Increased Risk of Proteinuria among a Cohort of Lead-exposed Pregnant Women

Environmental Health Perspectives, Volume 101, Number 5, October 1993

Increased Risk of Proteinuria among a Cohort of Lead-exposed Pregnant Women

Pam Factor-Litvak,^1,2 Zena Stein,^1,3 Joseph Graziano^2,4

¹Divisions of Epidemiology, ²Environmental Science, and ³The Sergievsky Center, Columbia School of Public Health, New York, NY 10032 USA; ⁴Department of Pharmacology, College of Physicians and Surgeons, Columbia University, New York, NY 10032 USA

Abstract

Long-term exposure to high concentrations of lead results in renal dysfunction. During a prospective study of environmental lead and pregnancy outcomes in 1502 women residing in two towns in Yugoslavia, we explored whether moderate exposure to lead results in increased rates of proteinuria. The geometric mean blood lead concentrations (BPb) were 17.1 and 5.1 µg/dl in the smelter and nonexposed towns, respectively. Increases in BPb were associated with increased odds ratios for both trace and >1+ proteinuria, measured using a urinary dipstick. Comparing the women in the upper 10th percentile of exposure to those in the lowest 10th percentile, the adjusted odds ratio for >1+ proteinuria was 4.5 (95% CI 1.5, 13.6). Similarly, the adjusted odds ratio for trace proteinuria was 2.3 (95% CI 1.3, 4.1). Similar to other studies showing associations between chronic exposure to lead and renal dysfunction, our data suggest that long-term exposure to environmental lead may be associated with proteinuria. Key words: epidemiology, lead exposure, pregnancy, proteinuria. Environ Health Perspect 101:000-000(1993).

Address correspondence to P. Factor-Litvak, Division of Epidemiology, Columbia School of Public Health, 600 W 168th Street, New York, NY 10032 USA.

This study was supported by grant R01-ES-03460 from NIEHS and by a grant from the Lucille Markey Charitable Trust. We acknowledge the helpful comments of Maureen Hatch, Jennie Kline, Judith Sackoff, and Pat Shrout on an earlier version of this manuscript. An earlier version of this paper was presented at the Society of Toxicology meetings in March 1993.

Received 5 April 1993; accepted 21 July 1993.

Introduction

Reports of renal dysfunction after exposure to lead have appeared in the literature for over a century (1,2). Excesses in mortality due to nephritis (3,4) and chronic hypertension (not associated with cardiovascular causes) (5) and renal diseases (6-10) have been observed among young adults surviving an epidemic of childhood lead encephalopathy in Australia and in workers in lead industries, respectively. Reductions in glomerular filtration rate (GFR) (11,12), creatinine clearance (13), and glucose transport (14) have been reported in adults with prolonged occupational exposure to lead. Together, these observations suggest that long-term, high-dose lead exposure is associated with increased mortality and morbidity due to renal dysfunction. Less is known, however, about the effect of more moderate environmental lead exposure on renal function. Exposure at environmental levels has been associated with small increases in both systolic and diastolic blood pressure (15-18) and with slight but statistically significant decreases in creatinine clearance (19).

During a prospective study of environmental lead and pregnancy outcome (20-22) in Kosovo, Yugoslavia, we measured proteinuria in lead-exposed pregnant women and a similar, but relatively unexposed, cohort. Leakage of protein into the urine may reflect functional and morphological changes in the kidney, findings associated with lead exposure in both animal and human models (23,24). Although a small amount of protein is normally present in the urine of healthy, nonpregnant persons, the physiologic changes of pregnancy normally increase this amount (25-27). This study explored whether environmental lead exposure is associated with an increased rate of proteinuria during pregnancy.

Kosovska Mitrovica (K. Mitrovica), the site of a lead smelter, refinery, and battery plant, is a lead-exposed area (28). Pristina, a town 25 miles to the south, has minimal lead exposure. Between May 1985 and December 1986, 1502 women, at approximately 12-20 weeks of gestation and making their first prenatal visit, were recruited for a study of pregnancy outcomes. The final sample sizes were 900 women in Pristina and 602 women in K. Mitrovica. A priori, we excluded women who reported at least one of the following conditions: previous history of hypertension (N = 22), previous history of gestational hypertension (N = 19), and use of antihypertensive medication (N = 6). The mean blood lead concentration (BPb) for the 37 excluded women did not differ from that of the women included; however, their average systolic and diastolic blood pressures were slightly higher.

Subjects were interviewed by trained, bilingual (Serbo-Croatian and Albanian) nurses, who were not privy to the hypotheses concerning renal function. Women were similar with respect to age, education, parity, and ethnicity, with more than 50% of the sample of Albanian origin (Table 1). Gestational age at entry to study was earlier in K. Mitrovica than in Pristina (17 versus 19 weeks, respectively). Cigarette smoking and alcohol use were reported more often in Pristina. More women were employed in Pristina.

A urine sample was obtained for dipstick analyses (Bili-Labstix, Ames, Miles Laboratories, Elkhart, Indiana) of protein. Dipstick analyses represent a qualitative assessment of protein in the urine. In the presence of protein, the reagent on the dipstick changes color; the color shade is proportional to the concentration of protein. Of the 1465 eligible women, 1447 (98.8%) provided urine samples. Proteinuria was trichotomized as none, trace, and >=1+, representing a modification of the recommendations of the Committee of Terminology of the American College of Obstetricians and Gynecologists (29,30).

The overall high prevalence of proteinuria (72%; Table 2) in this study may be attributed to normal physiological changes in renal function during pregnancy (25). GFR increases early in pregnancy and remains elevated until the month before delivery (26,27). These changes result in increased urinary excretion of glucose, amino acids, water-soluble vitamins, and protein. Trace proteinuria was detected in 89% of all women with proteinuria. Very few women had levels of 2+ or greater (i.e., levels that might indicate a clinically significant renal disorder).

Venous blood samples were obtained for the measurement of BPb (31), erythrocyte protoporphyrin (EP) (32), serum ferritin, (33), and hemoglobin. Blood specimens were refrigerated until transported to Columbia University, where the laboratory participates in the Centers for Disease Control BPb and EP quality control program and is certified for BPb analyses by the Occupational Safety and Health Administration. During the course of the study period, agreements with the CDC values for BPb and EP, measured by intraclass correlation coefficients, were 0.95 and 0.99, respectively.

Both BPb and EP were substantially higher in K. Mitrovica (geometric means of 17.1 µg/dl and 35.8 µg/dl, respectively) than in Pristina (geometric means of 5.1 µg/dl and 25.2 µg/dl, respectively; Table 3). The ranges of BPb were 3.0-56.7 in K. Mitrovica and 1.3-23.0 in Pristina.

We used logistic regression analyses (34,35) to estimate the association between lead and proteinuria, adjusting for potentially confounding variables (see Table 4). These variables were associated with proteinuria (p <0.25) or changed the coefficient of the exposure variable by more than 10% when included in the model. Data on acetaminophen use, a predictor of proteinuria, began to be collected midway through the study and was obtained only on a subset of women (N = 425). We examined these women separately and found essentially similar associations between lead and proteinuria.

Blood lead levels represented recent exposure. We first ranked the BPb levels of individual women from highest to lowest and divided them into exposure deciles. The estimated adjusted odds ratios for both trace and >1+ proteinuria appeared to increase in the highest exposure deciles. For >1+ proteinuria (Fig. 1), the increase appeared when BPb was greater than approximately 5.8 µg/dl; the estimated adjusted odds ratios (95% CI) above this value ranged from 2.1 (0.8, 6.4) to 8.4 (2.7, 26.5). For trace proteinuria (Fig. 2), the increases appeared when BPb was greater than approximately 6.7 µg/dl, after which the estimated adjusted odds ratios showed a monotonic increase ranging from 1.6 (1.0, 2.7) to 3.4 (1.8, 6.3).

Figure 1. Adjusted odds ratios (bars indicate 95% CIs) for >1+ proteinuria. Blood lead groups were defined by ranking BPb levels in ascending order and defining 10 groups with approximately equal numbers in each. Group 1 is always used as the reference. The odds ratios are adjusted for cigarette smoking (yes, no), height (linear and quadratic), age (linear and quadratic), average daily consumption of milk, gestational age (linear and quadratic), number of previous live births, average weekly consumption of meat, and hemoglobin.

Figure 2. Adjusted odds ratios (bars indicate 95% CIs) for trace proteinuria. Blood lead groups were defined by ranking BPb levels in ascending order and defining 10 groups with approximately equal numbers in each. Group 1 is always used as the reference. The odds ratios are adjusted for cigarette smoking (yes, no), ethnic group (Albanian, Serbian, other), age, average weekly consumption of meat (times per week), average daily consumption of milk (glasses per day), and hemoglobin.

Blood lead was then considered as a continuous variable. We report results for the logarithmic transformation of BPb; results for the untransformed metameter were essentially the same. For >1+ proteinuria, the estimated adjusted log odds was 1.8 (95% CI 1.0, 2.5) per log unit increase in BPb. Although lower than in the combined data, the log odds in K. Mitrovica and Pristina (Table 4) did not statistically differ from each other (p = 0.72). For trace proteinuria, the estimated adjusted log odds was 1.2 (95% CI 0.8, 1.6) per log unit increase in BPb. The log odds in K. Mitrovica (0.95) was above the null value and slightly, but not significantly, differed from that in Pristina (-0.2; p = 0.13).

We estimated cumulative exposure based on zone of residence and duration at that location. Zones were defined by a series of concentric circles around the smelter with radii progressively increasing by 2 km; thus, zone 1 was defined as the area within 2 km of the smelter, zone 2 as the area between 2 and 4 km from the smelter, etc. Zone 5 was the area between 8 km from K. Mitrovica and Pristina. Zone accounted for the largest proportion of the variance in midpregnancy BPb (20).

Neither trace nor >1+ proteinuria was associated with zone and duration of residence. Comparing zones 1-4 to zone 5, the estimated odds ratios for >1+ proteinuria were 0.7, 1.2, 1.0, and 0.2, respectively. Comparing zones 1-4 to zone 5, the estimated odds ratios for trace proteinuria were 1.1, 1.1, 1.2, and 1.7 for zones 1-4, respectively. Each of these odds ratios had wide 95% CI, which were consistent with the null value of 1.0.

These analyses suggest slight elevations in the risk of both trace and >1+ proteinuria at midpregnancy after exposure to environmental lead. Most of the exposed women lived in the smelter area for long periods of time, and exposure was probably higher earlier in life (28). In a subset of women from this study, we previously reported reduced serum erythropoietin (a renal hormone) among pregnant women with the highest BPb levels (36). Staessen et al. (19), in a sample of men and women from four areas in Belgium, found an inverse association between BPb and creatinine clearance, indicating a reduction in GFR after exposure to lead. Collectively, these observations suggest adverse renal effects of lead at more moderate exposures than those previously reported in the occupational literature.

The present analyses also suggest a threshold effect for trace proteinuria. The estimated log odds for BPb was elevated only in K. Mitrovica, where exposure levels are relatively high, suggesting that the BPb levels in Pristina may be below the no-observable effect level. Although the log odds in K. Mitrovica was not statistically significant, post hoc calculations revealed relatively low power (about 40%) for this outcome. Moreover, results from the grouped BPb suggested elevations in the odds ratio for trace proteinuria when BPb exceeded 6.4 µg/dl. Alternatively, however, both the narrow range of exposure in Pristina and the high overall prevalence of proteinuria may have obscured the result.

Similarly, results of the grouped BPb analysis did show statistically significant elevations in the odds ratio for >1+ proteinuria when BPb exceeded 5.3 µg/dl. Nevertheless, when BPb was considered as a continuous variable, the log odds in K. Mitrovica, although suggestive of an association, was not statistically significant. Again, relatively low post hoc power (about 14%) was found to detect the observed log odds. Our failure to find statistically significant results in the analyses of BPb as a continuous variable may thus be attributed to inadequate sample size and/or incorrect specification of the shape of the curve. Although we cannot exclude the null result, neither can we exclude an association between BPb and proteinuria.

For both >1+ and trace proteinuria, the log odds for the combined data are higher than for each town separately, suggesting an effect of town. Cadmium, emitted from the smelter at far lower concentrations than lead (37), is known to adversely affect renal function. We have previously demonstrated that placental cadmium concentrations in a subset of these women are 1000 times lower than placental lead (37). Nevertheless, studies in animals and in populations with high exposure to cadmium (38) suggest that cadmium affects both glomerular and tubular function, resulting in proteinuria, glucosuria, amino aciduria, and disorders involving impaired tubular resorption. In a population in Belgium, increased urinary cadmium excretion has been associated with a variety of markers of proximal tubule damage (39), including urinary retinol-binding protein and N-acetyl-ß-glucosaminidase. Thus, because trace quantities are emitted from the smelter in K. Mitrovica, cadmium cannot be excluded as a contributory factor.

We were unable to determine the mechanism of the proteinuria because urine dipsticks cannot assign protein molecular weight. Low-molecular-weight proteinuria would indicate damage to the proximal renal tubules, associated with the early stages of lead nephropathy, whereas proteinuria consisting of high-molecular-weight molecules would indicate increased glomerular permeability, indicative of later-stage lead nephropathy (23,24). Urinary dipsticks have been shown to be more sensitive to albumin than to other proteins (40,41) but are not specific to albumin. In conclusion, this study supports the hypothesis that lead exposure is associated with subclinical renal dysfunction during pregnancy.

References

1. Wedeen RP. Poison in the pot: the legacy of lead. Carbondale, IL:Southern Illinois University Press, 1984.

2. Dana SL [trans]. Lead diseases: a treatise from the French of L. Tanquerel des Planches, 1850.

3. Henderson DA. A follow-up of cases of plumbism in children. Aust Ann Med 3:219-224(1954).

4. Henderson DA. Chronic nephritis in Queensland. Aust Ann Med 4:163-177(1955).

5. Cooper WC, Gaffey WR. Mortality of lead workers. J Occup Med 17:100-107(1975).

6. Malcolm D, Barnett HAR. A mortality study of lead workers 1925-76. Br J Ind Med 39:404-410(1982).

7. McMichael AJ, Johnson HM. Long-term mortality profile of heavily exposed lead smelter workers. J Occup Med 24:375-378(1982).

8. Davies JM. Long term mortality study of chromate pigment workers who suffered lead poisoning. Br J Ind Med 41:170-178(1984).

9. Selevan SG, Landrigan PJ, Stern FB, Jones JH. Mortality of lead smelter workers. Am J Epidemiol 122:673-683(1985).

10. Cooper WC, Wong O, Kheifets L. Mortality among employees of lead battery plants and lead-producing plants 1947-1980. Scand J Work Environ Health 11:331-345(1985).

11. Wedeen RP, Maesaka JK, Weiner B, Lipat GA, Lyons MM, Vitale LF, Joselow MM. Occupational lead nephropathy. Am J Med 59:630-641(1975).

12. Wedeen RP, Mallik DK, Batuman V. Detection and treatment of occupational lead nephropathy. Arch Intern Med 139:53-57(1979).

13. Lilis R, Gavrilescu N, Nestorescu B, Dumitriu C, Roventa A. Nephropathy in chronic lead poisoning. Br J Ind Med 25:196-202(1968).

14. Hong CD, Hanenson IB, Lerner S, Hammond PB, Pesce AJ, Pollak VE. Occupational exposure to lead: effects on renal function. Kidney Int 18:489-494(1980).

15. Harlan WR, Landis R, Schmouder RL, Goldstein NG, Harlan LC. Blood lead and blood pressure. Relationship in the adolescent and adult US population. J Am Med Assoc 253:530-534(1985).

16. Neri LC, Hewitt D, Orser B. Blood lead and blood pressure: analysis of cross-sectional and longitudinal data from Canada. Environ Health Perspect 78:123-126(1988).

17. Pocock SJ, Shaper AG, Ashby D, Delves HT, Clayton BE. The relationship between blood lead, blood pressure, stroke, and heart attacks in middle-aged British men. Envrion Health Perspect 78:23-30(1988).

18. Schwartz J. Lead, blood pressure and cardiovascular disease in men and women. Environ Health Perspect 91:71-75(1991).

19. Staessen JA, Lauwerys RR, Buchet J-P, Bulpitt CJ, Rondia D, Vanrenterghem Y, Amery A, Cadmibel Study Group. Impairment of renal function with increasing blood lead concentrations in the general population. N Engl J Med 327:151-156(1992).

20. Graziano JH, Popovac D, Factor-Litvak P, Shrout P, Kline J, Murphy MJ, Zhao Y, Mehmeti A, Ahmedi G, Rajovic B, Zvicer Z, Nenezic DU, LoIacono NJ, Stein ZA. Determinants of elevated blood lead during pregnancy in a population surrounding a lead smelter in Kosovo, Yugoslavia. Environ Health Perspect 89:95-100(1990).

21. Murphy MJ, Graziano JH, Popovac D, Kline JK, Mehmeti A, Factor-Litvak P, Ahmedi G, Shrout P, Rajovic B, Nenezic D, Stein ZA. Past pregnancy outcomes among women living in the vicinity of a lead smelter in Kosovo, Yugoslavia. Am J Public Health 80:33-35(1990).

22. Factor-Litvak P, Graziano JH, Kline JK, Popovac D, Mehmeti A, Ahmedi G, Shrout P, Murphy M, Gashi E, Haxihu R, Rajovic L, Nenezic D, Stein ZA. A prospective study of birthweight and length of gestation in a population surrounding a lead smelter in Kosovo, Yugoslavia. Int J Epidemial 20:722-728(1991).

23. Choie DD, Richter GW. Effects of lead on the kidney. In: Lead toxicity (Singhal RL, Thomas JA, eds). Baltimore, MD:Urban and Schwarzenberg, 1980:187-212.

24. Goyer RA. Renal changes associated with lead exposure. In: Dietary and environmental lead: human health effects (Mahaffey K, ed). Amsterdam: Elsievier Science Publishers, 1985;315-338.

25. Lindheimer MD, Katz AK. The kidney in pregnancy. In: The kidney, 3rd ed. (Brenner BM, Rector FC, eds). Philadelphia, PA:WB Saunders,1986;1253-1296.

26. Davison JM, Dunlop W. Renal hemodynamics and tubular function during normal pregnancy. Kidney Int 18:152-161(1980).

27. Dunlop W. Serial changes in renal haemodynamics during normal human pregnancy. Br J Obstet Gynaecol 88:1-9(1981).

28. Popovac D, Graziano J, Seaman C, Kaul B, Colakovic B, Popovac R, Osmani I, Haxhiu M, Begraca M, Bozovic Z, Milic M. Elevated blood lead in a population near a lead smelter in Kosovo, Yugoslavia. Arch Environ Health 37:19-23(1982).

29. Davey DA, MacGillivray I. The classification and definition of the hypertensive disorders of pregnancy. Am J Obstet Gynecol 158:892-898(1988).

30. National High Blood Pressure Education Program Working Group. National high blood pressure education program working group report on high blood pressure in pregnancy. Am J Obstet Gynecol 163:1689-1712(1990).

31. Fernandez F, Hilligoss D. An improved graphite furnace method for the determination of lead in blood using matrix modification and the L'vov platform. Atomic Spectr 3:130-131(1982).

32. Piomelli S. A micromethod for free erythrocyte porphyrins: the FEP test. J Lab Clin Med 81:932-940(1973).

33. Miles LEM, Sipschitz DA, Bieter CP, Cook JD. Measurement of serum ferritin by a 2-site immunoradiometric assay. Anal Chem 61:209-224(1974).

34. Breslow NE, Day NE. Statistical methods in cancer research; volume I. The analysis of case-control studies. Lyon:International Agency for Research on Cancer, 1980.

35. Hosmer DW, Lemeshow S. Applied logistic regression. New York:John Wiley and Sons, 1989.

36. Graziano JH, Slavkovic V, Factor-Litvak P, Popovac D, Ahmedi X, Mehmeti A. Depressed serum erythropoietin in pregnant women with elevated blood lead. Arch Environ Health 46:347-350(1991).

37. LoIacono NJ, Graziano JH, Kline JK, Popovac D, Ahemdi X, Gashi E, Mehmeti A, Rajovic B. Placental cadmium and birthweight in women living near a lead smelter. Arch Environ Health 47:250-255(1992).

38. Kjellstrom T. Renal effects. In: Cadmium and health: a toxicological and epidemiological appraisal, vol II. Effects and response (Friberg L, Elinder CG, Kjellstrom T, Nordberg GF, eds). Boca Raton, FL:CRC Press, 1986;21-110.

39. Buchet JP, Lauwerys R, Roels H, Bernard A, Bruaux P, Claeys F, Ducoffre G, De Plaen P, Staessen J, Amery A, Lijnen P, Thijs L, Rondia D, Sartor F, Saint Remy A, Nick L. Renal effects of cadmium burden of the general population. Lancet 336:699-702(1990).

40. Evans GO, Parsons CE. Potential errors in the measurement of total protein in male rat urine using test strips. Lab Anim 20:27-31(1986).

41. Thysell H. A comparison between albustix, hema-combistix, labstix, the sulfosalicylic-acid test, heller's nitric-acid test, and a biuret method. Diagnosis of proteinuria. Acta Med Scand 185:401-407(1969).

[Table of Contents] [Citation in PubMed] [Related Articles]

Last Update: August 21, 1998