Cigarette smoking is associated with higher thyroid hormone and lower TSH levels: the PREVEND study

The study was performed in the frame of the PREVEND cohort study, which investigates vascular and renal damage among predominantly white inhabitants of the city of Groningen, The Netherlands, as reported in detail elsewhere [19‐23]. The study was approved by the Medical Ethics Committee of the University Medical Center Groningen and conformed to the principles outlined in the Declaration of Helsinki. All participants gave written informed consent. Briefly, in 1997–1998, all inhabitants of the city of Groningen, aged 28–75 years, were sent a short questionnaire on demographics and cardiovascular morbidity and a vial to collect an early morning urine sample (n = 85,421). Pregnant women and diabetic subjects using insulin were excluded. Altogether 40,856 subjects responded (47.8%), and their urinary albumin concentration was determined. Subjects with a urinary albumin concentration ≥10 mg/L (n = 7768) were invited to participate, of whom 6000 were enrolled. In addition, a randomly selected group with a urinary albumin concentration <10 mg/L (n = 3394) was invited to participate in the cohort, of whom 2592 were enrolled. The initial study population was comprised of 8592 subjects who completed the total screening program. For the present study, we used data from participants who completed the second screening round (2001–2003; n = 6894), excluding those with missing values for smoking behavior, thyroid function tests, as well as those who were not euthyroid (see Measurements and Definitions) and those using thyroid hormones, antithyroid drugs, amiodarone, or lithium carbonate. This left a cohort of 5766 participants with sufficient information for analysis (Fig. 1).

Euthyroidism was defined as TSH, FT4, and FT3 levels all within the euthyroid reference range (TSH: 0.27–4.20 mU/L; FT4: 12–22 pmol/L; FT3:3.1–6.8 pmol/L). Cigarette smoking was estimated by questionnaire, and was categorized as (i) never; (ii) former; (iii) current ≤20 cigarettes per day, and (IV) current >20 cigarettes per day. In addition, urinary cotinine concentrations were measured. Former smokers were subjects who were nonsmokers at the time of the second screening round but had ever smoked in their life; current smokers were those who reported smoking at this screening. Alcohol intake was estimated by using predefined categories covering a time period of one month [19, 22]. Alcohol intake was then calculated as alcoholic drinks per day with one drink being considered to be equivalent to 10 g of alcohol, regardless of the type of beverage [19]. Alcohol intake was categorized in (i) no/rarely; (ii) 0.1–10 g per day (occasional to light); (iii) 10–30 g per day (moderate), and (IV) >30 g per day (heavy).

The presence of a self-reported history of myocardial infarction, percutaneous coronary intervention, coronary artery bypass surgery, stroke, or the diagnosis of narrowing of one or both carotid arteries was defined as cardiovascular disease. Type 2 diabetes mellitus was defined as a fasting serum glucose concentration >7.0 mmol/L, a nonfasting serum glucose concentration >11.1 mmol/L, a self-report of a physician diagnosis, or the use of glucose-lowering drugs [23].

Information on medication use was combined with information from IADB.nl, a data base containing information of prescribed medication in public pharmacies in The Netherlands (http://​www.​iadb.​nl/​) [24]. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m). Blood pressure was measured using an automatic device (Dinamap XL model 9300 series; Johnson-Johnson Medical, Tampa, FL, USA). Urinary albumin excretion (UAE) was measured in two consecutive 24 h urine collections after oral and written instruction. The participants were asked to avoid heavy exercise as much as possible during the urine collection. Subjects were also instructed to postpone their urine collection in case of urinary tract infection, menstruation, or fever. The UAE results were averaged for analysis. Estimated glomerular filtration rate (eGFR) was calculated applying the combined creatinine cystatin C-based Chronic Kidney Disease Epidemiology Collaboration equation [25]. The participants were instructed to have their blood samples being taken in the morning after an overnight fasting period.

Venous blood samples were obtained after 15 min rest. Venous plasma and serum samples were obtained by centrifugation at 1400 × g for 15 min at 4 °C. Serum glucose was measured shortly after blood sampling. Ethylenediaminetetraacetic acid anticoagulated plasma and heparinized plasma and serum was obtained by centrifugation at 4 °C, and stored at −80 °C until analysis. Cotinine concentrations were measured with with Enzyme Multiplied Immunoassay Technique on the Abbott Architect c8000 system (Abbott Laboratories, Abbott Park, IL, USA). Serum TSH, FT4, FT3, and antithyroid peroxidase (anti-TPO) autoantibodies were determined using the Roche Modular E170 Analyzer electrochemiluminescent immunoassays (Roche Diagnostics, Mannheim, Germany). Anti-TPO autoantibodies were considered positive using a cut-off value as indicated by the supplier (≥35 kU/L). Twenty-four hour urinary cotinine levels were measured with Enzyme Multiplied Immunoassay Technique on the Abbott Architect c8000 system (Abbott Laboratories, Abbott Park, IL).

Total cholesterol was measured on an automatic analyzer (MEGA; Merck, Darmstadt, Germany) using the CHOD–PAP-method. Glucose was measured using Kodak Ektachem dry chemistry (Eastman Kodak, Rochester, New York, USA). HDL cholesterol and triglycerides were determined using the nuclear magnetic resonance (NMR)-derived LP4 lipoprotein profile deconvolution algorithm as described [26]. To this end NMR spectra were acquired on a Vantera^® Clinical Analyzer at LabCorp (Morrisville, NC, USA). Serum creatinine was measured by an enzymatic method on a RocheModular analyzer (Roche Diagnostics, Mannheim, Germany). Serum cystatin C was measured by Gentian Cystatin C Immunoassay (Gentian AS, Moss, Norway) on a Modular analyzer (Roche Diagnostics). Urinary albumin was measured by nephelometry (Dade Behring Diagnostic, Marburg, Germany) [19].

Data analysis was performed using IBM SPSS software (version 23.0, SPSS Inc., USA). Normally distributed data are given as mean ± SD and nonparametrically distributed data are presented as median (interquartile range). Categorical variables are given as numbers and %. Anti-TPO autoantibody titers (as continuous variable), TSH, triglycerides, urine cotinine, and UAE were natural logarithm (log_e) transformed in order to achieve approximately normal distributions. Differences in continuous variables between categories of cigarette smoking were determined by Analysis of Variance with subsequent Bonferroni procedure to correct for multiple comparisons. Differences in TSH, FT4, and FT3 according to anti-TPO antibody status were determined by unpaired T-tests. Proportions of dichotomous variables across categories of cigarette smoking were determined by χ² tests. Pearson correlation coefficients were calculated to determine univariate relationships. Multivariable linear regression analyses were used to discern the independent association of current cigarette smoking with TSH, FT4, and FT3. In these analyses, the category “never smoked” was used as the reference category. In these analyses, we also used categorized alcohol consumption as variable with the category “no/rarely” as reference. In addition, analyses were repeated with urinary cotinine data instead of self-reported smoking status. In these analyses, urinary cotinine concentrations of 0 were set at 0.01 to allow for log_e transformation. Standardized regression coefficients (β) were used to discern the strength of associations of current smoking (≤ and >20 cigarettes per day) with TSH, FT4, and FT3. Two-sided P-values < 0.05 were considered significant.