Association of Hypertension and Hyperthyroidism in a Subspecialty Clinic and a National Database

Ana M. Rivas, MD; Jeff Dennis, PhD; Camilo Pena, MD; Jonathan Kopel, BS; Kenneth Nugent, MD


South Med J. 2020;113(12):607-611. 

In This Article


After approval by the Texas Tech institutional review board, we conducted a retrospective review of records of adult patients 18 to 89 years old with an International Classification of Diseases, Ninth Revision (ICD-9) or ICD-10 diagnosis of hyperthyroidism or thyrotoxicosis evaluated at the endocrinology clinic in our department between January 1, 2011 and January 1, 2017.[11] The diagnosis of thyrotoxicosis was verified by reviewing clinical notes and laboratory data. A patient was considered hyperthyroid if he or she had a low TSH (<0.27 mIU/L) in the setting of elevated free triiodothyronine (FT3) and/or free thyroxine (FT4) levels (>4.2 pg/mL and >1.7 ng/dL, respectively). Patients were excluded from this study if they did not have both a blood pressure measurement and thyroid hormone measurements at the time of their visits or if they were taking thyroid replacement or antithyroidal medications at the time of their visits. Based on blood pressure cut defined by the Joint National Commission,[7] patients were classified as hypertensive if they had a systolic blood pressure >140 mm Hg and/or a diastolic blood pressure >90 mm Hg.[4,12] If they were not hypertensive but were taking medications that could lower their blood pressure at the time of their visits, then they were excluded from the data analysis.

Demographic, clinical information, and laboratory data were abstracted from the electronic medical records system. Means (standard deviations) and frequencies (percentages) were determined as appropriate for each group. Differences among groups were assessed using t tests or analysis of variance for continuous data and χ 2 tests for categorical data.

Data from a national sample came from NHANES, a nationally representative dataset collected biennially from a noninstitutionalized US population. Individuals younger than 18 years of age were excluded from analysis. Thyroid hormone levels were measured in the full sample in 2007–2008 and in one-third of the samples in 2009–2010 and in 2011–2012.[13] NHANES commonly conducts one-third samples of certain laboratory and examination measures to reduce the burden on respondents and to facilitate both scheduling and completion of these tests. These one-third samples use dedicated sample weights to maintain representativeness. We aggregated these 6 years to maximize the sample size of individuals with TSH, FT3, and FT4 tests. Most of the respondents also had their blood pressure checked and recorded, and additionally answered the question "Have you ever been told by a doctor or other health professional that had hypertension, also called high blood pressure?" A total of 9200 individuals had thyroid test values recorded and answered the self-report hypertension question; 8837 individuals had thyroid test values recorded and at least one blood pressure reading. The institutional review board did not require review of the study component using NHANES because the data are publicly available and deidentified.

NHANES records systolic and diastolic blood pressure readings for respondents between one and three times each, respectively. These values are averaged when two or three readings are present. Hypertension was defined as systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg. Hypertension requires a clinical diagnosis, so we identified these participants as "likely" hypertension diagnoses.

TSH, FT3, and FT4 cutoffs for low and high levels are established using NHANES laboratory manual definitions of the range of normal values. Normal ranges include TSH, 0.34 to 5.6 mIU/L; FT3, 2.5 to 3.9 pg/mL; and FT4, 0.6 to 1.6 ng/dL. The NHANES 2011–2012 TSH range differed slightly (0.24–5.4 mIU/L), and therefore this range was used to categorize those respondents as having low, normal, or elevated TSH.[14–17] Although the different ranges could cause a minor discrepancy analyzing TSH in continuous form, the creation of TSH categories circumvents this issue. Using the NHANES medication supplement, a variable identifying levothyroxine use was created to control for potential iatrogenic hyperthyroidism.

Analysis was performed using STATA 15.1 (StataCorp, College Station, TX) with population weighting to account for NHANES' complex sample design. χ 2 tests compared thyroid hormone levels with self-reported and measured hypertension. Ordinary least squares regression was used to predict the variation in systolic blood pressure levels. Statistical significance was set at P < 0.05.