Retrospective analysis of thyroid function in drug-free patients with bipolar disorder in China

Background Bipolar disorder is a common mental illness with serious consequences. Clinical studies have found that thyroid function may have an impact on patients with bipolar disorder, but there are few relevant studies in China. So this study explores the characteristics of thyroid function in patients with bipolar disorder in China. Methods Using retrospective cohort study and a cross-sectional study, thyroid function tests were performed in inpatients from September 2015 to January 2018 in the Second Affiliated Hospital of Xinxiang Medical University, who were diagnosed with bipolar disorder and were not treated with medications for at least three months before hospitalization. Results We found that the triiodothyronine (T3) and free triiodothyronine (FT3) levels were significantly higher in bipolar mania than in bipolar depression (P < 0.01). The thyroid stimulating hormone (TSH) levels were significantly lower in male than female patients, while the FT3, free thyroxine (FT4) levels were higher in male than female patients (P < 0.01). Patients were divided into two groups: those who used lithium and those who did not. Both groups had a tendency of hypothyroidism after treatment. In addition, the T3 and FT3 levels were significantly higher in manic than in depressive status in those patients who had a transition between mania and depression in 28 patients compared with previous hospitalization (P < 0.05). Patients bipolar disorder may thyroid at different and in male/female sexes.


Background
Bipolar disorder is a common mental illness with serious consequences. Clinical studies have found that thyroid function may have an impact on patients with bipolar disorder, but there are few relevant studies in China. So this study explores the characteristics of thyroid function in patients with bipolar disorder in China.

Methods
Using retrospective cohort study and a cross-sectional study, thyroid function tests were performed in inpatients from September 2015 to January 2018 in the Second Affiliated Hospital of Xinxiang Medical University, who were diagnosed with bipolar disorder and were not treated with medications for at least three months before hospitalization.

Results
We found that the triiodothyronine (T3) and free triiodothyronine (FT3) levels were significantly higher in bipolar mania than in bipolar depression (P < 0.01). The thyroid stimulating hormone (TSH) levels were significantly lower in male than female patients, while the FT3, free thyroxine (FT4) levels were higher in male than female patients (P < 0.01). Patients were divided into two groups: those who used lithium and those who did not. Both groups had a tendency of hypothyroidism after treatment. In addition, the T3 and FT3 levels were significantly higher in manic than in depressive status in those patients who had a transition between mania and depression in 28 patients compared with previous hospitalization (P < 0.05).

Conclusion
Patients with bipolar disorder may develop thyroid dysfunction at different stages and in male/female sexes.
Background Bipolar disorder is a chronic mood disorder characterized by mania, hypo-mania, alternating depression or mixed episodes (Grande et al. 2016). Bipolar disorder is a relatively common disease that is prone to relapse, has a heavy disease burden and has a high risk of suicide (Mitchell and  Wysokinski and Kloszewska 2014). Thyroid hormone plays an important role in the function and regulation of neural tissue activity. Therefore, any thyroid hormone synthesis, secretion, action and peripheral metabolic disorders can affect the normal function of nerve tissue (Bauer and Whybrow 2001). Patients with thyroid dysfunction are often associated with mood disorders. The most common psychiatric symptoms associated with thyroid dysfunction are mood disorders, cognitive dysfunction and anxiety (Bauer and Whybrow 2001). Gulseren et al. (Gulseren et al. 2006)found that patients had more severe symptoms of anxiety and depression during hypothyroidism, which recovered after thyroxine treatment. Another retrospective study of a large number of patients showed a higher incidence of bipolar disorder in patients diagnosed with hyperthyroidism (Hu et al. 2013).

Schrekenberger et al. conducted a cross-sectional study and it was performed between untreated
Graves' disease patients and the control group, and the differences between anxiety and depression levels and cerebral glucose metabolism were studied. They found that patients with hyperthyroidism had decreased glucose metabolism in the limbic system, activated foci in the posterior cingulate gyrus and inferior parietal lobe, and were associated with depression and anxiety (Schreckenberger et al. 2006). Bipolar disorder patients with thyroid dysfunction are often poorly treated when standardized treatment is used. Thyroid dysfunction increases the difficulty in the diagnosis and treatment of mood disorders (Bauer et al. 2014;Cole et al. 2002).
There is evidence that except for the presence of hyperthyroidism and hypothyroidism, more patients with bipolar disorder exhibit subclinical thyroid dysfunction, with small changes in thyroid function levels (Bauer et al. 2008). Subclinical or clinical hypothyroidism, such as lower levels of thyroxine (Rybakowski and Sowinski 1973), and elevated levels of thyrotropin are more common in patients with bipolar disorder (Ezzaher et al. 2011;Valle et al. 1999). In addition, the increase of thyroid antibodies (anti-TPO) in patients with bipolar disorder was significantly higher than that in the normal population (Kupka et al. 2002). A study found that thyroid dysfunction may be associated with poor treatment outcomes and increased risk of relapse in some patients with bipolar disorder (Cole et al. 2002). There may also be a gender difference in thyroid dysfunction in patients with bipolar disorder.
Some studies and reviews have found that there are also gender differences in thyroid function in the Chinese studies have found that TSH levels in patients with bipolar II depression were lower than those in healthy controls, and there was no significant difference in FT3, T3, FT4, and T4 between the two groups (Zhong et al. 2019). In addition, some studies have found that the incidence of goiter and average TSH level in Chinese patients after lithium salt use are higher than those in the control group (Lee et al. 1992). At present, the results of thyroid function in patients with bipolar disorder are inconsistent, and few studies have been conducted on the characteristics of thyroid function level in Chinese patients with bipolar disorder. Therefore, we investigated: 1. Differences in thyroid function in patients with bipolar disorder in the manic or depressive state; 2. Differences in thyroid function between sexes in patients with bipolar disorder; 3. Changes of thyroid function in patients with bipolar disorder after combined treatment.

Subjects
This study is a retrospective cohort study and a cross-sectional study, which was approved by the Ethics Committee of The Second Affiliated Hospital of Xinxiang Medical University. Diagnosis of the patient's condition by two professional clinicians, and the inclusion criteria were as follows: 1. Meeting the bipolar disorder diagnosis by the International Statistical Classification of Diseases and Related Health Problems Diagnostic (ICD-10); 2. No medication for at least three months before admission; 3.
No limit on age, sex. The following were exclusion criteria: 1. Alcohol or other substance dependence; 2. Have serious physical diseases; 3. Have a history of primary thyroid disease or other endocrine diseases; 4. Have other mental disorders other than bipolar disorder.
From September 2015 to January 2018, there were 20938 inpatients in the Second Affiliated Hospital of Xinxiang Medical University, including 3349 (15.99%) patients with bipolar disorder. Among these patients, we excluded 6 subjects with hyperthyroidism or hypothyroidism before treatment and 27 subjects whose bipolar disorder was not diagnosed clearly. Finally, 617 patients with bipolar disorder who were not treated for at least three months before admission and who underwent thyroid function test, were included in the sample study. There were 386 males (62.56%) and 231 females (37.44%).
Including 74 patients with bipolar disorder received at least two thyroid function tests during hospitalization (two tests before treatment and after treatment for four weeks.). Among the 74 followup patients, the patients were treated with combination therapy, divided into two groups according to whether lithium salt was used or not. 34 cases were treated with lithium salt and 40 cases were not.
There were no family history of bipolar disorder and other mental illness in groups, and gender, age, BMI, and thyroid function levels were matched before treatment. Patients mainly used mood stabilizer, anti-epileptic drugs, and second-generation antipsychotic treatment. The average lithium salt dosage was 0.6 g/day. In addition, there were 28 patients suffered from a shift in mood state between mania and depression during hospitalization (see Fig. 1).

Data collection and analysis
Demographic information was collected from patient records, such as gender, age, length of hospital stay, BMI, blood pressure, family history of mental illness. All patients had fasting phlebotomy measurements between 06:00 and 08:00 within 24 hours after admission. The laboratory test was performed by chemiluminescence immunoassay (Roche 411/Roche 601). The original reagents of Roche were used to determine the concentrations of TSH, T3, T4, FT3 and FT4 (TSH: thyroid stimulating hormone; T3: triiodothyronine; T4: thyroxine; FT3: free triiodothyronine; FT4: free thyroxine). The normal ranges of these measurements are: 0.27-4.20 mU/L for TSH, 1.30-3.10 nmol/L for T3, 66.00-181.00 nmol/L for T4, 3.10-6.80 pmol/L for FT3, and 12.00-22.00 pmol/L for FT4. Use Roche to control quality.

Statistical analysis
All data was analyzed using SPSS 22.0 statistical software. We used Shapiro-wilk to test whether the data was normally distributed. Differences in demographic data were compared using Mann-whitney U and chi-square test. Chi-square test was used to compare the prevalence of mania and depression in different seasons. Mann-whitney U was used to analyze the differences in thyroid function between gender or mood states in patients with bipolar disorder. The differences in thyroid function before and after treatment was compared by the Wilcoxon Signed rank-test and the differences in thyroid function between the two treatment groups was compared by Mann-whitney U. Statistical significance was P < 0.05.

Results
Clinical, socio-demographic and biological parameters between mania and depression in patients with bipolar disorder There were no differences in gender, age, length of hospital stay, BMI, family history of mental illness and thyroid dysfunction rate between mania and depression patients. The systolic/diastolic blood pressure in mania was significantly higher than that in depression patients, (P < 0.01) (see Table 1).
Since we need to count the changes in different seasons, we choose a whole year as the observation unit. From November 1, 2015 to October 31, 2016, there were 2741 patients with bipolar disorder.
There was no difference in the distribution of mania and depression in different seasons (P > 0.05) (  Thyroid function in patients with bipolar disorder in different mood states or genders T3 and FT3 in mania were significantly higher than those in depression (P < 0.01).Further, the TSH of male patients was significantly lower than that of female patients, while the FT3 and FT4 of male patients were significantly higher than those of female patients (P < 0.01), as shown in Table 3. Patients with bipolar disorder were divided into groups according to whether they used lithium.
Compared with those before treatment, TSH were significantly increased, but T3, T4, FT3, FT4 were significantly decreased after treatment in two groups (all P < 0.01). However, there was no significant differences in thyroid function between the two groups (all P > 0.05) (See Table 4). There was a shift in mood state between mania and depression in 28 patients compared with previous hospitalization. Table 5 showed that the T3 and FT3 of the manic patients were significantly higher than that of the depressed patients (both P < 0.05).

Discussion
In this study, 617 patients with bipolar disorder who had not been treated with drugs for at least three months before hospitalization were taken as the study objects, of which 442 were patients with manic episodes and 175 were with depressive episodes. There were no differences in general information such as gender, age, length of hospital stay, BMI, family history of mental illness and abnormal thyroid function rate among manic and depression patients who did not take drugs three months before hospitalization. Also, there was no difference in the distribution of emotional transition in different seasons. Only systolic/diastolic blood pressure in mania was higher than that in depression systolic/diastolic blood pressure. It is speculated that this may be related to the risk of irritability and high mood in patients with bipolar mania.
Our study found that T3 and FT3 in patients with mania were significantly higher than those in patients with depression, which was similar to the results of previous studies (Gordon et al. 1999;Weatherman 2007). Thyroid hormones (especially T3) are structurally similar to norepinephrine (NE) (Weatherman 2007). T3 is mainly concentrated in the nucleus and projection sites of the central NE energy system (Rozanov and Dratman 1996), which may be transmitted from the locus to its NE target through axonal transport (Gordon et al. 1999 all exist in key areas of the brain, and thyroid hormone receptors are widely distributed in the limbic system of the brain, which is related to the pathogenesis of emotional disorders, and the neurotransmitter system regulates emotions by regulating the activities of the limbic area and cortex (Bauer, Heinz, and Whybrow 2002). Therefore, thyroid hormone itself may act as a neurotransmitter or interact with the primary neurotransmitter system to influence the mood of patients (Bauer et al. 2008;Chakrabarti 2011;Zhang et al. 2015).
The results of this study showed that the TSH of male patients was significantly lower than that of female patients, and the FT3 and FT4 of male patients were significantly higher than those of female  This study has the following limitations: 1. Since it is an observational study, the reason cannot be explained. 2. In this study, we did not have the control group with lithium salt monotherapy for bipolar disorder patients. Thus, it was not possible to determine whether lithium salt would affect thyroid function itself. 3. The sample size in this study was limited. Moreover, due to the small number of patients with bipolar mixing, they were not included in the study. 4. In this natural real-world study, the choice of laboratory testing and treatment was based on the doctor's decision in clinical practice, which may have the follow-up bias and treatment drug selection bias.

Conclusions
In summary, we studied the level of thyroid function in patients with bipolar disorder in China. Thyroid hormones may play an important role in bipolar disorder. Patients with bipolar disorder had thyroid dysfunction in both mania and depression, and the change in thyroid function may also be related to gender and medication. Therefore, it is necessary to conduct a comprehensive thyroid assessment in patients with bipolar disorder. All patients should be regularly examined for thyroid function to prevent adverse reactions of hormonal imbalance.

Declarations
The data that support findings of this study are available from the corresponding author. The data are not publicly available due to in the need to safeguard the privacy of the participants.

Figure 1
Study design and sample selection