Association of Exposure to Ambient Air Pollution With Thyroid Function During Pregnancy

Key Points Question Is exposure to ambient air pollution in the first trimester associated with thyroid function throughout pregnancy? Findings Among 9931 pregnant women in 4 European cohorts and 1 US cohort, an increase of 5 μg/m3 in exposure to particulate matter with an aerodynamic diameter of 2.5 μm or less was associated with 20% higher odds of hypothyroxinemia. Meaning The findings of this study raise the possibility that exposure to particulate matter might disrupt thyroid function in pregnant women.


Rhea
Thyroid hormones and thyroid peroxidase antibody: IMMULITE 2000 immunoassay system (Siemens Healthcare Diagnostics, ILL 60015-0778, USA) (TPOAb positive if ≥35 IU/mL). Interassay and intra-assay coefficients of variation <12.5% and < 12.5% for TSH. Interassay and intra-assay coefficients of variation <7.1% and <7.8% for free T4. Interassay and intra-assay coefficients of variation <7.2% and 7.4% for TPOAb. Variables included in the imputation procedure for ABCD: Thyroid stimulating hormone (TSH) in pregnancy (mIU/l), Free thyroxine (T4) in pregnancy (μg/dl), particulate matter (PM)2.5 absorbance -averaged in the first trimester of pregnancy, PM2.5-10 -averaged in the first trimester of pregnancy, NO2 averaged in the first trimester of pregnancy, pregnant women's characteristics (education levels, country of birth, women´s age at enrollment, gestational age at thyroid measurement, history of smoking, alcohol intake during pregnancy, social class, marital status, parity, prepregnancy body mass index, and psychiatric symptoms.
Variables included in the imputation procedure for Generation R: TSH in pregnancy (mIU/l), free T4 in pregnancy (μg/dl), PM2.5 absorbance -averaged in the first trimester of pregnancy, PM2.5-10 -averaged in the first trimester of pregnancy, NO2 averaged in the first trimester of pregnancy, pregnant women's characteristics (education level, country of birth of parents, women´s age at enrollment, gestational age at thyroid measurement, history of smoking, alcohol intake during pregnancy, social class, marital status, parity, secondhand smoking, IQ, pre-pregnancy body mass index, and psychiatric symptoms, urinary iodine concentrations during pregnancy, urinary creatinine concentrations during pregnancy, and gestational age at iodine assessment.

Variables included in the imputation procedure for INMA:
TSH in pregnancy (mIU/l), free T4 in pregnancy (μg/dl), PM2.5 absorbance -averaged in the first trimester of pregnancy, PM2.5-10 -averaged in the first trimester of pregnancy, NO2 averaged in the first trimester of pregnancy, maternal characteristics (education levels, country of birth, women´s age at enrollment, gestational age at thyroid measurement, history of smoking, alcohol intake during pregnancy, social class, marital status, parity, secondhand smoking, IQ, pre-pregnancy body mass index, psychopathological symptoms, urinary iodine concentrations during pregnancy, urinary creatinine concentrations during pregnancy, iodine from diet (μg/day) and gestational age at iodine assessment (data on urinary iodine concentrations and gestational age at assessment were not available in the region of Asturias; data on PM was only available in Sabadell).
Variables included in the imputation procedure for Rhea: TSH in pregnancy (mIU/l), free T4 in pregnancy (μg/dl), PM2.5 absorbance -averaged in the first trimester of pregnancy, PM2.5-10 -averaged in the first trimester of pregnancy, pregnant women's characteristics (education levels, country of birth, women´s age at enrollment, gestational age at thyroid measurement, history of smoking, alcohol intake during pregnancy, social class, marital status, parity, secondhand smoking, IQ, pre-pregnancy body mass index, psychopathological symptoms, urinary iodine concentrations during pregnancy, urinary creatinine concentrations during pregnancy, iodine from diet (μg/day) and gestational age at iodine assessment Variables included in the imputation procedure for Project Viva: TSH in pregnancy (mIU/l), free T4 index in pregnancy, PM2.5 -averaged in the first trimester of pregnancy, Black carbon -averaged in the first trimester, NO2 -averaged in the second trimester, pregnant women's characteristics (education levels, country of birth, women´s age at enrollment, gestational age at thyroid measurement, history of smoking, alcohol intake during pregnancy, social class, marital status, parity, secondhand smoking, IQ, pre-pregnancy body mass index, psychopathological symptoms, urinary iodine concentrations during pregnancy, urinary creatinine concentrations during pregnancy, iodine levels from diet (μg/day), gestational age at iodine assessment Treatment of binary/categorical variables: logistic and multinomial models Statistical interactions included in imputation models: none Odds ratios (95%CI) were estimated using random-effects meta-analysis by cohort (ABCD, Generation R, INMA, Rhea, and Project Viva).
We defined hypothyroxinemia as free T 4 below the 5 th percentile of cohort distribution despite normal TSH. Higher TSH was defined as values > 95 th percentile. I 2 refers to the percentage of the total variability due to between-areas heterogeneity. P value of heterogeneity was estimated using the Cochran's Q test. Odds ratios (95%CI) were estimated using random-effects meta-analysis by cohort (ABCD, Generation R, INMA, Rhea, and Project Viva).
We defined hypothyroxinemia as free T 4 below the 5 th percentile of cohort distribution despite normal TSH. Higher TSH was defined as values > 95 th percentile. I 2 refers to the percentage of the total variability due to between-areas heterogeneity. P value of heterogeneity was estimated using the Cochran's Q test.
Models were adjusted for pregnant women's age at enrollment, educational level, country of birth, gestational age at thyroid measurement, history of smoking, alcohol intake during pregnancy, socioeconomic status, marital status, parity, and pre-pregnancy body mass index. In addition, analysis in INMA was adjusted for regions (Sabadell, Gipuzkoa, Valencia, and Asturias Odds ratios (95%CI) are shown for the associations between PM 2.5-10 exposure and hypothyroxinemia (a), PM 2.5 absorbance exposure and hypothyroxinemia (b), PM 2.5-10 exposure and high TSH (c), and PM 2.5 absorbance exospore and high TSH (d), estimated using random-effects metaanalysis by cohort (ABCD, Generation R, INMA, Rhea, and Project Viva).
We defined hypothyroxinemia as free T 4 below the 5 th percentile of cohort distribution despite normal TSH. High TSH was defined as values > 95 th percentile. I 2 refers to the percentage of the total variability due to between-areas heterogeneity. P value of heterogeneity was estimated using the Cochran's Q test.