Basnet, Sijan^a; Dhital, Rashmi^a; Tharu, Biswaraj^b; Ghimire, Sushil^a; Poudel, Dilli R.^a

Introduction

Disorders of thyroid, hyperthyroidism and hypothyroidism, are established risk factors for hypertension ^1,2. Comparison of these risk factors in terms of an association with hypertension has not been done.

Patients and methods

We used the National Inpatient Sample (NIS) database from 2012 to 2014 to explore the relationship between hypertension and two categories of thyroid hormone levels in comparison with hypertension without a thyroid diagnosis. NIS is the largest all-payer inpatient care database in the US and a part of Healthcare Cost Utilization Project. It contains discharge-level information on around 8 million hospital stays every year. It is sponsored by the Agency for Healthcare Research and Quality ³. We selected patients (≥18 years and nonpregnant) with diagnosis of hypertension in US hospitals from 2012 to 2014. Patients with hypertension, hyperthyroidism, and hypothyroidism were selected based on the International Classification of disease-9 (ICD-9) codes 401–405, 242, and 243–244, respectively. Multivariate logistic regressions were used to establish the association among all thyroid hormone level groups. We utilized STATA, version 13.0 (College Station, Texas, USA) to perform the analyses.

Results

The total number of discharge records in the NIS database (sample) from 2012 to 2014 was 21 477 520. Details of the selection process are depicted in Fig. 1. Our final study population was based on 15 633 745 (weighted count, N=78 168 721) inpatient hospital discharge records.

Fig. 1

Baseline characteristics of patients with hypertension and various thyroid states are presented in Table 1. Multivariate analysis with logistic regression (controlled for age, sex, race, smoking, obesity, dyslipidemia, diabetes, and Charlson comorbidity index) of hypertensive patients with hypothyroidism and hyperthyroidism in comparison with hypertensive patients with without a thyroid diagnosis was done (Table 2). It demonstrated that hypertension had a greater association with hyperthyroidism (odds ratio: 1.18; 95% confidence interval: 1.16–1.21, P<0.0001) than hypothyroidism (odds ratio: 1.06; 95% confidence interval: 1.06–1.07, P<0.0001). Both thyroid abnormalities had greater association with hypertension in comparison with those with hypertension but without a thyroid diagnosis.

Table 1

Table 2

Discussion

Hyperthyroidism contributes to increase in blood pressure by various mechanisms. T3 dilates peripheral arterioles and decreases systemic vascular resistance. This decrease in systemic vascular resistance stimulates renin secretion and increases effective arterial blood volume. Moreover, T3 increases heart rate and cardiac contractility with consequent increase in blood pressure ^2,4. On the contrary, hypothyroidism is usually associated with diastolic hypertension. It is thought to be secondary to increase in systemic vascular resistance ⁵. Thus, both states can contribute to increases in blood pressure and lead to hypertension. We did not find any previous studies comparing association of hyperthyroidism with hypothyroidism in cases with hypertension.

Strengths and limitations

The NIS database is the largest all-payer database representing the largest number of hospitalizations in the US. Thus, it is a more real-world representation of the inpatient care. It, however, is put together from discharge-level information which may be subject to coding errors. ICD-9 definitions do not specify what levels of thyroid constitute hypothyroidism or hyperthyroidism. It does not provide information on whether patient is receiving treatment for hypertension, hyperthyroidism, or hypothyroidism. This is due to limitations in ICD-9 coding.

Conclusion

Through our study, we can conclude that the risk of hypertension is greater with hyperthyroidism than with hypothyroidism. Whether this risk is improved with correction of abnormal thyroid states or not is a matter of future studies.

Acknowledgements

Conflicts of interest

There are no conflicts of interest.

References