ORIGINAL INVESTIGATION. Environmental Lead Exposure and Progressive Renal Insufficiency

Transcription

1 ORIGINAL INVESTIGATION Environmental Lead Exposure and Progressive Renal Insufficiency Ja-Liang Lin, MD; Dan-Tzu Tan, RN; Kuan-Huang Hsu, PhD; Chun-Chen Yu, MD Background: Several recent studies show that serum creatinine level or creatinine clearance is inversely associated with blood lead levels. However, the studies did not allow direct inferences about causality. Objective: To evaluate the relation between body lead burden (BLB) and progressive renal insufficiency in patients without previous heavy lead exposure. Design: A prospective, longitudinal study with a controlled clinical trial. Patients: One hundred ten patients with chronic renal insufficiency (serum creatinine level, µmol/l [ mg/dl]) and normal BLB (EDTA mobilization tests, 600 µg per 72-hour urine collection) and without a history of previous heavy lead exposure were divided into 2 groups according to BLB: the high-normal BLB group (BLB 80 µg and 600 µg) and the low BLB group (BLB 80 µg). Patients were prospectively followed up for 2 years. Main Outcome Measures: The primary outcome was a 1.5 times increase in the initial creatinine level. The secondary outcome was a change over time in the value of creatinine clearance. At the end of follow-up, a 3-month clinical trial with chelation therapy for patients with high-normal BLB was performed to clarify the role of environmental lead exposure in progressive renal insufficiency. Results: Fifteen patients (14 in the high-normal BLB group and 1 in the low BLB group) reached the primary outcome within 24 months. Renal outcome was significantly better in the low BLB group (P.001). From month 12 to month 24, renal function of high-normal BLB patients had a greater rate of progressive renal insufficiency than that of low BLB patients. In the Cox multivariate regression analysis, BLB was the most important risk factor for determining the progression of renal insufficiency. After chelation therapy, significant improvement in renal function was noted. In addition, the effect of improving renal function lasted for more than 12 months in these patients. Conclusions: Long-term low-level environmental lead exposure may subtly affect progressive renal insufficiency in the general population. Progressive renal insufficiency may be improved for at least 1 year after lead chelating therapy. Further investigations are needed to clarify this observation. Arch Intern Med. 2001;161: From the Poison Center and the Division of Nephrology, Chang Gung Memorial Hospital, Lin-Kou Medical Center, Medical College (Drs Lin and Yu and Ms Tan), and the School of Public Health and Epidemiology (Dr Hsu), Chang Gung University, Taipei, Taiwan, Republic of China. LEAD IS STILL one of the major environmental pollutants in the world. It is well known to induce nephropathy in persons with heavy or occupational lead exposure, 1-4 but few studies have attempted to evaluate the renal effects of environmental low-level lead exposure. Several recent epidemiological studies 5,6 showed that serum creatinine level or creatinine clearance was inversely associated with blood lead levels. The studies do not allow direct inferences about causality because they are only cross sectional. A longitudinal retrospective study 7 demonstrated an acceleration of age-related impairment of renal function in association with blood lead levels. However, the study was retrospective and, like previous cross-sectional epidemiological studies, did not control for many factors affecting the progression of renal insufficiency, such as daily protein intake, daily urinary protein level, and use of converting enzyme inhibitors. In addition, blood lead level reflects levels of exposure to lead during the recent weeks and months only and does not indicate body lead burden (BLB). The most reliable measurements of BLB are attained using bone radiograph fluorescence and EDTA mobilization tests. 8 In previous works, 9-13 investigators used the EDTA mobilization test to assess the BLB of patients with 264

2 PATIENTS, MATERIALS, AND METHODS We conducted a 3-year, prospective, longitudinal study. The protocol was approved by the Clinical Research Committee of Chang Gung Memorial Hospital, Taipei, Taiwan, and all the patients gave informed consent. STUDY PATIENTS Men and women aged 18 to 70 years who had CRI caused by various diseases were eligible for the study if they met the following criteria: (1) a serum creatinine concentration of 133 to 354 µmol/l ( mg/dl) and follow-up for at least 6 months in the outpatient department of Chang Gung Memorial Hospital to determine whether renal function was stable and blood pressure ( 140/90 mm Hg), hyperlipidemia (cholesterol level, 6.21 mmol/l [ 240 mg/ dl]), and daily protein intake ( 1 g/kg body weight) were well controlled; (2) without a known history of previous lead exposure, BLB had to be normal ( 600 µg) as measured by calcium disodium EDTA mobilization tests and 72-hour urine collections; and (3) stable renal function (changes in creatinine clearance less than 0.08 ml/s [ 5 ml/min] during the 6-month period) and a serum creatinine level of at least 133 µmol/l (1.5 mg/dl) for 6 months preceding study entry. Renal diseases were diagnosed based on the findings of history taking, laboratory evaluations, renal echogram, and radiological and renal histological examinations. To avoid the possibility that changes in the rate of progression of renal insufficiency might be related to the intrinsic diseases rather than the BLBs, we excluded patients with (1) potential reversible renal insufficiency, such as malignant hypertension, urinary tract infection, hypercalcemia, and drug-induced nephrotoxicity; (2) systemic diseases, such as connective tissue diseases or diabetes mellitus, and receiving drugs that might alter the natural history of renal disease, such as nonsteroidal anti-inflammatory drugs, corticosteroids, or immunosuppressive agents; (3) rapid progressive glomerulonephritis or severe daily urinary protein intake ( 10 g/d); (4) previous heavy lead exposure, including histories of lead poisoning or occupational lead exposure; and (5) a drug allergy history or no informed consent. Blood pressure was controlled with the use of diuretics and converting enzyme inhibitors given with or without calcium blocking agent therapy. These drugs were not changed but were dose adjusted during the study period. Patients without hypertension did not use any converting enzyme inhibitors. Because renal function deteriorated more quickly in patients with hypertension, hypercholesterolemia, and high protein intake, we were careful to ensure that blood pressure, cholesterol level, and protein intake were well controlled in all patients. Phosphate levels were controlled by administration of calcium carbonate. No patient received vitamin D 3 (calcitriol) therapy or erythropoietin treatment. Patients received dietary consultation and were advised to follow a normal protein diet. The diet required a daily intake of 0.8 to 1.0 g of protein per kilogram provided by foods containing most or all essential amino acids, such as meat, fish, chicken, and eggs. A nutritionist reviewed the dietary intake of each patient every 3 to 6 months. The 24-hour urine urea excretion was measured every 3 months to assess nitrogen balance and dietary compliance. 15 To assess whether there were any differences in the rate of progression of renal insufficiency, serum creatinine, cholesterol, urinary protein, and urea levels were checked every 3 months. MEASUREMENTS OF BLOOD LEAD LEVEL AND BLB Body lead burden was determined using the protocol of EDTA mobilization tests developed by Emmerson and modified by Batuman et al. 16 Every patient received an intravenous infusion of 1 g of calcium disodium EDTA mixed with 200 ml of normal saline solution over 2 hours. They collected 24-hour urine samples in 2-L bottles on 3 consecutive days to assess BLBs. Urine samples were collected by spontaneous voiding. All patients were hydrated orally with water sufficient to provide a steady urine flow of at least 1 ml/min. Patients who did not have an accurate urine collection ( 1 lost urine collection) or who had inadequate urine flow collected another urine sample after the next weekly EDTA therapy. Blood lead and urine lead excretion were quantified using an electrothermal atomic absorption spectrometer (model 5100PC; Perkin-Elmer, Norwalk, Conn) with Zeeman background correction and an L vov platform (Behringer method). All urinary lead determinations were performed at least in duplicate. The hospital laboratory participated in the external quality control program for lead measurement held by the College of American Pathologists for 5 years and obtained the excellent degree every year. Continued on next page chronic renal disease without previous lead exposure. They demonstrated that long-term low-level environmental lead exposure (1) may be associated with impaired renal function 9,10 in a small group of patients, (2) can subtly affect urate excretion in patients with chronic renal disease, 11 and (3) is associated with renal tubular and glomerular damage in a general population. 12,13 However, whether long-term low-level environmental lead exposure is associated with the progression of renal insufficiency remains unknown. In addition, although lead chelation therapy has been successfully introduced to treat chronic lead-related nephropathy 2,3 in persons with occupational exposure to lead and rats with long-term lowlevel lead exposure, 14 the efficacy of this therapy for patients with low-level environmental lead exposure is unknown. To determine whether long-term low-level environmental lead exposure plays a role in the progression of renal insufficiency in patients with chronic renal insufficiency (CRI), we performed a prospective, longitudinal study with a controlled clinical trial. RESULTS One hundred ninety-six patients with a serum creatinine level of 133 to 354 µmol/l ( mg/dl) were 265

3 STUDY PROTOCOL Sample Collection Period (Months 0-3) Blood lead levels, hemoglobin levels, BLBs, and biochemical data were checked in all eligible patients. Based on the findings of a previous work 13 that the mean BLB of healthy persons is 76.6 µg and of patients with CRI is 84.5 µg, we divided our patients into 2 groups: those with low BLB ( 80 µg) and those with high-normal BLB ( 80 µg and 600 µg). Clinical Observation Period (Months 4-27) Serum creatinine, cholesterol, and daily urinary urea levels were measured at the beginning and end and every 3 months during the 24-month clinical observation period with an autoanalyzer system (model 736; Hitachi, Tokyo, Japan). Two consecutive 24-hour urine collection samples and laboratory data were obtained from each patient, and the means of the 2 measurements were recorded. Patients who did not have an accurate urine collection ( 1losturine collection) or who had inadequate urine flow ( 1 ml/ min) collected another urine sample. At the end of this period, we compared the changes in renal function between the 2 groups and assessed the relation between BLB and the progression of renal insufficiency. The primary outcome measure was a 1.5 times increase in the initial serum creatinine level, confirmed 1 month later, or the need for hemodialysis. The secondary outcome measures were changes over time in the values of creatinine clearance, urine protein excretion, daily protein intake, mean arterial pressure, serum cholesterol, and body mass index. Chelation Clinical Trial (Months 28-30) and Follow-up (Months 31-42) The high-normal BLB group assessed their BLB again at the beginning of the period. Thirty-six patients with serum creatinine levels less than 371 µmol/l ( 4.2 mg/dl) were randomly assigned to either the control or the study group (1:2). The randomization was performed according to the random digital method in which digital numbers came from a computer. This study is a single-blind experiment. The control group received a weekly intravenous infusion of 1 vial (20 ml) of 50% glucose, as a placebo, mixed with 200 ml of normal saline solution over 2 hours. The study group received a weekly intravenous infusion of 1 vial (1 g) of calcium disodium EDTA mixed with 200 ml of normal saline solution over 2 hours. Because these drugs were prepared at the pharmacy, patients did not know which drugs they received. Patients collected 24-hour urine samples in 2-L bottles for 3 consecutive days to assess BLB every 2 weeks, and the treatment would be held if the EDTA mobilization test was less than 80 µg per 72-hour urine collection. At the end of this period, we compared the changes of renal function before and after the clinical trial. The same laboratory measurements were taken every 3 months for 12 months after chelation therapy to compare the changes in renal function. STUDY COMPLIANCE Patients were withdrawn from this study if they (1) did not complete the study or regularly check biochemical data; (2) had poor control of their hypertension ( 160/95 mm Hg), hyperlipidemia (cholesterol level, 6.21 mmol/l [ 240 mg/dl]), or protein intake greater than 1.5 g/kg per day for more than 6 months; and (3) had acute deterioration of renal function secondary to drug therapy or other etiologies, eg, trauma and hyperthermia, during the study periods. STATISTICAL ANALYSIS The primary outcome measure was analyzed with the logrank test, with the status of renal function determined as of the last day of the third year of treatment. For all patients who did not complete the 2-year study, data were censored after the last visit. Cox proportional hazards regression was used to determine the significance of the variables in predicting primary renal outcome. Progressive renal insufficiency of the 2 study groups was compared. The 2 test, t test, and Mann-Whitney U test were used to measure the differences between groups. The Mann-Whitney U test was used for data not distributed in a normal fashion. To protect against a type I error due to the one planned interim analysis, the significant level for outcome measures was set at P.01; P=.05 to.01 was considered possible significance. All P values were 2-tailed. All results are presented as mean±sd. An intent-to-treat analysis was performed. In addition, we conducted a sensitivity analysis that assigned the mean value of renal function in the study group to patients lost to follow-up in the control group and the mean value of renal function in the control group to patients lost to follow-up in the study group. screened for the study. One hundred ten patients with CRI were enrolled, and 96 completed the 24 months of observation. Fifty-five patients with high-normal BLB ( 80 µg and 600 µg, estimated by EDTA mobilization tests) and 55 with low BLB ( 80 µg) eligible for the study were assigned to 2 study groups according to their BLB. Six participants (3 were noncompliant and 3 dropped out for unknown reasons) in the high-normal BLB group and 8 (4 were noncompliant, 1 died, and 3 dropped out for unknown reasons) in the low BLB group were excluded from this study. CLINICAL OBSERVATION PERIOD Table 1 summarizes demographic data (age, sex, and body mass index), baseline chronic disease condition (serum creatinine level, prevalence of hypertension, hyperlipidemia, and underlying disease distribution), use of converting enzyme inhibitors in hypertensive patients, daily urinary urea and protein levels, and BLBs for participants in each group. No significant differences in these baseline values between the 2 groups were noted. Similarly, mean arterial pressure, serum 266

4 Table 1. Baseline Characteristics of 110 Patients With an Elevated Lead Burden at Study Entry* Variable High-Normal (n = 55) Body Lead Burden Group Low (n = 55) Age, y 57.8 ± 10.0 (38-70) 56.7 ± 12.9 (34-68).64 Sex, M/F 47/8 50/5.56 Body mass index, kg/m ± 3.0 ( ) 24.9 ± 3.2 ( ).32 Serum creatinine, µmol/l 186 ± 48 ( ) 183 ± 44 ( ).76 Creatinine clearance, ml/s 0.66 ± 0.20 ( ) 0.67 ± 0.22 ( ).78 Daily proteinuria, g 0.63 ± 0.72 ( ) 0.76 ± 0.89 ( ).39# Blood lead, µmol/l 0.32 ± 0.17 ( ) 0.19 ± 0.08 ( ).001# Blood lead burden, µg ± ( ) 37.9 ± 23.1 ( ).001# Hyperlipidemia, No Hypertension, No Using CEIs for hypertension, No. 34/43 34/38.67 Smoking, No Underlying renal disease, No. Chronic glomerulonephritis Hypertensive nephropathy Gouty nephropathy Obstructive uropathy Polycystic kidney disease Unknown *Data are given as mean ± SD (range) except where indicated otherwise. CEIs indicates converting enzyme inhibitors. To convert serum creatinine from micromoles per liter to milligrams per deciliter, divide micromoles per liter by To convert creatinine clearance from milliliters per second to milliliters per minute, divide milliliters per second by To convert blood lead from micromoles per liter to micrograms per deciliter, divide micromoles per liter by Serum cholesterol level 6.21 mmol/l ( 240 mg/dl) after diet control. Blood pressure 140/90 mm Hg for at least 2 measurements. Data were analyzed using the 2 test with the Fisher exact test. #Data were measured using the Mann-Whitney U test. Data were analyzed using the t test. P cholesterol, body mass index, and daily urinary protein excretion and protein intake did not differ significantly between the 2 groups during the observation period (Table 2). Table 3 compares the progression of renal insufficiency in the high-normal BLB and low BLB groups during the observation period. Creatinine clearance in the low BLB group was greater than that in the high-normal BLB group in months 18 to 24 of observation. Similar results were also noted in the sensitivity test (Table 4). Fifteen patients (14 in the high-normal BLB group and 1 in the low BLB group) reached the primary outcome within 24 months; all had a serum creatinine level 1.5 times the initial level, and none needed hemodialysis. Renal outcome was better in the low BLB group (P.001 by log-rank test) (Figure). In the Cox regression analysis, high-normal BLB was the most important risk factor for progressive renal insufficiency, even after adjusting for age, sex, smoking, hypertension, hyperlipidemia, daily urinary protein level, daily protein intake, body mass index, and underlying diseases (Table 5). In addition, age and body mass index also significantly predict progressive renal insufficiency. Similarly, if we entered BLB in the Cox multivariate regression analysis as a continuous measurement rather than as an arbitrary categorical variable, BLB remained the important risk factor (risk ratio, 1.006; 95% confidence interval, ; P=.004) after adjusting for other relating factors. CHELATION CLINICAL TRIAL AND FOLLOW-UP Thirty-six patients with high-normal BLB participated in the clinical trial, including 24 study group patients and 12 control group patients. The basic characteristics of both groups were similar (Table 6). Following 3 months of lead chelation therapy, the BLB of the study group decreased to 39.2±29.4 µg (range, µg). The average therapeutic dose of EDTA was 5 g (range, 3-13 g). The improvement in renal function in the study group was greater than that of the control group after chelation therapy. In addition, during follow-up after chelation therapy, the effect of improving renal function in the study group persisted for at least 12 months (Table 7). Two study group patients and 1 control group patient were lost to follow-up for unknown reasons during the 12-month study. The sensitivity analysis shows similar results (Table 8). COMMENT In this prospectively longitudinal study, we first demonstrated that patients with a high-normal BLB had more rapidly progressive renal insufficiency during 2-year follow-up; even the ranges of their BLBs and blood lead levels were within reference limits. The mean blood lead level of our patients was only 0.26 µmol/l (5.4 µg/dl), which is far less than the upper limit of the reference range (0.97 µmol/l [20 µg/dl]) 7 and between those of European (

5 µmol/l [11.4 µg/dl]) 5 and American (0.13 µmol/l [2.7 µg/dl]) general populations. 17 The mean BLB of our patients was only µg, which is far less than toxic ( 1000 µg) 2 and subtle poisoning ( 600 µg) levels. 16 Table 2. Intent-to-Treat Analysis of Body Mass Index, Mean Arterial Pressure (MAP), Serum Cholesterol Level, Daily Urine Protein Level, and Daily Protein Intake in Study Patients During the 24-Month Observation Period* Body Lead Burden Group Variable High-Normal Low P Month 0 (n = 55) (n = 55) Body mass index, kg/m ± ± MAP, mm Hg 97.3 ± ± Cholesterol, mmol/l 5.20 ± ± h urinary protein, g 0.63 ± ± Daily protein intake, g/kg 1.00 ± ± Month 6 (n = 55) (n = 55) Body mass index, kg/m ± ± MAP, mm Hg 96.3 ± ± Cholesterol, mmol/l 5.10 ± ± h urinary protein, g 0.67 ± ± Daily protein intake, g/kg 0.97 ± ± Month 12 (n = 54) (n = 53) Body mass index, kg/m ± ± MAP, mm Hg 96.5 ± ± Cholesterol, mmol/l 5.10 ± ± h urinary protein, g 0.67 ± ± Daily protein intake, g/kg 0.96 ± ± Month 18 (n = 53) (n = 50) Body mass index, kg/m ± ± MAP, mm Hg 97.2 ± ± Cholesterol, mmol/l 5.25 ± ± h urinary protein, g 0.70 ± ± Daily protein intake, g/kg 0.98 ± ± Month 24 (n = 49) (n = 47) Body mass index, kg/m ± ± MAP, mm Hg 95.7 ± ± Cholesterol, mmol/l 5.13 ± ± h urinary protein, g 0.78 ± ± Daily protein intake, g/kg 0.98 ± ± *Data are given as mean ± SD. Data were analyzed using the t test. Data were analyzed using the Mann-Whitney U test. To convert cholesterol from millimoles per liter to milligrams per deciliter, divide millimoles per liter by Hence, environmental lead exposure might not be innocent in the progression of renal insufficiency in the general population. Although the pathogenesis of progressive renal insufficiency is unknown, results of experimental studies in animals and some studies in humans have suggested that progression of different types of renal disease might largely be due to hemodynamic and metabolic factors rather than to the activity of underlying diseases. On the basis of the observations of clinical studies, blood pressure control, a restricted daily protein diet, control of hyperlipidemia, and reduction of daily urinary protein excretion are effective in retarding the development and progression of renal insufficiency. In the Cox multivariate regression analysis, we also first demonstrated that a BLB greater than or equal to 80 µg and less than 600 µg is the most important risk factor for determining the progression of renal insufficiency, even after adjusting for possible factors affecting progression of renal insufficiency. In addition, renal function of patients with CRI and highnormal BLB ( 80 µg and 600 µg) significantly improved after their BLBs were reduced to less than 80 µg by chelation therapy. The effects lasted for at least 12 months. At the end of 12-month follow-up, increments of renal function up to 10.2% in the study group and decrements of re- Table 4. Sensitivity Analysis of Renal Function During Months 12 to 24 of the Observation Period* Observation Period Creatinine Clearance, Mean ± SD, ml/s High-Normal BLB Group (n = 55) Low BLB Group (n = 55) P (95% CI) Month ± ± ( 9.3 to 0.5) Month ± ± ( 9.6 to 0.8) Month ± ± ( 12.0 to 1.5) Month ± ± ( 11.1 to 1.2) Month ± ± ( 12.0 to 2.0) *For more information about the sensitivity analysis, see the Statistical Analysis subsection of the Patients, Materials, and Methods section. BLB indicates body lead burden; CI, confidence interval. To convert creatinine clearance from milliliters per second to milliliters per minute, divide milliliters per second by Data were analyzed using the t test. Table 3. Intent-to-Treat Analysis of Renal Function During 24 Months of Observation* Creatinine Clearance, Mean ± SD, ml/s Observation Period High-Normal BLB Group Low BLB Group P (95% CI) Month ± 0.20 (n = 55) 0.67 ± 0.22 (n = 55).78 ( 5.5 to 4.1) Month ± 0.23 (n = 55) 0.67 ± 0.23 (n = 55).50 ( 7.0 to 3.4) Month ± 0.23 (n = 55) 0.67 ± 0.23 (n = 55).54 ( 6.9 to 3.6) Month ± 0.22 (n = 55) 0.66 ± 0.24 (n = 55).26 ( 8.0 to 2.2) Month ± 0.20 (n = 54) 0.68 ± 0.23 (n = 53).07 ( 9.7 to 0.4) Month ± 0.22 (n = 54) 0.69 ± 0.24 (n = 52).08 ( 10.2 to 0.6) Month ± 0.22 (n = 53) 0.72 ± 0.25 (n = 50).007 ( 13.3 to 2.2) Month ± 0.22 (n = 51) 0.70 ± 0.24 (n = 49).006 ( 13.0 to 2.2) Month ± 0.22 (n = 49) 0.70 ± 0.24 (n = 47).001 ( 15.0 to 3.8) *BLB indicates body lead burden; CI, confidence interval. To convert creatinine clearance from milliliters per second to milliliters per minute, divide milliliters per second by Data were analyzed using the t test. P.05 is significant. 268

6 nal function up to 11.1% in the control group were noted. Results of the clinical trial further confirm the initial findings of the 2-year follow-up and imply that environmental lead exposure might play an important role in the progression of CRI. Based on the findings of the present study and previous epidemiological studies, of the relations between environmental low-level lead exposure and renal function, we suggest that long-term low-level environmental lead exposures might subtly affect progressive renal insufficiency in the general population. After EDTA chelation therapy, the amount of BLB in the study group decreased from µg to 39.2 µg, and their renal function increased 8%. Three months later, renal function increased to 12.8%. Twelve months later, the increments of renal function remained at 10.2%. The effect of improving renal function lasted for at least 1 year after chelation therapy. The findings are similar to those of previous short-term studies in chelation therapy of lead workers 2,3 and rats with long-term low-level lead exposure. 14 In addition, the results are also in agreement with those of a previous study 9 in a small group of patients with CRI. Hence, it seems reasonable to conclude that long-term low-level environmental lead exposure might play a role in progressive renal insufficiency. The EDTA chelating agent is safe in treating patients with CRI with a smaller dose and a longer interval. 3,9,12,24 Wedeen et al 25 extensively used EDTA as test doses in large numbers of patients and as therapy in a smaller number, yet they never encountered any evidence of nephrotoxicity. The same findings were also noted in previous studies. 2-4,9-13,16 Hence, the EDTA chelation therapy might be a better alternative to treat patients with progressive renal insuffi- Patients Not Reaching the Primary Outcome High-Normal BLB Group Event Times of High-Normal BLB Group Low BLB Group Event Times of Low BLB Group Months Kaplan-Meier estimates of primary renal outcome in patients with chronic renal insufficiency who had high-normal ( 80 µg and 600 µg) and low ( 80 µg) body lead burdens (BLBs) (P.001 by log-rank test). Table 5. Cox Regression Analysis of Overall Risk of Progressive Renal Insufficiency in All Patients According to Baseline Prognostic Factors and Primary Outcome* Variable Relative Risk (95% CI) P Age 1.09 ( ).01 Female sex 2.73 ( ).18 Basal body mass index 0.71 ( ).01 Hypertension 4.17 ( ).10 Hyperlipidemia 2.38 ( ).14 Smoking 2.05 ( ).33 High-normal body lead ( ).002 burden ( 80 µg) Basal daily proteinuria 1.92 ( ).08 Basal daily protein intake 0.35 ( ).42 Chronic glomerulonephritis 0.48 ( ).27 *CI indicates confidence interval. Table 6. Baseline Characteristics of 36 Patients With High-Normal Body Lead Burden ( 80 µg and 600 µg) at Study Entry* Variables Study Group (n = 24) Control Group (n = 12) P Age, y 58.0 ± 6.9 ( ) 54.6 ± 10.7 ( ).32 Sex, M/F 21/3 9/3.38 Body mass index, kg/m ± 8.5 ( ) 24.6 ± 3.4 ( ).56 Serum creatinine, µmol/l 216 ± 57 ( ) 208 ± 45 ( ).74 Creatinine clearance, ml/s 0.63 ± 0.26 ( ) 0.58 ± 0.20 ( ).59 Blood lead, µmol/l 0.25 ± 0.12 ( ) 0.26 ± 0.13 ( ).90 Body lead burden, µg ± 83.9 ( ) ± ( ).70# Hyperlipidemia, No Hypertension, No Using CEIs for hypertension, No. 17/18 7/7.99 Smoking, No Underlying renal disease, No. Chronic glomerulonephritis Hypertensive nephropathy Gouty nephropathy Polycystic kidney disease Unknown *Data are given as mean ± SD (range) except where indicated otherwise. CEIs indicates converting enzyme inhibitors. To convert serum creatinine from micromoles per liter to milligrams per deciliter, divide micromoles per liter by To convert creatinine clearance from milliliters per second to milliliters per minute, divide milliliters per second by To convert blood lead from micromoles per liter to micrograms per deciliter, divide micromoles per liter by Serum cholesterol level 6.21 mmol/l ( 240 mg/dl) after diet control. Blood pressure 140/90 mm Hg for at least 2 measurements. Data were analyzed using the t test. Date were analyzed using the 2 with the Fisher test. #Data were measured using the Mann-Whitney U test. 269

7 Table 7. Intent-to-Treat Analysis of Increments of Renal Function After Chelation Therapy and During 12-Month Follow-up Compared With Renal Function Before Chelation Therapy* Follow-up After Chelation Therapy Ccr, Mean ± SD, ml/s Study Group Control Group P (95% CI) Basal (n = 24) (n = 12) Ccr 0.63 ± ± ( 13.1 to 7.6) 0 mo (n = 24) (n = 12) Ccr 0.67 ± ± ( 18.0 to 4.4) Increments 0.05 ± ± ( 6.6 to 4.3) 3 mo (n = 24) (n = 12) Ccr 0.71 ± ± ( 21.9 to 1.9) Increments 0.08 ± ± ( 10.2 to 4.3) 6 mo (n = 23) (n = 12) Ccr 0.69 ± ± ( 22.8 to 2.1) Increments 0.08 ± ± ( 12.5 to 4.3) 9 mo (n = 22) (n = 11) Ccr 0.69 ± ± ( 25.0 to 0.2) Increments 0.08 ± ± ( 12.5 to 5.0) 12 mo (n = 22) (n = 11) Ccr 0.68 ± ± ( 25.0 to 0.2) Increments 0.07 ± ± ( 12.6 to 5.0) *Ccr indicates creatinine clearance; CI, confidence interval. To convert creatinine clearance from milliliters per second to milliliters per minute, divide milliliters per second by Data were analyzed using the t test. Data were analyzed using the Mann-Whitney U test. Table 8. Sensitivity Analysis of Increments of Renal Function After Chelation Therapy and During 12-Month Follow-up Compared With Renal Function Before Chelation Therapy* Follow-up After Chelation Therapy 6mo Ccr, Mean ± SD, ml/s Study Group (n = 24) Control Group (n = 12) P (95% CI) Ccr 0.69 ± ± ( 22.3 to 2.1) Increments 0.08 ± ± ( 12.2 to 3.9) 9mo Ccr 0.68 ± ± ( 25.0 to 0.2) Increments 0.06 ± ± ( 11.2 to 3.4) 12 mo Ccr 0.67 ± ± ( 25.0 to 0.2) Increments 0.06 ± ± ( 11.3 to 3.5) *For more information about the sensitivity analysis, see the Statistical Analysis subsection of the Patients, Materials, and Methods section. CI indicates confidence interval. To convert creatinine clearance from milliliters per second to milliliters per minute, divide milliliters per second by Data were analyzed using the t test. Data were analyzed using the Mann-Whitney U test. ciency. The results of the present study shed further light on the treatment of CRI. Further investigation is needed to confirm our observations. The mechanism by which chelation therapy improves renal function is unknown. The effect of improving renal function after chelation therapy might theoretically relate to removal of essential trace elements, such as zinc and copper, which act as cofactors for enzymes that affect synthesis of vasoactive hormones. 18,19 For example, zinc is a cofactor for neutral metalloendopeptidase, a kidney enzyme of degrading atrial natriuretic peptide, 26 and for angiotensin-converting enzyme. 27 A depletion of zinc might cause an elevation in atrial natriuretic peptide and a reduction in angiotensin II formation together with a decline in blood pressure and an increase in renal blood flow. However, the dimercaptosuccinic acid chelating agent, which also improves renal function in rats with long-term low-level lead exposure, does not alter the zinc and copper content of the kidney cortex. 14 Hence, the hypotheses are against the results of an animal study 14 of dimercaptosuccinic acid therapy. The other possible explanation might come from the studies of cellular biology. Long-term low-level lead exposure, but not long-term high-level lead exposure, might affect endothelium-derived relaxing factor and endothelin 3. 28,29 The improvement in renal function in our patients after EDTA treatment could therefore be a consequence of removal of excessive body lead. Further evaluation is needed to clarify a definite mechanism. We used serum creatinine and creatinine clearance to assess renal function because these were the only quantitative measurements of renal function available for prospective analysis in our patients. Furthermore, resource constraints prohibited us from measuring inulin or isotopic clearance 30 in this work. Our study is limited by using these methods to assess changes in renal function because whether this method accurately reflects changes in glomerular filtration has not been tested rigorously. Other limitations of our study include the lack of a doubleblind placebo design, the inability to mask patients and providers, the relatively small sample size, and the questionable generalizability of our findings to patients with multifactorial etiologies of renal insufficiency. In conclusion, long-term low-level environmental lead exposure may be associated with chronic renal disease. Progressive renal insufficiency may be improved after lead chelating therapy. Further investigation is needed to confirm our observations and to clarify the mechanism by which long-term, low-level lead exposure might affect renal function. Accepted for publication July 20, This study was supported in part by grant NSC B-182A-022 from the National Science Council Foundation, Taipei, Taiwan, Republic of China. Corresponding author and reprints: Ja-Liang Lin, MD, Poison Center and Division of Nephrology, Chang Gung Memorial Hospital, 199, Tung Hwa North Road, Taipei, Taiwan, Republic of China ( jllin99@ hotmail.com). 270

8 REFERENCES 1. Henderson DA. The etiology of chronic nephritis in Queensland. Med J Aust. 1958; 1: Wedeen RP, Maesaka JK, Weiner B, Mallick DK. Occupational lead nephropathy. Am J Med. 1975;59: Wedeen RP, Mallick DK, Batuman V. Detection and treatment of occupational lead nephropathy. Arch Intern Med. 1979;139: Nuyts GD, Daelemans RA, Jorens PhG, Elseviers MM, Van de Vyver VD, De Broe ME. Does lead play a role in the development of chronic renal disease? Nephrol Dial Transplant. 1991;6: Staessen JA, Lauwerys RR, Buchet JP, et al. Impairment of renal function with increasing blood lead concentrations in the general population. N Engl J Med. 1992;327: Payton M, Hu H, Sparrow D, Weiss ST. Low-level lead exposure and renal function in the Normative Aging Study. Am J Epidemiol. 1994;140: Kim R, Rotnitzky A, Sparrow D, Weiss ST, Wager C, Hu H. A longitudinal study of low-level lead exposure and impairment of renal function: the Normative Aging Study. JAMA. 1996;275: Pollock CA, Ibels LS. Lead nephropathy: a preventable cause of renal failure. Int J Artif Organs. 1988;11: Lin JL, Ho HH, Yu C-C. Chelation therapy for paients with elevated body lead burden and progressive renal insufficiency: a randomized, clinical trial. Ann Intern Med. 1999;130: Lin JL, Lim PS. Does lead play a role in the development of renal insufficiency in some patients with essential hypertension? J Hum Hypertens. 1994;8: Lin JL, Huang PT. Body lead stores and urate excretion in men with chronic renal disease. J Rheumatol. 1994;21: Lin JL, Lim PS. Disappearance of immune deposits with EDTA chelation therapy in a case of IgA nephropathy. Am J Nephrol. 1992;12: Lin JL, Yeh KH, Chen WY, Lai HH, Lin YC. Urinary N-acetyl-glucosaminidase excretion and environmental lead exposure. Am J Nephrol. 1993;13: Khalil-Manesh F, Gonick HC, Cohen A, Bergamaschi E, Mutti A. Experimental model of lead nephropathy, II: effect of removal from lead exposure and chelation treatment with dimercaptosuccinic acid (DMSA). Environ Res. 1992;58: Isaksson B. Urinary nitrogen output as a validity test in dietary surveys. Am J Clin Nutr. 1980;33: Batuman V, Maesaka JK, Haddad B, Tepper E, Landy E, Weeden RP. The role of lead in gouty nephropathy. N Engl J Med. 1981;304: Pirkle JL, Brody DJ, Gunter EW, et al. The decline in blood lead levels in the United States: the National Health and Nutrition Examination Surveys (NHANES). JAMA. 1994;272: Hostetter TH, Olson JL, Rennke HG, Venkatachalam MA, Brenner BM. Hyperfiltration in remnant nephrons: a potential adverse response to renal ablation. Am J Physiol. 1981;241: Brenner BM, Meyer TW, Hostetter TH. Dietary protein intake and the progressive nature of kidney disease: the role of hemodynamically mediated glomerular injury in the pathogenesis of progressive glomerular sclerosis in aging, renal ablation, and intrinsic renal disease. N Engl J Med. 1982;307: Obrador GT, Arora P, Kausz AT, Pereira BJ. Pre end-stage renal disease care in the United States: a state of disrepair. J Am Soc Nephrol. 1998;9:S44-S Henrich WL. Approach to volume control, cardiac preservation, and blood pressure control in the pre end stage renal disease patients. J Am Soc Nephrol. 1998; 9:S63-S Mitch WE. Dietary protein restriction in patients with chronic renal failure. Kidney Int. 1991;29: Hunsicker LG, Adler S, Caggiula A, et al. Predictors of the progression of renal disease in the Modification of Diet in Renal Disease Study. Kidney Int. 1997;51: Batuman V. Lead nephropathy, gout, and hypertension. Am J Med Sci. 1993; 305: Wedeen RP, Batuman V, Landy E. Safety of EDTA lead mobilization test. Environ Res. 1983;30: Carretero OA, Scicli AG. Zinc metallopeptidase inhibitors: an oral antihypertensive treatment. Hypertension. 1991;19: Maguire GA, Price CP. A continuous monitoring spectrophotometric method for measurement of angiotensin-converting enzyme in human serum. Ann Clin Biochem. 1985;22: Khalil-Manesh F, Gonick HC, Weiler EW, Prins B, Weber MA, Purdy RE. Leadinduced hypertension: possible role of endothelial factors. Am J Hypertens. 1993; 6: Hermann M, Schulz E, Ruschitzka F, Muller GA. Preventive stategies in endothelininduced renal failure. Kidney Int. 1998;54(suppl 67):S202-S Levey AS, Perrone RD, Madias NE. Serum creatinine and renal function. Annu Rev Med. 1988;39: