Abnormal thyroid function predicts mortality in patients receiving long-term peritoneal dialysis: A case-controlled longitudinal study : Journal of the Chinese Medical Association

Secondary Logo

Journal Logo

Original Article

Abnormal thyroid function predicts mortality in patients receiving long-term peritoneal dialysis: A case-controlled longitudinal study

Lin, Yen-Chunga,e; Lin, Yi-Chunb,d; Chen, Tzen-Wena,e; Yang, Wu-Changc,d; Lin, Chih-Chingc,d,*

Author Information
Journal of the Chinese Medical Association: February 2012 - Volume 75 - Issue 2 - p 54-59
doi: 10.1016/j.jcma.2011.12.006


    1. Introduction

    The peritoneal dialysis (PD) survival rate has been reported to be good in Asian countries.1–3 The 5-year survival rate of PD patients is 70% in Japan. Type 2 diabetes mellitus (DM) is a common cause of end-stage renal disease (ESRD) and is a predictor of PD patient survival.4 Cardiovascular disease (CVD) is also the leading cause of mortality in patients with ESRD. Coronary artery calcification is being increasingly recognized in PD patients with multiple risk factors, including inflammation and malnutrition.5 In addition, dialysis patients also show low free triiodothyronine (T3) levels, which are caused by a diminished peripheral conversion of thyroxine (T4) to T3 and are linked with micro-inflammation.6 A previous epidemiology study has shows that increased prevalence of hypothyroidism in patients with ESRD requires hemodialysis.7 In addition, hypothyroidism is increased in patients with reduced glomerular filtration rate, from 5.3% to 20% in patients with estimated glomerular filtration rate <60 mL/minute/1.73 m2. The presence of goiter is also correlated with duration of treatment of PD patients.8

    Thus, the aim of the present study was to evaluate the following aspects in long-term PD patients: (1) factors affecting patient survival; and (2) the relationships between thyroid hormone levels, micro-inflammation status, and cardiac function.

    2. Methods

    2.1. Subjects

    We included those patients who had received PD therapy for >8 years between January 1, 1986 and September 30, 2010 in Taipei Veterans General Hospital, Taiwan. All the patients chose PD as the first modality for renal replacement therapy. The exclusion criteria included absence of data of thyroid function, hyperthyroidism, or having received thyroid hormone replacement. Finally, seven patients who were missing data of thyroid function were not included, and a total of 46 patients were enrolled in this retrospective study. None of the patients had residual renal function or psychiatric illness, or were receiving specific medication, including dopamine, glucocorticoid or phenytoin, that might affect thyroid function. Those who discontinued PD therapy after 8 years by shifting to hemodialysis (HD) or receiving a kidney transplant were censored. The research project was approved by the Institutional Review Board of Taipei Veterans General Hospital.

    2.2. Demographics and blood sampling

    To analyze the overall patient survival rate during the 25-year period, the following data were collected: sex, patient age at initial PD, presence of underlying renal disease, comorbidity, follow-up duration, cause of death, left and right ventricle ejection fractions, cardiothoracic ratio (CRT), rate of peritonitis (number of episodes/patient–months), and serum biochemical data. To assess underlying renal disease, patient history of chronic glomerular disease or interstitial disease was reviewed. To assess comorbidity, patient history of DM, CVD, and hypertension was examined. CVD was defined as a history of coronary, cerebrovascular, or peripheral vascular disease. Free T4, thyroid stimulating hormone (TSH), T3, left and right ventricle ejection fraction, and plain chest film were checked 3 months after the initiation of PD and followed by annual check-up; all these tests were parts of the routine in our Nephrology Department. The first single measurement of thyroid function was recorded as the data for statistical analysis. Peritoneal equilibration test was performed 1 month after the initiation of PD. The causes of discontinuation of PD or death were also recorded.

    2.3. Laboratory tests

    Abnormal thyroid function was defined as the presence of hypothyroidism: (1) subclinical hypothyroidism (TSH > 4.0 μIU/mL and normal levels of free T4 or free T4 < 0.59 ng/dL and normal TSH); or (2) overt hypothyroidism (free T4 < 0.59 ng/dL and TSH > 4 μIU/mL); or sick euthyroid syndrome: (1) low T4 syndrome (T4 < 4.5 μg/dL); or (2) low T3 syndrome (T3 < 95 ng/dL and normal levels of free T4 and TSH). The serum biochemical analysis included assessment of: hematocrit, serum albumin, serum creatinine, intact parathyroid hormone, normalized protein catabolic rate, peritoneal urea clearance (Kt/V urea), C-reactive protein (CRP), total T4 (SPAC T4 RIA Kit, Daiichi, Tokyo, Japan), total T3 (SPAC T3 RIA Kit; Incstar Corp., Stillwater, MN, USA), free T4 [GammaCoat Free T4 (two-step) 125I RIA Kit; Incstar Corp.], and TSH (GammaCoat hTSH 125I IRMA Kit; Incstar Corp.). The intra- and inter-assay coefficients of variation were 1.5% and 4.8% for TSH, 5.9% and 6.6% for T3, 3.3% and 5.0% for free T4, and 4.5% and 5.1% for T4, respectively. The sensitivities were 0.03 μIU/mL for TSH, 0.5 pg/mL for free T4, 0.025 μg/dL for T4, and 7 ng/dL for T3, respectively. Serum intact parathyroid hormone (i-PTH) level was measured using the N-tact PTH SP Kit (Incstar Corp.). The CTR was calculated as the ratio of the largest horizontal diameter of the heart divided by the largest internal diameter of the thorax wall on chest radiography. The ejection fraction of left and right ventricles was measured via radionuclide ventriculography. The reference ranges of the parameters were as follows: T4 = 4.5–11.0 μg/dL; T3 = 95–205 ng/dL; free T4 = 0.59–1.81 ng/dL; TSH = 0.25–4.0 μIU/mL; i-PTH = 10–57 pg/mL; ferritin = 30–400 ng/mL; high sensitivity CRP < 0.5 mg/dL; and CTR < 0.5. Serum biochemical parameters were measured using a computerized automated analyzer (Hitachi 736-60, Tokyo, Japan).

    2.4. Statistical analysis

    We performed statistical analyses using SPSS version 17.0 (SPSS Inc., Chicago, IL, USA). Pearson's χ2 test was used to analyze the categorical variables between the groups. An independent-samples t test was used to analyze the clinical and biochemical parameters of the groups, and the Kaplan–Meier method was used to analyze patient survival rate. Relationships between paired parameters were evaluated via Pearson correlation coefficients. We analyzed the hazard ratios using backward logistic regressions of the Cox proportional hazards method. Independent variables were selected for multivariate analysis if they had a p value < 0.15 in the univariate analysis. The level of significance was set at 0.05 for all tests.

    3. Results

    3.1. Patient characteristics

    A total of 46 patients were enrolled for this study. The mean age at initiation of PD was 44 years (standard deviation, 13.11 years; median, 44 years; range, 14–68 years). The mean duration of PD was 147.8 months (standard deviation, 48.3 months; median, 132 months; range, 96–298 months). Nineteen of the 46 (41.3%) patients had abnormal thyroid function tests. The frequency of abnormal thyroid function in these 46 patients was 20.9% for low T3, 25.5% for high TSH, 11.6% for low free T4, and 9.3% for low T4. Presence of hypothyroidism including subclinical hypothyroidism accounted for 28.2%, and sick euthyroid syndrome including free T3 or T4 syndrome accounted for 13.0%. Basic demographic data were not statistically different between the study and control groups regarding age at initiation of PD, duration of PD therapy, sex, and underlying disease (Table 1).

    Table 1:
    Baseline demographic characteristics in PD patients according to the status of thyroid function (n = 46).

    3.2. Patients' clinical parameters

    In the laboratory data, T3, T4, and TSH levels were significant lower in the patients with abnormal thyroid function (p = 0.02, 0.02, and 0.01, respectively). Left ventricular ejection fraction (LVEF) and peritoneal Kt/V were not significantly different between the two groups (p = 0.05). There were also no significant differences between the two groups when considering all other clinical parameters including levels of albumin, hemoglobin, ferritin, right ventricle ejection, and peritonitis rate (Table 2).

    Table 2:
    Comparison of baseline clinical parameters in PD patients according to the status of thyroid function* (n = 46).

    3.3. Patient survival analysis

    The cumulative survival rate in patients with abnormal thyroid function was shown by the Kaplan–Meier method (Fig. 1). Abnormal thyroid function was a significant predictor of patient survival (log rank, p = 0.02). In Cox proportional hazards analysis and logistical regression analysis, as shown in Table 3, age, DM, CVD and abnormal thyroid function were all significant predictors of cumulative patient mortality. Abnormal thyroid function (hazard ratio = 7.633, 95% confidence interval = 1.328–43.889, p = 0.02) was an independent predictive factor in logistic regression. The subgroup analysis of patients with hypothyroidism, including subclinical or overt, and sick euthyroid syndrome, including low T3 or T4 syndrome, by univariate Cox proportional hazards analysis showed no significance (p = 0.41 and 0.10, respectively; data not shown).

    Fig. 1:
    Cumulative patient survival determined using the Kaplan–Meier method. Cumulative survival of peritoneal dialysis patients (n = 46) with normal thyroid function (line I, dotted line; n = 27) and abnormal thyroid function (including subclinical hypothyroidism or sick euthyroid syndrome) (line II, black line; n = 19).
    Table 3:
    Cox proportional hazard model for the analysis of factors associated with mortality in PD patients (n = 46).

    3.4. Factors related to free T4 levels and LV systolic function

    Bivariate correlation analyses revealed that free T4 levels were inversely associated with LVEF (r = –0.245; p = 0.06), right ventricular ejection fraction (r = –0.258; p = 0.06), CRP levels (r = –0.33; p = 0.02), and CTR (r = 0.266; p = 0.04; Fig. 2). Only CRP levels and CTR showed significant correlation with free T4 levels. In multivariate linear regression, CRP levels (β = –0.547; p < 0.01) showed significant inverse correlation with free T4 levels, after controlling for age, DM, CVD, albumin, LVEF and CRP (Table 4).

    Fig. 2:
    Factors related to free thyroxine levels. Left ventricle ejection fraction (A; r = –0.245, p = 0.06); right ventricle ejection fraction (B; r = –0.258, p = 0.06); C-reactive protein (C; r = –0.33, p = 0.02); and cardiothoracic ratio (D; r = 0.266, p = 0.04).
    Table 4:
    Multivariate linear regression for free T4.

    3.5. Patient outcome

    By the end of the study, 28 (65.1%) patients were still alive, and 15 (34.9%) patients had died. Septic shock accounted for 10 of 15 (66.7%) of the cases of mortality, sudden cardiac death accounted for three of 15 (20%), and the other two patients died from respiratory failure or peripheral arterial occlusive disease. All of the patients were still undergoing PD therapy when they died. Seventeen of 28 living patients continued PD therapy, but the other 11 discontinued PD. Eight of 11 (72%) patients had shifted to HD and three (28%) had undergone kidney transplantation. Only one (1.9%) patient developed encapsulating peritoneal sclerosis after 13 continuous years of PD therapy.

    4. Discussion

    This case-controlled longitudinal study demonstrated that PD patients with abnormal thyroid function had a shorter cumulative survival in a long-term observation period. In our study group, age, sex, duration of PD, comorbidity, Kt/V, peritoneal equilibrium test (PET), left ventricular ejection fraction (LVEF), and serum CRP and albumin levels showed no significant difference, but thyroid function tests did differ significantly (Tables 1 and 2). Compared with another cohort study in Korea,3 our patients were younger and had lower peritonitis rate. A high percentage (41.3%) of abnormal thyroid dysfunction was demonstrated in our study. However, Kang et al9 have reported that the percentage of subclinical hypothyroidism in continuous ambulatory PD (CAPD) patients was 27.5%, and the total percentage of hypothyroidism (excluding sick euthyroid syndrome) in our study was similar at 26.0%. In addition, there was an increasing frequency of having low T3 for patients in a later stage of chronic kidney disease (CKD); up to 20% in CKD stage 3.7 The frequency of abnormal thyroid function in this long-term study was similar to that in other studies.7,8

    Patients diagnosed with ESRD had a higher chance of developing primary hypothyroidism with lower free T4 levels than the normal population had.10 Primary subclinical hypothyroidism with high TSH levels but normal T4 or T3 levels is relatively common in patients with CKD but not requiring dialysis.11 The reason for the high percentage of high TSH or low T3 levels among the PD patients was not clear, but abdominal protein loss through PD exchange may be one reason.12 Furthermore, low levels of T3 in ESRD patients showed highly significant correlation with the level of inflammatory markers interleukin-6 and CRP, and both were independent predictors of death in HD patients.13 The reason for patients with low T3 syndrome but without TSH elevation is not dysfunction of the hypothalamo-pituitary axis because truly hypothyroid patients with renal failure can still mount a high TSH response.14 Inflammation may precipitate CKD or ESRD patients into the development of abnormal thyroid function. Low T3 levels were correlated with all-cause mortality and cardiac disease mortality in patients with biochemical euthyroid status. In addition to low T3 syndrome, subclinical hypothyroidism was also found to be significantly correlated with patient survival. This finding could be explained by the negative inotropic effect of hypothyroidism on LV systolic function in PD patients, and it could be supported by the report by Kang et al,9 who have also demonstrated that TSH is negatively correlated with LVEF.

    Low-T3 syndrome has commonly been interpreted by the medical community as sick euthyroid syndrome. The etiology of low T3 syndrome is decreased conversion of T4 to T3 (the active form) by acute inhibition of I 5′ deiodinase due to acute illness. Low T3 syndrome is also a strong predictor of death in patients with heart disease. In patients selected for coronary angiography, free T3 levels are inversely correlated with the presence of coronary artery disease and adverse prognosis.15 Low T3 is associated with left ventricular dysfunction and left ventricular hypertrophy in ESRD patients.16 In addition, the presence of inflammation and malnutrition was associated with valvular calcification, and prolonged PD may enhance aortic valve calcification.17 Furthermore, CRP is an important precipitating factor associated with the abdominal calcification index in HD patients.18 Christ-Crain et al19 reported that CRP levels were insignificantly correlated with free T4 levels; a trend of elevated CRP in subclinical or overt hypothyroidism patients was detected. Demirci et al20 found high CTR and CRP levels in fluid-overloaded PD patients, which means that a higher CRP may jeopardize cardiac function. In our study, free T4 levels were inversely correlated with CRP, and higher free T4 levels showed an insignificant trend towards depressed LVEF; a finding different from the results of others studies of L-thyroxine supplement in subclinical hypothyroidism.21,22

    It is surprising that we found that a much higher mortality was caused by septic shock (66.7%) than cardiovascular events (20%). In another large retrospective study, immediate cause of death from cardiovascular events accounted for 41.2% and peritonitis was directly implicated in 15.2% of the overall deaths.23 This discrepancy could be explained as follows. First, lower free T4 levels significantly correlated with higher CRP levels in our study, but not LVEF. That means that abnormal thyroid function may not cause deterioration of heart function. Second, these patients who had undergone long-term PD were younger and may have had lower frequency of vascular calcification or atherosclerosis than the total group of PD patients.

    There were several limitations to this study. First, this was a single-center case-controlled study. Also, we selected younger patients with lower peritonitis rates and cardiovascular complications and good nutritional status. A larger population and multicenter cohort study is warranted to obtain a clearer picture regarding the factors involved in the long-term outcomes of PD. Second, only systolic but not diastolic heart function was evaluated. We know that diastolic dysfunction is also an important cause of heart failure in CKD patients. Third, we did not analyze inflammatory markers other than CRP (such as interleukin-6 or intercellular adhesion molecule-1), and we did not set a clear definition of inflammatory status for our patients.

    In conclusion, this study showed that abnormal thyroid function in CAPD patients predicted a shorter cumulative patient survival. Lower thyroid hormone levels were significantly associated with higher CRP levels. In addition, compared with cardiovascular events, sepsis was an even more important issue with regard to the mortality of long-term PD patients. Our study demonstrates the association of abnormal thyroid function with CRP for PD patients in a long-term case-controlled study. Clinical nephrologists should be alert to the change in thyroid function parameters (including free T4, TSH, T3 and T4 levels) and detect the possible underlying mechanism of abnormal thyroid function such as proinflammatory status in CAPD patients as early as possible.


    1. Nakamoto H, Kawaguchi Y, Suzuki H. Is technique survival on peritoneal dialysis better in Japan? Perit Dial Int. 2006;26:136-143.
    2. Shiao CC, Kao TW, Hung KY, Chen YC, Wu MS, Chu TS, et al. Seven-year follow-up of peritoneal dialysis patients in Taiwan. Perit Dial Int. 2009;29:450-457.
    3. Han SH, Lee JE, Kim DK, Moon SJ, Kim HW, Chang JH, et al. Long-term clinical outcomes of peritoneal dialysis patients: single-center experience from Korea. Perit Dial Int. 2008;28:S21-S26.
    4. Gokal R, Oreopoulos DG. Is long-term technique survival on continuous ambulatory peritoneal dialysis possible? Perit Dial Int. 1996;16:553-555.
    5. Wang AY. Vascular and other tissue calcification in peritoneal dialysis patients. Perit Dial Int. 2009;29:S9-14.
    6. Enia G, Panuccio V, Cutrupi S, Pizzini P, Tripepi G, Mallamaci F, et al. Subclinical hypothyroidism is linked to micro-inflammation and predicts death in continuous ambulatory peritoneal dialysis. Nephrol Dial Transplant. 2007;22:538-544.
    7. Song SH, Kwak IS, Lee DW, Kang YH, Seong EY, Park JS. The prevalence of low triiodothyronine according to the stage of chronic kidney disease in subjects with a normal thyroid-stimulating hormone. Nephrol Dial Transplant. 2009;24:1534-1538.
    8. Lin CC, Yang WC, Ng YY, Chou YH, Tarng DC, Chen TW. Association of anion gap with thyroid dysfunction and nodular goiter in CAPD patients. Perit Dial Int. 2002;22:394-399.
    9. Kang EW, Nam JY, Yoo TH, Shin SK, Kang SW, Han DS, et al. Clinical implications of subclinical hypothyroidism in continuous ambulatory peritoneal dialysis patients. Am J Nephrol. 2008;28:908-913.
    10. Kaptein EM, Quion-Verde H, Chooljian CJ, Tang WW, Friedman PE, Rodriquez HJ, et al. The thyroid in end-stage renal disease. Medicine(Baltimore). 1988;67:187-197.
    11. Chonchol M, Lippi G, Salvagno G, Zoppini G, Muggeo M, Targher G. Prevalence of subclinical hypothyroidism in patients with chronic kidney disease. Clin J Am Soc Nephrol. 2008;3:1296-1300.
    12. Robey C, Shreedhar K, Batuman V. Effects of chronic peritoneal dialysis on thyroid function tests. Am J Kidney Dis. 1989;13:99-103.
    13. Zoccali C, Tripepi G, Cutrupi S, Pizzini P, Mallamaci F. Low triiodothyronine: a new facet of inflammation in end-stage renal disease. J Am Soc Nephrol. 2005;16:2789-2795.
    14. Lim VS. Thyroid function in patients with chronic renal failure. Am J Kidney Dis. 2001;38:S80-S84.
    15. Coceani M, Iervasi G, Pingitore A, Carpeggiani C, L'Abbate A. Thyroid hormone and coronary artery disease: from clinical correlations to prognostic implications. Clin Cardiol. 2009;32:380-385.
    16. Zoccali C, Benedetto F, Mallamaci F, Tripepi G, Cutrupi S, Pizzini P, et al. Low triiodothyronine and cardiomyopathy in patients with end-stage renal disease. J Hypertens. 2006;24:2039-2046.
    17. Gruppen MP, Groothoff JW, Prins M, van der Wouw P, Offringa M, Davin JC, et al. Cardiac disease in young adult patients with end-stage renal disease since childhood: a Dutch cohort study. Kidney Int. 2003;63:1058-1065.
    18. Yamada K, Fujimoto S, Nishiura R, Komatsu H, Tatsumoto M, Sato Y, et al. Risk factors of the progression of abdominal aortic calcification in patients on chronic haemodialysis. Nephrol Dial Transplant. 2007;22:2032-2037.
    19. Christ-Crain M, Meier C, Guglielmetti M, Huber PR, Riesen W, Staub JJ, et al. Elevated C-reactive protein and homocysteine values: cardiovascular risk factors in hypothyroidism? A cross-sectional and a double-blind, placebo-controlled trial. Atherosclerosis. 2003;166:379-386.
    20. Demirci MS, Demirci C, Ozdogan O, Kircelli F, Akcicek F, Basci A, et al. Relations between malnutrition–inflammation–atherosclerosis and volume status: the usefulness of bioimpedance analysis in peritoneal dialysis patients. Nephrol Dial Transplant. 2011;26:1708-1716.
    21. Erkan G, Erkan AF, Cemri M, Karaahmetoglu S, Cesur M, Cengel A. The evaluation of diastolic dysfunction with tissue Doppler echocardiography in women with subclinical hypothyroidism and the effect of L-thyroxine treatment on diastolic dysfunction: a pilot study. J Thyroid Res. 2011;2011:654304.
    22. Monzani F, Di Bello V, Caraccio N, Bertini A, Giorgi D, Giusti C, et al. Effect of levothyroxine on cardiac function and structure in subclinical hypothyroidism: a double-blind, placebo-controlled study. J Clin Endocrinol Metab. 2001;86:1110-1115.
    23. Pérez Fontan M, Rodríguez-Carmona A, García-Naveiro R, Rosales M, Villaverde P, Valdés F. Peritonitis-related mortality in patients undergoing chronic peritoneal dialysis. Perit Dial Int. 2005;25:274-284.

    C-reactive protein; hypothyroidism; long-term peritoneal dialysis; sick euthyroid syndrome

    © 2012 by Lippincott Williams & Wilkins, Inc.