Risk of Depression, Chronic Morbidities, and l-Thyroxine Treatment in Hashimoto Thyroiditis in Taiwan: A Nationwide Cohort Study.

Risk of Depression, Chronic Morbidities, and

l-Thyroxine

Treatment in Hashimoto Thyroiditis in Taiwan

A Nationwide Cohort Study

I-Ching Lin, MD, PhD, Hsin-Hung Chen, MD, MS, Su-Yin Yeh, MS,

Cheng-Li Lin, MSc, and Chia-Hung Kao, MD

INTRODUCTION

A

1-year prevalence of depression among the patients was 23%, whereas that in healthy control patients it was 3.2%. Compared with healthy control patients, the risk of depression was 2-fold higher in patients with diabetes (DM), hypertension (HTN), and coronary artery disease (CAD) and 3-fold higher in patients with end-stage renal failure and cerebrovascular disease.2

Hashimoto thyroiditis (HT) is an autoimmune thyroid disease that causes primary hypothyroidism by destroying thyroid tissue.3 Hypothyroidism or subclinical hypothyroidism may

cause coronary heart disease through hyperlipidemia, HTN, DM, or obesity.4–7 Previous studies have reported that thyroid

hormones are critical in the development of mood disturbances, cognitive impairment, and other neuropsychiatric manifestations.

8 A previous study revealed that patients older than

55 years and with depression exhibited high prevalence rates of thyroid disease.9 Furthermore, another study indicated that

patients with major depressive disorder had high prevalence rates of thyroid disease.10 However, 1 study reported that

patients with depression did not exhibit high prevalence rates of thyroid disease.11 These inconsistent results necessitate

conducting additional clinical or basic studies to confirm the relationships between depression and thyroid dysfunction. Therefore, the present study investigated the risk of depression

and other chronic morbidities among patients with HT and examined the interaction between depression and the effect of

L_{-thyroxine therapy period on patients in Taiwan.}

METHODS

Data Source

In this retrospective, nationwide cohort study, we retrieved data from the Longitudinal Health Insurance Database 2000 (LHID 2000). The LHID 2000 was established through the

cooperation of the Taiwan National Health Insurance Administration and the National Health Research Institute (NHRI).

It contains original inpatient and outpatient claims data of 1,000,000 randomly sampled beneficiaries of the National Health Insurance (NHI) program in the year 2000 (23.75 million citizens). The NHI program is a nationwide, single-payer

program that was established in March 1995, and it has covered nearly 99% of Taiwan residesnts.12 The LHID 2000 research

database contains patient medical orders, operative procedures, and clinical diagnoses, and the diagnostic codes are based on the International Classification of Diseases, Ninth Revision,

Clinical Modification Code (ICD-9-CM). All patient data in this database have been anonymized, and the NHRI approves access to the data. This study was approved to fulfill the condition for exemption by the Institutional Review Board (IRB) of China Medical University (CMUH104-REC2-115). The IRB also specifically waived the consent requirement.

Sampled Patients

We identified patients ages 20 years and older and newly diagnosed with HT (ICD-9-CM code 245.2) according to the data in the LHID2000 from 2000 to 2011. The date of HT diagnosis was defined as the index date. Patients with a history of depression (ICD-9-CM codes 296.2–296.3, 300.4, 311) before the index date were excluded. The non-HT cohort was formed by randomly selecting patients without a history of HT and depression from the LHID 2000. The selected patients were frequency matched with the patients with HT according to sex, age (in 5-year bands), and index year at a ratio of 1:4.

Outcomes and Comorbidities

All study patients were followed-up until they were given a diagnosis of depression or censored because of loss to followup,

withdrawal from the insurance program, death, or the end of

2011. Furthermore, we defined the following baseline comorbidities: HTN (ICD-9-CM codes 401–405), DM (ICD-9-CM

code 250), hyperlipidemia 9-CM code 272), stroke (ICD-9-CM codes 430–438), CAD (ICD-(ICD-9-CM codes 410–414), and cancer (ICD-9-CM codes 140–208).

Statistical Analysis

The distributions of sex, age, and comorbidities between the HTand non-HT cohorts were compared and examined using the chi-squared test. The mean ages (standard deviations, SDs) and follow-up periods (SDs) were measured and examined using a t test. The incidence rates of depression (per 1000 person-year) were calculated in the 2 cohorts. Univariate and multivariate Cox proportional hazards regression models were used to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for assessing the effects of HT on the risk of depression. The multivariate models were simultaneously adjusted for age and sex and the comorbidities of HTN and CAD. Further analysis was conducted to assess whether

L_{-thyroxine treatment affected depression outcomes. Cumulative}

incidence curves of depression were computed using the Kaplan–Meier method, and the differences in the cumulative incidence curves between the 2 cohorts were tested using a log-rank test. All analyses were executed using SAS statistical software for Windows (Version 9.4, SAS Institute Inc., Cary, NC). A 2-tailed probability level of P<0.05 was considered statistically significant.

RESULTS

In this retrospective cohort study, we collected data on 1220 patients with HT and 4880 patients without HT. In the HT cohort, 89.6% of the patients were women and 71.4% of the patients were ages 49 years or younger (Table 1). The mean age of the study patients was 42.7_13.8 years for the HT cohort and 42.5_14.1 years for the non-HT cohort. Compared with the non-HT cohort, the HT cohort exhibited a higher prevalence of DM, hyperlipidemia, and CAD. The mean follow-up periods for the HT and non-HT cohorts were 5.77 and 5.83 years, respectively. The Kaplan–Meier analysis results revealed that

during the follow-up periods, the HT cohort had a higher cumulative incidence of depression than did the non-HT cohort (log-rank test, P¼0.002; Figure 1).

After adjustments for age, sex, and comorbidities (namely HTN and cancer), the overall incidence of depression was greater in the HT cohort than that in the non-HT cohort (8.67 and 5.49 per 1000 person-year, crude HR¼1.58, 95% CI¼1.18–2.13), with an adjusted HR (aHR) of 1.55 (95% CI¼1.16–2.09; Table 2). In both cohorts, the women had a higher depression incidence than the men did. Furthermore, the sex-specific depression aHR was greater for the women

(aHR¼1.62, 95% CI¼1.20–2.19) compared with that for the men in both cohorts.

Except for patients older than 50 years, the increase in the risk of depression in patients with HT was greater for 2 age groups (i.e., _34 years: aHR¼1.82, 95% CI¼1.07–3.11; 35– 49 years: aHR¼1.63, 95% CI¼1.02–2.60) than it was for

those without HT. Patients with HT and with no comorbidity had a higher risk of depression than did those without HT and

with no comorbidity (aHR¼2.09, 95% CI¼1.46–2.99). In the multivariate model, the risk of depression was 2.34-fold higher in women compared with that in men (95% CI¼1.24–4.44). The risk of developing depression was higher in patients with the HTN comorbidity (HR¼1.55, 95% CI¼1.03–2.31;

Table 3). Table 4 shows the relationship between L_-thyroxine

treatment and the risk of depression. Compared with those in the non-HT cohort, patients in the HT cohort without L_-thyroxine

treatment were associated with a higher risk of depression (those without L_{-thyroxine treatment: aHR¼2.00, 95%}

CI¼1.37–2.91). By contrast, the risk of depression was decreased after treatment with thyroxine and did not differ from that of the non-HT cohort (aHR¼1.23, 95% CI¼ 0.82–1.84).

DISCUSSION

HT and Chronic Morbidities

HT is an autoimmune thyroid disease that causes primary

hypothyroidism.3 Previous studies have reported that hypothyroidism,

higher incidence of DM and hyperlipidemia.13–15 Another

study mentioned that HT caused coronary heart disease.16 In

the present study, the HT cohort exhibited a higher prevalence of DM, hyperlipidemia, and CAD compared with that in the non-HT cohort. Previous studies have reported that in addition inducing hypothyroidism, HT causes thyroid autoimmunity. HT

is considerably influenced by hereditary factors, such as insulindependent DM and pernicious anemia.17 A previous study

reported that HT altered the cell-mediated immunity through a genetic defect, resulting in a defective suppression of T cell functions,18 which causes patients with HT to produce various

cytokines such as interferon-g and tumor necrosis factor-a.19

The interferon-g and tumor necrosis factor-a cytokines can regulate fat inflammation or promote lipogenesis in mice.20

These cytokines may also cause weight gain and lipolysis in humans.20

HT,

-Thyroxine Treatment, and Depression

Our results revealed that the overall incidence of depression was greater in the HT cohort and the aHR was 1.55 after adjustments for age, sex, and comorbidities (HTN and cancer). Several studies have examined the association between depression and hypothyroidism.21,22 HT is an autoimmune

thyroid disease that causes primary hypothyroidism.3 However,

thyroid immunity has been reported to have a higher association with mood disorders in populations of patients with endocrine and psychiatric disorders compared with that in general populations.

23–25 Except for patients older than 50 years, the increase

in the risk of depression for patients with HT was more significant in 2 age groups (i.e., _34 years: aHR¼1.82, 95% CI¼1.07–3.11; 35–49 years: aHR¼1.63, 95%

CI¼1.02–2.60) than that for patients without HT. Our finding is consistent with those of other studies that have reported that the prevalence of hypothyroidism increased with age and was higher among patients older than 65 years.26,27 In the present

study, patients with HT and without comorbidities had a higher risk of depression than did those without HT and comorbidities; thus, HT is a possible independent risk factor for depression in younger Asian populations. In the multivariate model, the risk of depression was 2.34-fold higher in women compared with

that in men, a result that is consistent with that of a previous study.27 Studies have commonly reported hypercortisolism in

patients with depression. This disorder results in the modification of the HPT axis, which may reduce the secretion of serum

thyroid-stimulating hormone (TSH) and cause a weakened TSH response to thyrotrophin-releasing hormone (TRH).28,29 Cortisol

can also diminish the production of thyroid hormones by inhibiting the pathway of T4 conversion to active T3.30 A

previous study indicated that compared with people in the general population, patients with depression had a lower level of T3.31 This possibly explains the L_{-thyroxine treatment results}

for patients with HT. Studies on the preventive effects of using

L_{-thyroxine treatment in patients with HT to prevent CAD or}

reduce mortality have reported inconsistent results. Some studies have shown that L_{-thyroxine treatment could improve}

coronary flow or reduce hyperlipidemia in patients with HT.32,33 By contrast, another study reported that L_-thyroxine

treatment increased coagulation factor levels and inhibited

fibrinolysis, increasing the risk of ischemic stroke.34 In the present study, an insufficient

period of L_{-thyroxine treatment}

affected the outcome, whereas L_{-thyroxine treatment reduced}

the risk of depression in Taiwan.

CONCLUSION

In Taiwan, patients with HT demonstrated a higher prevalence of DM, hyperlipidemia, and CAD. Furthermore, in the HT

cohort, younger patients exhibited a higher overall incidence of depression compared with that in other patients. L_-thyroxine

treatment can reduce the risk of depression.

LIMITATIONS

There are certain study limitations: No individual subject’s information including the tobacco use, alcohol amount, stress evaluation, obesity status, exercise habit, income data, and family history of thyroid disease. The NHIRD could not provide the patients’ thyroid functions (such as TSH, TRH, T4, or T3) and images (such as thyroid echo or scans). These records in the NHIRD are majorly from billing of insurance uses, not for study uses, although all insurance claims in the NHIRD were already strictly surveyed by specialists’ peer review under the standard

regulations. Because the identification for every patient in the NHIRD is anonymous under the laws for the personal information protection in Taiwan, we could not directly check