The association between tobacco smoking and thyroid function is incompletely understood.
In a cross-sectional, population-based study conducted between August 15, 1995, and June 18, 1997, of 20 479 women and 10 355 men without previously known thyroid disease, we calculated the geometric mean serum concentration of thyrotropin and the prevalence of hypothyroidism and hyperthyroidism among current, former, and never smokers.
Among women, the mean thyrotropin level was lower in current (1.33 mIU/L; 95% confidence interval [CI], 1.29-1.36 mIU/L) and former smokers (1.61 mIU/L; 95% CI, 1.56-1.65 mIU/L) compared with never smokers (1.66 mIU/L; 95% CI, 1.63-1.70 mIU/L). Similarly, among men, the mean thyrotropin level was lower in current (1.40 mIU/L; 95% CI, 1.36-1.44 mIU/L) and former smokers (1.61 mIU/L; 95% CI, 1.57-1.66 mIU/L) compared with never smokers (1.70 mIU/L; 95% CI, 1.66-1.75 mIU/L). In former smokers, thyrotropin levels increased gradually with time since smoking cessation (P for trend < .001). Among current smokers, moderate daily smoking was associated with higher thyrotropin levels than heavier smoking. In women, the prevalence of overt hypothyroidism was lower in current smokers compared with never smokers (odds ratio, 0.60; 95% CI, 0.38-0.95), whereas the prevalence of overt hyperthyroidism was higher among current smokers (odds ratio, 2.37; 95% CI, 1.34-4.20). The associations related to subclinical thyroid dysfunction were similar to those for overt thyroid disease.
These findings indicate that smoking is negatively associated with hypothyroidism but positively associated with hyperthyroidism. The associations with smoking cessation suggest that smoking may have reversible effects on thyroid function. Notably, we report for the first time, to our knowledge, a lower prevalence of overt hypothyroidism among current smokers.
The relation between tobacco smoking and thyroid function is not well understood. Most population-based studies show that smokers have lower levels of thyrotropin1-4 and higher levels of thyroid hormones,2-4 and smoking appears to increase the risk of Graves hyperthyroidism.5-8 However, smoking cessation may reduce the risk of Graves hyperthyroidism,5 suggesting that hyperthyroid effects of smoking may be reversed in people who quit.
For hypothyroid disease, studies have shown either no association6,7 or an increased risk8,9 of hypothyroidism associated with smoking. Recent studies4,10 suggest that smokers are less likely to have elevated thyrotropin levels, and some studies10,11 have shown that smokers are less likely to have thyroid peroxidase antibodies, which may suggest that autoimmune thyroid disease could be less common in smokers. In a large, cross-sectional, population-based study from an iodine-sufficient area in Norway,12 we studied the association between smoking habits and thyroid function in people without previously known thyroid disease.
All inhabitants 20 years or older of Nord-Trøndelag County in Norway were invited to participate in the Nord-Trøndelag Health Study (HUNT) between August 15, 1995, and June 18, 1997. A total of 92 936 individuals were eligible to participate, and 66 140 (71.2%) attended. The study has been described in detail elsewhere.13
Briefly, the participants were asked to complete a self-administered questionnaire, which included questions about thyroid disease and smoking habits. They were asked if they had ever had hyperthyroidism, hypothyroidism, goiter, or disease in the thyroid gland; if they had been receiving treatment with thyroxine, carbimazole, or radioactive iodine; or if they had undergone thyroid surgery. They were also asked if they currently smoked cigarettes, cigars, or a pipe daily, if they had never smoked daily; or, if they had previously been daily smokers, when they stopped smoking. Current and former smokers were asked at what age they started smoking, the average number of cigarettes per day, and how many years they had been smoking. On the basis of these questions, we classified the participants as never smokers, former smokers, or current smokers.
A nonfasting venous serum sample was obtained from each individual in the study, and analysis of thyrotropin was performed in subsamples, including all women older than 40 years and 50% of men older than 40 years. In addition, thyrotropin levels were measured in 5% random samples of men and women 20 to 40 years of age. In total, 34 851 individuals from these samples were selected for thyrotropin analysis. If the thyrotropin level was lower than 0.20 mIU/L, free thyroxine (FT4) and total triiodothyronine (T3) levels were also measured, and if the thyrotropin level was higher than 4.0 mIU/L, the FT4 level was measured. Among the 34 851 individuals who were selected for thyrotropin analysis, we excluded people with previously known thyroid disease (n = 2904) and people with missing information on thyrotropin level or smoking status (n = 1113), leaving 30 834 people (20 479 women and 10 355 men) eligible for the present study.
Serum concentrations of thyrotropin, FT4, and total T3 were analyzed at the Hormone Laboratory, Aker University Hospital, Oslo, using DELFIA hTSH Ultra (sensitivity, 0.03 mIU/L; total analytical variation, < 5%), DELFIA FT4 (total analytical variation, < 7%), and AutoDELFIA T3 (total analytical variation, < 5%), respectively, all from Wallac Oy, Turku, Finland. Reference ranges for thyrotropin from this population have been published previously.14 On the basis of these results, the reference range for thyrotropin in the present study was defined as 0.50 to 3.5 mIU/L. The laboratory's reference ranges were 0.62 to 1.55 ng/dL (to convert to picomoles per liter, multiply by 12.871) for FT4 and 77.9 to 175.3 ng/dL (to convert to nanomoles per liter, multiply by 0.0154) for total T3.
In a general linear model, we calculated geometric mean thyrotropin for never smokers, current smokers, and former smokers. In former smokers, we assessed whether mean thyrotropin levels differed according to time since smoking cessation (0-1, 2-3, 4-7, 8-12, 13-17, 18-22, and ≥ 23 years). In current smokers, we assessed whether mean thyrotropin levels differed according to the average number of cigarettes smoked per day (0-3, 4-7, 8-12, 13-17, and ≥ 18 cigarettes) or according to what age they started smoking (< 16, 16-20, 21-25, and ≥ 26 years). When reporting the number of years since quitting or the number of cigarettes smoked per day, the participants tended to round off the numbers to 5, 10, 15, and so on; we therefore used categories of smoking that were centered around these numbers. Serum thyrotropin concentrations were log-transformed because of nonnormal distribution.
Using a logistic regression model, we calculated the odds ratios (ORs) for thyroid dysfunction in current and former smokers compared with never smokers. Thus, overt hypothyroidism was defined as a thyrotropin level higher than 4.0 mIU/L combined with an FT4 level less than 0.62 ng/dL, and subclinical hypothyroidism was defined as a thyrotropin level higher than 4.0 mIU/L combined with an FT4 level of 0.62 ng/dL or higher. Overt hyperthyroidism was defined as a thyrotropin level lower than 0.20 mIU/L combined with an FT4 or total T3 level above the reference range, and subclinical hyperthyroidism was defined as a thyrotropin level lower than 0.20 mIU/L combined with neither an FT4 level nor a total T3 level above the reference range. In former smokers, we also assessed whether the prevalence of hypothyroidism or hyperthyroidism differed according to time since smoking cessation (0-3, 4-12, and ≥ 13 years).
We analyzed women and men separately, and all results were adjusted for age. The data were analyzed using SPSS statistical software, version 14.0, for Windows (SPSS Inc, Chicago, Illinois).
The HUNT Study is a collaborative effort of the Faculty of Medicine, the Norwegian University of Science and Technology, the Norwegian Institute of Public Health, and Nord-Trøndelag County Council. The study was approved by the regional committee for medical research ethics and by the Norwegian Data Inspectorate.
Among women, the mean thyrotropin level was lower in current (1.33 mIU/L; 95% confidence interval [CI], 1.29-1.36 mIU/L) and former smokers (1.61 mIU/L; 95% CI, 1.56-1.65 mIU/L) compared with never smokers (1.66 mIU/L; 95% CI, 1.63-1.70 mIU/L). Among men, the mean thyrotropin level was similarly lower in current (1.40 mIU/L; 95% CI, 1.36-1.44 mIU/L) and former smokers (1.61 mIU/L; 95% CI, 1.57-1.66 mIU/L) compared with never smokers (1.70 mIU/L; 95% CI, 1.66-1.75 mIU/L) (Table 1).
In former smokers, the thyrotropin level increased gradually with time since smoking cessation (P for linear trend across categories < .001 in both women and men). Thus, women who had quit smoking approximately 5 to 10 years ago had similar mean thyrotropin levels as never smokers, whereas in men, the mean thyrotropin level was similar to never smokers among former smokers who had quit 18 or more years ago (Figure 1).
Among current smokers, moderate daily smoking was associated with higher thyrotropin levels compared with heavier smoking. Thus, the mean thyrotropin level was higher in women who reported smoking fewer than 4 cigarettes per day (1.46 mIU/L; 95% CI, 1.35-1.59 mIU/L) compared with those who reported smoking 8 to 12 cigarettes per day (1.30 mIU/L; 95% CI, 1.26-1.34 mIU/L). In men, the mean thyrotropin level was higher among those who reported smoking fewer than 4 (1.53 mIU/L; 95% CI, 1.39-1.68 mIU/L) or 4 to 7 cigarettes (1.48 mIU/L; 95% CI, 1.40-1.56 mIU/L) per day compared with those who reported smoking 8 to 12 cigarettes per day (1.38 mIU/L; 95% CI, 1.33-1.44 mIU/L). In these data, smoking more than 12 cigarettes per day was not related to further reduction in the concentration of thyrotropin (Figure 2). In current smokers, age when they started smoking was not associated with thyrotropin (data not shown).
Current smoking in women was related to nearly half the prevalence of overt (OR, 0.60; 95% CI, 0.38-0.95) and subclinical (OR, 0.54; 95% CI, 0.45-0.66) hypothyroidism compared with never smokers. In men, the ORs for overt and subclinical hypothyroidism were 0.51 (95% CI, 0.15-1.73) and 0.37 (95% CI, 0.26-0.52) among current smokers compared with never smokers (Table 2).
In relation to hyperthyroidism, the prevalence of overt (OR, 2.37; 95% CI, 1.34-4.20) and subclinical (OR, 1.83; 95% CI, 1.10-3.06) disease among women was approximately twice as high in current smokers compared with never smokers. For men, there were too few individuals with hyperthyroidism to yield meaningful estimates (Table 3).
In former smokers, the prevalence of thyroid dysfunction was not substantially different from that of never smokers. Analyses related to time since smoking cessation suggest that the prevalence of thyroid dysfunction gradually approached that of never smokers after smoking cessation (Tables 2 and 3). Notably, the lower prevalence of overt hypothyroidism that we found among current smokers was not present among former smokers who had recently quit smoking.
In a separate analysis, we included people with thyrotropin levels of 3.6 to 4.0 mIU/L or 0.20 to 0.49 mIU/L as subclinically hypothyroid or hyperthyroid, respectively. The associations with smoking were similar to the associations in the original analysis, except that the greater number of individuals allowed analysis of subclinical hyperthyroidism among men. Compared with never smoking men, the prevalence of subclinical hyperthyroidism was higher among current smokers (OR, 2.10; 95% CI, 1.39-3.17) and among former smokers who had quit smoking 4 to 12 years ago (OR, 1.95; 95% CI, 1.11-3.44).
The estimates were not substantially different when we restricted the analyses to individuals older than 40 years.
In this large population-based study of people without previously known thyroid disease, current smokers had lower levels of thyrotropin, a lower prevalence of hypothyroidism, and a higher prevalence of hyperthyroidism compared with never smokers. Our findings indicate essentially similar associations between tobacco smoking and thyroid function in women and men, except that there were too few men with hyperthyroidism to allow meaningful estimates for that group.
The cross-sectional design may not allow us to draw conclusions for causality, but we consider an effect of tobacco smoking on thyroid function biologically more plausible than an effect of thyroid function on smoking habits. Previous population-based studies have also reported lower thyrotropin levels among current smokers,1-4 and some of these studies found correspondingly higher FT4 levels,3,4 free thyroxine indices,2 or free T3 levels,3 suggesting that smokers may have increased levels of thyroid hormones that are not mediated by thyrotropin.
The results of population-based studies have consistently shown that smokers have a lower prevalence of elevated thyrotropin levels,4,10 which is consistent with our finding related to subclinical hypothyroidism. We also found that current smoking was associated with a lower prevalence of overt hypothyroidism, which is at variance with previous studies that have reported either no association6,7 or an increased risk8,9 of hypothyroidism among smokers. Also, in a study of patients with Hashimoto thyroiditis, smoking was associated with a higher prevalence of overt hypothyroidism.15
Current smoking has been shown to increase the risk of Graves hyperthyroidism,6-8 and this evidence was recently strengthened by a prospective study among women.5 Also, the investigators of a large cross-sectional study found that smokers were more likely to have relatively low thyrotropin concentrations.10 Thus, the higher prevalence of overt and subclinical hyperthyroidism that we found among current smokers is consistent with previous results.
Neither the components in tobacco that may cause the thyroid effects nor their mechanisms of action are clear.4,10,16-18 Knudsen et al4 found that the associations between smoking and thyrotropin and FT4 levels disappeared after adjustment for thyroid volume and thyroid nodularity and suggested that the differences in thyrotropin and FT4 levels were secondary to, or parallel with, changes in thyroid structure. Recently, a lower prevalence of thyroid autoantibodies was associated with current smoking,10,11 suggesting that tobacco smoke may reduce the risk of chronic autoimmune thyroiditis, which is the most common cause of hypothyroidism in iodine-sufficient areas. Other studies have suggested that tobacco smoking is negatively associated with autoimmune diabetes19 and ulcerative colitis.20
The dose of smoking, assessed by serum cotinine concentration, has been negatively associated with thyrotropin and also with the presence of thyroid autoantibodies,10 but others have failed to confirm any association between number of cigarettes per day and thyrotropin levels.2,3 Our results showed a gradual decline in thyrotropin level related to moderate smoking (from 0 to 12 cigarettes per day), but we observed no further decline related to heavier smoking.
We found gradually higher levels of thyrotropin with time since smoking cessation, and approximately 10 to 20 years after quitting, the thyrotropin level in former smokers did not differ from that of never smokers. Also, those who had quit smoking long ago had a prevalence of hypothyroidism or hyperthyroidism that did not differ from that of never smokers. For overt hypothyroidism, the lower prevalence among current smokers was not present in former smokers, not even among those who had recently quit. These results suggest that the effects of smoking that appear to influence thyroid function may be reversed if the smoking habit is discontinued. This possibility is supported by the results of other studies1,2 that have reported lower thyrotropin levels among current compared with former smokers. A prospective study5 found a decreasing risk of Graves hyperthyroidism related to time since smoking cessation. Nonetheless, our findings related to smoking cessation should be further tested in prospective studies.
In conclusion, we found that among people without previously known thyroid disease, smoking is associated with lower concentrations of thyrotropin, a lower prevalence of hypothyroidism, and a higher prevalence of hyperthyroidism. Our findings indicate that smoking tobacco influences thyroid function, that smoking is related to the risk of developing both hypothyroid and hyperthyroid disease, and that the thyroid effects of smoking may be reversible. The lower prevalence of overt hypothyroidism among current smokers has not previously been reported.
Correspondence: Bjørn O. Åsvold, MD, Department of Public Health, Faculty of Medicine, N-7489 Trondheim, Norway (firstname.lastname@example.org).
Accepted for Publication: March 14, 2007.
Author Contributions: Dr Åsvold had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Åsvold, Bjøro, and Vatten. Acquisition of data: Bjøro and Vatten. Analysis and interpretation of data: Åsvold, Bjøro, Nilsen, and Vatten. Drafting of the manuscript: Åsvold and Vatten. Critical revision of the manuscript for important intellectual content: Bjøro, Nilsen, and Vatten. Statistical analysis: Nilsen and Vatten.Obtained funding: Bjøro. Study supervision: Bjøro, Nilsen, and Vatten.
Financial Disclosure: None reported.
Funding/Support: The study was financially supported by the Norwegian University of Science and Technology and by the Central Norway Regional Health Authority.
Additional Contributions: A special thanks to the Hormone Laboratory, Aker University Hospital, Oslo, which analyzed all thyroid function tests with financial support from Wallac Oy, Turku, Finland. The research is done independently of any of the funding sources. Data were provided by the HUNT Research Centre.
F Effects of cigarette smoking on thyroid function and the prevalence of goitre, thyrotoxicosis and autoimmune thyroiditis. J Intern Med
67- 71PubMedGoogle ScholarCrossref
H Cigarette smoking and thyroid hormone levels in males. Int J Epidemiol
972- 977PubMedGoogle ScholarCrossref
J Serum TSH levels in smokers and non-smokers: the 5th Tromso study. Exp Clin Endocrinol Diabetes
343- 347PubMedGoogle ScholarCrossref
T High occurrence of thyroid multinodularity and low occurrence of subclinical hypothyroidism among tobacco smokers in a large population study. J Endocrinol
571- 576PubMedGoogle ScholarCrossref
RD Smoking and other lifestyle factors and the risk of Graves' hyperthyroidism. Arch Intern Med
1606- 1611PubMedGoogle ScholarCrossref
et al. Smoking as a risk factor for Graves' disease, toxic nodular goiter, and autoimmune hypothyroidism. Thyroid
69- 75PubMedGoogle ScholarCrossref
G Smoking: a risk factor for hypothyroidism. J Endocrinol Invest
129- 131PubMedGoogle ScholarCrossref
PW Smoke exposure is associated with a lower prevalence of serum thyroid autoantibodies and thyrotropin concentration elevation and a higher prevalence of mild thyrotropin concentration suppression in the third National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab
6077- 6086PubMedGoogle ScholarCrossref
WM Risk factors for and prevalence of thyroid disorders in a cross-sectional study among healthy female relatives of patients with autoimmune thyroid disease. Clin Endocrinol (Oxf)
396- 401PubMedGoogle ScholarCrossref
L Excretion of iodine in the urine: a study from 6 different Norwegian districts in 1985 [in Norwegian]. Tidsskr Nor Laegeforen
1320-1321, 1317PubMedGoogle Scholar
et al. The Nord-Trøndelag Health Study 1995-97 (HUNT 2): objectives, contents, methods and participation. Norsk Epidemiol
19- 32Google Scholar
et al. Prevalence of thyroid disease, thyroid dysfunction and thyroid peroxidase antibodies in a large, unselected population: The Health Study of Nord-Trondelag (HUNT). Eur J Endocrinol
639- 647PubMedGoogle ScholarCrossref
M Relationship between cigarette smoking and hypothyroidism in patients with Hashimoto's thyroiditis. J Endocrinol Invest
607- 612PubMedGoogle ScholarCrossref
JJ Impaired action of thyroid hormone associated with smoking in women with hypothyroidism. N Engl J Med
964- 969PubMedGoogle ScholarCrossref
V Smoking is associated with an increased risk of type 2 diabetes but a decreased risk of autoimmune diabetes in adults: an 11-year follow-up of incidence of diabetes in the Nord-Trondelag study. Diabetologia
1953- 1956PubMedGoogle ScholarCrossref
J Tobacco and IBD: relevance in the understanding of disease mechanisms and clinical practice. Best Pract Res Clin Gastroenterol
481- 496PubMedGoogle ScholarCrossref