Effects of Thyrotrophin, Thyroid Hormones and Thyroid Antibodies on Metabolic Parameters in a Euthyroid Population With Obesity

Sule Temizkan; Bilgken Balaforlou; Aysenur Ozderya; Mehmet Avci; Kadriye Aydin; Selin Karaman; Mehmet Sargin


Clin Endocrinol. 2016;85(4):616-623. 

In This Article


Study Population

In this study, we retrospectively evaluated 5300 consecutive subjects with obesity (aged 18–65 years) admitted to the Obesity Outpatient Clinic of Kartal Dr. Lutfi Kirdar Training and Research Hospital in Istanbul between January 2013 and June 2015. We excluded 4025 subjects with chronic illnesses and subjects using medication [cancer history (n = 52); chronic renal failure; chronic liver disease; pulmonary disease (n = 73); psychiatric disease and drug use (antidepressants, lithium, and other psychiatric drugs) (n = 337); rheumatological disease and drug use (n = 70), diabetes mellitus (n = 1034) and/or metformin use (n = 781); subclinical or overt thyroid dysfunction (n = 113); levothyroxine (LT4) use (n = 137); cardiac disease and drug use (n = 179); lipid-lowering drugs (n = 242), antihypertensive drug use (n = 1277), oral contraceptive drug use (n = 75) and pregnancy].Two hundred and fifty-five subjects were excluded for missing data (thyroid antibodies). Therefore, 1275 euthyroid subjects with obesity were eligible for the study. The study was conducted in agreement with the Declaration of Helsinki II. Kartal Dr. Lutfi Kirdar Training and Research Hospital Ethical Committee approved the study protocol. Informed consent was not required because of the retrospective nature of our study design.

Measurement of Anthropometric and Biochemical Parameters

Physical and biochemical test records of the subjects at first admission to the obesity outpatient clinic were examined. Subjects were evaluated for specific data at first admission: age, gender, history (presence of comorbidities and medication use) and physical examination [weight (kg), height (m)] and body mass index (BMI; kg/m2). Waist circumference was not available. Bioelectrical impedance (Jawon GAIA 359 Plus, Body Composition Analyser, Kyungsan, Korea) was performed on all subjects at first clinic admission to measure percentage body fat mass (PBF).

Blood tests were performed after overnight fasting. Biochemical and hormonal parameters [fasting blood glucose (FBG), glycosylated haemoglobin (HbA1C), fasting insulin (FI), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglyceride (TG), uric acid levels, creatinine, thyroid-stimulating hormone (TSH), free triiodothyronine (FT3), free thyroxine (FT4), antithyroglobulin (anti-TG) and antithyroid peroxidase (anti-TPO) antibodies, c-reactive protein (CRP) and white blood cell (WBC) counts] were simultaneously measured at first clinic admission.

Obesity was defined as BMI ≥30 kg/m2; grade I obesity was defined as BMI 30–34·9 kg/m2 and grade III (morbid) as BMI ≥40 kg/m2. Euthyroidism was defined as TSH values >0·4 and <4·5 μIU/ml and FT3 and FT4 in the normal reference range. Antibody positivity was defined as one or both thyroid antibody values being greater than the upper limit of the reference value. Insulin resistance (IR) was calculated using HOMA-IR. HOMA-IR = [fasting plasma insulin (μIU/ml) × fasting plasma glucose (mg/dl)]/405.[23] HOMA-IR ≥ 2·7 was accepted as subjects having IR.[24] Atherogenic dyslipidaemia was defined as TC/HDL-C ≥ 5.[25]

Laboratory Analysis

TSH was measured by a chemiluminescence immunoanalysis method (third-generation hypersensitive hTSH test) (Beckman Coulter Inc., Fullerton, CA, USA). The intra- and interassay coefficient of variation (CV%) was below 10%. The reference interval for TSH was 0·34–5·6 μIU/ml. FT4 and FT3 were measured by a chemiluminescence immunoanalysis method (Beckman Coulter Inc.). For both, the intra- and interassay CV% was below 10%. Reference values were 7·7 16 pmol/l and 3·1–5·8 pmol/l, respectively. Anti-TPO and anti-TG were measured by a chemiluminescence immunoanalysis method (Beckman Coulter Inc.). For both, the intra- and interassay CV% was below 10%. Upper range values for anti-TG and anti-TPO antibodies were 4·0 IU/ml and 9·0 IU/ml, respectively. Plasma venous glucose was measured using the hexokinase method. HbA1C was assayed by a HPLC method. Serum insulin levels were measured by immunoassay (Abbott Diagnostics, Santa Clara, CA, USA). Serum uric acid, TC, HDL-C and TG levels were measured by enzymatic calorimetric methods (Beckman Coulter Inc.). LDL-C was calculated using the Friedewald Formula [LDL-C = TC-(TG/5 + HDL-C)]. High-sensitivity CRP was assayed by the nephelometric method.

Statistical Analysis

Data are presented as mean ± standard deviation (SD) for continuous variables or median (25% and 75% interquartiles) for non-normally distributed variables. Normality of data distribution was assessed by the Kolmogorov–Smirnov test. Statistical significance was evaluated using the independent sample t-test for normally distributed variables, Mann–Whitney U-test for non-normally distributed variables and chi-squared test for categorical variables. In Table 2, analysis of covariance (ancova) was used to adjust results for age, gender, smoking status, thyroid antibody positivity (positive anti-TPO and/or anti-TG) and WBC and in Table 3 to adjust results for age, gender, smoking status, thyroid antibody positivity, FT3 and FT4. Pearson or Spearman's correlation was used according to the distribution of data (Table 4). In Table 5, logistic regression analysis was performed to determine independent predictors of IR (HOMA-IR ≥ 2·7) and atherogenic dyslipidaemia (TC/HDL-C ≥ 5) in all study groups. The Hosmer–Lemeshow goodness-of-fit statistics were used to assess the model fit. A 5% type 1 error level defined statistical significance.