Prevalence of hypothyroidism in different occupational groups of Bangladeshi population

Background and aims: Hypothyroidism is a common global endocrine disorder. The magnitude of hypothyroidism at community level in Bangladesh is unknown except some clinic-based studies. The present study was undertaken to determine the prevalence of hypothyroidism in different occupational groups of Bangladeshi population and to assess the risks related to it. Study design: Three occupational groups (house-wives, college students, rickshaw-pullers) of native Bangladeshi population were purposively selected. Investigations included socio-demography, anthropometry, blood pressure and biochemistry [fasting blood glucose, lipids, thyroid stimulating hormone (TSH) and free thyroxin (FT4)]. Laboratory tests were done only on a randomized sample of participants. Results: Overall, 626 (M/F=123 / 503) participants with a mean age of 35.9 (34.75 – 37.02) years volunteered. The mean values of all participant for TSH and FT4 were 2.08 (95%CI: 1.72 – 2.45) μiu/ml and 13.04 (95CI:12.86 – 13.22) pmol/L respectively. The third percentile of TSH ranged from 0.42 to 0.46 μiu/ml and 97th percentile ranged from 5.16 to 5.24 μiu/ml. For FT4, the 3rd and the 97th percentile were 10.3 and 16.41 pmol/L, respectively. The prevalence of hypothyroidism in both sexes was 7.0% (M/F=4.1/8.3%). Occupational groups, sex and increasing age, obesity, blood pressure, and lipids showed no association with hypothyroidism. Hyperglycemia was proved to be a significant risk for hypothyroidism (prevalence in diabetic vs. non-diabetic was12.9% vs. 5.5%, p = 0.04; FBG was correlated with TSH, r = 0.138, p <0.001). Conclusions: It is concluded that the prevalence of hypothyroidism was almost equal to other studies. Hypothyroidism was not related to increasing age, obesity, blood pressure and lipids. It was found to affect all sexes, all social classes and all occupational groups. Hyperglycemia was evidently found as significant risk for hypothyroidism. IMC J Med Sci 2019; 13(2): 002. EPub date: 24 July 2019


Introduction
Hypothyroidism is a common endocrine problem that is encountered everywhere in the world, be it in developed or developing countries [1][2][3]. This endocrine disorder affects health from utero to childhood and even extends to adulthood [4,5]. Hypothyroidism in pregnancy may lead to premature delivery or even its loss [6]. Additionally, hypothyroxemia in pregnancy may be associated with gestational diabetes [7] and hypertension [8]. It has long been known that normal growth and development was affected by hypothyroidism if not diagnosed and intervened in early life [9]. An interesting observation reported from Bangladesh was that dyslipidemia, hyperuricemia and impaired renal function were related to hypothyroidism [10]. Congenital hypothyroidism has been reported to be 1.5 per thousand population in southern region of Bangladesh [11]. If these cases are not detected at an early stage, then they become physically and mentally handicapped [12]. Again, hypothyroidism has been reported to be associated with coronary artery disease [13]. Although there are many studies on hypothyroidism, there are few population based reports on the prevalence of hypothyroidism. This study aimed to determine the prevalence of hypothyroidism in different occupational groups of native Bangladeshi population and to assess the risks related to it.

Participants
Three occupational (housewives, students, rickshawpullers) groups of native Bangladeshi people were selected purposively. The housewives of a suburb community were approached through local female health workers. This population represented the females of a Bangladeshi community maintaining traditional lifestyle. The second group, medical college students, represented young urban affluent community. The third group, rickshaw-pullers, represented non-affluent, hard-working young men of rural origin temporarily living in urban slum.
In different geographical site, enlistment of each participant was confirmed after discussion in detail. The objectives of the study and procedural steps were explained. If the participant agreed to volunteer then a token noting identification number was given and they were advised to attend an adjacent investigation site after an overnight fast. Gonoshasthya Kendra (GSK), a local community hospital was selected for the suburb housewives. The Department of Community Medicine of Ibrahim Medical College was appointed for the medical college students. A garage (local office of Rickshaw Sromik Union) at Nandigram, Dhaka was selected for the rickshaw-pullers with the help of their union leaders.

Interview and investigations
The interview included socio-demographic information (on contact address, age, sex, family income, education, and occupation) as well as clinical history (of present and past illness, medication, family history of hypertension, diabetes, stroke, thyroid diseases, and physical activity). Anthropometric measures included height, weight, waist and hip circumference. Body mass index (BMI: weight in kg / height in meter sq.), waist-to-hip ration (WHR: waist / hip) and waist-to-height ration (WHtR: waist / height) were calculated. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) was taken using a mercury sphygmomanometer after ten min physical rest and mental relaxation.
Taking an aseptic measure five ml venous blood was taken after an overnight fast and centrifuged. Serum was separated and immediately transported to biochemistry laboratory for estimation of fasting blood glucose (FBG), total cholesterol (T-Chol), triglycerides (TG), high-density lipoproteins (HDL) and low-density lipoproteins (LDL). Separated aliquots were sent to Bangladesh Institute of Research and Rehabilitation for Diabetes, Endocrine and Metabolic Disorders (BIRDEM) for assaying thyroid stimulating hormone (TSH) and Free Thyroxin (FT4) using chemiluminescent microparticle immunoassay technology.

Operational definition
Hypothyroidism was diagnosed when the TSH value exceeded 4.0 µIU /L and the FT4 value was normal [14,15].

Statistical analysis
The quantitative values were expressed in mean with standard deviation (SD) or mean with 95% confidence interval (CI). The prevalence rates were given in percentages. Unpaired t-tests were used to determine the differences between groups. Chi-sq estimated the associations between two or more variables including trends. One-way ANOVA used for multiple comparisons among the three occupational groups. We used SPSS version 20. The level significance was <0.05.
Table-3 shows the prevalence of hypothyroidism according to sex, social class, obesity, hypertension, diabetes and occupation. The overall prevalence was 7.0% in the total study population. The prevalence of hypothyroidism was significantly higher in diabetic than non-diabetic participants (12.9% vs. 5.5%, p = 0.04). The difference of prevalence of hypothyroidism was not significant among the different sex, grades of obesity or occupational group; neither there were any differences between participants with and without systolic and diastolic hypertension. Table-4 shows the comparison of participants having hypothyroidism (TSH > 4.0 µiu /L) with that of without hypothyroidism (TSH ≤ 4.0 µiu /L). Obesity, blood glucose and lipids did not differ significantly; whereas, diastolic blood pressure was significantly lower in the hypothyroid group (74.4 ± 9.8 vs 79.4 ± 11.6 mmHg, p = 0.016).
Correlations of TSH and FT4 with other variables (BMI, WHR, WHtR, blood pressure, and blood glucose) are shown in Table-5. TSH showed, as expected, significant negative correlation with FT4 (r = -0.304, p <0.001) and significant positive correlation with FBG (r = 0.138, p <0.001). Systolic blood pressure showed significant positive correlation with obesity related variables (BMI, WHR, WHtR; for all p<0.001) and also with FBG (p<0.001). Neither TSH nor FT4 showed significant association with obesity variables and blood pressure.
One-way ANOVA analyzed the multiple comparisons (BMI, WHR, WHtR, SBP, DBP, FBG, TSH, FT4) taking occupational group as factor (Table 6a and 6b). These tables clearly depict that the rickshaw pullers had lowest BMI and WHtR, which were significant (for both, p<0.001). The house wives had significantly higher BMI, WHtR, SBP and DBP than the other two occupational groups. In contrast, TSH and FT4 did not differ among the three occupational groups.       Figure-1. The measures of central tendencies and variability of thyroid stimulating hormone (TSH) and free thyroxin (FT4) are shown in Table-7. The prevalence did not increase significantly with increasing age, BMI, WHtR; whereas, the trend was significant for increasing level of FBG (p=0.04).

Discussions
Hypothyroidism is based only on the circulating blood level of TSH despite normal FT4 level, and the clinical manifestations are usually not evident. This study is unique in the sense that it addressed the prevalence of subclinical hypothyroidism at community level. Additionally, it investigated whether the risk factors, so far known, are associated with hypothyroidism in our population. Simultaneously, this study included different occupational groups for comparison of prevalence rates and the associated risk factors acting upon the occupational groups. The study could propose the values of TSH and FT4 at 3 rd and 97 th percentile (Table-1). This finding may help to compare or to determine future reference range of TSH and FT4.
The prevalence of hypothyroidism observed in this study is somehow lower than that of India [3]. In India, overall prevalence was 9.4% (men /women = 6.2% /11.4%); whereas, the prevalence of this study was 7% (men / women = 4.1% / 8.3%). Jeannine et al. reported that the prevalence of hypothyroidism varied 3% to 11% depending on the diagnostic cut-off of TSH, geographical site and ethnicity [15]. A clinic based study in Assam reported the prevalence as 13.1% [16]. Most of the studies opined that the prevalence among women is higher than that of men [2,3,16,17]. We also found higher prevalence in women than men (8.2% vs. 4.1%, p = 0.09), but the difference was not significant.
Regarding diabetes, the prevalence of hypothyroidism was significantly higher among the diabetic than among the non-diabetic group (Table-3). Correlation was also found significant between FBG and TSH (Table-5). Additionally, we found that the trend of hypothyroidism increased significantly with increasing fasting blood glucose (Figure-1). The associations between hypothyroidism and diabetes have been reported in other studies and in other forms of diabetes [6,7,[18][19][20][21][22]. A study found a higher TSH level in patients with metabolic syndrome suggesting that hypothyroidism may be a risk factor for it [23]. In subclinical hypothyroidism, insulin resistance may result from diminished rate of insulin stimulated glucose transport caused by perturbed expression of glucose transporter type 2 genes (GLUT 2). There is also impaired insulin stimulated glucose utilization in peripheral tissues [24].
In Table-4, hypothyroid group had significantly lower diastolic blood pressure than their normal thyroid counterpart. It is not clear why hypothyroid group had significantly lower diastolic blood pressure. Possibly, the hypothyroid group has also diabetes as mentioned above, leading to autonomic neuropathy resulting in diastolic dysfunction and lower diastolic blood pressure. On the contrary, some investigators found association of hypothyroidism with hypertension [25]. There is a plausible explanation that hypothyroidism reduces dopaminergic activity in central nervous system, which in turn increases norepinephrine leading to hypertension. Several statistical analyses were undertaken based on this theory, but we found no significant association between hypothyroidism and hypertension (Table-3, 4, 5).
As regards to age, it has been observed that TSH increased with age [5,26] though with much genetic variation. We found that age had no significant correlations with TSH and FT4 (Table-5, Figure 1). Neither, we found any significant difference of age between participants with and without hypothyroidism (Table-4).
The study has some limitations. Had we clinically examine those who had high TSH level we could have identified the common signs or symptoms related to hypothyroidism which could help physician to look into clinical features cautiously. Secondly, we could have assayed free triiodothyronine (FT3), thyroid peroxidase antibody (anti-TPO) and reverse thyroxine (rt3) for more reliable thyroid dysfunction.