Urinary Heavy Metals and Longitudinal Changes in Blood Pressure in Midlife Women: the Study of Women’s Health Across the Nation (SWAN)

Abstract

Environmental exposure to heavy metals may contribute to increased blood pressure, however, evidence from midlife women who are at greater risk of cardio-metabolic disease, is limited. We evaluated the associations of urinary concentrations of arsenic, cadmium, mercury, and lead with longitudinal changes in blood pressure in the Study of Women’s Health Across the Nation Multi-Pollutant Study. The sample included 1,317 White, Black, Chinese and Japanese women, aged 45–56 years at baseline (1999–2000), whose systolic and diastolic blood pressure (SBP and DBP) were measured annually or biannually through 2017. Urinary metal concentrations were determined at baseline. Longitudinal changes in SBP and DBP were modeled using linear mixed effects models by tertiles of metal concentrations. After multivariable adjustment, estimated annualized increases (95% CI) in SBP in the highest and lowest tertiles were 0.93 (0.85, 1.01) mmHg and 0.74 (0.66, 0.82) mmHg for arsenic, 0.82 (0.75, 0.90) mmHg and 0.72 (0.65, 0.80) mmHg for mercury, and 0.86 (0.78, 0.93) mmHg and 0.72 (0.64, 0.79) mmHg for lead, respectively. Similar results were observed for associations of arsenic, mercury, lead with DBP. Urinary cadmium was associated with a greater rate of increase in SBP only among never smokers. Women with higher concentrations of all four metals had greater annualized increases in SBP and DBP than those with lower concentrations. Our findings suggest that exposure to heavy metals may accelerate the increase in blood pressure in midlife women, supporting the need for continued efforts to reduce these environmental exposures.

Introduction

High blood pressure remains a significant public health concern in the United States (U.S.).¹ People with high blood pressure are at a higher risk of developing cardiovascular disease, chronic kidney disease, and mortality.^2,3 Although obesity, smoking, unhealthy diet, and physical inactivity are important contributors to high blood pressure, growing evidence suggests that exposure to environmental toxicants may also play a role.⁴

Heavy metals, including arsenic, cadmium, mercury, and lead are well-known environmental toxicants that may exert adverse effects on blood pressure through the generation of reactive oxygen species (ROS) and induction of oxidative stress.⁵ Animal studies have demonstrated that exposure to these metals is associated with an increase in blood pressure.⁶ Epidemiologic studies have reported a consistent positive association between lead and blood pressure,^7,8 whereas results have been mixed for other metals. Both arsenic and mercury concentrations have been associated with elevated blood pressure,^9,10 but these associations are largely based on studies in populations exposed to moderate-to-high levels of these metals. The few studies that have examined low-level exposure, for example, in the United States, have been inconclusive.^9,10 Similarly, findings for cadmium and blood pressure have been inconsistent.¹¹ Additionally, these previous studies have been predominantly cross-sectional, suggesting a need to investigate the association between heavy metals and blood pressure, especially in a well-characterized prospective cohort study.

Women become more susceptible to cardio-metabolic diseases, including high blood pressure, in their midlife because women at this stage experience the menopausal transition marked by a dramatic shift in women’s sex hormone profile owing to permanent changes in the ovarian functions, which are further linked with increased risks of cardio-metabolic diseases in postmenopause.¹² Menopause is also considered a risk factor for oxidative stress attributed to the estrogen depletion over the menopausal transition.¹³ Due to this increased sensitivity, as a key environmental source of oxidative stress, heavy metals may be an important risk factor for high blood pressure, particularly for women during this life stage. However, the evidence of the associations between heavy metals and blood pressure in midlife women is extremely limited.

Thus, to inform the hypothesis that exposure to heavy metals may increase blood pressure prospectively, we investigated the associations of urinary concentrations of heavy metals including arsenic, cadmium, mercury, and lead with longitudinal changes in blood pressure over 17 years of follow-up in the Study of Women’s Health Across the Nation (SWAN).

Methods

SWAN provides access to public use datasets that include data from SWAN screening, the baseline visit and follow-up visits (https://agingresearchbiobank.nia.nih.gov/). To preserve participant confidentiality, some, but not all, of the data used for this manuscript are contained in the public use datasets. A link to the public use datasets is also located on the SWAN web site: http://www.swanstudy.org/swan-research/data-access/. Investigators who require assistance accessing the public use dataset may contact the SWAN Coordinating Center at the following email address: swanaccess@edc.pitt.edu.

Study population

SWAN is an on-going, multi-site, multi-ethnic, community-based prospective study of midlife women. The study goal is to assess psychosocial changes that occur during the menopausal transition and their effects on subsequent health endpoints. In 1996 and 1997, 3,302 women were enrolled from seven study sites: Boston, MA; Chicago, IL; Southeast Michigan; Los Angeles, CA; Newark, NJ; Oakland, CA; and Pittsburgh, PA. Each site enrolled White women and women from one minority group (Black women from Boston, Chicago, Southeast Michigan, and Pittsburgh; Chinese women from Oakland; Japanese women from Los Angeles; and Hispanic women from Newark). Eligibility criteria for enrollment into the SWAN cohort included the following: age 42 to 52 years, intact uterus and at least one ovary, no use of exogenous hormones affecting ovarian function in the past 3 months, at least one menstrual period in the previous 3 months, and self-identification with a site’s designated racial/ethnic groups. These women returned for regular approximately examinations annually with approximately 75% of still-living participants completed the 15^th SWAN follow-up visit (2015–2017). Institutional review board approval was obtained at each study site, and all participants provided signed informed consent at each study visit.

Participants in the current analysis were from the women enrolled in the SWAN Multi-Pollutant Substudy (SWAN-MPS),^14–16 which provided urine samples from the SWAN Repository collected at the third SWAN follow-up visit (V03, 1999–2000). Of 2,694 women enrolled at V03, we excluded women from Chicago (n = 368) and Newark (n = 278) because urine samples were not available in these two sites. We additionally excluded 648 women with insufficient urine samples at V03, yielding the 1400 women with urinary metal concentrations measured in the SWAN-MPS. For this analysis, we further excluded 65 participants who had no information on key covariates, and 18 participants with missing information on blood pressure throughout the follow-up, yielding a final analytic sample of 1,317 women with 12,637 observations from 1999 to 2017.

Blood pressure

This analysis considers two outcomes, systolic blood pressure (SBP) and diastolic blood pressure (DBP). At each in-person clinic visit, SBP and DBP were measured in the morning once the participants visited the study clinic after a 12-hour fast by trained and certified technicians according to a standardized protocol. The participant was asked to refrain from smoking or consuming any caffeinated beverage within 30 minutes of blood pressure measurement. Readings were taken on the right arm, with the participant seated and feet flat on the floor for at least 5 minutes before measurement. Appropriate cuff size was determined based on arm circumference. A standard mercury sphygmomanometer was used to record readings at the first and fifth phase Korotkoff sounds. Two sequential measurements were taken with a minimum 2-minute rest period between measurements. The average of the two SPB and DBP measures was used in this analysis.

Heavy metals

At the follow-up visit in 1999–2000, SWAN participants provided a morning spontaneously voided urine sample. These samples were sent to the Applied Research Center of NSF International (Ann Arbor, Michigan) for metal assessments. Urinary concentrations of arsenic, cadmium, mercury, and lead were measured using high-resolution inductively coupled plasma-mass spectrometry (ICP-MS) (Thermo Scientific iCAP RQ, Waltham, MA). The laboratory methods and quality control procedures have been described previously.¹⁴ The detection rate was 100% for arsenic, 94.3% for cadmium, 99.8% for mercury, and 97.8% for lead. For participants with metal concentrations below the limit of detection, a value equal to the limit of detection divided by the square root of 2 was assigned.

Covariates

Information on age, self-defined race/ethnicity (White, Black, Chinese, and Japanese), study site (southeast Michigan, Boston, Oakland, Los Angeles, and Pittsburgh), education (high school or less, some college, and college degree or higher), smoking status (never smoked, former smoker, and current smoker), alcohol drinking (<1 drink/month, >1 drink/month and ≤1/week, and >1 drink/week), physical activity, and menopausal status (premenopausal, postmenopausal, and unknown due to hormone therapy use) were obtained from standardized interviews. Physical activity was assessed based on a modified version of the Kaiser Physical Activity Survey and summarized into a score ranging from 3 to 15 with 15 indicating the highest level of activity.¹⁷ Body mass index (BMI) was calculated as measured weight in kilograms divided by the square of measured height in meters. Dietary seafood and rice intake was collected using a detailed semi-quantitative food frequency questionnaire (FFQ) adopted from the Block FFQ.¹⁴ Urinary specific gravity was determined using a handheld digital refractometer (ATAGO model PAL-10S, Tokyo, Japan) as a marker of urine dilution.

Statistical analysis

We used linear mixed effects models with random intercepts to estimate longitudinal changes in SBP and DBP in relation to each tertile of urinary metal concentration. We added a constant of 10 mmHg and 5 mmHg to participants’ SBP and DBP, respectively, at timepoints when they reported current use of anti-hypertensive medications, according to an established method to adjust for medication use.¹⁸ Urinary metal concentrations were categorized into tertiles. To test whether metals affected the rate of change in blood pressure over time, multiplicative interaction terms between time elapsed from the SWAN-MPS baseline and tertiles of metal concentrations were included in each model. Time was modeled using a linear term of year. We decided not to add a quadratic term of time into the model or use linear splines of time due to the worse model fitting performance based on the Bayesian information criterion. The model is shown as

$$B{P}_{ij}={\beta }_0+{\beta }_1{\mathit{\text{Metal}}}_{\mathit{\text{Tertile}}{2}_i}+{\beta }_2{\mathit{\text{Metal}}}_{\mathit{\text{Tertile}}{3}_i}+{\beta }_{3}Tim{e}_{ij}+{\beta }_{4}Tim{e}_{ij}\times {\mathit{\text{Metal}}}_{\mathit{\text{Tertile}}{2}_i}+{\beta }_{5}Tim{e}_{ij}\times {\mathit{\text{Metal}}}_{\mathit{\text{Tertile}}{3}_i}+b_{0j}+\beta {\mathit{\text{Covariates}}}_{ij}$$

where BP_ij represents SBP or DBP for the i^th participant with the j^th observation; ${\mathit{\text{Metal}}}_{\mathit{\text{Tertile}}{2}_i}$ and ${\mathit{\text{Metal}}}_{\mathit{\text{Tertile}}{3}_i}$ represents the second and third tertiles of metal concentrations, respectively, comparing with the first tertile (reference group). Time_ij indicates for follow-up time (time-on-study, in years). b_0j is the random intercept. The annualized rate of changes in blood pressure equals β₃, β₃ + β₄, and β₃ + β₅ in the first, second, and third tertile of the metal concentration, respectively. We adjusted P values for multiple comparisons at a false discovery rate (FDR) of 0.05 using the Benjamini–Hochberg Method.¹⁹

Potential confounders were adjusted progressively. The base model included age (baseline), race/ethnicity (baseline), study site (baseline), and specific gravity (log-transformed) (baseline). The same terms were included in the fully adjusted model along with education level (baseline), BMI (baseline), smoking (time-varying), alcohol drinking (time-varying), physical activity (time-varying), and menopausal status (time-varying). We also adjusted for seafood and rice intake at baseline in the full model as we have identified these dietary components to be important determinants of heavy metal concentrations in a previous study.¹⁴ Multiplicative interaction terms between race/ethnicity and follow-up time was also included in the full model for accounting for its potential impact on longitudinal changes in blood pressures.²⁰ BMI was not included as a time-varying covariate in our analysis because of its role as a risk factor for high blood pressure and the fact that it could be affected by metal exposures at baseline.²¹ As smoking is considered a primary source of cadmium for the general population and as the association between cadmium and blood pressure could be complicated by smoking status,²² we further evaluated the association of cadmium with blood pressure stratified by smoking status (never smoker vs. former or current smoker) at baseline. To quantify differences in longitudinal changes in blood pressure associated with exposure to metal mixtures, we also included tertiles of all four metals and their interaction terms with follow-up time simultaneously in a final adjusted model.

Several sensitivity analyses were conducted to evaluate the robustness of our findings. First, because adjustment for seafood intake may not sufficiently control for the less toxic organic arsenic in evaluation of the association between arsenic and blood pressure,²³ we evaluated the association in a subpopulation with seafood intake less than 1.5 times/week (median level). Second, menopause has been suggested to play an important role in the mobilization of lead from bone into the circulation due to an increased bone turnover rate.²⁴ Participants who were approaching menopause at the time when metal exposure was assessed may have elevated urinary lead concentrations due to the mobilization of lead from bone,²⁵ thus the observed associations could be confounded by increased bone resorption. As a sensitivity analysis, we used urinary N-telopeptide (NTX) adjusted urinary lead concentration. Finally, we used raw values of SBP and DBP and adjusted for use of anti-hypertensive medications in regression analyses. All analyses were conducted using R, version 4.0.2 (www.R-project.org).

Results

The median age of the 1,317 women included in the current analysis was 49.4 years with an interquartile range (IQR) of 47.4 to 51.5 years at baseline (Table S1). Of those, 50.2% were White, 21.5% were Black, 12.9% were Chinese, and 15.4% were Japanese. Most women had never smoked (63.2%) and were pre-menopausal (70.5%) at SWAN MPS baseline. The medians (IQRs) of blood pressure at baseline were 112 (103, 125) mmHg for SBP and 73 (67, 81) mmHg for DBP. The distributions of urinary metal concentrations are shown in Table S2.

Baseline SBP did not differ by tertiles of metals, except for mercury, where SBP levels were higher for participants in the lowest tertile (Table S3). As shown in Table 1, SBP consistently increased with time, as was observed with a positive annualized change estimate in each tertile for each metal examined. Statistically significantly higher rates of increase in SBP were observed in higher tertiles of arsenic (P for interaction=0.0003), mercury (P for interaction=0.01), and lead (P for interaction=0.007). To adjust for multiple comparisons, a significance level of α=0.01 was used, which corresponded to an FDR of 5% using the Benjamini–Hochberg Method. The estimated annualized increase in SBP was 0.93 (95% CI: 0.85, 1.01) mmHg/year in the highest and 0.74 (95% CI: 0.66, 0.82) mmHg/year in the lowest tertile of arsenic; 0.82 (95% CI: 0.75, 0.90) mmHg/year in the highest and 0.72 (95% CI: 0.65, 0.80) mmHg/year in the lowest tertile of mercury; and 0.86 (95% CI: 0.78, 0.93) mmHg/year in the highest and 0.72 (95% CI: 0.64, 0.79) mmHg/year in the lowest tertile of lead. At the end of the follow-up period when participants had reached ages between 62 and 73 years, the mean SBP was estimated to be 126.2 (95% CI: 124.2, 128.1) mmHg in the highest and 124.3 (95% CI: 122.4, 126.2) mmHg in the lowest tertile of arsenic; 123.6 (95% CI: 121.7, 125.6) mmHg in the highest and 125.3 (95%: 123.3, 127.2) mmHg in the lowest tertile of mercury; and 125.7 (95% CI: 123.6, 127.7) mmHg in the highest and 123.1 (95% CI: 121.2, 125.0) mmHg in the lowest tertile of lead.

Table 1.

Estimated annualized changes in systolic blood pressure (mmHg) by tertiles of urinary heavy metal concentrations.

Tertiles of metal concentration	Annual change (95% CI) in systolic blood pressure (mmHg)			P for interaction
	Tertile 1	Tertile 2	Tertile 3
Arsenic
Base model*	0.68 (0.61, 0.75)	0.70 (0.63, 0.77)	0.87 (0.80, 0.94)	0.0004
Full model†	0.74 (0.66, 0.82)	0.76 (0.68, 0.83)	0.93 (0.85, 1.01)	0.0003
Cadmium
Base model*	0.70 (0.63, 0.77)	0.75 (0.69, 0.82)	0.80 (0.73, 0.87)	0.13
Full model†	0.75 (0.68, 0.83)	0.81 (0.73, 0.89)	0.85 (0.78, 0.93)	0.12
Mercury
Base model*	0.67 (0.60, 0.74)	0.81 (0.75, 0.88)	0.77 (0.70, 0.84)	0.01
Full model†	0.72 (0.65, 0.80)	0.87 (0.79, 0.94)	0.82 (0.75, 0.90)	0.01
Lead
Base model*	0.66 (0.59, 0.73)	0.79 (0.72, 0.85)	0.80 (0.74, 0.87)	0.006
Full model†	0.72 (0.64, 0.79)	0.84 (0.77, 0.92)	0.86 (0.78, 0.93)	0.007

^*Adjusted for age at baseline, race/ethnicity, study site, specific gravity (log-transformed).

^†Adjusted for age at baseline, race/ethnicity, study site, specific gravity (log-transformed), education, smoking status, alcohol drinking, physical activity, body mass index at baseline, menopausal status, seafood and rice intake at baseline, and interaction term between race/ethnicity and follow-up time.

DBP at baseline did not differ across tertiles of metals (Table S4). Similar to SBP, a positive annualized change estimate in each tertile for each metal examined was observed for DBP (Table 2), with the rates of increase greater in the higher tertiles of arsenic (P for interaction=0.004), mercury (P for interaction=0.0004), and lead (P for interaction=0.004). A significance level of α=0.004 was used, corresponding to an FDR of 5%. In the fully adjusted model, the estimated annualized increase in DBP was 0.20 (95% CI: 0.15, 0.25) mmHg/year in the highest and 0.09 (95% CI: 0.04, 0.14) mmHg/year in the lowest tertile of arsenic; 0.17 (95% CI: 0.13, 0.22) mmHg/year in the highest and 0.05 (95% CI: 0.01, 0.11) mmHg/year in the lowest tertile of mercury; and 0.18 (95% CI: 0.13, 0.23) mmHg/year in the highest and 0.07 (95% CI: 0.03, 0.12) mmHg/year in the lowest tertile of lead. At the end of the follow-up period, the mean DBP was estimated to be 76.0 (95% CI: 74.9, 77.2) mmHg in the highest and 74.7 (95%: 73.6, 75.9) mmHg in the lowest tertile of arsenic; 75.4 (95% CI: 74.2, 76.5) mmHg in the highest and 74.9 (95% CI: 73.7, 76.1) mmHg in the lowest tertile of mercury; and 76.1 (95% CI: 74.9, 77.3) mmHg in the highest and 74.0 (95% CI: 72.9, 75.2) mmHg in the lowest tertile of lead.

Table 2.

Estimated annualized changes in diastolic blood pressure (mmHg) by tertiles of urinary heavy metal concentrations.

Tertiles of metal concentration	Annualized change (95% CI) in diastolic blood pressure (mmHg)			P for interaction
	Tertile 1	Tertile 2	Tertile 3
Arsenic
Base model*	0.08 (0.04, 0.13)	0.10 (0.05, 0.14)	0.19 (0.14, 0.23)	0.004
Full model†	0.09 (0.04, 0.14)	0.11 (0.06, 0.16)	0.20 (0.15, 0.25)	0.004
Cadmium
Base model*	0.11 (0.07, 0.16)	0.12 (0.08, 0.17)	0.13 (0.09, 0.18)	0.80
Full model†	0.12 (0.07, 0.17)	0.13 (0.08, 0.18)	0.14 (0.09, 0.19)	0.78
Mercury
Base model*	0.05 (0, 0.09)	0.15 (0.11, 0.20)	0.17 (0.12, 0.21)	0.0003
Full model†	0.05 (0.01, 0.11)	0.16 (0.11, 0.21)	0.17 (0.13, 0.22)	0.0004
Lead
Base model*	0.07 (0.02, 0.11)	0.13 (0.08, 0.17)	0.17 (0.13, 0.22)	0.004
Full model†	0.07 (0.03, 0.12)	0.14 (0.09, 0.18)	0.18 (0.13, 0.23)	0.004

^*Adjusted for age at baseline, race/ethnicity, study site, specific gravity (log-transformed).

^†Adjusted for age at baseline, race/ethnicity, study site, specific gravity (log-transformed), education, smoking status, alcohol drinking, physical activity, body mass index at baseline, menopausal status, seafood and rice intake at baseline, and interaction term between race/ethnicity and follow-up time.

For cadmium, we conducted stratified analyses by smoking status (Table S5). Significantly greater annualized increases in SBP were observed in upper tertiles of cadmium only among never smokers (P for interaction=0.03). In never smokers, estimated annualized increase in SBP was 0.89 (95% CI: 0.80, 0.97) mmHg/year in the highest and 0.74 (95% CI: 0.64, 0.83) mmHg/year in the lowest tertile. No association between cadmium and DBP was observed.

In the model simultaneously incorporating four metals, higher rates of increases in both SBP and DBP were observed when a participant had higher concentrations of all four metals (Table 3 and Table S6). After adjustment for potential confounders, the estimated annualized increase in SBP was 0.91 (95% CI: 0.81, 1.01) mmHg/year in the highest and 0.67 (95% CI: 0.57, 0.76) mmHg/year in the lowest tertiles of all four metals. The estimated annualized increase in DBP was 0.19 (95% CI: 0.13, 0.26) mmHg/year in the highest and 0.07 (95% CI: 0.01, 0.13) mmHg/year in the lowest tertiles of all four metals. The associations for each metal from these multi-metal models are shown in Table S7.

Table 3.

Estimated annualized changes in systolic and diastolic blood pressures (mmHg) by tertiles of urinary metal concentrations in the model incorporating all metals.

Tertiles of all metal concentrations	Annualized change (95% CI) in blood pressure (mmHg)*			P for interaction‖
	Tertile 1†	Tertile 2‡	Tertile 3§
Systolic blood pressure	0.67 (0.57, 0.76)	0.84 (0.73, 0.95)	0.91 (0.81, 1.01)	0.06
Diastolic blood pressure	0.07 (0.01, 0.13)	0.13 (0.05, 0.20)	0.19 (0.13, 0.26)	0.0003

^*Models were adjusted for age at baseline, race/ethnicity, study site, specific gravity (log-transformed), education, smoking status, alcohol drinking, physical activity, body mass index at baseline, menopausal status, seafood and rice intake at baseline, and interaction term between race/ethnicity and follow-up time.

^†Annualized changes in blood pressure when arsenic, cadmium, mercury, and lead concentrations were at their tertile 1.

^‡Annualized changes in blood pressure when arsenic, cadmium, mercury, and lead concentrations were at their tertile 2.

^§Annualized changes in blood pressure when arsenic, cadmium, mercury, and lead concentrations were at their tertile 3.

^‖P for interaction was calculated using likelihood ratio test between the model incorporating interaction terms between tertiles of metals and follow-up time and the model without interaction terms.

In the sensitivity analysis, associations between urinary arsenic and blood pressure were evaluated in a subpopulation of 657 participants with seafood intake less than 1.5 times/week. In this subpopulation, higher rates of increase in SBP were observed in the participants with higher arsenic concentrations, however, the interaction terms between arsenic and time were no longer statistically significant (Table S8), possibly due to a reduced statistical power in accordance with the smaller sample size in this analysis. In the sensitivity analysis assessing associations between urinary NTX adjusted lead and blood pressure, similar associations were observed as in the primary analysis (Table S9). Similar associations were also observed when raw values of SBP and DBP were used after adjustment for use of anti-hypertensive medications in regression models (Table S10).

Discussion

This study examined associations between heavy metal exposures and longitudinal change in blood pressure in midlife women as they aged through the menopausal transition. In this multi-site, multi-ethnic cohort study of women at midlife, urinary concentrations of arsenic, mercury, and lead were associated with a faster rate of increase in both SBP and DBP. Urinary cadmium was associated with a greater rate of increase in SBP among never smokers only. Higher concentrations of all heavy metals were also associated with greater annualized increases in both SBP and DBP.

Even a small increase in blood pressure associated with heavy metal exposures may have substantial public health consequences, given the widespread exposure to metals and the predominance of high blood pressure related health outcomes in the U.S. For example, across the 17 years of follow-up, the mean SBP in the highest tertile of arsenic climbed by 15.8 mmHg, while SBP increased by 12.6 mmHg in the lowest tertile. These elevations in blood pressure are associated with higher risks of cardiovascular disease, stroke, and mortality. Population-based studies suggest that a 1 mmHg increase in SBP is associated with a 1% increase in age-adjusted stroke mortality²⁶ and a 2–4% increase in cardiovascular mortality.²⁷ Thus, the impact of metal exposure on blood pressure differences translates to significant differences in long-term cardiovascular endpoints. Our findings suggest that heavy metals may be an underappreciated contributing factor to accelerated deterioration of cardiovascular health, especially during the potentially sensitive period of midlife.

Arsenic is a ubiquitous industrial and naturally occurring environmental toxicant. Inorganic arsenic is a toxicant and its primary sources are drinking water and certain foods including rice.¹⁴ After absorption from the gastrointestinal tract, inorganic arsenic is rapidly excreted together with its metabolites primarily through the urine. A systematic review including 11 cross-sectional studies, mostly conducted in areas with high exposure to arsenic in drinking water such as Bangladesh and Taiwan, showed a pooled odds ratio of 1.27 (95% CI: 1.09, 1.47) for high blood pressure when comparing the highest versus lowest category of inorganic arsenic in drinking water.⁹ In contrast, arsenic exposure at the low-to-moderate levels typical of the U.S. general population was not associated with blood pressure or prevalence of hypertension.²⁸ Our study, leveraging a prospective cohort design, found a positive association between arsenic and longitudinal changes in SBP, suggesting arsenic exposure at low-to-moderate levels in the United States may also incur adverse effects.¹⁴ Experimental studies suggest that arsenic may elevate blood pressure via pathways related to inflammation, oxidative stress, and endothelial dysfunction.⁹ Arsenic has also been shown to disrupt renal function, leading to an increased risk of high blood pressure.²⁹

Mercury is a ubiquitous and persistent toxicant that has different forms.³⁰ General populations can be exposed to methylmercury (organic form) primarily through seafood and be exposed to inorganic mercury through occlusal surfaces of teeth that are filled with mercury-containing amalgams³¹ and skin lightening products with inorganic mercury compounds.³² In humans, urinary mercury mainly reflects inorganic mercury.³⁰ It has also been positively associated with seafood intake in SWAN.¹⁴ Epidemiologic studies in population settings and occupational cohorts have produced mixed results. A recent meta-analysis based on 29 studies reported a borderline positive association between mercury and hypertension (pooled OR=1.35, 95% CI: 0.99, 1.83, comparing the highest and lowest categories of mercury concentrations measured in hair, blood, or urine) for populations with relative high exposures.¹⁰ In contrast, no significant association was observed for populations with relatively low exposures (pooled OR=1.12, 95% CI: 0.82, 1.52).¹⁰ We demonstrated an association of mercury, particularly its inorganic from, with accelerated increases in SBP and DBP. In animal models, acute or chronic administration of both inorganic mercury and methylmercury have been shown to induce hypertension.¹⁰ Possible mechanisms underlying mercury’s effect include oxidative stress, inflammation, changes in vascular smooth muscle, endothelial and renal dysfunction, dyslipidemia, and immune and mitochondrial dysfunction.³³ In our study, an inverse association was observed between urinary mercury and SBP at baseline, even after further adjustment for seafood intake. To the best of our knowledge, no biological evidence supports mercury itself being beneficial for blood pressure control. The observed inverse association at baseline may indicate that mercury concentration is a surrogate measure of dietary seafood intake in our population.¹⁴

Lead toxicity is acknowledged as a prevalent and persistent public health problem. In adults, more than 95% of the total lead body burden is found in bone.³⁴ Bone lead is, therefore, considered a biomarker of cumulative lead exposure.^35–37 Bone lead has been consistently associated with high blood pressure.⁷ Bone lead undergoes constant interchange with lead in the circulation and soft tissue. Urinary lead adjusted for urine dilution rather than whole blood lead has been found to closely reflect lead mobilized from the bone.²⁵ In our study, urinary lead was associated with a faster rate of increase in both SBP and DBP. Consistent results were found when urinary NTX-adjusted lead concentrations were used in the sensitivity analysis. These findings support the hypothesis that long-term lead accumulation may increase blood pressure, particularly in midlife for women, attributed in part to a greater mobilization of lead from bone into the circulation during the menopausal transition. Our finding is also consistent with the association between lead and blood pressure in postmenopausal women from the U.S. general population.³⁸ Impaired renin-angiotensin system function, oxidative stress, inflammation, and neurotoxic effects interfering with autonomic nervous system function may serve as potential mechanisms underlying the association between lead and blood pressure.⁷

A positive association between urinary cadmium and annual rate of change in SBP was observed among women who never smoked, which accords with previous findings. A stronger association between blood cadmium and blood pressure among never smokers compared to former/current smokers has been observed in U.S. adults using data from NHANES 1999–2004.³⁹ The positive association between urinary cadmium and blood pressure has also been found among never smokers but not former/current smokers in a recent study of American Indian adults.²² Importantly, as smoking is a primary source of cadmium, we did observe higher urinary concentrations of cadmium in smokers in SWAN.¹⁴ Given this, our finding suggests that the association between cadmium and blood pressure is unrelated to other toxicants in cigarette smoke. Alternatively, it is possible that other sources of cadmium exposure exist in our cohort. Co-exposure to other toxicants which are correlated with cadmium in those sources may also contribute to the elevation in SBP. More investigations into the potential sources of cadmium exposure and the association between cadmium and blood pressure in populations with diverging context of exposure levels are needed in the future.

The primary strength of the current study is that is used a large cohort of community-based midlife women from four racial/ethnic groups and followed them for up to 17 years. To our knowledge, this is the first investigation of associations of heavy metal exposures with trajectories of blood pressure in midlife women. Standard annualized measures also provided reliable estimates of changes in blood pressure and ensure temporality between exposure and outcome.

Our study also has several limitations. First, we measured metal concentrations in urine and urinary concentrations may not consistently reflect metals in all forms and from all exposure sources. Second, urinary metal concentrations were measured at a single time point at the SWAN-MPS baseline. Existing evidence demonstrates the long-term constancy in arsenic exposure.⁴⁰ Cadmium is not rapidly excreted and has half-lives of years to decades, and urinary cadmium has suggested a biomarker for long-term exposure. Urinary lead adjusted for urine dilution has been found to closely reflect lead mobilized from the bone,²⁵ where bone lead is a biomarker of long-term exposure and has been suggested as an endogenous and primary source of lead in mid-and late life.⁴¹ Third, only total arsenic was measured in urine samples and information on arsenic metabolism was not available. Recent studies suggested that the pattern of arsenic metabolisms including monomethylated and dimethylated compounds was associated with cardiovascular disease and diabetes risks.⁴² Additional measurements of arsenic metabolism would improve our understanding of arsenic exposure and its associated health risks. Finally, we cannot eliminate residual confounding due to the observational nature of the study, although we have controlled for many known confounders.

Perspectives

In this prospective cohort study, we observed that higher urinary concentrations of arsenic, mercury, and lead accelerated the rate of elevation of blood pressure in midlife women, during the menopausal transition, a life stage when women are at greater risk of cardio-metabolic diseases. Urinary cadmium was associated with accelerated elevation of SBP among never smokers. High exposure to all heavy metals was also associated with greater elevation of blood pressure. Our findings have potentially important public health implications. They suggest that heavy metal exposure may exacerbate blood pressure profiles. Continued efforts to reduce heavy metal exposures are needed, especially in midlife women.

Novelty and Significance.

What is New?

• This is the first prospective cohort study of associations of heavy metal exposures with trajectories of blood pressure in midlife women.

What is Relevant?

• Women become more susceptible to cardio-metabolic diseases. Exposure to heavy metals including arsenic, cadmium, mercury, and lead may accelerate the increase in blood pressure during midlife for women.

Summary

• In the SWAN, higher urinary concentrations of arsenic, mercury, and lead accelerated the elevation of blood pressure in midlife women. Urinary cadmium was associated with accelerated elevation of SBP among never smokers. Thus, this analysis of SWAN suggests that heavy metal exposure may exacerbate blood pressure profiles in midlife women.