The definition of chronic kidney disease (CKD) is based

Transcription

1 Abnormal Thyroid-Stimulating Hormone and Chronic Kidney Disease in Elderly Adults in Taipei City Mei-hsing Chuang, MD,* Kuo-Meng Liao, PhD, Yao-Min Hung, MPH, Paul Yung-Pou Wang, MD, k Yi-Chang Chou, MPH, and Pesus Chou, DrPH From the *Division of Family Medicine, Department of Community Medicine, Taipei City Hospital; Division of Endocrinology and Metabolism, Department of Internal Medicine, Taipei City Hospital; Department of Education and Research, Taipei City Hospital, Taipei; Department of Emergency Medicine, Kaohsiung Veterans General Hospital, Kaohsiung; Institute of Public Health and Community Medicine Research Center, National Yang-Ming University, Taipei, Taiwan; and k Division of Nephrology, Kaiser Permanente Baldwin Park Medical Center, Baldwin Park, California. The copyright line for this article was changed on 28 June 2016 after original online publication. Address correspondence to Pesus Chou, Institute of Public Health and Community Medicine Research Center, National Yang-Ming University, No. 155, Section 2, Linong Street, Taipei 11221, Taiwan. pschou@ym.edu.tw DOI: /jgs OBJECTIVES: To examine whether older people with abnormal thyroid function are more likely to develop chronic kidney disease (CKD) over a 5-year follow-up period. DESIGN: Retrospective cohort study. SETTING: Health examination data from the Taipei Databank for Public Health Analysis. PARTICIPANTS: Individuals aged 65 and older (N = 41,454). MEASUREMENTS: Thyroid-stimulating hormone (TSH) levels were repeatedly measured, and subjects were categorized into four thyroid function groups (hyperthyroid, euthyroid, subclinical hypothyroid, overt hypothyroid). The risk of incident CKD was evaluated using a stepwise Cox proportional hazards regression model adjusted for sex, baseline age, hypertension, diabetes mellitus (DM), dyslipidemia, hyperuricemia, anemia, obesity, liver function, smoking, and alcohol. RESULTS: Higher TSH levels were associated with greater risk of subsequent CKD. Individuals with subclinical hypothyroidism (hazard ratio (HR) = 1.15, 95% confidence interval (CI) = ) and those with overt hypothyroidism (HR = 1.27, 95% CI = ) were more likely than those who were euthyroid to have CKD. Women were more likely to have CKD than men (HR = 1.11, 95% CI = ). When stratified by gender, subclinical hypothyroidism in women was associated with an increased risk of developing CKD (HR = 1.22; 95% CI = ). When stratified by DM, subclinical hypothyroidism and overt hypothyroidism were associated with an increased risk of developing CKD in nondiabetics (HR = 1.19; 95% CI = ; and HR = 1.34; 95% CI = , respectively). CONCLUSION: This cohort study of elderly persons in Taipei City found a significant association between hypothyroidism and development of CKD in women and individuals without DM. J Am Geriatr Soc 64: , Key words: thyroid-stimulating hormone; chronic kidney disease; glomerular filtration rate; hypothyroidism; proteinuria The definition of chronic kidney disease (CKD) is based on information including glomerular filtration rate (GFR), urinalysis, radiological studies, and occasionally kidney biopsy. The prevalence of CKD ranges from 8% to 16% worldwide. 1 CKD has important public health consequences. Individuals with CKD are at greater risk of all-cause and cardiovascular mortality, 2 and CKD is associated with greater healthcare use. 3 Individuals with end-stage renal disease (ESRD) use a disproportionately large amount of healthcare resources. 4 The causes of CKD include diabetes mellitus (DM), hypertension, environmental toxins, pesticides, analgesic abuse, herbal medications, and the use of unregulated food additives. 5 The influence of thyroid function on CKD has not been widely investigated. Thyroid hormone is secreted from the thyroid gland and is controlled by the thyroid-stimulating hormone (TSH) of the pituitary gland. Thyroid hormones affect kidney function by mediating effects on cardiac output and renal blood flow, contributing to changes in GFR. Thyroid hormones also have direct effects on the kidney by influencing GFR, tubular secretory and reabsorptive functions, and electrolyte homeostasis. Hypothyroidism leads to high serum creatinine as a result of the reduction in GFR JAGS 64: , The Authors. The Journal of the American Geriatrics Society published by Wiley Periodicals, Inc. on behalf of The American Geriatrics Society /16/$15.00 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

2 1268 CHUANG ET AL. JUNE 2016 VOL. 64, NO. 6 JAGS Recent studies have suggested that there may be an association between thyroid function and CKD. Several case reports have demonstrated that renal impairment secondary to hypothyroidism is reversible after thyroxin replacement Several cross-sectional studies have found subclinical and clinical hypothyroidism to be associated with low estimated GFR (egfr) and CKD. Low thyroid function, even within the clinically normal range, is associated with low GFR. 14,15 One large prospective cohort study in South Korea found that high levels of TSH were modestly associated with greater risk of CKD in euthyroid individuals. 16 However, another longitudinal study questioned the causal relevance of low thyroid function in the decline of renal function in older adults. 17 Few longitudinal studies have evaluated the association between hypothyroidism and CKD. The purpose of this study was to examine whether older adults with abnormal thyroid function were more likely to develop CKD over 5 years of follow-up. METHODS Data Source The Taipei City government offers free annual health examinations for citizens aged 65 and older. They can receive examinations at several public and private hospitals approved by the Department of Health of Taipei City. The Taipei City Hospital, in cooperation with the Department of Health, Taipei City, established a public health database (Taipei Databank for Public Health Analysis (DoPHA)). The Taipei City Hospital can provide data including demographic characteristics, height, weight, medical history, medication history, alcohol consumption, smoking status, blood pressure, serum biochemistry, TSH, complete blood count, and routine urine measurements, to researchers who have obtained ethical approval. TSH levels were measured using a third-generation assay and provided every other year. Study Subjects This was a retrospective cohort study based on the Taipei Databank for Public Health Analysis to estimate the risk of CKD according to thyroid function over 5 years of follow-up. The study cohort included 101,137 elderly adults who had received health examinations between 2005 and Individuals who had CKD at baseline (n = 26,995), had only one examination from 2005 to 2010 (n = 30,053), received medication for thyroid disease (n = 907), had missing serum TSH data (n = 421), were younger than 65 (n = 275) or older than 100 (n = 8), and had abnormal TSH levels that returned to normal (n = 1,024) were excluded (n = 59,683) (Figure 1), leaving 41,454 subjects for the analysis. The institutional review board of Taipei City Hospital approved this study. CKD Definition CKD is defined as an egfr of less than 60 ml/min per 1.73 m 2 or 60 ml/min per 1.73 m 2 or greater with proteinuria of 1+ or greater. GFR levels were estimated using Figure 1. Flowchart of study subjects. Hyperthyroid: thyroidstimulating hormone (TSH) miu/l, euthyroid: TSH miu/l, subclinically hypothyroid: TSH 5 10 miu/l, overtly hypothyroidism: TSH miu/l. CKD = chronic kidney disease. the four-variable version of the Modification of Diet in Renal Disease formula as follows: e-gfr (ml/min per 1.73 m 2 ) = serum creatinine age (if female). 18 Possible confounders in the relationship between TSH and CKD include hypertension, DM, dyslipidemia, hyperuricemia, abnormal liver function, anemia, obesity, smoking, and alcohol drinking. 2,19 Hypertension was defined as systolic blood pressure of 140 mmhg or greater, diastolic blood pressure of 90 mmhg or greater, or receiving antihypertensive medications, or based on history. Diabetes mellitus was defined as fasting serum glucose of 126 mg/ dl or greater, receiving antidiabetic medication, or based on history. Dyslipidemia was defined as low high-density lipoprotein cholesterol (HDL-C) (<40 mg/dl for men, <50 mg/dl for women), high low-density lipoprotein cholesterol (LDL-C) (>130 mg/dl), or high triglycerides ( 200 mg/dl). Hyperuricemia was defined as uric acid 7.9 mg/dl or greater. Abnormal liver function was defined as glutamic-pyruvic transaminase (GPT) greater than 42 U/L. Anemia was defined as <13 g/dl for men and <12 g/dl for women. Body mass index (BMI) was calculated as weight divided by height squared (kg/m 2 ). Obesity was defined as BMI 27 kg/m 2. History of smoking was defined as never, occasional, or frequent smoking. History of drinking was defined as never, occasional, or frequent drinking. Thyroid Function Status Thyroid status was defined according to general clinical consensus Normal plasma TSH levels were defined as between 0.1 miu/l and 5 miu/l, abnormally low as 0.1 miu/l or less, and abnormally high TSH as 5 miu/l or greater.

3 JAGS JUNE 2016 VOL. 64, NO. 6 ABNORMAL TSH AND CKD 1269 Based on plasma TSH levels, subjects were categorized as hyperthyroid if they had one or more TSH value of 0.1 miu/l, euthyroid if they had persistent TSH values of 0.1 less than 5 miu/l, subclinically hypothyroid if they had one or more TSH value of 5 10 miu/l, and overtly hypothyroid if they had one or more TSH value of miu/l. The euthyroid group was the reference group. Statistical Analysis Statistical analysis was conducted using SAS version 9.3 (SAS Institute, Inc., Cary, NC). Two-sided P <.05 was considered statistically significant. The categorical variables are presented as frequencies and percentages, and continuous variables are presented as means and standard deviations. One-way analysis of variance and the chisquare test were used to compare differences in mean ages and categorical variables between the thyroid function groups. In univariate analysis, differences in means of continuous variables and percentages of categorical variables were examined using two-sample t-tests and chi-square tests. Stepwise Cox proportional hazards regression adjusted for sex, age, hypertension, DM, dyslipidemia (low HDL-C, high LDL-C, hypertriglyceridemia), hyperuricemia, abnormal liver function, anemia, obesity, smoking, and alcohol drinking were used to compare differences between these groups in the development of CKD. Table 1. Baseline Participant Characteristics According to Thyroid Status (N = 41,454) Characteristic Euthyroidism (TSH 0.1 <5 miu/l), n = 39,267 (94.7%) Hyperthyroidism (TSH 0 <0.1 miu/l), n = 179 (0.4%) Subclinical Hypothyroidism (TSH 5 <10 miu/l), n = 1,700 (4.1%) Overt Hypothyroidism (TSH miu/l), n = 308 (0.8%) P-Value Age, mean standard deviation <.01 Sex, n (%) Male 20,494 (95.2) 59 (0.3) 807 (3.7) 171 (0.8) <.01 Female 18,773 (94.2) 120 (0.6) 893 (4.5) 137 (0.7) Hypertension, n (%) Yes 11,500 (94.5) 64 (0.5) 512 (4.2) 94 (0.8).20 No 27,767 (94.8) 115 (0.4) 1,188 (4.1) 214 (0.7) Diabetes mellitus, n (%) Yes 5,118 (94.5) 32 (0.6) 239 (4.4) 28 (0.5).02 No 34,149 (94.8) 147 (0.4) 1,461 (4.0) 280 (0.8) Total cholesterol 200 mg/dl, n (%) Yes 16,359 (94.9) 59 (0.4) 710 (4.1) 109 (0.6).01 No 22,908 (94.6) 120 (0.5) 990 (4.1) 199 (0.8) Triglycerides 200 mg/dl, n (%) Yes 3,364 (94.0) 9 (0.3) 172 (4.8) 33 (0.9).02 No 35,903 (94.8) 170 (0.5) 1,528 (4.0) 275 (0.7) High-density lipoprotein cholesterol <40 mg/dl for men, <50 mg/dl for women, n (%) Yes 9,290 (93.9) 57 (0.6) 466 (4.7) 82 (0.8) <.001 No 29,977 (95.0) 122 (0.4) 1,234 (3.9) 226 (0.7) Low-density lipoprotein cholesterol 130 mg/dl, n (%) Yes 12,025 (94.9) 40 (0.3) 517 (4.1) 88 (0.7).09 No 27,242 (94.6) 139 (0.5) 1,183 (4.1) 220 (0.8) Uric acid 7.9 mg/dl, n (%) Yes 1,968 (94.5) 8 (0.4) 90 (4.3) 16 (0.8).90 No 37,299 (94.7) 171 (0.4) 1,610 (4.1) 292 (0.8) Glutamic-pyruvic transaminase >42 U/L, n (%) Yes 2,347 (94.2) 12 (0.5) 106 (4.2) 27 (1.1).20 No 36,920 (94.8) 167 (0.4) 1,594 (4.1) 281 (0.7) Hemoglobin <13 g/dl men, <12 g/dl women, n (%) Yes 6,183 (92.8) 36 (0.6) 362 (5.4) 82 (1.2) <.01 No 33,084 (95.1) 143 (0.4) 1,338 (3.9) 226 (0.6) Body mass index 27 kg/m 2, n (%) Yes 6,563 (94.6) 22 (0.3) 312 (4.5) 44 (0.6).07 No 32,704 (94.8) 157 (0.4) 1,388 (4.0) 264 (0.8) Smoking, n (%) Nonsmoker 36,619 (94.7) 164 (0.4) 1,624 (4.2) 287 (0.7) <.01 Frequent smoker 2,648 (95.9) 15 (0.5) 76 (2.8) 21 (0.8) Alcohol drinking, n (%) Nondrinker 31,531 (94.5) 159 (0.5) 1,410 (4.2) 261 (0.8).002 Occasional 6,968 (95.6) 16 (0.2) 261 (3.6) 43 (0.6) Frequent drinker 768 (95.4) 4 (0.5) 29 (3.6) 4 (0.5) P-values were calculated using chi-square tests for categorical covariates and analysis of variance for continuous covariates. TSH = thyroid-stimulating hormone.

4 1270 CHUANG ET AL. JUNE 2016 VOL. 64, NO. 6 JAGS RESULTS Median follow-up was 3 years. Of 41,454 participants, 39,267 (97.4%) were classified as euthyroid, 179 (0.4%) as hyperthyroid, 1,700 (4.1%) as subclinically hypothyroid, and 308 (0.8%) as overtly hypothyroid (Table 1). The overtly hypothyroid group had a higher mean age ( ) than the other three groups (P <.001). Women were more likely to be hyperthyroid (0.6% vs 0.3%) and subclinically hypothyroid (4.5% vs 3.7%) than men (P <.001). Ten thousand eighty of 41,454 subjects (24.3%), 9,458 of 39,267 euthyroid subjects (24.1%), 40 of 179 hyperthyroid subjects (22.4%), 485 of 1,700 subclinically hypothyroid subjects, (28.5%), and 97 of 308 overtly hypothyroid subjects (31.5%) developed CKD (Table 2). Subjects with DM (30.2% vs 23.4%), dyslipidemia (hypertriglyceridemia: 30.0% vs 23.8%; low HDL-C: 27.4% vs 23.4%), smoking (26.1% vs 24.4%), or obesity (28.4% vs 23.5%) had a greater risk of developing incident CKD (Table 2). Women had a greater risk of developing incident CKD than men (HR = 1.11, 95% CI = ). Age was associated with developing incident CKD (HR = 1.02, 95% CI = ). The adjusted HRs for incident CKD in the different thyroid function groups are shown in Table 3. Individuals with hyperthyroidism (HR = 1.07, 95% CI = ), subclinical hypothyroidism (HR = 1.15, 95% CI = ), and overt hypothyroidism (HR = 1.27, 95% CI = ) were more likely to develop CKD than euthyroid individuals (P for trend test <.001). When stratified by sex, subclinical hypothyroidism was associated with an increased risk of developing CKD in women (HR = 1.22, 95% CI = ). When stratified by DM, subclinical hypothyroidism and overt hypothyroidism were associated with an increased risk of developing CKD in nondiabetics (HR = 1.19, 95% CI = ; HR = 1.34, 95% CI = , respectively). DISCUSSION In this retrospective observational study of a Taipei City-based cohort of older adults, higher TSH levels, which represent subclinical hypothyroidism and overt hypothyroidism, were associated with greater risk of developing CKD. No significant association was found between hyperthyroidism and development of CKD. It was also found that sex and DM may modify the relationship between hypothyroid status and CKD. Women and persons without DM were more susceptible to the effects of hypothyroidism. In several previous studies of individuals with renal function impairment secondary to hypothyroidism, all participants had improvement in renal function after thyroid hormone replacement therapy. These findings suggested that impairment of renal function might be an overlooked manifestation of hypothyroidism. Several cross-sectional studies found an association between high TSH levels and low GFR. 14,15 Studies conducted in Australia and Norway have demonstrated that low thyroid function, even within the clinically normal range, is associated with higher odds of CKD in adults. Table 2. Univariate Analysis of Predictors of Chronic Kidney Disease (CKD) Characteristic Total, N = 41,454 CKD, n = 10,080 P-Value Age, meansd <.01 Sex, n (%) Male 21,531 (51.9) 5,368 (24.9) <.01 Female 19,923 (48.1) 4,712 (23.7) Thyroid function (thyroid-stimulating hormone level, miu/l) Euthyroidism 39,267 (94.7) 9,458 (24.1) <.01 (0.1 <5) Hyperthyroidism 179 (0.4) 40 (22.4) (0 <0.1) Subclinical 1,700 (4.1) 485 (28.5) hypothyroidism (5 <10) Overt 308 (0.8) 97 (31.5) hypothyroidism (10 99) Hypertension, n (%) Yes 12,170 (29.4) 2,898 (23.8).15 No 29,284 (70.6) 7,182 (24.5) Diabetes mellitus, n (%) Yes 5,417 (13.1) 1,636 (30.2) <.01 No 36,037 (86.9) 8,444 (23.4) Total cholesterol 200 mg/dl, n (%) Yes 17,237 (41.6) 3,984 (23.1) <.01 No 24,217 (58.4) 6,096 (25.2) Triglycerides 200 mg/dl, n (%) Yes 3,578 (8.6) 1,072 (30.0) <.01 No 37,876 (91.4) 9,008 (23.8) High-density lipoprotein cholesterol <40 mg/dl for men, <50 mg/dl for women, n (%) Yes 9,895 (23.9) 2,706 (27.4) <.01 No 31,559 (76.1) 7,374 (23.4) Low-density lipoprotein cholesterol 130 mg/dl, n (%) Yes 12,670 (30.6) 2,926 (23.1) <.01 No 28,784 (69.4) 7,154 (24.9) Uric acid 7.9 mg/dl, n (%) Yes 2,082 (5.0) 904 (43.4) <.01 No 39,372 (95.0) 9,176 (23.3) Glutamic-pyruvic transaminase >42 U/L, n (%) Yes 2,492 (6.0) 673 (27) <.01 No 38,962 (94.0) 9,407 (24.1) Hemoglobin <13 g/dl men, <12 g/dl women, n (%) Yes 6,663 (16.1) 2,209 (33.2) <.01 No 34,791 (83.9) 7,871 (22.6) Body mass index 27 kg/m 2, n (%) Yes 6,941 (16.7) 1968 (28.4) <.01 No 34,513 (83.3) 8,112 (23.5) Smoking, n (%) Nonsmoker 38,694 (93.3) 9,682 (24.4).02 Frequent smoker 2,760 (6.7) 720 (26.1) Alcohol drinking, n (%) Nondrinker 33,361 (80.5) 8,189 (24.6).06 Occasional 7,288 (17.6) 1,693 (23.2) Frequent drinker 805 (1.9) 198 (24.6) P-values were calculated using chi-square tests for categorical covariates, and two-sample t-tests for continuous covariates. The Norwegian study also showed that subclinical and overt hypothyroidism were associated with greater prevalence of CKD. Subclinical and overt hypothyroidism might be associated with prevalent CKD, but there is little evidence

5 JAGS JUNE 2016 VOL. 64, NO. 6 ABNORMAL TSH AND CKD 1271 Table 3. Multivariate Cox Regression Model for Predictors of Chronic Kidney Disease Hazard Ratio (95% Confidence Interval) Sex DM Variable Total Male Female Non-DM DM Age 1.02 ( ) 1.02 ( ) 1.03 ( ) 1.02 ( ) 1.02 ( ) Female 1.11 ( ) 1.12 ( ) Thyroid function (reference euthyroidism) Hyperthyroidism 1.07 ( ) 0.99 ( ) 1.08 ( ) Subclinical hypothyroidism 1.15 ( ) 1.22 ( ) 1.19 ( ) Hypothyroidism 1.27 ( ) 1.31 ( ) 1.34 ( ) Hypertension 1.17 ( ) 1.18 ( ) 1.15 ( ) 1.17 ( ) 1.17 ( ) Diabetes mellitus 1.28 ( ) 1.29 ( ) 1.29 ( ) High triglycerides 1.22 ( ) 1.23 ( ) 1.24 ( ) 1.20 ( ) 1.28 ( ) Low high-density lipoprotein cholesterol 1.11 ( ) 1.17 ( ) 1.09 ( ) 1.19 ( ) High low-density lipoprotein cholesterol 1.08 ( ) Hyperuricemia 2.05 ( ) 2.00 ( ) 2.17 ( ) 2.09 ( ) 1.92 ( ) High glutamic-pyruvic transaminase 1.10 ( ) 1.17 ( ) 1.15 ( ) Obesity 1.18 ( ) 1.18 ( ) 1.19 ( ) 1.19 ( ) 1.18 ( ) Anemia 1.41 ( ) 1.42 ( ) 1.42 ( ) 1.39 ( ) 1.47 ( ) Frequent smoker 1.21 ( ) 1.17 ( ) 1.24 ( ) 1.19 ( ) 1.32 ( ) Alcohol drinking (reference nondrinker) Occasional drinker 0.79 ( ) Frequent drinker 1.10 ( ) Stepwise Cox proportional hazard regression used. that hypothyroidism is independently associated with CKD. One large prospective study conducted in South Korea found that there was a modestly greater risk of incident CKD in euthyroid persons with high-normal TSH levels. 16 Another recent longitudinal study conducted in the Netherlands found no association between low thyroid function and low renal function in older adults. 17 The current study aimed to explore the association between abnormal thyroid function and onset of CKD. A statistically significant association between hypothyroidism and incident CKD was found. Low thyroid function affects renal function by decreasing cardiac output and increasing systemic vascular resistance, which leads to low renal blood flow. 8 10,25 Low thyroid function also causes low creatinine clearance and high serum creatinine concentrations. 26,27 Hyperthyroidism has effects on renal function. Initially, high thyroid function may increase renal blood flow and GFR, but it ultimately results in high serum creatinine and low GFR through several pathways, including intraglomerular hypertension, proteinuria, and oxidative stress The current study revealed that the association between hyperthyroidism and CKD was not significant. It is likely that this observation is the result of the small number of cases of incident CKD in the hyperthyroid group. Men were at a lower risk of developing CKD than women after adjusting for covariates. Women with hypothyroidism were at even higher risk of developing incident CKD than men. Several surveys in the United States, Japan, Australia, and the Netherlands have shown that women have a greater prevalence of CKD than men. The prevalence of hypothyroidism in women was greater than that of men A previous study showed that, at physiological free thyroxine (free T 4 ) concentrations, men had higher median TSH levels than women. 21 The response of TSH to changes in free T 4 was also different between men and women. The mechanisms are not well understood. Because men have higher TSH levels than women for equivalent free T 4 levels, the hypothyroid group of men might contain fewer individuals with overt hypothyroidism than the hypothyroid group of women in the study. This observation might explain sex disparities in the association between hypothyroidism and onset of CKD. The current findings suggest that age is a major risk factor for CKD. Earlier studies have demonstrated that TSH progressively increases over time in older individuals. 21,23,38,39 TSH distribution shifted to a higher concentration with age. 21 Another longitudinal study revealed similar results in older adults, and the higher TSH levels were not associated with greater mortality. 38 These findings have suggested that the prevalence of hypothyroidism may be overestimated in elderly adults and that the effects of comorbidities and age might supersede the influence of hypothyroidism on CKD in the oldest age group. The current study found a significant association between hypothyroidism and onset of CKD. In the fully adjusted model in the hypothyroid group, the association between hypothyroidism and onset of CKD in the group without DM was significant. The study showed that the effects of comorbidities and DM can supersede the influence of hypothyroidism on CKD in individuals with DM. The study was repeated using TSH levels that were at least twice the value of abnormally high or abnormally low values. The number of participants was 33,602. The results were similar.

6 1272 CHUANG ET AL. JUNE 2016 VOL. 64, NO. 6 JAGS Strengths and Limitations This study has several strengths. First, participants were classified based on repeated measures of TSH, and those with abnormal TSH values that spontaneously returned to normal were excluded. Second, the large sample of community-dwelling older people increased the validity of this study. Third, extensive clinical data were collected to control for multiple confounders of CKD. Several limitations of the study must be considered. First, the databank provided only TSH values. Free T 4 was not measured. The lack of free T 4 measurements prohibited determination of whether the greater risk was present with overt hypothyroidism or with subclinical hypothyroidism. Second, several factors are associated with decline in kidney function. The data source did not contain detailed prior history of other kidney diseases, such as polycystic kidney disease and glomerular disease, that might induce the progressive decline in GFR over time. Third, several types of medications, such as amiodarone, may interfere with thyroid function and several types of medications, such as nonsteroidal anti-inflammatory drugs, may interfere with renal function, but data regarding participants medication history were limited. Fourth, the study sample was obtained from community-dwelling older people but was restricted to those who volunteered to undergo follow-up examinations. The lower participation rate of sicker elderly adults could bias the estimates. The facts that several institutionalized adults participated in the annual health examinations, that 5.1% of the subjects were in the welfare program, and that approximately 0.73% of subjects were disabled might add diversity to the sample and may not truly represent the common population sample. Finally, because of the observational nature of the data, a causal relationship between hypothyroidism and CKD could not be inferred. CONCLUSION This cohort study of Taipei City elderly adults found an association between hypothyroidism and development of CKD in women and participants without DM. This finding suggests that thyroid function should be periodically evaluated in elderly adults to detect those individuals with abnormal thyroid function. Additionally, persons with poor renal function should have thyroid laboratory tests routinely because they are more likely to develop hypothyroidism. Thyroid and renal function should be carefully monitored in persons with mildly abnormal TSH levels because they are at risk of deterioration of renal function. ACKNOWLEDGMENTS This study is based on data from the Taipei City Public Health Database provided by the Department of Health, Taipei City Government, and managed by the Databank for Public Health Analysis. The interpretation and conclusions contained in the present study do not represent those of the Department of Health, Taipei City Government, or the Databank for Public Health Analysis. Conflict of Interest: All of the authors declare that they have no competing interests to disclose. Author Contributions: Chuang: study concept and design, data analysis and interpretation, manuscript preparation. Liao: study concept and design, data analysis and interpretation, critical revision of manuscript. Hung: study concept and design, data interpretation, critical revision of manuscript. Wang: study concept, data interpretation, critical revision of manuscript. Y. Chou: data acquisition, analysis, and interpretation; critical revision of manuscript. P. Chou: study concept and design, data analysis and interpretation, revision of manuscript for accuracy of intellectual content, final approval of manuscript. Sponsor s Role: None. REFERENCES 1. Jha V, Garcia-Garcia G, Iseki K et al. Chronic kidney disease: Global dimension and perspectives. Lancet 2013;382: Wen CP, Cheng TY, Tsai MK et al. All-cause mortality attributable to chronic kidney disease: A prospective cohort study based on 462,293 adults in Taiwan. Lancet 2008;371: Khan SS, Kazmi WH, Abichandani R et al. Health care utilization among patients with chronic kidney disease. Kidney Int 2002;62: Wish J, Schulman K, Law A et al. Healthcare expenditure and resource utilization in patients with anaemia and chronic kidney disease: A retrospective claims database analysis. Kidney Blood Press Res 2009;32: Jha V. End-stage renal care in developing countries: The India experience. Ren Fail 2004;26: Basu G, Mohapatra A. Interactions between thyroid disorders and kidney disease. Indian J Endocrinol Metab 2012;16: den Hollander JG, Wulkan RW, Mantel MJ et al. Correlation between severity of thyroid dysfunction and renal function. Clin Endocrinol (Oxf) 2005;62: Mohamedali M, Reddy Maddika S, Vyas A, et al. Thyroid disorders and chronic kidney disease. Int J Nephrol 2014;2014: Iglesias P, Diez JJ. Thyroid dysfunction and kidney disease. Eur J Endocrinol 2009;160: Mariani LH, Berns JS. The renal manifestations of thyroid disease. J Am Soc Nephrol 2012;23: Mooraki A, Broumand B, Neekdoost F et al. Reversible acute renal failure associated with hypothyroidism: Report of four cases with a brief review of literature. Nephrology 2003;8: Connor A, Taylor JE. Renal impairment resulting from hypothyroidism. NDT Plus 2008;1: Hataya Y, Igarashi S, Yamashita T et al. Thyroid hormone replacement therapy for primary hypothyroidism leads to significant improvement of renal function in chronic kidney disease patients. Clin Exp Nephrol 2013;17: Gopinath B, Harris DC, Wall JR et al. Relationship between thyroid dysfunction and chronic kidney disease in community-dwelling older adults. Maturitas 2013;75: Asvold BO, Bjoro T, Vatten LJ. Association of thyroid function with estimated glomerular filtration rate in a population-based study: The HUNT study. Eur J Endocrinol 2011;164: Zhang Y, Chang Y, Ryu S et al. Thyroid hormone levels and incident chronic kidney disease in euthyroid individuals: The Kangbuk Samsung Health Study. Int J Epidemiol 2014;43: Meuwese CL, Gussekloo J, Craen AJd et al. Thyroid status and renal function in older persons in the general population. J Clin Endocrinol Metab 2014;99: National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: Evaluation, classification, and stratification. Am J Kidney Dis 2002;39:S1 S Taal MW, Brenner BM. Predicting initiation and progression of chronic kidney disease: Developing renal risk scores. Kidney Int 2006;70: Bensenor IM, Olmos RD, Lotufo PA. Hypothyroidism in the elderly: Diagnosis and management. Clin Interv Aging 2012;7: Hadlow NC, Rothacker KM, Wardrop R et al. The relationship between TSH and free T(4) in a large population is complex and nonlinear and differs by age and sex. J Clin Endocrinol Metab 2013;98: Rosario PW. Natural history of subclinical hyperthyroidism in elderly patients with TSH between 0.1 and 0.4 miu/l: A prospective study. Clin Endocrinol (Oxf) 2010;72:

7 JAGS JUNE 2016 VOL. 64, NO. 6 ABNORMAL TSH AND CKD Surks MI, Hollowell JG. Age-specific distribution of serum thyrotropin and antithyroid antibodies in the U.S. population: Implications for the prevalence of subclinical hypothyroidism. J Clin Endocrinol Metab 2007;92: Surks MI, Boucai L. Age- and race-based serum thyrotropin reference limits. J Clin Endocrinol Metab 2010;95: Villabona C, Sahun M, Roca M et al. Blood volumes and renal function in overt and subclinical primary hypothyroidism. Am J Med Sci 1999;318: Karanikas G, Schutz M, Szabo M et al. Isotopic renal function studies in severe hypothyroidism and after thyroid hormone replacement therapy. Am J Nephrol 2004;24: Montenegro J, Gonzalez O, Saracho R et al. Changes in renal function in primary hypothyroidism. Am J Kidney Dis 1996;27: Coresh J, Selvin E, Stevens LA et al. Prevalence of chronic kidney disease in the United States. JAMA 2007;298: Iseki K. Chronic kidney disease in Japan. Intern Med 2008;47: Chadban SJ, Briganti EM, Kerr PG et al. Prevalence of kidney damage in Australian adults: The AusDiab kidney study. J Am Soc Nephrol 2003;14: S131 S Wetzels JF, Kiemeney LA, Swinkels DW et al. Age- and gender-specific reference values of estimated GFR in Caucasians: The Nijmegen Biomedical Study. Kidney Int 2007;72: Schindler AE. Thyroid function and postmenopause. Gynecol Endocrinol 2003;17: Vanderpump MP, Tunbridge WM, French JM et al. The incidence of thyroid disorders in the community: A twenty-year follow-up of the Whickham Survey. Clin Endocrinol (Oxf) 1995;43: Empson M, Flood V, Ma G et al. Prevalence of thyroid disease in an older Australian population. Intern Med J 2007;37: Wiener R, Utiger RD, Lew R et al. Age, sex, and serum thyrotropin concentrations in primary hypothyroidism. Acta Endocrinol (Copenh) 1991;124: Hollowell JG, Staehling NW, Flanders WD et al. Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 2002;87: Canaris GJ, Manowitz NR, Mayor G et al. The Colorado Thyroid Disease Prevalence Study. Arch Intern Med 2000;160: Waring AC, Arnold AM, Newman AB et al. Longitudinal changes in thyroid function in the oldest old and survival: The Cardiovascular Health Study All-Stars Study. J Clin Endocrinol Metab 2012;97: Over R, Mannan S, Nsouli-Maktabi H et al. Age and the thyrotropin response to hypothyroxinemia. J Clin Endocrinol Metab 2010;95: