Herbal Therapy Is Associated With the Risk of CKD in Adults Not Using Analgesics in Taiwan

Herbal Therapy Is Associated With the Risk of CKD in Adults Not Using Analgesics in Taiwan

Herbal Therapy Is Associated With the Risk of CKD in Adults Not Using Analgesics in Taiwan Jinn-Yuh Guh, MD,1 Hung-Chun Chen, MD,1 Jung-Fa Tsai, MD,1 ...

116KB Sizes 0 Downloads 11 Views

Herbal Therapy Is Associated With the Risk of CKD in Adults Not Using Analgesics in Taiwan Jinn-Yuh Guh, MD,1 Hung-Chun Chen, MD,1 Jung-Fa Tsai, MD,1 and Lea-Yea Chuang, PhD2 Background: Taiwan has the greatest incidence rate of end-stage renal disease in the world. Several cases of Chinese herb nephropathy were reported in Taiwan. Therefore, we studied the association between herbal therapy and chronic kidney disease (CKD) in Taiwan. Study Design: Cross-sectional survey. Setting & Participants: 1,740 adults in the Nutrition and Health Survey in Taiwan (1993 to 1996). Predictor: Herbal and analgesic therapy. Outcomes & Measurements: CKD after adjustment for potential confounding variables. Results: Among medication users, prevalences of herbal therapy and analgesic use were 21.6% and 13.2%, respectively. The prevalence of CKD was 9.9%. Participants with CKD were older and had more analgesic use, diabetes, hypertension, and cardiovascular disease. Analgesic use was associated independently and positively with CKD (odds ratio, 2.2; 95% confidence interval, 1.4 to 3.5; P ⫽ 0.003) and CKD stage (odds ratio, 2.3; 95% confidence interval, 1.4 to 3.6; P ⫽ 0.003). Conversely, herbal therapy was associated independently and positively with CKD (odds ratio, 1.39; 95% confidence interval, 1.2 to 1.7; P ⫽ 0.002) and CKD stage (odds ratio, 1.38; 95% confidence interval, 1.1 to 1.7; P ⫽ 0.004) only in participants who did not use analgesics. Limitations: Because this was a cross-sectional study, cause and effect could not be ascertained. Conclusions: Herbal therapy was associated with CKD in adults in Taiwan who did not use analgesics. Am J Kidney Dis 49:626-633. © 2007 by the National Kidney Foundation, Inc. INDEX WORDS: Chronic kidney disease; herbs; analgesics; epidemiology.

C

hronic kidney disease (CKD) and resultant end-stage renal disease (ESRD) are threatening to reach epidemic proportions during the next decade worldwide.1 The major risk factors for CKD are diabetes mellitus (DM), hypertension, cardiovascular disease, hyperlipidemia, obesity, analgesic use, and smoking.1,2 From the Departments of 1Internal Medicine and 2Biochemistry, Faculty of Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan. Received September 2, 2006; accepted in revised form February 13, 2007. Originally published online as doi:10.1053/j.ajkd.2007.02.259 on March 27, 2007. Support: This study was supported by grant no. DOH93TD-D-113-021(2) from the Department of Health, Taiwan. Data analyzed in this article were collected by the research project Nutrition and Health Survey in Taiwan (NAHSIT) sponsored by the Department of Health in Taiwan (DOH FN8202, DOH-83-FS-41, DOH-84-FS-11, DOH-85-FS-11, DOH-86-FS-11). Potential conflicts of interest: None. Address reprint requests to Lea-Yea Chuang, PhD, Department of Biochemistry, Kaohsiung Medical University, 100 Zihyou 1st Rd, Kaohsiung, Taiwan 807. E-mail: [email protected] nsysu.edu.tw © 2007 by the National Kidney Foundation, Inc. 0272-6386/07/4905-0009$32.00/0 doi:10.1053/j.ajkd.2007.02.259 626

The incidence of ESRD was 384 per million population in 2003 in Taiwan, which was the greatest incidence in the world.3 Moreover, a recent study found that the prevalence of persons with decreased glomerular filtration rate (GFR; ⬍60 mL/min/1.73 m2 [⬍1.0 mL/s/ 1.73m2]) was 6.9% in Taiwan.4 The very high prevalence of CKD and incidence of ESRD in Taiwan makes the study of risk factors for CKD a high priority issue in Taiwan’s health care system. There are 3 health care systems in Taiwan: modern medicine, traditional Chinese medicine, and folk medicine.5 As of 1992, the ratio of modern medicine physicians to traditional Chinese medicine physicians in Taiwan was 8.57.6 Many people in Taiwan who access traditional Chinese medicine and folk medicine habitually use prescribed or over-the-counter herbal remedies to treat diseases.7 Unfortunately, there have been several case reports of Chinese herb nephropathy in Taiwan.8-11 Therefore, we studied the association between herbal therapy and CKD in Taiwan by using the Nutrition and Health Survey in Taiwan (NAHSIT; 1993 to 1996) database.

American Journal of Kidney Diseases, Vol 49, No 5 (May), 2007: pp 626-633

Herb Use and CKD

627 METHODS

Study Population NAHSIT was a complex survey with stratified multistage probability sampling.12,13 Details of the study procedure were described in our previous study.14 Briefly, Taiwan (with ⬃21 million inhabitants in 1993 to 1996) was stratified into 7 strata, and in each stratum, 3 townships were selected with the selection probability proportional to the population size of the township. Among the 7 strata, there were 9 cities (population ⱖ 100,000) and 12 rural areas (population ⬍ 100,000). A total of 9,961 persons aged 4 to 96 years were enrolled. All NAHSIT enrollees signed the informed consent form. By design, only enrollees in the first and third year of NAHSIT (N ⫽ 6,594) underwent serum creatinine measurement. Of these enrollees, only nonpregnant adults aged 20 to 96 years (target population; N ⫽ 3,292) were included in this study. This study was approved by the institutional review board. Enrollees with the following conditions were excluded: not undergoing physical examination or phlebotomy (N ⫽ 1,198) and missing data for smoking or alcohol drinking (N ⫽ 96), medication use (N ⫽ 28), DM or hypertension (N ⫽ 46), serum creatinine (N ⫽ 333), or blood glucose (N ⫽ 178). Thus, data from 1,740 participants were available for analysis in this study. Note that some enrollees met more than 1 exclusion criteria.

Interview Details of the interview and physical examination were described in our previous study.14 Briefly, the household interview and physical examination were performed by trained technicians. Smoking was coded as pack-years (pack/ day ⫻ years of smoking; 20 cigarettes/pack). Alcoholic drinks in Taiwan were classified into 9 categories according to alcohol concentration,15 and alcohol drinking was coded as drink-years (drinks/day ⫻ years). Note that a “drink” is defined as the amount of an alcoholic drink that contains one-half ounce (13.7 g) of alcohol.16 The question for DM was “Have you been diagnosed as having DM (also called “a disease with sugar in urine” in lay terms) by a physician?” The question for hypertension was “Have you been diagnosed as having hypertension (also called “high blood pressure” in lay terms) by a physician?” Questions for cardiovascular disease were “Do you have physician-diagnosed heart disease?” and “Do you have physician-diagnosed stroke?” The question for medication use was “What is (are) your medication(s) on a regular basis?” Analgesic therapy was coded as “yes/no (specify)” and identified by the specified medication names, which were composite names (specifying indications for the relief of pain). Herbal therapy was coded as “others (specify)” and identified by the specified herbal names, which were composite brand names of multiple herbs. Note that disclosure of medication ingredients was not mandatory in 1993 to 1996 in Taiwan,17 and most herbal therapies, including those of this study, are combinations of multiple herbs with unknown ingredients.18 Moreover, there are 18 pharmacological groups of herbs in traditional Chinese medicine.19 However, many of these groups have no equivalents in modern medicine.19 Thus, we classified these herbs according to their indications.

Additionally, dietary supplements (vitamins, minerals, or herbs intended to supplement the diet) were not counted as “herbal therapy.” Body weight was measured by using a weighing scale, with subjects wearing light clothing, to the nearest 0.1 kg.14 Body mass index (BMI) was calculated as body weight (kg)/body height (m)2. Obesity was defined as BMI of 27 kg/m2 or greater according to the official definition of obesity in Taiwan.20 Blood pressure was measured by using a mercury sphygmomanometers after participants rested for 5 minutes in the supine position.20 Systolic and diastolic blood pressures were recorded as the first and fifth phases of Korotkoff’s sound, respectively. Two measurements were made 30 seconds apart. If the 2 measurements differed by more than 10 mm Hg, a third measurement was made and the 2 closest blood pressure readings were averaged. Hypertension was defined as blood pressure of 140/90 mm Hg or greater or known hypertension (physician-diagnosed hypertension or current use of antihypertensive agents). DM was defined as fasting (whole) blood glucose level of 110 mg/dL or greater or known DM (physician-diagnosed DM or current use of hypoglycemic agents).

Measurements Fasting whole blood glucose was measured by means of the glucose oxidase method (YSI portable model 23A; YSI Co, Taipei, Taiwan) immediately after blood was drawn. Fasting morning blood samples were drawn and centrifuged on site. Serum was stored at ⫺70°C until cholesterol and creatinine were measured by using the Hitachi 747 autoanalyzer (Hitachi, Tokyo, Japan). The first freshly voided morning urine was collected. Two milliliters of collected urine was aliquotted in test tubes and measured using immediate semiquantitative urine protein dipstick tests (graded negative, trace, 1, 2, 3, and 4 levels). Proteinuria was defined as the presence of at least level 1 proteinuria. GFR was calculated as the Cockcroft-Gault formula– derived creatinine clearance21: GFR (mL/min/1.73 m2) ⫽ [(140 ⫺ age) ⫻ body weight (kg) ⫻ 0.85 if female]/[serum creatinine (mg/dL) ⫻ 72] ⫻ 1.73/body surface area (m2), where body surface area was calculated using the DuBois equation.21 Participants with GFR greater than 200 mL/min/ 1.73 m2 (⬎3.33 mL/s/1.73 m2) were assigned a maximum of 200 mL/min/1.73 m2 (3.33 mL/s/1.73 m2).22 Note that the Cockcroft-Gault formula was validated in the Chinese population21 and shown to be more accurate than the Modification of Diet in Renal Disease formula in Asians.23 CKD was defined as the presence of proteinuria or GFR less than 60 mL/min/1.73 m2 (⬍1.0 mL/s/1.73 m2) as defined by the Kidney Disease Outcomes Quality Initiative of the National Kidney Foundation.1 CKD was divided further into 2 levels of severity: CKD stages 1 to 2 (GFR ⱖ 60 mL/min/1.73 m2 [ⱖ1.0 mL/s/1.73 m2] with proteinuria) and stages 3 to 5 (GFR ⬍ 60 mL/min/1.73 m2 [⬍1.0 mL/s/1.73 m2]).

Statistics Stata statistical package (version 8.2; Stata Corp, College Station, TX) was used. Serum creatinine was log-trans-

628

Guh et al

formed to obtain the normal distribution before applying statistical tests. Data are expressed as mean ⫾ SEM. Statistical significance is defined as P less than 0.05. The weighted “svy” or “robust” commands were used to account for the probability sampling weight, complex survey design, and stratification in NAHSIT. Unpaired t-tests were used to compare differences between 2 continuous variables. Chi-square tests were used to compare differences between 2 categorical variables. Logistic regression and ordinal logistic regression analysis were used to test for the association of herbal therapy with CKD and the 2 severity levels of CKD after adjustment for age, sex, obesity, DM, hypertension, cardiovascular disease, smoking, analgesic use, alcohol drinking, and total cholesterol level. Interactions24 between herbal therapy and sex and between herbal therapy and analgesic use were prespecified and tested in multiple logistic regression analyses based on the following reasons. First, the focus of this study is the effect of herbal therapy on renal function. Second, there was a significant difference in sex between participants and nonparticipants. Finally, analgesic use is a well-known cause of CKD and they often are used with herbs.25 The trend test was performed by means of orthogonal contrasts in multiple logistic regression analyses.14

RESULTS

Demographic Characteristics by Participation

The participation rate was 52.9% (1,740 of 3,292). As listed in Table 1, there was a significantly smaller proportion of men in participants than nonparticipants. Thus, men and women were analyzed separately when there were significant interactions between sex and herbal therapy. Conversely, there were no significant differences in residence area (city versus rural area), age, smoking (pack-years), alcohol drinking (drink-years), herbal therapy, analgesic use, known hypertension, known DM, and cardiovascular diseases between participants and nonparticipants.

It was not possible to compare BMI, fasting blood glucose level, blood pressure, CKD, serum creatinine level, GFR, and proteinuria between participants and nonparticipants because most nonparticipants did not have these data. However, a recent nationally representative study (n ⫽ 5,936) in Taiwan26 found that mean BMI was 23.7 ⫾ 3.7 kg/m2 in Taiwanese, which was not significantly different from mean BMI (22.8 ⫾ 0.2 kg/m2) in NAHSIT participants (P ⫽ 0.89). Demographic Characteristics by Herbal Therapy

The overall prevalence of medication use was 26.7%. In medication users, prevalences of herbal therapy and analgesic use were 21.6% and 13.2%, respectively. As listed in Table 2, there were no differences in sex, BMI, alcohol drinking, analgesic use, DM, hypertension, cardiovascular disease, proteinuria, GFR, or serum creatinine and cholesterol levels between participants who did and did not use herbal therapy. Conversely, herbal therapy users smoked more than nonusers. Additionally, women herbal therapy users were older than women nonusers. Indications for Herbal Therapy

Indications for herbal therapy are listed in Table 3. Prevalence of Proteinuria by 5 Levels of GFR

The overall prevalence of proteinuria was 3.3%, similar to that (3.2%) of a recent study in Hong Kong.27 As listed in Table 4, there was no linear trend in the prevalence or level of protein-

Table 1. Demographic Characteristics of NAHSIT Participants by Participation Participants

No. of subjects City (%) Age (y) Men (%) Smoking (pack-y) Alcohol drinking (drink-y) Herbal therapy (%) Analgesic use (%) Known hypertension (%) Known DM (%) Cardiovascular disease (%) Note: N ⫽ 3,292.

1,740 53.7 43 ⫾ 0.6 49 4.2 ⫾ 0.2 16 ⫾ 3 5.7 3.5 12.5 3.6 4.1

Nonparticipants

1,552 54.9 42.6 ⫾ 0.6 55 4.4 ⫾ 0.3 18 ⫾ 5 3.8 2 14.5 2.8 4.08

P

— 0.93 0.32 ⬍0.001 0.43 0.1 0.18 0.12 0.21 0.34 0.46

Herb Use and CKD

629

Table 2. Demographic Characteristics of NAHSIT Participants by Herbal Therapy

No. of subjects Age, men (y) Age, women (y) Men (%) BMI (kg/m2) Smoking (pack-y) Alcohol drinking (drink-y) Analgesic use (%) DM (%) Hypertension (%) Cardiovascular disease (%) Serum creatinine (mg/dL) GFR (mL/min/1.73 m2) Proteinuria (%) Level of proteinuria (dipstick) in participants with proteinuria Serum cholesterol (mg/dL)

Herbal Therapy

No Herbal Therapy

P

69 39 ⫾ 3 50 ⫾ 2 66 22.7 ⫾ 0.7 6⫾1 20 ⫾ 6 1.9 4.2 22 3.6 0.89 ⫾ 0.03 92 ⫾ 2 2.8 1 (0)* 199 ⫾ 8

1,671 43 ⫾ 1 41 ⫾ 1 49 23 ⫾ 0.2 4 ⫾ 0.2 16 ⫾ 5 3.6 5.1 25 4.6 0.86 ⫾ 0.01 95 ⫾ 1 3.3 1 (0)* 191 ⫾ 3

— 0.34 0.001 0.052 0.84 0.047 0.3 0.45 0.73 0.49 0.44 0.23 0.3 0.1 0.99 0.45

Note: N ⫽ 1,740. To convert serum creatinine in mg/dL to ␮mol/L, multiply by 88.4; GFR in mL/min to mL/s, multiply by 0.01667; serum cholesterol in mg/dL to mmol/L, multiply by 0.0259. *Interquartile range.

uria by the 5 levels of GFR. However, the prevalence of proteinuria was significantly greater in those with a GFR of 30 to 59 mL/min/1.73 m2 (0.5 to 0.98 mL/s/1.73 m2) than in those with a GFR of 60 to 89 mL/min/1.73 m2 (1.0 to 1.48 mL/s/1.73 m2; 8.4% versus 2.4%, P ⫽ 0.002). Demographic Characteristics by CKD

The overall prevalence of CKD was 9.9% (men versus women, 11.4% ⫾ 2% versus 8.5% ⫾ Table 3. Indications for Herbal Therapy No. of Subjects

Pain

9

Circulatory disorders

8

Gastrointestinal disorders Respiratory disorders Neuropsychiatric Ear, nose, and throat disorders Diuretics Menstrual disorders Miscellaneous

8

Note: N ⫽ 69.

3 3 3 2 1 32

Comments

Sciatica, arthritis, rheumatism, unspecified pain Hyperlipidemia, sports injury Digestives, dyspepsia, constipation, hepatitis Asthma, cough Numbness, insomnia Allergic rhinitis, epistaxis Urinary tract diseases Irregular menstruation Tonics, growth retardation, unspecified symptoms

2%; P ⫽ 0.2). As listed in Table 5, there were no significant differences in sex, obesity, herbal therapy, alcohol drinking, or serum cholesterol levels between participants with and without CKD. Conversely, participants with CKD were older and had more analgesic use, DM, hypertension, and cardiovascular disease. Additionally, men with CKD smoked more than men without CKD; there were no differences in smoking between women with and without CKD. Determinants of CKD

CKD was associated positively with analgesic use (odds ratio, 2.2; 95% confidence interval, 1.4 to 3.5; P ⫽ 0.003) and there was a Table 4. Prevalence of Proteinuria in NAHSIT Participants by 5 GFR Levels

GFR (mL/min/ 1.73 m2)

ⱖ90 60-89 30-59 15-29 ⬍15 P for trend

Prevalence of Each GFR

Prevalence of Proteinuria in Each GFR

60.4% (748/1,740) 32.4% (754/1,740) 6.84% (226/1,740) 0.354% (11/1,740) 0.006% (1/1,740) —

3.1% (27/748) 2.4% (29/754) 8.4%* (16/226) 0.4% (1/11) 0% (0/1) 0.73

Note: N ⫽ 1,740. Prevalences were weighted estimates. To convert GFR in mL/min to mL/s, multiply by 0.01667. *P ⫽ 0.002 versus GFR of 60 to 89 mL/min/1.73 m2.

630

Guh et al

Table 5. Demographic Characteristics of NAHSIT Participants by CKD CKD

No CKD

P

No. of subjects 292 1,448 — Age (y) 62 ⫾ 3 40 ⫾ 1 ⬍0.001 Men (%) 57 49 0.2 Obesity (%) 14 9.7 0.07 Smoking, men (pack-y) 16 ⫾ 3 7 ⫾ 0.5 0.03 Smoking, women (pack-y) 0.07 ⫾ 0.05 0.15 ⫾ 0.07 0.34 Alcohol drinking (drink-y) 19 ⫾ 3 16 ⫾ 6 0.36 Herbal therapy (%) 6.6 5.6 0.59 Analgesic use (%) 6.7 3.1 0.003 DM (%) 15 4 0.001 Hypertension (%) 49 22 ⬍0.001 Cardiovascular disease (%) 8.6 4.6 0.001 Cholesterol (mg/dL) 204 ⫾ 9 190 ⫾ 2 0.12 Note: N ⫽ 1,740. To convert serum cholesterol in mg/dL to mmol/L, multiply by 0.0259.

significant interaction between herbal therapy and analgesic use (P ⫽ 0.01). Thus, determinants of CKD were analyzed separately for participants with and without analgesic use. As listed in Table 6, CKD was not associated independently with herbal therapy in participants with analgesic use (P ⫽ 0.5). Conversely, CKD was associated independently and positively with herbal therapy in participants without analgesic use (odds ratio, 1.39; 95% confidence interval, 1.2 to 1.7; P ⫽ 0.002). Note there was no significant interaction between sex and herbal therapy.

Determinants of the 2 Severity Levels of CKD

Overall prevalences of CKD stages 1 and 2 and stages 3 to 5 were 2.7% and 7.2%, respectively. The 2 severity levels of CKD were associated positively with analgesic use (odds ratio, 2.3; 95% confidence interval, 1.4 to 3.6; P ⫽ 0.003) and there was a significant interaction between herbal therapy and analgesic use (P ⬍ 0.001). Thus, determinants of the 2 severity levels of CKD were analyzed separately for participants with and without analgesic use. As listed in Table 7, the 2 severity levels of CKD were not associated independently with herbal therapy in participants with analgesic use (P ⫽ 0.48). Conversely, the 2 severity levels of CKD were associated independently and positively with herbal therapy in participants without analgesic use (odds ratio, 1.38; 95% confidence interval, 1.1 to 1.7; P ⫽ 0.004). Note there was no significant interaction between sex and herbal therapy. Interestingly, CKD apparently was not associated with cardiovascular disease. However, CKD stages 3 to 5 were associated independently and positively with cardiovascular disease (odds ratio, 2; 95% confidence interval, 1.1 to 3.7; P ⫽ 0.02). This finding is similar to previous studies.28 DISCUSSION

This is the first demonstration on a population basis that herbal therapy is associated independently with CKD in adults who were not using analgesics. Moreover, the prevalence of CKD was as high as 9.9% in Taiwan. These findings

Table 6. Determinants of CKD in NAHSIT Participants by Analgesic Use Analgesic Use (N ⫽ 136)

No Analgesic Use (N ⫽ 1,604)

CKD

Odds Ratio (95% confidence interval)

P

Odds Ratio (95% confidence interval)

P

Age (y) Male sex Obesity Smoking (pack-y) Alcohol drinking (drink-y) Herbal therapy DM Hypertension Cardiovascular disease Cholesterol (mg/dL)

1.5 (1.2-1.9) 25 (3-62) 1.3 (0.2-10) 0.97 (0.93-1.02) 0.99 (0.98-1.01) 0.13 (0.001-3.3) 0.83 (0.05-13) 0.6 (0.1-3.5) 0.6 (0.06-5.6) 1.01 (0.98-1.06)

0.002 0.004 0.8 0.2 0.45 0.5 0.89 0.54 0.6 0.49

1.09 (1.05-1.14) 1.3 (0.7-2.6) 0.96 (0.5-1.8) 1.01 (0.99-1.02) 0.99 (0.97-1.02) 1.39 (1.2-1.7) 1.26 (0.4-3.8) 1.05 (0.8-1.4) 1.4 (0.8-2.4) 1.0 (0.99-1.01)

0.001 0.35 0.89 0.23 0.6 0.002 0.66 0.72 0.25 0.96

Note: To convert serum cholesterol in mg/dL to mmol/L, multiply by 0.0259.

Herb Use and CKD

631

Table 7. Determinants of the 2 Severity Levels of CKD in NAHSIT Participants by Analgesic Use Analgesic Use (N ⫽ 136)

No Analgesic Use (N ⫽ 1,604)

Severity of CKD

Odds Ratio (95% confidence interval)

P

Odds Ratio (95% confidence interval)

P

Age (y) Male sex Obesity Smoking (pack-y) Alcohol drinking (drink-y) Herbal therapy DM Hypertension Cardiovascular disease Cholesterol (mg/dL)

1.6 (1.2-2.3) 24 (3-56) 0.58 (0.1-3) 0.97 (0.92-1.01) 0.99 (0.98-1.01) 0.1 (0.001-3) 0.38 (0.03-5.4) 0.7 (0.1-4.5) 0.8 (0.08-8) 1.02 (0.98-1.07)

0.008 0.002 0.49 0.15 0.51 0.48 0.45 0.68 0.87 0.31

1.1 (1.06-1.2) 1.3 (0.7-2.6) 0.93 (0.5-1.8) 1.01 (0.99-1.03) 0.99 (0.98-1.001) 1.38 (1.1-1.7) 1.2 (0.4-3.5) 1.1 (0.8-1.5) 1.5 (0.85-2.5) 1.001 (0.99-1.01)

⬍0.001 0.37 0.82 0.21 0.63 0.004 0.74 0.54 0.15 0.98

Note: To convert serum cholesterol in mg/dL to mmol/L, multiply by 0.0259.

have important epidemiological implications because Taiwan is number 1 in the world in the incidence of ESRD.3 Some of the major CKD risk factors were confirmed, whereas others were not confirmed in univariate analysis in this study. However, multivariate analyses are more appropriate to account for so many risk factors simultaneously. Thus, herbal therapy was associated independently and positively with CKD in participants not using analgesics in multiple logistic regression analyses. Use of herbal therapy has increased dramatically in recent years worldwide.29 For example, a World Health Organization survey indicated that 70% to 80% of the world populations rely mainly on herbal therapies.18 Kidney syndromes caused by herbs30 consist of acute tubular necrosis/toxicity, acute interstitial nephritis, papillary necrosis, hypertension, kidney stones, urinary retention, chronic tubulointerstitial nephritis with fibrosis, and urinary tract carcinoma. There are several mechanisms for explaining herb-induced adverse effects31: inherent toxicity, drug-herb interactions, adulteration, contamination, coexisting diseases, idiosyncrasy, substitution/ misidentification, improper processing and preparation, variability in ingredients, and overdoses. Inherent toxicity is present in many herbs.30,32 For example, licorice, ephedra, ginseng, dong quai, and yohimbe have hypertensive effects.33,34 However, the most prominent example of herbinduced nephrotoxicity is Chinese herb nephropathy, caused by aristolochic acid contamination of some Chinese herbs.29,32 Moreover, herbs can interact with analgesics, whereas some herbdrug interactions are nephrotoxic.25,30,35

Many herbs in Taiwan are adulterated (up to 23.7%).31,36 The most common adulterants (listed in order of frequency) include caffeine, acetaminophen, indomethacin, hydrochlorothiazide, and so on.31,36 Caffeine increases blood pressure.37 Hydrochlorothiazide can be nephrotoxic by inducing prerenal azotemia, acute interstitial nephritis, or hypokalemia.38,39 Acetaminophen (analgesic) and indomethacin (nonsteroidal anti-inflammatory drug) are nephrotoxic.38 Thus, we speculate that analgesic-adulterated herbs account for our finding that herbal therapy was associated with CKD only in participants not using analgesics. However, we cannot assume that all herbs for pain relief were analgesic adulterated and simply classify them as classic “analgesics” and classify herbs not for pain relief as “not analgesics” based on the following reasons. First, some herbs (without adulterants) can relieve pain.40 Second, herbs for pain relief are not more likely adulterated with classic analgesics than herbs not for pain relief. For example, a previous study in Taiwan found that of the 618 adulterated herbs, only 301 (49%) were used for pain relief.17 Nephrotoxic heavy metals, such as lead and mercury, contaminate many herbs.30,41 For example, several studies showed that low-level lead exposure contributed to a considerable number of patients with CKD in Taiwan.42-44 Moreover, two previous studies showed that herbal therapy was a risk factor for high blood levels of lead in Taiwan.45,46 Coexisting diseases can affect pharmacological characteristics and toxicity of herbs.31 Additionally, improper processing and preparation may affect pharmacological characteristics and

632

Guh et al

toxicity of Chinese herbs.31 Substitution/misidentification is another cause of herbal toxicity. Stephania tetrandra substituted with Aristolochia fangchit was the cause of the first reported case series of Chinese herb nephropathy.47 Hypertension and DM, 2 of the well-known risk factors for CKD, were associated positively with CKD in univariate analysis. However, they were not associated with CKD in multiple logistic regression analyses. This result is similar to 2 previous studies of south Asians48 and Australians.49 The first study found that DM and hypertension were not associated with CKD in multivariate analyses. Note that although the investigators interpreted the adjusted odds ratio for hypertension as P less than 0.05, the 95% confidence interval (1.0 to 6.81) was not statistically significant because it encompassed 1.0.50 The second study found that only hypertension (not DM) was associated with CKD.49 Being a secondary analysis, there are several limitations of this study. First, sampling and measurement issues are inevitable. However, NAHSIT was a nationally representative study, whereas all relevant measurements were made using standard methods. Second, missing data were not missing at random. However, concerning the factors impacting on renal function, there was no selection bias associated with participation other than sex, which had been adjusted for in multivariate analyses. Third, types, ingredients, and cumulative doses of herbal therapy, analgesics, and other medications were not known. Therefore, the possible mechanism of herb-induced nephrotoxicity can only be speculated or inferred from previous studies. Finally, being a cross-sectional study, causality cannot be established. In conclusion, herbal therapy is an independent risk factor for CKD in adults in Taiwan who are not using analgesics. ACKNOWLEDGEMENT NAHSIT was carried out by the Institute of Biomedical Sciences of Academia Sinica and Department of Biochemistry, College of Medicine of National Taiwan University, directed by Dr Wen-Harn Pan and Dr Po-Chao Huang. The Office of Survey Research of Academia Sinica is responsible for data distribution. The assistance provided by the institutes and aforementioned individuals is greatly appreciated. The views expressed herein are solely those of the authors.

REFERENCES 1. El Nahas AM, Bello AK: Chronic kidney disease: The global challenge. Lancet 365:331-340, 2005 2. McMurray JJ: Chronic kidney disease in patients with cardiac disease: A review of evidence-based treatment. Kidney Int 68:1419-1426, 2005 3. US Renal Data System: USRDS 2005 Annual Data Report: International comparisons. Am J Kidney Dis 47: S215-S226, 2006 (suppl 1) 4. Hsu CC, Hwang SJ, Wen CP, et al: High prevalence and low awareness of CKD in Taiwan: A study on the relationship between serum creatinine and awareness from a nationally representative survey. Am J Kidney Dis 48:727738, 2006 5. Liu CT: Health care systems in transition. II. Taiwan, Part I. A general overview of the health care system in Taiwan. J Public Health Med 20:5-10, 1998 6. Chi C, Lee JL, Lai JS, et al: The practice of Chinese medicine in Taiwan. Soc Sci Med 43:1329-1348, 1996 7. Kaphle K, Wu LS, Yang NY, Lin JH: Herbal medicine research in Taiwan. Evid Based Complement Alternat Med 3:149-155, 2006 8. Yang CS, Lin CH, Chang SH, Hsu HC: Rapidly progressive fibrosing interstitial nephritis associated with Chinese herbal drugs. Am J Kidney Dis 35:313-318, 2000 9. Yang HY, Lin JL, Chen KH, et al: Aristolochic acidrelated nephropathy associated with the popular Chinese herb Xi Xin. J Nephrol 19:111-114, 2006 10. Chang CH, Wang YM, Yang AH, Chiang SS: Rapidly progressive interstitial renal fibrosis associated with Chinese herbal medications. Am J Nephrol 21:441-448, 2001 11. Hong YT, Fu LS, Chung LH, et al: Fanconi’s syndrome, interstitial fibrosis and renal failure by aristolochic acid in Chinese herbs. Pediatr Nephrol 21:577-579, 2006 12. Pan WH, Yeh WT, Chang HY, Hwu CM, Ho LT: Prevalence and awareness of diabetes and mean fasting glucose by age, sex, and region: Results from the Nutrition and Health Survey in Taiwan, 1993-1996. Diabet Med 20:182-185, 2003 13. Pan WH, Kao MD, Tzeng MS, et al: Nutrition and Health Survey in Taiwan (NAHSIT) 1993-1996: Design, contents, and operations. Nutr Sci J 24:1-10, 1999 14. Guh JY, Chuang LY, Chen HC: Betel quid use is associated with the risk of metabolic syndrome in adults. Am J Clin Nutr 83:1313-1320, 2006 15. Lin YC, Yen LL, Chen SY, et al: Prevalence of overweight and obesity and its associated factors: Findings from National Nutrition and Health Survey in Taiwan, 1993-1996. Prev Med 37:233-241, 2003 16. Lichtenstein AH, Appel LJ, Brands M, et al: Diet and lifestyle recommendations revision 2006: A scientific statement from the American Heart Association Nutrition Committee. Circulation 114:82-96, 2006 17. Huang WF, Wen KC, Hsiao ML: Adulteration by synthetic therapeutic substances of traditional Chinese medicines in Taiwan. J Clin Pharmacol 37:344-350, 1997 18. Chan K: Some aspects of toxic contaminants in herbal medicines. Chemosphere 52:1361-1371, 2003 19. Cheng JT: Review: Drug therapy in Chinese traditional medicine. J Clin Pharmacol 40:445-450, 2000

Herb Use and CKD 20. Pan WH, Flegal KM, Chang HY, et al: Body mass index and obesity-related metabolic disorders in Taiwanese and US whites and blacks: Implications for definitions of overweight and obesity for Asians. Am J Clin Nutr 79:31-39, 2004 21. Zuo L, Ma YC, Zhou YH, et al: Application of GFR-estimating equations in Chinese patients with chronic kidney disease. Am J Kidney Dis 45:463-472, 2005 22. Coresh J, Astor BC, Greene T, Eknoyan G, Levey AS: Prevalence of chronic kidney disease and decreased kidney function in the adult US population: Third National Health and Nutrition Examination Survey. Am J Kidney Dis 41:112, 2003 23. Jafar TH, Schmid CH, Levey AS: Serum creatinine as marker of kidney function in South Asians: A study of reduced GFR in adults in Pakistan. J Am Soc Nephrol 16:1413-1419, 2005 24. Fitzmaurice G: The meaning and interpretation of interaction. Nutrition 16:313-314, 2000 25. Abebe W: Herbal medication: Potential for adverse interactions with analgesic drugs. J Clin Pharm Ther 27:391401, 2002 26. Hwang LC, Bai CH, Chen CJ: Prevalence of obesity and metabolic syndrome in Taiwan. J Formos Med Assoc 105:626-635, 2006 27. Li PK, Kwan BC, Leung CB, et al: Prevalence of silent kidney disease in Hong Kong: The screening for Hong Kong Asymptomatic Renal Population and Evaluation (SHARE) program. Kidney Int Suppl 94:S36-S40, 2005 28. Go AS, Lo JC: Epidemiology of non-dialysisrequiring chronic kidney disease and cardiovascular disease. Curr Opin Nephrol Hypertens 15:296-302, 2006 29. Isnard Bagnis C, Deray G, Baumelou A, Le Quintrec M, Vanherweghem JL: Herbs and the kidney. Am J Kidney Dis 44:1-11, 2004 30. Colson CR, De Broe ME: Kidney injury from alternative medicines. Adv Chronic Kidney Dis 12:261-275, 2005 31. Ko RJ: A U.S. perspective on the adverse reactions from traditional Chinese medicines. J Chin Med Assoc 67:109-116, 2004 32. Wojcikowski K, Johnson DW, Gobe G: Medicinal herbal extracts—Renal friend or foe? Part one: The toxicities of medicinal herbs. Nephrology (Carlton) 9:313-318, 2004 33. van Uum SH: Liquorice and hypertension. Neth J Med 63:119-120, 2005 34. Richard CL, Jurgens TM: Effects of natural health products on blood pressure. Ann Pharmacother 39:712-720, 2005 35. Magee KA: Herbal therapy: A review of potential health risks and medicinal interactions. Orthod Craniofac Res 8:60-74, 2005

633 36. Huang WF, Wen KC, Hsiao ML: Adulteration by synthetic therapeutic substances of traditional Chinese medicines in Taiwan. J Clin Pharmacol 37:344-350, 1997 37. Hartley TR, Sung BH, Pincomb GA, et al: Hypertension risk status and effect of caffeine on blood pressure. Hypertension 36:137-141, 2000 38. Choudhury D, Ahmed Z: Drug-associated renal dysfunction and injury. Nature Clin Pract Nephrol 2:80-91, 2006 39. Markowitz GS, Perazella MA: Drug-induced renal failure: A focus on tubulointerstitial disease. Clin Chim Acta 351:31-47, 2005 40. Wirth JH, Hudgins JC, Paice JA: Use of herbal therapies to relieve pain: A review of efficacy and adverse effects. Pain Manag Nurs 6:145-167, 2005 41. Garvey GJ, Hahn G, Lee RV, Harbison RD: Heavy metal hazards of Asian traditional remedies. Int J Environ Health Res 11:63-71, 2001 42. Lin JL, Ho HH, Yu CC: Chelation therapy for patients with elevated body lead burden and progressive renal insufficiency. A randomized, controlled trial. Ann Intern Med 130:7-13, 1999 43. Lin JL, Lin-Tan DT, Hsu KH, Yu CC: Environmental lead exposure and progression of chronic renal diseases in patients without diabetes. N Engl J Med 348:277-286, 2003 44. Yu CC, Lin JL, Lin-Tan DT: Environmental exposure to lead and progression of chronic renal diseases: A fouryear prospective longitudinal study. J Am Soc Nephrol 15:1016-1022, 2004 45. Liou SH, Wu TN, Chiang HC, et al: Blood lead levels in Taiwanese adults: Distribution and influencing factors. Sci Total Environ 180:211-219, 1996 46. Chu NF, Liou SH, Wu TN, Ko KN, Chang PY: Risk factors for high blood lead levels among the general population in Taiwan. Eur J Epidemiol 14:775-781, 1998 47. Vanherweghem JL, Depierreux M, Tielemans C, et al: Rapidly progressive interstitial renal fibrosis in young women: Association with slimming regimen including Chinese herbs. Lancet 341:387-391, 1993 48. Domrongkitchaiporn S, Sritara P, Kitiyakara C, et al: Risk factors for development of decreased kidney function in a southeast Asian population: A 12-year cohort study. J Am Soc Nephrol 16:791-799, 2005 49. Chadban SJ, Briganti EM, Kerr PG, et al: Prevalence of kidney damage in Australian adults: The AusDiab Kidney Study. J Am Soc Nephrol 14:S131-S138, 2003 (suppl 2) 50. Shakespeare TP, Gebski VJ, Veness MJ, Simes J: Improving interpretation of clinical studies by use of confidence levels, clinical significance curves, and risk-benefit contours. Lancet 357:1349-1353, 2001