Subclinical hypothyroidism and the risk of coronary heart disease and mortality

Abstract

Context: Data regarding the association between subclinical hypothyroidism and cardiovascular disease outcomes are conflicting among large prospective cohort studies. This might reflect differences in participants' age, sex, thyroid-stimulating hormone (TSH) levels, or preexisting cardiovascular disease.

Objective: To assess the risks of coronary heart disease (CHD) and total mortality for adults with subclinical hypothyroidism.

Data sources and study selection: The databases of MEDLINE and EMBASE (1950 to May 31, 2010) were searched without language restrictions for prospective cohort studies with baseline thyroid function and subsequent CHD events, CHD mortality, and total mortality. The reference lists of retrieved articles also were searched.

Data extraction: Individual data on 55,287 participants with 542,494 person-years of follow-up between 1972 and 2007 were supplied from 11 prospective cohorts in the United States, Europe, Australia, Brazil, and Japan. The risk of CHD events was examined in 25,977 participants from 7 cohorts with available data. Euthyroidism was defined as a TSH level of 0.50 to 4.49 mIU/L. Subclinical hypothyroidism was defined as a TSH level of 4.5 to 19.9 mIU/L with normal thyroxine concentrations.

Results: Among 55,287 adults, 3450 had subclinical hypothyroidism (6.2%) and 51,837 had euthyroidism. During follow-up, 9664 participants died (2168 of CHD), and 4470 participants had CHD events (among 7 studies). The risk of CHD events and CHD mortality increased with higher TSH concentrations. In age- and sex-adjusted analyses, the hazard ratio (HR) for CHD events was 1.00 (95% confidence interval [CI], 0.86-1.18) for a TSH level of 4.5 to 6.9 mIU/L (20.3 vs 20.3/1000 person-years for participants with euthyroidism), 1.17 (95% CI, 0.96-1.43) for a TSH level of 7.0 to 9.9 mIU/L (23.8/1000 person-years), and 1.89 (95% CI, 1.28-2.80) for a TSH level of 10 to 19.9 mIU/L (n = 70 events/235; 38.4/1000 person-years; P <.001 for trend). The corresponding HRs for CHD mortality were 1.09 (95% CI, 0.91-1.30; 5.3 vs 4.9/1000 person-years for participants with euthyroidism), 1.42 (95% CI, 1.03-1.95; 6.9/1000 person-years), and 1.58 (95% CI, 1.10-2.27, n = 28 deaths/333; 7.7/1000 person-years; P = .005 for trend). Total mortality was not increased among participants with subclinical hypothyroidism. Results were similar after further adjustment for traditional cardiovascular risk factors. Risks did not significantly differ by age, sex, or preexisting cardiovascular disease.

Conclusions: Subclinical hypothyroidism is associated with an increased risk of CHD events and CHD mortality in those with higher TSH levels, particularly in those with a TSH concentration of 10 mIU/L or greater.

Figures

Figure 1. Subclinical Hypothyroidism vs Euthyroidism for Coronary Heart Disease (CHD) Events, CHD Mortality, and Total Mortalitya

aThe sizes of the data markers are proportional to the inverse variance of the hazard ratios (HRs). CI indicates confidence interval; HUNT, Nord-Trøndelag Health Study; HR, hazard ratio. bForty-six participants from the Whickham survey and 3 participants from the Busselton Health Study were not included because follow-up data were only available for death. cNine participants were excluded from the analysis because of missing cause of death. The Brazilian Thyroid Study was not included in this analysis because of unreliable estimates based on the small number of CHD deaths (n=10).

Figure 2. Hazard Ratios (HRs) for Coronary Heart Disease (CHD) Events, CHD Mortality, and Total Mortality According to Elevated Thyroid-Stimulating Hormone (TSH) Categories and Subclinical Hypothyroidism Stratified by Age vs Euthyroidisma

aThe sizes of the filled square data markers are proportional to the inverse variance of the HRs. The unfilled squares indicate the reference categories. For the analyses stratified by age, the HRs for CHD events, CHD mortality, and total mortality were adjusted for sex and age as a continuous variable to avoid residual confounding within age strata. CI indicates confidence interval. bData were available from 7 studies. cData were available from 10 studies. The Brazilian Thyroid Study was not included because of unreliable estimates based on the small number of CHD deaths (n=10). Nine participants were excluded from the analysis because of missing cause of death. dData were available from 11 studies.