Chronic kidney disease was demonstrated to increase all cause and cardiovascular disease mortality. Prior report revealed the negative
correlation between thyroid stimulating hormone(TSH) and estimated glomerular filtration ratio(eGFR). Our aim was to demonstrate the relationship
between elevated TSH levels and the prevalence of CKD among elder Taiwan Chinese.
Results
A total of 195 people were selected into analysis. Subjects with higher TSH were 3.05±1.75 nIU/ml and those with lower TSH were 0.96±0.37
nIU/ml. There were 52 people corresponding to CKD definition. LnTSH was negatively associated to eGFR after adjusting possible confounders. The
odds ratios of CKD among higher TSH group was 2.31(1.05-5.11) after adjustment with age, gender, BMI, WC, social habits, chronic disease,
compared to lower TSH group. And this effect was more prominent among high WC subgroup.
S.-L.Hu
1,2,5
, C.-S.Liu
2,3
, C.-H..Lin
1,2
, C.-I.Li
3
, C.-C.Lin
1,3,4
, T.-C.Li
3,4
, W.-Y.Lin
2,3
胡松林、劉秋松、林志學、李佳霙、林正介、李采娟、林文元
1
中國醫藥大學暨附設醫院老年醫學科、
2
家庭醫學科、
3
醫學研究部；
4
中國醫藥
大學生物統計所、
5
醫學院老化醫學博士學位學程
4
Background
ELEVATED TSH WAS POSITIVELY CORRELATED TO THE
PREVALENCE OF CKD AMONG ELDER TAIWAN CHINESE
We recruited 200 elder people who consecutively attended the hospital for regular health exam. Basic data and laboratory report were collected.
Besides, thyroid function as thyroid stimulating hormone and free thyroxine(FT4) were recorded. Chronic kidney disease were defined as urine
protein more than 30 md/dL and/or eGFR less than 60 ml/min/1.73m2. eGFR was calculated by the Modification of Diet in Renal Disease equation.
FT4 between 0.54-1.40 ng/dL were selected. TSH was divided into higher TSH and lower TSH group according to equal number. WC was divided
into two group with cutoff value of 90 cm in men, and 80 cm in women. Student t was used for continuous values. Logistic linear regression was used
for predicting the odds ratio of CKD between higher and lower TSH groups.
Conclusions
Table 1 Baseline demographic characteristics according to TSH
level
＃CMUH
Higher TSH levels were related to the prevalence of CKD. Adequately controlling the insult of CKD was necessary.
Materials and methods
Keyword: thyroid stimulating hormone, chronic kidney disease, elder Taiwan Chinese
Lower TSH Group (N=97) Higher TSH Group (N=98) P§ Male (n,%) 44(45.4%) 48(49%) 0.613 Age (years) 76.3±6.1 76.9±6.6 0.518 Height (cm) 159.8±8.4 158.8±7.9 0.432 Body weight (kg) 59.6±11.0 61.9±13.0 0.197 WC(cm) 85.9±10.3 88.3±11.2 0.113 BMI (kg/m2) 23.27±3.33 24.39±3.85 0.032 SBP (mmHg) 135.5±18.9 138.2±15.7 0.277 DBP (mmHg) 76.7±11.6 77.3±10.8 0.711 MAP (mmHg) 96.3±13.6 97.6±11.6 0.473 ALT(U/L) 24.7±18.1 25.7±18.4 0.708 AST(U/L) 26.9±16.5 27.0±12.4 0.984 FPG (mg/dL) 108.4±25.1 106.2±20.8 0.491 TCHOL (mg/dL) 198.6±52.7 185.6±43.5 0.062 Triglycerides (mg/dL) 101.8±53.8 128.6±88.7 0.012 HDL-C(mg/dL) 55.0±14.3 49.8±11.2 0.005 Creatinine (mg/dL) 0.92±0.51 1.01±0.61 0.254 eGFR(ml/min) 78.9±22.3 72.5±21.3 0.042 Urine protein(mg) 8.8±30.5 17.1±64.9 0.251 TSH(nIU/mL) 0.96±0.37 3.05±1.75 0.000 Free T4(ng/dL) 0.97±0.15 0.89±0.14 0.000 Smoking (n,%) 0.643 Current 4(4.1%) 2(2%) Former 2(2.1%) 3(3.1%) Never 91(93.8%) 93(94.9%) Alcohol consumption (n, %) 0.509 Current 1(1.0%) 2(2.0%) Former 1(1.9%) 3(3.1%) Never 95(97.9%) 93(94.9%) Exercise (n, %) 0.801 Current 78(80.4%) 75(76.5%) Former 8(8.2%) 10(10.2%) Never 11(11.3%) 13(13.3%) HTN (n, %) 72 (74.2%) 75(76.5%) 0.709 T2DM (n, %) 91 (93.8%) 93(94.9%) 0.743 Dyslipidemia(n, %) 90 (92.8%) 95(96.9%) 0.188
Data are presented as mean ± SD or n (%)
Abbreviation: BMI, body mass index ; MAP, mean arterial pressure; SBP, systolic blood pressure; DBP, diastolic blood pressure; ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase; FPG, fasting plasma glucose, TCHOL, total cholesterol; HDL-C, high-density lipoprotein
cholesterol; Cr, creatinine; eGFR, estimated glomerular filtration rate; TSH: thyroid stimulating hormone
＃TSH level was divided into two groups according to population numbers. §Student’s test was used for comparing mean values of continuous variables between groups and Pearson χ† test was used for categorical data.
Table 2 Baseline demographic characteristics according to
chronic kidney disease
＃Normal (n=145) CKD (n=52) P§ Male (n,%) 44(45.4%) 48(49%) 0.613 Age (years) 75.6±6.1 79.1±6.2 0.000 Height (cm) 158.6±8.1 161.5±7.9 0.024 Body weight (kg) 58.8±11.8 66.2±11.4 0.000 BMI (kg/m2) 23.29±3.46 25.33±3.67 0.000 WC(cm) 85.6±10.0 91.4±11.8 0.001 SBP (mmHg) 135.1±16.3 141.5±19.3 0.021 DBP (mmHg) 76.1±10.4 79.1±13.0 0.100 MAP (mmHg) 95.8±11.8 99.9±14.1 0.042 ALT(U/L) 23.5±13.8 29.9±26.4 0.102 AST(U/L) 25.3±9.5 31.3±22.9 0.071 FPG (mg/dL) 104.9±19.7 113.3±29.7 0.064 TCHOL (mg/dL) 194.5±36.0 193.3±63.6 0.869 Triglycerides (mg/dL) 105.8±55.1 139.3±108.7 0.037 HDL-C(mg/dL) 54.3±12.4 48.2±12.4 0.003 Creatinine (mg/dL) 0.79±0.18 1.47±0.88 0.000 eGFR (ml/min) 83.6±16.3 53.4±21.0 0.000 Urine protein(mg) 3.0±6.4 42.4±93.3 0.004 TSH(nIU/mL) 1.99±2.39 3.05±4.34 0.032 Free T4(ng/dL) 0.93±0.14 0.91±0.17 0.444 Smoking (n,%) 0.643 Current 4(4.1%) 2(2.0%) Former 2(2.1%) 3(3.1%) Never 91(93.8%) 93(94.9%) Alcohol consumption (n,%) 0.509 Current 1(1.0%) 2(2.0%) Former 1(1.0%) 3(3.1%) Never 95(97.9%) 93 (94.9%) Exercise (n, %) 0.801 Current 78(80.4%) 75(76.5%) Former 8(8.2%) 10(10.2%) Never 11(11.3%) 13(13.3%) HTN (n, %) 111(76.6%) 37(71.2%) 0.440 T2DM (n, %) 140(96.6%) 46(88.5%) 0.029 Dyslipidemia(n, %) 135(93.1%) 51(98.1%) 0.180 Data are presented as mean ± SD or n (%)
Abbreviation: WC, waist circumference; BMI, body mass index ; MAP, mean arterial pressure; SBP, systolic blood pressure; DBP, diastolic blood pressure; ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase; FPG, fasting plasma glucose, TCHOL, total cholesterol; HDL-C, High-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; CKD, chronic kidney disease; TSH: thyroid stimulating hormone
＃CKD was defined as eGFR below 60 mL/min and/or urine protein
equal to or more than 30mg/dL. eGFR was calculated from Modification of Diet in Renal Disease equation.
§Student’s test was used for comparing mean values of continuous
variables between groups and Pearson χ test was used for categorical data Variables Model 1 (R2=0.129) Model 2δ (R2=0.217) Model 3δ,＃ (R2=0.319) Model 4δ,＃ (R2=0.239) Ln TSH -2.584±1.672 -3.274±1.851 1.023±3.220 -5.136±2.382* Age -0.770±0.244 † -0.735±0.254† -0.489±0.407 -0.872±0.368* Gender 9.465±3.081 † 9.700±3.642† 14.252±6.550* 10.868±5.540 BMI - -1.034±0.622 -1.531±1.004 -0.530±0.898 WC - 0.218±0.214 0.565±0.445 0.042±0.371 MAP - -0.144±0.124 -0.261±0.195 -0.101±0.177 Ln FPG - -4.733±9.040 -30.197±20.433 -1.395±10.800 Total cholesterol - 0.026±0.040 0.122±0.082 -0.014±0.049 Ln triglycerides - -8.914±3.894* -15.681±6.457* -4.633±5.500 HDL-C - -.0.35±0.156 -0.055±0.236 -0.076±0.223 Free T4 - -25.594±11.332* -6.402±17.126 -38.007±15.776*
Table 3 Multiple linear regression models showing regression coefficient
(β±S.E.) with eGFR as dependent variable, and the other listed variables as
independent variables.
*: p <0.05; †: p <0.01; ‡: p <0.001
Abbreviation: BMI, body mass index ; WC: waist circumference; MAP, mean arterial pressure; FPG, Fasting plasma glucose; TCHOL, total cholesterol; HDL-C, High-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; TSH: thyroid stimulating hormone; Free T4: free thyroxine
δModel 2, 3 & 4 were additionally adjusted with social habit of cigarette smoking,
alcohol consumption and exercise.
＃Model 3 & 4: divided into low WC group and high WC group with cutoff value of
90 cm in men, and 80 cm in women.
Total Model 1 Model 2 Model 3
Lower TSH 1.00(Reference) 1.00(Reference) 1.00(Reference)
Higher TSH 2.78(1.37-5.64) † 2.30(1.10-4.83) * 2.31(1.05-5.11) *
Low WCε
Lower TSH 1.00(Reference) 1.00(Reference) 1.00(Reference)
Higher TSH 1.50(0.50-4.56) 1.42(0.41-4.90) 1.25(0.34-4.54)
High WCε
Lower TSH 1.00(Reference) 1.00(Reference) 1.00(Reference)
Higher TSH 5.12(1.84-14.20) † 5.38(1.76-16.41) † 6.96(1.95-24.81) †
Table 4. Odds ratio (95% confidence interval) for chronic kidney
disease in different models derived from a stepwise multiple logistic
regression analysis using TSH group as independent variables,
adjusted for potential confounders.
*: p <0.05; †: p <0.01
Model 1: adjusted for age and gender
Model 2: model 1 + adjusted for social history (smoking, alcohol consumption, and exercise habit) body mass index, and waist circumference
Model 3: model 2 + adjusted for chronic disease (hypertension, diabetes, dyslipidemia)
εWC was divided into two group with cutoff value of 90 cm in men, and 80 cm
