 2	1.00	1.10 (0.80 to 1.51)	0.57	1.22 (0.89 to 1.67)	0.21	0.99 (0.72 to 1.36)	0.96	0.96	1.01 (0.95 to 1.07)	0.76
Subclinical hyperthyroidism
Sample size	19,136	17,810		18,213		18,741
No of cases	94	75		82		74
Crude model	1.00	1.05 (0.70 to 1.57)	0.82	1.10 (0.69 to 1.75)	0.69	0.86 (0.59 to 1.25)	0.43	0.40	0.98 (0.92 to 1.05)	0.61
Model 1	1.00	1.09 (0.73 to 1.62)	0.68	1.10 (0.68 to 1.76)	0.71	0.87 (0.60 to 1.27)	0.47	0.40	0.98 (0.92 to 1.06)	0.63
Model 2	1.00	1.17 (0.78 to 1.74)	0.45	1.24 (0.77 to 1.98)	0.37	1.05 (0.71 to 1.55)	0.81	0.87	1.02 (0.94 to 1.10)	0.65
Overt hypothyroidism
Sample size	19,136	17,810		18,213		18,741
No of cases	259	179		213		193
Crude model	1.00	0.82 (0.63 to 1.07)	0.14	1.32 (1.03 to 1.68)	0.03	1.05 (0.80 to 1.38)	0.73	0.27	1.03 (0.98 to 1.08)	0.19
Model 1	1.00	0.88 (0.68 to 1.15)	0.36	1.29 (1.01 to 1.66)	0.04	1.08 (0.82 to 1.42)	0.58	0.27	1.03 (0.98 to 1.08)	0.18
Model 2	1.00	0.93 (0.71 to 1.21)	0.59	1.40 (1.10 to 1.78)	0.01	1.23 (0.94 to 1.61)	0.13	0.04	1.05 (1.00 to 1.11)	0.03
Subclinical hypothyroidism
Sample size	19,136	17,810		18,213		18,741
No of cases	2788	2457		2643		2532
Crude model	1.00	0.83 (0.76 to 0.91)	<0.001	1.14 (1.05 to 1.24)	0.003	1.13 (1.04 to 1.23)	0.004	<0.001	1.02 (1.00 to 1.03)	0.01
Model 1	1.00	0.86 (0.79 to 0.94)	0.001	1.13 (1.04 to 1.24)	0.004	1.15 (1.05 to 1.25)	0.002	<0.001	1.02 (1.01 to 1.04)	0.01
Model 2	1.00	0.86 (0.78 to 0.94)	0.001	1.10 (1.01 to 1.20)	0.03	1.10 (1.01 to 1.21)	0.04	<0.001	1.01 (1.00 to 1.03)	0.15
Graves' disease
Sample size	19,136	17,810		18,213		18,741
No of cases	107	89		97		89
Crude model	1.00	1.16 (0.80 to 1.69)	0.44	1.15 (0.80 to 1.66)	0.45	0.93 (0.64 to 1.34)	0.70	0.54	0.99 (0.92 to 1.06)	0.79
Model 1	1.00	1.17 (0.80 to 1.71)	0.41	1.17 (0.81 to 1.69)	0.40	0.93 (0.65 to 1.35)	0.72	0.56	0.99 (0.92 to 1.06)	0.82
Model 2	1.00	1.26 (0.88 to 1.82)	0.21	1.32 (0.91 to 1.92)	0.15	1.12 (0.76 to 1.64)	0.56	0.69	1.03 (0.95 to 1.11)	0.50
Autoimmune thyroiditis
Sample size	19,136	17,810		18,213		18,741
No of cases	2787	2445		2609		2677
Crude model	1.00	1.02 (0.94 to 1.11)	0.60	1.09 (1.01 to 1.18)	0.03	1.06 (0.98 to 1.14)	0.14	0.10	1.01 (0.99 to 1.02)	0.44
Model 1	1.00	1.07 (0.99 to 1.16)	0.11	1.10 (1.02 to 1.19)	0.02	1.08 (1.00 to 1.17)	0.05	0.08	1.01 (0.99 to 1.02)	0.32
Model 2	1.00	1.04 (0.95 to 1.13)	0.39	1.12 (1.03 to 1.21)	0.01	1.09 (1.00 to 1.18)	0.04	0.03	1.01 (1.00 to 1.03)	0.15
Thyroid peroxidase antibodies positivity
Sample size	19,136	17,810		18,213		18,741
No of cases	2071	1789		1819		1880
Crude model	1.00	1.01 (0.93 to 1.11)	0.76	1.05 (0.96 to 1.15)	0.26	0.98 (0.90 to 1.07)	0.68	0.69	0.99 (0.98 to 1.01)	0.44
Model 1	1.00	1.06 (0.96 to 1.16)	0.24	1.06 (0.96 to 1.16)	0.24	1.00 (0.91 to 1.09)	0.94	0.72	0.99 (0.98 to 1.01)	0.52
Model 2	1.00	1.03 (0.94 to 1.13)	0.55	1.08 (0.98 to 1.18)	0.11	1.02 (0.93 to 1.12)	0.69	0.68	1.00 (0.98 to 1.02)	0.96
Thyroglobulin antibodies positivity
Sample size	19,136	17,810		18,213		18,741
No of cases	1812	1648		1830		1939
Crude model	1.00	1.05 (0.96 to 1.16)	0.30	1.13 (1.03 to 1.23)	0.01	1.17 (1.07 to 1.28)	<0.001	<0.001	1.02 (1.00 to 1.04)	0.01
Model 1	1.00	1.10 (1.00 to 1.21)	0.05	1.15 (1.05 to 1.26)	0.003	1.20 (1.10 to 1.31)	<0.001	<0.001	1.02 (1.01 to 1.04)	0.01
Model 2	1.00	1.05 (0.95 to 1.15)	0.34	1.14 (1.04 to 1.25)	0.005	1.17 (1.07 to 1.29)	<0.001	<0.001	1.02 (1.00 to 1.04)	0.01

Particles of diameter ≤ 2.5 μm. Crude model: unadjusted model. Model 1: adjusted for age and sex. Model 2: Model 1 + adjusted for ethnicity, body mass index, urine iodine concentration, education, average annual household income, smoking history, family history of thyroid disorders and urbanization.

Fig. 2 shows the exposure–response curve depicting the association between thyroid disorders and PM_2.5 levels. The analysis revealed that PM_2.5 concentrations in the third quartile range (60.18 to 73.78 μg/m³) were associated with the highest risks of overt hypothyroidism, subclinical hypothyroidism, and elevated TSH levels. However, across the entire range of PM_2.5 concentrations, we observed a linear concentration–response correlation between the PM_2.5 concentration and both autoimmune thyroiditis and TgAb positivity.

FIG. 2.

Multivariable adjusted odds ratios for all thyroid disorders according to levels of PM_2.5 on a continuous scale. Note: Solid purple lines are multivariable-adjusted odds ratios, with shaded areas lines showing confidence intervals derived from restricted cubic spline regressions with three knots. Analyses were adjusted for age, sex, race/ethnicity, body mass index, urine iodine concentration, education, average annual household income, smoking history, family history of thyroid disorders, and residential addresses. ORs, odds ratios; PM_2.5, particles of diameter ≤2.5 μm; TgAb, thyroglobulin antibody; TPOAb, thyroid peroxidase antibody; TSH, thyrotropin.

Subgroup and sensitivity analyses

The Supplement presents subgroup analyses, sensitivity analyses, and multiple comparisons related to fluctuations in thyroid disease prevalence associated with PM_2.5 levels. None of the sensitivity analyses yielded significant deviations from our primary model outcomes (Supplementary Tables S1–S5). In both our primary analysis (Table 4) and sensitivity analyses (Supplementary Tables S1–S5), there was no evidence supporting an association between subclinical hyperthyroidism and exposure to PM_2.5. The stratification results for the risk factors are detailed in the appendix (Supplementary Tables S1–S5).

Discussion

To our knowledge, this study is the first to estimate the relationship between long-term exposure to PM_2.5 at a spatial resolution of 1 × 1 km and the prevalence of thyroid disorders in the general adult population. Our findings reveal the associations between prolonged exposure to elevated levels of PM_2.5 and increased risks of overt hypothyroidism, subclinical hypothyroidism, and autoimmune thyroiditis. Even within subgroup and sensitivity analyses, these associations remained robust.

In the present study, the average concentration of PM_2.5 was 66.41 μg/m³, nearly seven times greater than that recommended by the WHO air quality guideline (10 μg/m³).² The analysis revealed a 5% increase in the risk of overt hypothyroidism with every 10 μg/m³ increase in the PM_2.5 concentration. Consequently, compared with the PM_2.5 level recommended in the WHO standard, the risk among the Chinese general adult population exposed to the average PM_2.5 concentration is estimated to increase by more than 30%.

This study underscores the associations between long-term exposure to PM_2.5 and increased risks of overt hypothyroidism and subclinical hypothyroidism among adults in the general population. This finding is consistent with the findings of previous studies conducted on pregnant women and neonates.^10,13,19,20 Hypothyroidism has been the most extensively studied pollutant-associated thyroid disorder in recent years. A previous study revealed that preconception exposure to PM_2.5 increased the risk of thyroid dysfunction during pregnancy.²⁰ Zhao et al. demonstrated that during the first and second trimesters, there was a 28% increase (OR 1.28 [CI 1.05–1.57]) and 23% increase (1.23 [1.00–1.51]) in the risk of maternal hypothyroidism for each increase of 10 μg/m³ in the PM_2.5 concentration, respectively.¹² In addition, a European cohort study yielded similar conclusions regarding exposure to PM_2.5 in early pregnancy and mild thyroid dysfunction throughout pregnancy.¹⁰ PM_2.5 exposure affects pregnant mothers and can also impact the thyroid function of subsequent generations. A study based on a Chinese national database revealed a positive association between maternal exposure to PM_2.5 and an increased risk of congenital hypothyroidism in offspring.¹³ The present study differed from previous studies that focused on thyroid function in pregnant women and fetuses; instead, we studied the general adult population, expanding the scope of PM_2.5 exposure within our research. A Mendelian randomization study provided evidence for the causal relationship between PM_2.5 exposure and the risk of thyroid dysfunction. The study revealed an association between elevated concentrations of PM_2.5 and an increased risk of thyroid dysfunction. Hence, air pollution control may be important for preventing thyroid dysfunction.²¹

Previous studies have rarely reported the associations between air pollution exposure and autoimmune thyroid diseases. A study from Bosnia and Herzegovina failed to demonstrate the significant effects of air pollution on the incidence of autoimmune thyroiditis.²² However, our findings indicated that a greater risk of autoimmune thyroiditis was associated with long-term exposure to high levels of PM_2.5. These discrepancies in the conclusions might be attributed to varying pollution levels, differing methods of pollution estimation, and variations in exposure duration among studies, which led to differences in the levels of exposure among the participants involved in the analysis. Moreover, there was a substantial difference in the sample sizes analyzed; this study had a smaller sample size (n = 174), which could have limited its statistical power and the interpretability of the results.

Previous findings suggested that pregnant women who were TPOAb positive were more susceptible to the effects of air pollution, yet TPOAb status did not influence the relationship between air pollution exposure and fT4 levels.¹² Similarly, the present study did not find a statistically significant difference between PM_2.5 exposure and TPOAb positivity. However, we observed a linear relationship between long-term exposure to PM_2.5 and TgAb positivity. TgAb levels are important for the adjunctive diagnosis and treatment of autoimmune thyroiditis.²³ However, there are no reported studies on the relationship between PM_2.5 exposure and TgAb positivity.

Currently, conclusive evidence linking air pollution exposure to thyroid disruption is lacking. Animal experiments suggest that exposure to PM_2.5 may disrupt thyroid hormone homeostasis in female rats by affecting the negative feedback regulation of thyroid secretion controlled by the hypothalamic–pituitary–thyroid axis.²⁴ Perturbations in the gut–thyroid axis could also be one of the causes of thyroid toxicity induced by PM_2.5 exposure.²⁵ PM_2.5-induced thyroid toxicity leads to alterations in the gut microbiota and metabolites in rats, which can in turn influence thyroid function. Furthermore, compounds within air pollutants might disrupt estrogen receptor signaling.²⁶

Exposure to PM_2.5 has been shown to impact human immune function,²⁷ and these effects are also evident in the thyroid. This study revealed a significant association between long-term exposure to PM_2.5 and increased susceptibility to autoimmune overt hypothyroidism and autoimmune subclinical hypothyroidism. This association was notably stronger than that for overt and subclinical hypothyroidism (Supplementary Table S6). These findings reaffirmed that chronic inflammation due to prolonged exposure to PM_2.5 disrupts the balance of the immune system, consequently increasing the risk of immune-related hypothyroidism.

The toxicity of pollutants may directly cause tissue damage or induce a cascade of pathophysiological effects through oxidative stress²⁸ and the activation of pro-inflammatory signals.²⁹ Individuals with preexisting conditions are more susceptible to the impact of particulate air pollution;^30,31 however, the detailed underlying mechanisms involved remain incompletely understood and warrant further investigation.

Strengths and limitations of this study

Compared with previous studies, this study has several advantages. First, this was a large-scale cross-sectional study based on a population comprising 31 provinces across China, with PM_2.5 levels ranging from 17.58 μg/m³ to 120.40 μg/m³, thereby ensuring nationwide representativeness. Second, the assessment of PM_2.5 exposure levels is important. In contrast to satellite-derived data, which possess temporal constraints and relatively coarse spatial resolutions (3–10 km), our use of the STET model yielded more accurate PM_2.5 data (1 × 1 km). This approach also enables more precise analyses of urban areas and the history of pollution. Third, our study included six thyroid disorders and two thyroid antibody positivity outcomes while simultaneously adjusting for various potential risk factors, including the UIC. This comprehensive approach allowed for a more holistic evaluation of the association between the prevalence of thyroid disorders and PM_2.5 exposure.

Moreover, there are limitations to this study. First, our research had a cross-sectional study design, thereby restricting causal inferences due to the constraints of an observational study design, necessitating prospective cohort studies for validation. Second, our statistical analyses did not consider other air pollutants, heavy metals, environmental noise, wildfires, or similar contaminants, limiting our access to higher quality exposure data. Furthermore, we were unable to assess the interactions among pollutants. Third, we collected environmental PM_2.5 exposure data based solely on participants’ residential addresses, without accounting for their daily activity durations, occupational exposure periods, specific locations, or indoor air pollution exposure. These factors might have led to an overestimation or underestimation of participant exposure levels.

Conclusion

In summary, this study revealed an association between long-term exposure to relatively high concentrations of PM_2.5 and an increased risk of thyroid disorders. However, further prospective cohort studies are needed to corroborate these findings, assess effect modifications and interactions, and examine the potential biological mechanisms involved. This research is crucial to guide future longitudinal research, environmental monitoring, and health policy development in heavily polluted low- and middle-income nations, such as China. The potential effect of air pollution exposure on thyroid disorders requires further attention.

Footnotes

Acknowledgments

The authors thank the participants of this study. For continuous support, assistance, and cooperation, the authors thank the investigators for the Thyroid Disorders, Iodine Status and Diabetes Epidemiological Survey Group.

Authors’ Contributions

Y.L., Z.S., and W.T. are joint corresponding authors. Y.L., Z.S., and W.T. conceived and designed the study. K.Y., K.C., Y.L., Z.S., and W.T. supervised the study. K.Y. and C.L. performed the statistical analysis. The Thyroid Disorders, Iodine Status and Diabetes Epidemiological Survey Group conducted the epidemiological survey. All the authors contributed to the acquisition, analysis, or interpretation of the data. K.Y. and Y.L. drafted the article. All authors revised the report and approved the final version before submission. Y.L., Z.S., and W.T. are the guarantors and attest that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

Data Sharing

All the data will be made available upon request to the corresponding author. Proposals will be reviewed and approved by the sponsor, investigator, and collaborators on the basis of scientific merit. After approval of a proposal, data will be shared through a secure online platform after the signing of a data access agreement.

Author Disclosure Statement

All authors declare no competing interests.

Funding Information

This work was supported by the Clinical Research Fund of the Chinese Medical Association (Grant No. 15010010589), the Research Fund for Public Welfare from the National Health and Family Planning Commission of China (Grant No. 201402005), the National Natural Science Foundation of China (Grant No. 82470826), and the China Postdoctoral Science Foundation (Grant No. 2021MD703910 and 2023T160724). The funder of the study had no role in the study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Supplementary Material

Supplementary Data S1

Supplementary Figure S1

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4

Supplementary Table S5

Supplementary Table S6

Supplementary Table S7

References

Shaddick

, Thomas

, Amini

, et al. Data integration for the assessment of population exposure to ambient air pollution for global burden of disease assessment. Environ Sci Technol, 2018; 52(16):9069–9078; doi: 10.1021/acs.est.8b02864

World Health Organization. Ambient air pollution. Available from: https://www.who.int/health-topics/air-pollution#tab=tab_1

Landrigan

, Fuller

, Acosta

NJR

, et al. The lancet commission on pollution and health. Lancet, 2018; 391(10119):462–512; doi: 10.1016/s0140-6736(17)32345-0

The global burden of cancer attributable to risk factors, 2010-19: A systematic analysis for the global burden of disease study 2019. Lancet, 2022; 400(10352):563–591; doi: 10.1016/s0140-6736(22)01438-6

, Xu

, Shan

, et al. Association between air pollution and type 2 diabetes: An updated review of the literature. Ther Adv Endocrinol Metab, 2019; 10:2042018819897046; doi: 10.1177/2042018819897046

Taylor

, Albrecht

, Scholz

, et al. Global epidemiology of hyperthyroidism and hypothyroidism. Nat Rev Endocrinol, 2018; 14(5):301–316; doi: 10.1038/nrendo.2018.18

Chaker

, Bianco

, Jonklaas

, et al. Hypothyroidism. Lancet, 2017; 390(10101):1550–1562; doi: 10.1016/s0140-6736(17)30703-1

, Teng

, Ba

, et al. Efficacy and safety of long-term universal salt iodization on thyroid disorders: Epidemiological evidence from 31 provinces of Mainland China. Thyroid, 2020; 30(4):568–579; doi: 10.1089/thy.2019.0067

Wang

, Liu

, Zhang

, et al. Evaluation of maternal exposure to PM(2.5) and Its components on maternal and neonatal thyroid function and birth weight: A cohort study. Thyroid, 2019; 29(8):1147–1157; doi: 10.1089/thy.2018.0780

10.

Ghassabian

, Pierotti

, Basterrechea

, et al. Association of exposure to ambient air pollution with thyroid function during pregnancy. JAMA Netw Open, 2019; 2(10):e1912902; doi: 10.1001/jamanetworkopen.2019.12902

11.

Kuehn

. Air pollution’s effects on newborns’ thyroid hormones. JAMA, 2021; 326(7):592; doi: 10.1001/jama.2021.13123

12.

Zhao

, Cao

, Li

, et al. Air pollution exposure in association with maternal thyroid function during early pregnancy. J Hazard Mater, 2019; 367:188–193; doi: 10.1016/j.jhazmat.2018.12.078

13.

Shang

, Huang

, Yang

, et al. Maternal exposure to PM(2.5) may increase the risk of congenital hypothyroidism in the offspring: A national database based study in China. BMC Public Health, 2019; 19(1):1412; doi: 10.1186/s12889-019-7790-1

14.

Zhang

, Wang

, Qin

, et al. Six air pollutants associated with increased risk of thyroid nodules: A study of 4.9 million Chinese adults. Front Endocrinol (Lausanne), 2021; 12:753607; doi: 10.3389/fendo.2021.753607

15.

Wei

, Li

, Cribb

, et al. Improved 1 km resolution PM2.5 estimates across China using enhanced space–time extremely randomized trees. Atmos Chem Phys, 2020; 20(6):3273–3289; doi: 10.5194/acp-20-3273-2020

16.

, Teng

, Shi

, et al. Prevalence of diabetes recorded in mainland China using 2018 diagnostic criteria from the American diabetes association: National cross sectional study. Bmj, 2020; 369:m997; doi: 10.1136/bmj.m997

17.

National Bureau of Statistics of China. Tabulation on the 2010 population census of the People’s Republic of China. 2010. Available from: http://www.stats.gov.cn/sj/zxfb/202303/t20230301_1919254.html

18.

Wei

, Li

, Lyapustin

, et al. Reconstructing 1-km-resolution high-quality PM2.5 data records from 2000 to 2018 in China: Spatiotemporal variations and policy implications. Remote Sensing of Environment, 2021; 252:112136; doi: 10.1016/j.rse.2020.112136

19.

Harari-Kremer

, Calderon-Margalit

, Korevaar

TIM

, et al. Associations between prenatal exposure to air pollution and congenital hypothyroidism. Am J Epidemiol, 2021; 190(12):2630–2638; doi: 10.1093/aje/kwab187

20.

Sun

, Chen

, Ye

, et al. Association of hypothyroidism during pregnancy with preconception and early pregnancy exposure to ambient particulate matter. Environ Sci Pollut Res Int, 2023; 30(37):88084–88094; doi: 10.1007/s11356-023-28683-7

21.

Zhang

, Liu

, Wang

, et al. Causal relationship between particulate matter 2.5 and hypothyroidism: A two-sample Mendelian randomization study. Front Public Health, 2022; 10:1000103; doi: 10.3389/fpubh.2022.1000103

22.

Izic

, Husejnovic

, Caluk

, et al. Urban air pollution associated with the incidence of autoimmune thyroid diseases. Med Arch, 2022; 76(2):115–121; doi: 10.5455/medarh.2022.76.115-121

23.

Constantinescu

, Hospel

, Daumerie

, et al. Significance of thyroperoxydase and thyroglobulin antibodies in medically treated Graves’ disease. Eur Thyroid J, 2023; 12(6); doi: 10.1530/etj-23-0193

24.

Dong

, Wu

, Yao

, et al. PM(2.5) disrupts thyroid hormone homeostasis through activation of the Hypothalamic-Pituitary-Thyroid (HPT) axis and induction of hepatic transthyretin in female rats 2.5. Ecotoxicol Environ Saf, 2021; 208:111720; doi: 10.1016/j.ecoenv.2020.111720

25.

Dong

, Yao

, Deng

, et al. Alterations in the gut microbiota and its metabolic profile of PM(2.5) exposure-induced thyroid dysfunction rats. Sci Total Environ, 2022; 838(Pt 3):156402; doi: 10.1016/j.scitotenv.2022.156402

26.

Klein

, Hodge

, Diamond

, et al. Gas-phase ambient air contaminants exhibit significant dioxin-like and estrogen-like activity in vitro . Environ Health Perspect, 2006; 114(5):697–703; doi: 10.1289/ehp.8496

27.

Johnson

, Hoffmann

, Behlen

, et al. Air pollution and children’s health-a review of adverse effects associated with prenatal exposure from fine to ultrafine particulate matter. Environ Health Prev Med, 2021; 26(1):72; doi: 10.1186/s12199-021-00995-5

28.

Ferrari

, Fallahi

, Elia

, et al. Thyroid autoimmune disorders and cancer. Semin Cancer Biol, 2020; 64:135–146; doi: 10.1016/j.semcancer.2019.05.019

29.

Schraufnagel

, Balmes

, Cowl

, et al. Air pollution and noncommunicable diseases: A review by the forum of international respiratory societies’ environmental committee, part 1: The damaging effects of air pollution. Chest, 2019; 155(2):409–416; doi: 10.1016/j.chest.2018.10.042

30.

Kasai

. Chemistry-based studies on oxidative DNA damage: Formation, repair, and mutagenesis. Free Radic Biol Med, 2002; 33(4):450–456.

31.

, Dorans

, Wilker

, et al. Residential proximity to major roadways, fine particulate matter, and adiposity: The framingham heart study. Obesity (Silver Spring), 2016; 24(12):2593–2599; doi: 10.1002/oby.21630

Association Between Long-Term Exposure to Environmental Fine Particulate Matter and the Prevalence of Thyroid Disorders: A National Cross-Sectional Study in China

Abstract

Background:

Methods:

Results:

Conclusions:

Introduction

Methods

Sampling and study population

Air pollution exposure assessment

Diagnostic criteria

Statistical analyses

Results

Characteristics of participants

Associations between PM2.5 and thyroid disorders

Subgroup and sensitivity analyses

Discussion

Strengths and limitations of this study

Conclusion

Footnotes

Acknowledgments

Authors’ Contributions

Data Sharing

Author Disclosure Statement

Funding Information

Supplementary Material

References

Associations between PM_2.5 and thyroid disorders