Lead is a multitargeted toxicant, affecting cardiovascular, renal and nervous systems, and may contribute to morbidity and mortality through its adverse impacts on these systems 12.

An association between lead and mortality has been observed in both occupational and community based cohorts 3. Results from the second National Health and Nutrition Examination Survey (NHANES II, 1976–1980) showed that blood lead concentration was an important predictor of mortality 4. Individuals with baseline blood lead concentrations of 20 to 29 μg/dL [0.96 to 1.39 μmol/L (To convert μg/dL of blood lead into μmol/L, multiply by 0.048)] experienced 46% increased all cause mortality, relative to those with blood lead concentrations less than 10 μg/dL(0.48 μmol/L) 5. In NHANES III (1988–1994) an increased risk of death from all causes, cardiovascular disease, and cancer was associated with much lower blood lead concentrations of 5–9 μg/dL(0.24–0.43 μmol/L) as compared to those with < 5.0 μg/dL 6. Furthermore Menke et al documented a 25% increased all cause and 55% increased cardiovascular mortality in NHANES III (1988–1994) at considerably lower blood lead concentrations: > 3.62 μg/dL (0.17 μmol/L) as compared to those with < 1.94 μg/dL (0.09 μmol/L) 7. However they did not observe an association between blood lead and cancer mortality in this range of exposure.

Environmental exposures to lead have been associated with hypertension and the incidence of clinical cardiovascular endpoints such as coronary heart disease, stroke, and peripheral artery disease 8. Cardiac abnormalities such as left ventricular hypertrophy 9 and alteration in cardiac rhythm 10 have also been documented with lead exposure. Higher blood concentrations have been associated with cognitive and neuromuscular decline 111213, and renal effects 2141516 all of which could contribute to an increased risk of mortality. The effects of blood lead concentrations on cancer mortality however, are poorly understood. In the current analysis we prospectively examined the association of blood lead concentrations and mortality in a cohort of 533 white women with mean age of 72.5 (± 4.4) (range: 68–89) years and mean blood lead concentrations of 5.3 μg/dL (± 2.3 SD) (range: 1–21). We hypothesized that woman with blood lead concentrations above a threshold will experience higher total and cause specific mortality.

The women in lead ancillary study were compared to the rest of SOF study participants. Lead study cohort was younger in age from the rest of the SOF participants (Table 1). A lower proportion of women in lead cohort was hypertensive, or walked for exercise. Proportion of diabetics was higher in the lead study. However, BMI, education, alcohol use, smoking, use of estrogen, bone density in total hip were comparable.

Mean blood lead concentration was 5.3 ± 2.3 μg/dL (range 1–21) [0.25 ± 0.11 μmol/L (range 0.05–1.008)]. A total of 123 (23%) women died over a mean follow up of 12.0 (± 3.0) years. Women with ≥ 8 μg/dL(0.384 μmol/L), blood lead concentration had higher alcohol intake, were more likely to smoke, and had 8% lower total hip BMD (Table 2).

Women who died had 7% higher mean (± SD) blood lead concentration 5.6 (3) μg/dL, [0.27(± 0.14) μmol/L] than survivors: 5.17(± 2.0) [0.25(± 0.1) μmol/L] μg/dL (p = 0.09, Table 3). As compared to survivors, women who died were older, more likely to smoke and to have hypertension. A lower proportion of women who died reported walking for exercise. Compared to women whose blood lead concentrations were < 8 μg/dL, (< 0.384 μmol/L) survival decreased more rapidly in women with blood lead concentration ≥ 8 μg/dL (≥ 0.384 μmol/L) (Figure 1). Age, clinic, smoking, hypertension, and total hip BMD were significantly associated with mortality in women with blood lead concentration ≥ 8 μg/dL (≥ 0.384 μmol/L) (Table 4).

Women with baseline blood lead concentration of ≥ 8 μg/dL (≥ 0.384 μmol/L) had a 73% increased risk of dying. (Age and clinic adjusted hazard ratio [HR], 1.73; 95% confidence interval [CI], 1.12–2.68) (p = 0.014) compared to women in blood lead < 8 μg/dL (0.38 μmol/L). With further adjustment for covariates women with ≥ 8 μg/dL (0.384 μmol/L) still had 59% higher risk of all cause mortality (HR = 1.59; 1.02–2.49) (p = 0.041), compared to women with < 8 μg/dL (0.38 μmol/L) of blood lead. Although the multivariate adjusted hazards ratio (95% CI) for CVD mortality for women who had ≥ 8 μg/dL (≥ 0.384 μmol/L) versus the < 8 μg/dL (< 0.38 μmol/L) blood lead concentrations was not significant; 1.78(95% CI, 0.92–3.45), p = 0.089, women in higher lead group experienced 3 fold higher risk of mortality due to coronary heart disease 3.08(95% CI, 1.23–7.70), p < 0.016. There was no association of blood lead and mortality from stroke, cancer and other causes (Table 5 and Figure 2).

Blood lead was an important predictor of all cause mortality in this cohort of community dwelling older women. Mortality was significantly higher in women with blood lead concentrations ≥ 8 μg/dL as compared to those with lower blood lead concentrations. Our results are consistent with earlier studies based on occupational cohorts 3242526 and the general population: NHANES II and III 4567. To our knowledge, this is the first study to look at the association between blood lead concentrations and mortality in older women with median age of 71 years (range 65–87).

Despite declines in blood lead concentrations during the past 30 years, environmental lead exposure continues to be a public health concern 2. The use of organic lead as gasoline additive, initiated during the 1920s, was phased out in the US since 1976. Once released as combustion exhaust, particulate lead persists in air, water, and soil. Mean blood lead concentrations in the U.S. population in the NHANES III Phase 2 collected during 1991 to 1994 ages 20 to 49 years were 2.1 μg/dL1. The mean blood lead concentration for ages > 1 year has declined further to 1.45 μg/dL(0.07 μmol/L) as reported from the most recent survey (NHANES IV, 2001–2002)27.

The skeleton is repository for 95% of absorbed lead and can serve as an endogenous source for many years after exposure. Lead may be mobilized from skeleton during conditions of high bone turn over, such as pregnancy, lactation, menopause and aging 28. A 25% adjusted increase in median blood lead concentration was reported for post menopausal women, compared to premenopausal women in NHANES II 29. Furthermore in NHANES III, blood lead concentrations were highest in ages 70 years and older [3.4 μg/dL,(0.16 μmol/L) ] 27.

A multitargeted toxicant, lead effects cardiovascular, renal and nervous systems and may contribute to morbidity and mortality through its adverse impacts on these systems. The increased cardiovascular mortality risk may reflect an effect on sub-clinical risk factors for disease. The evidence for this association is supportive 27. As part of NHANES III, an increased risk of peripheral arterial disease, hypertension and renal dysfunction was observed in populations with an average blood lead concentration of 2 μg/dL (0.10 μmol/L) 2142730. For example, the odds ratio of diastolic hypertension was 8.1 comparing women with a blood lead concentration of 4.0–31.1 μg/dL (0.19–1.5 μmol/L) to women with lower blood lead concentrations of 0.5–1.6 μg/dL(0.02–0.08 μmol/L). Other analyses support an association between blood lead and renal function impairment 143132, and increased blood pressure 3334, a biologically plausible relationship 35.

Lead contributes to nephrotoxicity, even at blood concentration below 5 ug/dL (0.24 μmol/L) 32. Increase in blood pressure and an association with renal damage have also been observed after lead exposure in rodent models 3637. Alterations in signal transduction that involve renal pathways (eg, angiotensin and vasopressin) were reported in rat models 383940. Other mechanisms by which lead may increase cardiovascular risk include effects on neuromuscular and neuro-humoral regulation of vascular function, alteration in sodium transport, and alterations in calcium regulation 711273041424344454647.

Blood lead was associated with almost three fold risk in coronary heart disease (CHD) mortality (HR = 3.08) in our study. Our results are consistent with findings in which

bone

The pathogenesis of CHD is multifactorial; lead may be one of the mediators by two causal pathways; i.e., mediation through higher blood pressure 8 and by atherogenic process 49. Lead-related atherosclerosis could be explained by several mechanisms, impairment of renal function 32, induction of oxidative stress 50 and endothelial dysfunction 50.

Previous studies have linked lead as low as 3.62 μg/dL (0.17 μmol/L)with an increased risk of stroke mortality 78. In our study, we found no significant association with stroke, although the mean lead concentration in participants who died due to stroke was 22% higher as compared to the rest of the cohort (6.33 ug/dL vs. 5.21 ug/dL, p < 0.028)(0.30 vs.0.25 μmol/L)perhaps reflecting the small number of stroke deaths in this cohort (Table 3).

Lead is a toxic metal and categorized as probably carcinogenic to humans (Group 2A IARC 2004) [51 Monograph]. Associations between occupational lead exposure and cancers of brain, stomach, kidney and lung have been reported 525354. However among non occupational cohorts, there has been inconsistent evidence of an association between blood lead and cancer 5. Individuals with mean blood lead concentrations 10–19 μg/dL (0.48–0.91 μmol/L) in the NHANES II cohort (1976–1980) did not have increased risk of cancer mortality, when compared to those with blood lead concentrations < 10 μg/dL (0.48 μmol/L).

Our results are consistent with this observation as the median value in the "≥ 8 μg/dL" lead group was 9 μg/dL (0.43 μmol/L). A higher risk of cancer deaths was only observed in with blood lead concentration > 20 μg/dL(0.96 μmol/L) 4. Similarly, individuals with blood lead concentrations ≥ 3.62 μg/dL (0.17 μmol/L) in the NHANES III (1988–1994) did not have increased risk of cancer mortality when compared to those with < 1.94 μg/dL(0.09 μmol/L) 7. In contrast, another analysis from the same population survey reported 44% increased risk of cancer death at blood lead concentrations ≥ 5 μg/dL (0.24 μmol/L) when compared to those with < 5 μg/dL 6.

Bone loss accelerates after menopause and bone demineralization may release bone lead into circulation 28. Inverse association has been reported between mortality, (predominantly stroke deaths) and bone density 55. In our study women with ≥ 8 μg/dL (≥ 0.384 μmol/L) blood lead had 8% lower total hip bone mineral density at baseline as compared to women with lower lead concentrations. We also observed that women who died had 7% higher blood lead concentration than survivors. In another analysis in older women, each SD increase in BMD loss at the hip was associated with a 1.3-fold increase in total mortality, adjusted for age, baseline BMD, diabetes, hypertension, incident fractures, smoking, physical activity, health status, weight loss, and calcium use. In particular, hip BMD loss was associated with increased mortality from coronary heart disease (relative hazard [RH] = 1.3 per SD; 95% CI, 1.0–1.8) 56.

Alternatively, osteoporosis and atherosclerosis may result from elevated concentrations of homocysteine, an amino acid whose normal metabolism depends on folate and vitamin B₁₂ as cofactors. Lead and homocysteine both are associated with cardiovascular disease and cognitive dysfunction57. In subjects 50–70 years of age, blood lead and homocysteine concentrations were correlated (Pearson's r = 0.27, p < 0.01), homocysteine concentration increased 0.35 μmol/L per 1.0 μg/dL (0.05 μmol/L) increase in blood lead (p < 0.01). Homocysteine is an example of plausible mechanism that may mediate the affect of lead on the cardiovascular 58 and central nervous systems 59. Taken together, more research is clearly needed to further our understanding of the mechanism of lead toxicity and these multisystem outcomes.

Compared to the rest of SOF participants, the lead study cohort was of comparatively younger age, and had lower proportion with hypertension 60. The proportion of women with type 2 diabetes was higher in the lead cohort, we adjusted for diabetes in secondary analysis and the association of lead, and mortality remained significant. The number ofwomen who are older and atrisk is growing. Cardiovascular disease is the leading cause of mortality worldwide 61 and in the United States 8. It kills nearlyhalf a million women in the United States every year, more than the nextfivecauses of death combined and nearly twice as many as all forms of cancer, including breast cancer 62. Environmental toxicants such as lead may explain part of the burden of CVD.

There are several strengths to our study: our follow up was more than 95% complete and we adjudicated all mortality events. We followed women for more than12 years after the blood lead measures were obtained. We controlled for a number of covariates and cardiovascular risk factors. However, this study has several limitations; participation was limited to older Caucasian women, and the findings may not apply to men or nonwhite women. We did not determine co-contaminants such as cadmium that might be associated with cardiovascular disease through its known effects on kidney function 6364. There are factors that differ by lead concentrations and we could not measure may have confounded our results. For example we had no measure of renal function, homocysteine or lipid concentrations and thus we could not examine whether these measures influenced the association between lead and mortality. We relied on death certificates and discharge summaries were only available for 33% of participants which may result in some misclassification of cause of death 65. Use of death certificates may be problematic for assigning a single cause of death, especially among the oldest women who often have multiple medical problems.

Our study extends the findings of higher mortality associated with blood lead concentrations from NHANES III surveys to community dwelling older women. An increased mortality risk, especially coronary heart disease was found at blood lead concentrations ≥ 8 μg/dL (0.384 μmol/L). Our results add to the existing evidence of adverse affects of lead on health as seen in an older cohort who experienced greater historic environmental lead exposure.

BMD: Bone mineral density; CHD: Coronary Heart Disease; CVD: Cardiovascular Disease; DXA: Dual energy X-ray absorptiometry; HR: Hazard Ratio; ICD: International Classification of Diseases, Ninth, revision, Clinical Modification; μg/dL: microgram per deciliter; NHANES: National Health and Nutritional Examination Survey; SOF: Study of Osteoporotic Fractures; VDR: vitamin D receptor gene.

JAC has received research support from Merck & Company, Eli Lilly & Company, Pfizer Pharmaceuticals, and Novartis Pharmaceuticals. She has also received consulting fees from Eli Lilly & Company, and Novartis Pharmaceuticals. She is on the speaker's bureau for Merck and Company. SRC receives research support from Amgen, Pfizer, Novartis, Eli Lilly and Co. and consulting fees or honoraria from Eli Lilly and Co., Zelos, Merck and Co., Novartis, GlaxoSmithKline, Procter & Gamble, and Aventis. NK, JWW, EOT, LAM, MCH, TAH, and SBM had no conflicts.

NK carried out the analysis and drafted the manuscript; JAC participated in the conceptual design, draft of the manuscript. SBM conceived the study and performed data acquisition; JWW participated in the analysis plan, and design of the study. EOT, LAM, SRC, MCH, TAH, participated in its design, coordination, and review of the study. All authors read and approved the final manuscript.