Micronutrients and lifestyles in Taiwanese patients with stage 3 to 5 chronic kidney disease.

(1)

Applied nutritional investigation

Micronutrients and lifestyles in Taiwanese patients with stage 3 to 5

chronic kidney disease

Shou-Shan Chiang, M.D.

a

, Cheng-Wei Tai, M.S.

b

, Chi-Jung Chung, M.S.

b

,

Horng-Sheng Shiue, M.D.

c

, Jin-Bor Chen, M.D.

d

, Chien-Tien Su, M.D.

e

,

and Yu-Mei Hsueh, Ph.D.

f,

*

a

Department of Internal Medicine/Nephrology, Shin Kong Wu Ho-Su Memorial Hospital, Taipei, Taiwan

b_{School of Public Health, Taipei Medical University, Taipei, Taiwan}

c_{Department of Chinese Medicine, Chang Gung Memorial Hospital, and Graduate Institute of Medical Sciences, College of Medicine,}

Taipei Medical University, Taipei, Taiwan

d_{Nephrology Division, Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of Medicine, Kaohsiung, Taiwan} e_{Department of Family Medicine, Taipei Medical University Hospital, Taipei, Taiwan}

f_{Department of Public Health, School of Medicine, Taipei Medical University, Taipei, Taiwan}

Manuscript received November 16, 2008; accepted April 25, 2009.

Abstract Objective: Lycopene is an antioxidant that reduces oxidative stress. Analgesics are commonly used and may impair kidney function. However, the associations among plasma lycopene, analgesic use, and chronic kidney disease (CKD) are unknown. A hospital-based, case–control study was conducted to determine the association among plasma lycopene, analgesic use, and CKD.

Methods: Two hundred one patients with CKD and 313 controls were recruited, and CKD was defined as an estimated glomerular filtration rate <60 mL/min per 1.73 m2, as calculated by the Modification of Diet in Renal Disease formula. Plasma antioxidants were measured by high-performance liquid chro-matography.

Results: This study showed that the higher the plasma lycopene was, the lower the CKD risk. Specif-ically, in subjects with a plasma lycopene level17.97 or 7.72–17.97 mg/dL, the adjusted odds ratio of CKD was 0.32 (95% confidence interval 0.18–0.58) or 0.49 (95% confidence interval 0.29–0.83), re-spectively, compared with subjects with a plasma lycopene level <7.72 mg/dL, independent of age, gender, level of education, paternal and maternal ethnicities, cigarette smoking, analgesic use, hyper-tension, and diabetes history. In contrast, the higher the plasma retinol level, the higher the risk of CKD. A significantly higher risk was demonstrated in analgesic users than in non-users (odds ratio 3.83, 95% confidence interval 1.75–8.40), but a significantly lower risk was shown in subjects who used analge-sics on an as-needed basis than in non-users. Plasma lycopene tended to interact additively with anal-gesic consumption in modifying the CKD risk; however, the interactions were statistically insignificant. Conclusion: This is the first study showing that a low plasma lycopene level is associated with CKD risk.Ó 2010 Elsevier Inc. All rights reserved.

Keywords: Micronutrients; Lycopene; Analgesics; Chronic kidney disease

Introduction

Much epidemiologic and clinical evidence has shown a link between several factors and the initiation and

progression of chronic kidney disease (CKD). Apart from well-known risk factors for CKD, including age[1], genetic

[2]or familial predisposition[3], hypertension[4], and diabe-tes, obesity, and metabolic syndrome[5], the identification of other risk factors related to CKD is important for early detec-tion and treatment.

The association between habitual analgesic use and kid-ney function has been inconsistent[6,7]. One study has found that the metabolic process of analgesics produces oxidative The study was supported by grant SKH-TMU-95-23 from Shin Kong Wu

Ho-Su Memorial Hospital and Taipei Medical University in Taipei, Taiwan. *Corresponding author. Tel.:2-2736-1661, ext. 6513; fax: þ886-2-2738-4831.

E-mail address:ymhsueh@tmu.edu.tw(Y.-M. Hsueh).

(2)

stress[8]. Oxidative stress plays a role in the pathogenesis and complications of CKD [9]. Recent studies have also indicated that increased adiposity may amplify the oxidative stress and inflammation that accompany moderate to severe CKD[10], although a recent study has reported that lycopene supplementation reduces apoptosis but does not affect oxi-dant-responsive heme oxygenase-1 in human lymphocytes

[11]. However, lycopene and vitamin C from tomato juice have been shown to have an influence on the biomarkers of oxidative stress and inflammation[12]. Plasma water-soluble antioxidants, such as vitamin C, deteriorate within a very short time unless specimens are acid-stabilized and then fro-zen[13], and in this study frozen plasma specimens were not acid-stabilized before storage. Lycopene is a lipid-soluble an-tioxidant and has been inversely associated with lipid perox-idation, including low-density lipoprotein oxidation[14]and reduced oxidative stress and inflammation[12]. Low plasma levels of lycopene or analgesic consumption may be a risk factor for CKD. However, the associations among lycopene, analgesic consumption, and CKD are unknown. We con-ducted the present case–control study to determine the asso-ciations among plasma lycopene, analgesic consumption, and CKD in Taiwanese subjects.

Material and methods

Study subjects and questionnaire interview

Two hundred one patients with clinically proven CKD (age range 22–88 y) were recruited from the Department of Internal Medicine/Nephrology of Shin Kong Wu Ho-Su Memorial Hospital in Taipei, Taiwan, from September 2005 to December 2007. The glomerular filtration rate (GFR) is traditionally considered the best overall index of renal function in health and disease. We used the abbreviated equation from the Modification of Diet in Renal Disease Study [15] to estimate the GFR as 186.3 3 (serum creatinine)1.1543 (age)0.2033 (0.742 for females) and defined five stages of CKD. In this study, subjects with stage 3 to 5 renal disease (GFR <60 mL/min per 1.73 m2) for 3 mo were diagnosed with CKD. A total of 313 age-matched con-trol subjects with no evidence of CKD (GFR60 mL/min per 1.73 m2) were accrued from a hospital-based pool, including those receiving senior citizen health examinations at Taipei Medical University Hospital and those receiving adult health examinations at Taipei Municipal Wan Fang Hospital.

Well-trained personnel carried out standardized personal interviews based on a structured questionnaire. The informa-tion collected included demographic and socioeconomic characteristics, general potential risk factors for CKD (lifestyle, alcohol consumption, and cigarette smoking in quantified detail), exposure to potential occupational and environmental carcinogens (hair dyes and pesticides, long-term medication history, and consumption of conventional

and alternative medicines), and personal and family histories of hypertension, diabetes, and CKD. Frequent alcohol drinkers were defined as those who consumed alcohol >3 d/wk for6 mo; those who consumed less than this level were classified as occasional drinkers. Pesticide users were farmers who used pesticides for agricultural purposes. The research ethics committee of Taipei Medical University (Taipei, Taiwan) approved the study. All patients provided informed consent forms before sample and data collection. The study was consistent with the World Medical Associa-tion DeclaraAssocia-tion of Helsinki. A 10-mL blood sample was also collected from subjects on recruitment by use of vacuum syringes treated with ethylenediaminetetra-acetic acid and disposable needles. Plasma samples were centrifuged at 3000 rpm for 15 min at room temperature, separated into aliquots, and stored at80_{C until analyzed.}

Determination of plasma antioxidant micronutrient levels The levels of b-carotene, lycopene, a-tocopherol, and retinol of plasma samples were measured by high-performance liquid chromatography according to the proce-dure described previously [16]. Analysis was carried out using reverse phase high-performance liquid chromatogra-phy (Hitachi, Naka, Japan) with the mobile phase of metha-nol:acetonitrite:chloroform (47:47:6) and multiwave length monitoring. Retinol was detected at 325 nm, a-tocopherol was detected at 280 nm, and lycopene and b-carotene were detected at 466 nm. The plasma samples for each case– control set were thawed from the 80_{C refrigerator in}

dim light at room temperature and assayed on the same day to ensure that temporal variability in the laboratory as-says would affect cases and controls equally. All laboratory personnel were unaware of the disease status of the subjects from whom the plasma samples were obtained. The recovery rates for b-carotene, lycopene, a-tocopherol, and retinol were 90% to 100% at the highest concentration and 90% to 107% at the lowest concentration of the standard solution. The precision (coefficient of variance) of b-carotene, lyco-pene, a-tocopherol, and retinol was 1.0% to 6.0%. We also used a-tocopherol acetate as an internal control to reduce systematic error; the coefficient of variance for a-tocopherol acetate was 2.5%.

Statistical analysis

Continuous variables are expressed as mean 6 standard error. Student’s t test was used to compare differences in the plasma antioxidants between case subjects and controls. Multiple logistic regression models were used to estimate the multivariate-adjusted odds ratio (OR) and the 95% confi-dence interval (CI). Cutoff points for continuous variables were the respective tertiles of the controls. Significance tests for linear trend among ORs across exposure strata were cal-culated by categorizing exposure variables and treating scored variables as continuous. For the joint-effect analysis,

(3)

the cutoff points for the antioxidant micronutrients were the respective medians of the controls. The synergy index pro-posed by Rothman[17]was computed to assess the additive interaction relation between antioxidant micronutrients and analgesic consumption on CKD risk. An observed synergy index value that departs substantially from the expected additive null, i.e., synergy index not equal to 1, suggests an additive interaction effect. The OR values and the vari-ance–covariance matrix were then used to calculate values for the synergy index and 95% CIs[18].

Results

The sociodemographic characteristics of the cases and controls are presented inTable 1. Subjects who had higher educational levels had a significantly lower risk than those with lower educational levels. Mainland Chinese or aborigi-nal and Hakka Taiwanese had a significantly lower CKD risk than the Fukien Taiwanese. Subjects with diabetes or hyper-tension had a significantly higher CKD risk than those with-out diabetes or who were normotensive (ORs 2.10 and 2.15, 95% CIs 1.38–3.35 and 1.48–3.55, respectively) when ad-justed for multiple variables.

Table 2presents the association among lifestyle, chemical contact, analgesic use, and CKD. Cigarette smokers had

a significantly lower CKD risk than non-smokers. Alcohol or coffee consumption, pesticide exposure, and paint or dye use did not affect the risk of CKD. The significantly higher risk was shown in analgesic users than in non-users (OR 3.83, 95% CI 1.75–8.40). In contrast, subjects who used analgesics as needed had a significantly lower risk than non-users (OR 0.29, 95% CI 0.14–0.60) when adjusted for multiple variables.

Figure 1shows the association between plasma micronu-trients and GFR in CKD cases. Plasma retinol was not related to the GFR, but a-tocopherol, lycopene, and b-carotene were significantly related to the GFR.

Based on trend analysis of micronutrient strata in tertiles, with the exception of a-tocopherol and b-carotene, retinol was shown to be significantly associated with CKD risk in a dose–response relation, especially in subjects with a plasma retinol level162.96 mg/dL (OR 8.06, 95% CI 4.25–15.17) compared with those with a retinol level <107.57 mg/dL. Conversely, higher lycopene levels produced a lower CKD risk after multivariate analysis (Table 3). In subjects with a plasma lycopene level 17.97 or 7.72–17.97 mg/dL, the OR and 95% CI of CKD was 0.32 (0.18–0.58) or 0.49 (0.29–0.83) compared with those with a level <7.72 mg/dL, independent of age, gender, educational level, paternal and maternal ethnicities, cigarette smoking, analgesic use, and hypertension and diabetes history.

Table 1

Sociodemographic characteristics of chronic kidney disease and healthy controls

Variables Cases (%) Healthy controls (%) Odds ratios*(95% CI) Odds ratioy(95% CI)

Gender

Male 100 (49.8) 181 (44.4) 1.00 1.00

Female 101 (50.2) 227 (55.6) 0.87 (0.62–1.21)z 0.47 (0.29–0.74)jj

Age (y), mean 6 SD 59.18 6 13.79 59.73 6 13.36 1.00 (0.99–1.01)x 0.97 (0.96–1.02)

Educational level

Illiterate/elementary school 103 (51.2) 104 (25.5) 1.00 1.00

Junior /senior high school 57 (28.4) 134 (32.7) 0.34 (0.22–0.53){ 0.36 (0.22–0.59){

College and above 41 (20.4) 170 (41.3) 0.15 (0.09–0.26){ 0.18 (0.10–0.34){

Paternal ethnicity Fukien Taiwanese 178 (88.6) 237 (57.6) 1.00 1.00 Hakka Taiwanese 8 (3.9) 49 (12.5) 0.25 (0.13–0.51){ 0.27 (0.12–0.60){ Mainland Chinese/aboriginal 15 (7.5) 122 (29.9) 0.19 (0.11–0.32){ 0.28 (0.15–0.52){ Maternal ethnicity Fukien Taiwanese 180 (89.6) 250 (60.7) 1.00 1.00 Hakka Taiwanese 8 (3.9) 47 (12.0) 0.28 (0.14–0.57){ 0.29 (0.13–0.66){ Mainland Chinese/aboriginal 13 (6.5) 111 (27.3) 0.19 (0.11–0.34){ 0.26 (0.14–0.51){ Diabetes No 133 (73.9) 370 (91.8) 1.00 1.00 Yes 47 (26.1) 33 (8.2) 2.37 (1.34–4.18){ 2.10 (1.38–3.35)jj Hypertension No 104 (57.8) 306 (75.9) 1.00 1.00 Yes 76 (42.2) 97 (24.1) 2.30 (1.48–3.56){ 2.15 (1.48–3.55)jj

CI, confidence interval * Adjusted for age and gender.

y

Adjusted for age, gender, cigarette smoking, coffee drinking, analgesic usage, educational level, paternal and maternal ethnicities, and hypertension and diabetes histories, except its main variable in the model.

z_{Adjusted for age.} x_{Adjusted for gender.} jj_{P < 0.01.}

(4)

Because both micronutrients (retinol and lycopene) and analgesic use affect the CKD risk, further analyses were carried out to assess the joint effects of the two risk factors (Table 4). Trend analysis revealed progressively increased risks through exposure to no risk factors, one of the factors, or both risk factors. Although plasma lycopene or retinol tended to interact additively with analgesic consumption in modifying the CKD risk, the interactions were statistically insignificant, as shown by the absence of a substantial devi-ation from 1 in the synergy index.

Discussion

Many studies[19,20]have reported that hypertension and diabetes are important risk factors for CKD. We also found two times the CKD risk for patients with hypertension or diabetes compared with those with normal blood pressure and without diabetes in this study. Diabetes and hypertension may activate the renin-angiotensin-aldosterone system, accompanied by a loss of glycemic control[21], or enhance oxidative stress or inflammation[4]with progression to ne-phropathy. In addition, the present study showed that women had a significantly lower CKD risk than men, in disagreement with a previous study[22]. A recent study has found that moderate wine consumption may enhance kidney antioxidant

defense against renal injury[23]; however, alcohol consump-tion did not affect the risk of CKD in this study. We did not determine why cigarette smokers had a significantly lower risk of CKD in this study compared with non-smokers; per-haps controls are less concerned about their lifestyle because they are healthy; another possible reason may be the preva-lent cases were advised to discontinue cigarette smoking. Subjects with a low level of education had a higher CKD risk than those with a high level of education, as reported in another Taiwanese study[24]. Paternal and maternal eth-nicities as Hakka Taiwanese or Mainland Chinese or aborig-inal had a lower CKD risk than Fukien Taiwanese; this disparity of ethnicity on CKD prevalence remains unex-plained, in much the same way that blacks have a higher CKD prevalence than whites in the United States[25].

Results from this study have shown that high plasma lyco-pene levels were independently and inversely associated with CKD. This postulate was supported by several findings. First, study subjects with a high plasma lycopene level had a lower prevalence of CKD than those with a low plasma lycopene level. Second, the highest or higher plasma lycopene groups were related to a 0.32- or 0.49-fold decreased risk of CKD than those with a low plasma lycopene level, independent of age, gender, level of education, paternal or maternal eth-nicity, cigarette smoking, analgesic usage, and hypertension Fig. 1. Association between micronutrients and GFR in patients with chronic kidney disease. GFR, glomerular filtration rate.

(5)

and diabetes history. Third, the subjects with low plasma lycopene levels and habitual analgesic use had a further in-creased risk for CKD. These findings may be explained by several hypotheses. First, lycopene is a potent antioxidant shown to quench reactive oxygen species in vitro[26]and to diminish oxidative stress biomarkers of serum thiobarbitu-ric acid-reactive substances in vitro[27]. Second, lycopene can attenuate inflammatory cytokine-stimulated endothelial cell adhesion molecule expression [28]. It is possible that high plasma lycopene levels are inversely associated with CKD through lycopene decreasing the oxidative stress and attenuating the inflammation.

In this study, there was no correlation between the plasma retinol level and GFR; however, it was found that study sub-jects with high plasma retinol had a higher risk of CKD than those with a low plasma retinol. In addition, the OR of CKD was significantly increased to 8.06 or 2.05 for subjects with the highest or higher plasma retinol levels than those with a low retinol level after adjusted multiple risk factors. The el-evated plasma retinol in patients with CKD is consistent with a previous study in which it was found that increased plasma retinol and retinol-binding protein (RBP) levels in patients with chronic renal failure [29,30] may be due to reduced vitamin excretion and decreased conversion of retinol to ret-inoic acid[31]. An animal study also observed an increase of serum retinol in the nephrectomized rats, suggesting the increase in retinol is associated with RBP, coupled with the

thyroxine-binding protein, and the upregulation of the hepatic release mechanism in renal disease[32]. It has been suggested that plasma RBP might be a biomarker of nephrop-athy[33–35], and there is a strong correlation between the plasma retinol and RBP levels[29]. Based on these findings, high plasma retinol may be a biomarker related to the CKD risk. However, we cannot exclude both findings regarding the association between retinol and lycopene levels with low GFR in this study as the result rather than the cause for a low GFR.

The possibility that long-term ingestion of analgesics leads to renal pathology and renal failure is controversial. There is evidence of a relation between the ingestion of phen-acetin-containing analgesics and kidney disease [36]. An early study reported that lifetime regular acetaminophen use was related to CKD[37], and another study found that subjects who often take acetaminophen or non-steroidal anti-inflammatory drugs have an increased risk of end-stage renal disease, but not for those who often take aspirin[38]. We cannot explain why subjects who used analgesics on an as-needed basis had a lower CKD risk than never-users; per-haps the appropriate use of analgesics can prevent CKD[39]. In this study we found that analgesic users had a significantly higher CKD risk than non-users. Long-term use of analgesics leads to renal injury through chronic interstitial fibrosis with or without papillary necrosis[40], damage is generally irre-versible, and progression to renal failure occurs. Table 2

Association among lifestyle, chemical contact, analgesic usage, and chronic kidney disease

Variables Cases (%) Controls (%) Odds ratio*(95% CI) Odds ratioy(95% CI)

Cigarette smoking No 166 (82.6) 305 (74.8) 1.00 1.00 Yes 35 (17.4) 103 (25.2) 0.48 (0.29–0.79)z 0.40 (0.23–0.70)z Alcohol drinking No 186 (92.5) 357 (87.5) 1.00 1.00 Yes 15 (7.5) 51 (12.5) 0.61 (0.33–1.14) 1.28 (0.62–2.64) Coffee dinking No 109 (54.2) 191 (46.9) 1.00 1.00 Yes 45 (22.4) 95 (23.4) 1.78 (1.07–2.96)z 1.48 (0.84–2.61) Occasional 47 (23.4) 121 (29.7) 1.02 (0.64–1.60) 1.07 (0.65–1.76) Dye contact No 117 (96.7) 403 (99.3) 1.00 1.00 Yes 4 (3.3) 3 (0.7) 1.01 (0.55–1.87) 0.33 (0.21–1.51) Paint contact No 186 (92.5) 381 (93.8) 1.00 1.00 Yes 15 (7.5) 25 (6.2) 1.06 (0.68–1.65) 1.06 (0.52–2.15) Pesticide contact No 197 (98.0) 396 (97.5) 1.00 1.00 Yes 4 (2.0) 10 (2.5) 0.96 (0.50–1.85) 1.14 (0.70–1.85) Analgesic usage No 157 (78.1) 309 (75.1) 1.00 1.00 Yes as usual 24 (11.9) 14 (3.4) 3.25 (1.56–6.76)x 3.83 (1.75–8.40)x

Yes as the need arises 20 (10.0) 83 (20.5) 0.58 (0.34–1.00) 0.29 (0.14–0.60)x

y_{Adjusted for age, gender, educational level, paternal and maternal ethnicities, and hypertension and diabetes histories.} z_{P < 0.05.}

(6)

Diclofenac-induced non-steroidal anti-inflammatory drug nephrotoxicity involves oxidative stress, which translates into orderly fragmentation of genomic DNA leading to apo-ptotic death in the kidney[41].

Our study has some important limitations that need to be considered when interpreting our results. First, selection bias was minimized even though cases and controls were

recruited from two different hospitals, because these hospi-tals belonged to medical centers and were located in Taipei. Furthermore, the majority of cases and controls lived in Tai-pei and were similar to each other with respect to dietary habits. Possible selection bias may have occurred because the recruited controls were more likely women, with less el-ementary school education, and less often Fukien Taiwa-nese than the CKD cases. However, in a large-scale screening program, it has been reported that men had a higher rate of CKD than women, and subjects with a low level of education had a higher CKD risk than sub-jects with a high level of education in Taiwan[24]. Second, we could not validate the self-reported analgesic use be-cause of the lack of independent records for these medica-tions. In addition, we did not collect information on the kind of analgesics and the duration of their use. Therefore, we could not evaluate the different kinds of analgesics and the consumption of certain cumulative amounts of an anal-gesic for the CKD risk. Many people in Taiwan access tra-ditional Chinese medicine and folk medicine; however, we had no information about herbal remedies in this study, and we cannot exclude the possibility that analgesic-adulterated herbs may have accounted for our findings, at least in part. Third, the accuracy of one-spot evaluation of plasma anti-oxidants may be in doubt. However, the values might be re-liable with no accompanying changes in lifestyle in all subjects. Fourth, because of the small sample, statistical sig-nificance should be interpreted with caution.

In conlusion, this is, to our knowledge, the first study showing that a low plasma lycopene level is associated with CKD risk.

Acknowledgments

The authors thank Dr. Ying-Chin Lin of the Health Man-agement Center, Taipei Medical University Municipal Wan Fang Hospital, for recruitment of the healthy controls. Table 4

Multiple logistical regression analysis effects of combination of antioxidant micronutrient levels and analgesic use status on chronic kidney disease

Variables Analgesic use status Cases/controls Odds ratio*(95% CI) Synergy index

Retinol (mg/dL) 1.15 (0.41–3.11)

135.8 No/yes as the need arises 142/193 1.00

Yes as usual 21/8 3.46 (1.36–8.04)y

< 135.8 No/yes as the need arises 34/199 0.22 (0.13–0.36)z

Yes as usual 5/6 1.53 (0.33–7.09)

P for trend <0.0001

Lycopene (mg/dL) 6.09 (0.11–216.01)

12.7 No/yes as the need arises 49/176 1.00

Yes as usual 10/24 4.54 (1.31–15.71)y

< 12.7 No/yes as the need arises 125/183 2.97 (1.76–5.01)z

Yes as usual 19/23 12.82 (3.71–44.33)z

P for trend <0.0001

CI, confidence interval

* Adjusted for age, gender, educational level, paternal and maternal ethnicities, cigarette smoking, coffee drinking, and hypertension and diabetes histories.

y_{P < 0.05.} z_{P < 0.001.}

Table 3

Dose–response relation between chronic kidney disease risk and plasma level of antioxidant micronutrients

Variables Cases/controls Odds ratio* (95% CI) Odds ratioy (95% CI) Retinol (mg/dL) < 107.6 21/136 1.00 1.00 107.6–163.0 46/136 2.15 (1.22–3.81)x 2.05 (1.03–4.03)z 163.0 134/136 6.35 (3.79–10.67)jj 8.06 (4.25–15.17)jj P for trend <0.0001 <0.0001 a-Tocopherol (mg/dL) < 818.7 65/136 1.00 1.00 818.7–1153.9 88/136 1.47 (0.98–2.13) 1.60 (0.94–2.52) 1153.9 48/136 0.81 (0.51–1.27) 1.06 (0.61–1.79) P for trend 0.40 0.77 Lycopene (mg/dL) < 7.7 108/136 1.00 1.00 7.7–18.0 52/136 0.46 (0.29–0.72)jj 0.49 (0.29–0.83)x 18.0 41/136 0.35 (0.22–0.56)jj 0.32 (0.18–0.58)jj P for trend <0.0001 <0.0001 b-Carotene (mg/dL) < 17.2 61/136 1.00 1.00 17.2–40.3 66/136 1.14 (0.74–1.75) 1.10 (0.66–1.85) 40.3 74/136 1.31 (0.85–2.01) 1.29 (0.77–2.15) P for trend 0.22 0.33

y_{Adjusted for age, gender, educational level, paternal and maternal}

ethnicities, cigarette smoking, coffee drinking, analgesic usage, and hypertension and diabetes histories.

z_{P < 0.05.} x

P < 0.01.

jj

(7)

References

[1] Lee YT, Chiu HC, Su HM, Yang JF, Voon WC, Lin TH, et al. Lower hemoglobin concentrations and subsequent decline in kidney function in an apparently healthy population aged 60 year and older. Clin Chim Acta 2008;389:25–30.

[2] Hsu CC, Bray MS, Kao WH, Pankow JS, Boerwinkle E, Coresh J. Ge-netic variation of the renin-angiotensin system and chronic kidney dis-ease progression in black individuals in the atherosclerosis risk in communities study. J Am Soc Nephrol 2006;17:504–12.

[3] Satko SG, Sedor JR, Iyengar SK, Freedman BI. Familial clustering of chronic kidney disease. Semin Dial 2007;20:229–36.

[4] Higashikuni Y, Ishizaka N, Ishizaka Y, Toda E, Nagai R, Yamakado M. Relationship between blood pressure and chronic kidney disease in the Japanese population: the lower the better even in individuals without hy-pertension? Hypertens Res 2008;31:213–9.

[5] Haroun MK, Jaar BG, Hoffman SC, Comstock GW, Klag MJ, Coresh J. Risk factors for chronic kidney disease: a prospective study of 23,534 men and women in Washington County, Maryland. J Am Soc Nephrol 2003;14:2934–41.

[6] Agodoa LY, Francis ME, Eggers PW. Association of analgesic use with prevalence of albuminuria and reduced GFR in US adults. Am J Kidney Dis 2008;51:573–83.

[7] Curhan GC, Knight EL, Rosner B, Hankinson SE, Stampfer MJ. Life-time nonnarcotic analgesic use and decline in renal function in women. Arch Intern Med 2004;164:1519–24.

[8] Olaleye MT, Rocha BT. Acetaminophen-induced liver damage in mice: effects of some medicinal plants on the oxidative defense system. Exp Toxicol Pathol 2008;59:319–27.

[9] Vaziri ND. Roles of oxidative stress and antioxidant therapy in chronic kidney disease and hypertension. Curr Opin Nephrol Hypertens 2004; 13:93–9.

[10] Ramos LF, Shintani A, Ikizler TA, Himmelfarb J. Oxidative stress and inflammation are associated with adiposity in moderate to severe CKD. J Am Soc Nephrol 2008;19:593–9.

[11] Markovitch D, Tyrrell RM, Tauler P, Frystyk J, Stokes K, Thompson D. Lycopene supplementation (passata sauce) reduces apoptosis but does not affect oxidant-responsive heme oxygenase-1 in human lympho-cytes. Nutrition 2009;25:668–75.

[12] Jacob K, Periago MJ, Bohm V, Berruezo GR. Influence of lycopene and vitamin C from tomato juice on biomarkers of oxidative stress and inflammation. Br J Nutr 2008;99:137–46.

[13] Basu TK, Schorah CL. Vitamin C reserves and requirements in health and disease. In: Basu TK, Schorah CS, editors. Vitamin C in health and disease. London: Croom Helm; 1982, p. 62–92.

[14] Hadley CW, Clinton SK, Schwartz SJ. The consumption of processed tomato products enhances plasma lycopene concentrations in associa-tion with a reduced lipoprotein sensitivity to oxidative damage. J Nutr 2003;133:727–32.

[15] National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002;39(suppl 1):S1–266.

[16] Miller KW, Lorr NA, Yang CS. Simultaneous determination of plasma retinol, alpha-tocopherol, lycopene, alpha-carotene, and beta-carotene by high-performance liquid chromatography. Anal Biochem 1984; 138:340–5.

[17] Rothman KJ. Modern epidemiology. Boston, MA: Litter Brown and Company; 1986, p. 322–6.

[18] Hosmer DW, Lemeshow S. Confidence interval estimation of interac-tion. Epidemiology 1992;3:452–6.

[19] Chou CY, Cheng SY, Liu JH, Cheng WC, Kang IM, Tseng YH, et al. Association between betel-nut chewing and chronic kidney disease in men. Public Health Nutr 2008;12:1–5.

[20] Chou CY, Lin CH, Lin CC, Huang CC, Liu CS, Lai SW. Association between waist-to-hip ratio and chronic kidney disease in the elderly. In-tern Med J 2008;38:402–6.

[21] Snyder RW, Berns JS. Use of insulin and oral hypoglycemic medica-tions in patients with diabetes mellitus and advanced kidney disease. Semin Dial 2004;17:365–70.

[22] USRDS. United States renal data system—excerpts from the 2003 data report. Am J Kidney Dis 2003;42(suppl 5):S1–230.

[23] Presti RL, Carollo C, Caimi G. Wine consumption and renal diseases: new perspectives. Nutrition 2007;23:598–602.

[24] Wen CP, Cheng TY, Tsai MK, Chang YC, Chan HT, Tsai SP, et al. All-cause mortality attributable to chronic kidney disease: a prospective co-hort study based on 462 293 adults in Taiwan. Lancet 2008; 371:2173–82.

[25] Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, Eggers P, et al. Prevalence of chronic kidney disease in the United States. JAMA 2007;298:2038–47.

[26] Bohm F, Tinkler JH, Truscott TG. Carotenoids protect against cell membrane damage by the nitrogen dioxide radical. Nat Med 1995; 1:98–9.

[27] Rao AV, Agarwal S. Bioavailability and in vivo antioxidant properties of lycopene from tomato products and their possible role in the preven-tion of cancer. Nutr Cancer 1998;31:199–203.

[28] Martin KR, Wu D, Meydani M. The effect of carotenoids on the expres-sion of cell surface adheexpres-sion molecules and binding of monocytes to human aortic endothelial cells. Atherosclerosis 2000;150:265–74. [29] Abahusain MA, Al-Nahedh NN. The biochemical status of vitamin

A and alpha-tocopherol during different stages of renal disease and its relationship to diabetes. Saudi J Kidney Dis Transpl 2002; 13:18–23.

[30] Jaconi S, Saurat JH, Siegenthaler G. Analysis of normal and trun-cated holo- and apo-retinol-binding protein (RBP) in human serum: altered ratios in chronic renal failure. Eur J Endocrinol 1996; 134:576–82.

[31] Glover J, Moxley L, Muhilal H, Weston S. Micromethod for fluorimet-ric assay of retinol-binding protein in blood–plasma. Clin Chim Acta 1974;50:371–80.

[32] Bernard A, Vyskocyl A, Mahieu P, Lauwerys R. Effect of renal-insuf-ficiency on the concentration of free retinol-binding protein in urine and serum. Clin Chim Acta 1988;171:85–93.

[33] Cabre A, Lazaro I, Girona J, Manzanares J, Marimon F, Plana N, et al. Retinol-binding protein 4 as a plasma biomarker of renal dysfunction and cardiovascular disease in type 2 diabetes. J Intern Med 2007; 262:496–503.

[34] Ziegelmeier M, Bachmann A, Seeger J, Lossner U, Kratzsch J, Bluher M, et al. Serum levels of adipokine retinol-binding protein-4 in relation to renal function. Diabetes Care 2007;30:2588–92. [35] Papavasileiou V, Liakopoulos V, Sakkas GK, Hadjigeorgiou GM,

Koukoulis G, Stefanidis I. Serum levels of adipokine retinol-binding protein-4 in relation to renal function: response to Ziegelmeier et al. Di-abetes Care 2008;31:e23.

[36] Henrich WL, Clark RL, Kelly JP, Buckalew VM, Fenves A, Finn WF, et al. Non–contrast-enhanced computerized tomography and analgesic-related kidney disease: report of the national analgesic nephropathy study. J Am Soc Nephrol 2006;17:1472–80.

[37] Sandler DP, Smith JC, Weinberg CR, Buckalew VM Jr, Dennis VW, Blythe WB, Burgess WP. Analgesic use and chronic renal disease. N Engl J Med 1989;320:1238–43.

[38] Perneger TV, Whelton PK, Klag MJ. Risk of kidney failure associated with the use of acetaminophen, aspirin, and nonsteroidal antiinflamma-tory drugs. N Engl J Med 1994;331:1675–9.

[39] Lipton RB, Raskin N, Ruoff G, Saper J, Sheftell F, Solomon S. Over-the-counter analgesics and kidney disease. Am J Kidney Dis 1996; 27:917–8.

[40] Choudhury D, Ahmed Z. Drug-associated renal dysfunction and injury. Nat Clin Pract Nephrol 2006;2:80–91.

[41] Hickey EJ, Raje RR, Reid VE, Gross SM, Ray SD. Diclofenac induced in vivo nephrotoxicity may involve oxidative stress-mediated massive genomic DNA fragmentation and apoptotic cell death. Free Radic Biol Med 2001;31:139–52.