Health & Medical Environmental

Arsenic Exposure and Blood Pressure Changes in Pregnancy

Arsenic Exposure and Blood Pressure Changes in Pregnancy

Discussion


To our knowledge, our study is the first prospective study to examine the association between arsenic and BP in the context of pregnancy and among the few studies on the cardiovascular effects of arsenic exposure in the United States. Because pregnancy is a vulnerable window of susceptibility to adverse BP changes, by focusing on a cohort of pregnant women we found that higher levels of urinary arsenic during pregnancy prospectively related to greater increases in SBP and PP over the course of pregnancy.

Arsenic has been associated with a range of cardiovascular outcomes in populations with appreciable levels of chronic exposure, such as in Bangladesh and Taiwan, including increased risks of fatal and nonfatal cardiovascular disease, as well as intermediary factors, such as increased carotid intima-media thickness and metabolic syndrome (Chen CJ et al. 1995; Chen Y et al. 2007a, 2011, 2013; Kwok et al. 2007; Wang et al. 2007; Wu et al. 1989). Arsenic has been associated with hypertension in a number of cross-sectional studies, from which a meta-analysis derived a pooled odds ratio for hypertension of 1.27 (95% CI: 1.09, 1.47) for high versus low arsenic exposure (Abhyankar et al. 2012). Although potential causal mechanisms for the association between arsenic and BP increase during pregnancy have not yet been explored, many of the mechanisms hypothesized to explain associations with other cardiovascular outcomes could be involved. Arsenic exposure has been related to increased plasma markers of inflammation and endothelial damage (Burgess et al. 2013; Chen et al. 2007b; Wu et al. 2012), suggesting that arsenic may act in part by promoting endothelial dysfunction, pathologic vascular remodeling, and atherosclerosis. Thus, while speculative, arsenic exposure could impact the pregnancy-related hemodynamic adaptations that increase blood volume and maintain placental perfusion, which is critical to fetal nutrient and oxygen supply.

BP normally increases toward the latter part of pregnancy, with increases in SBP generally tending to be somewhat more pronounced than those in DBP (Cunningham et al. 2010; Miller et al. 2007; Thompson et al. 2007). A prospective study of longitudinal BP during pregnancy reported average increases of about 3.7 mmHg and 2.2 mmHg between the first and third trimesters for SBP and DBP, respectively (Miller et al. 2007). Abnormal increases pose a serious risk of complications during pregnancy such as preterm birth, low birth weight, fetal growth restriction, and perinatal mortality (Ray et al. 2001; Xiong and Fraser 2004; Zhang et al. 2007) and the deleterious effects of gestational hypertension (defined as new onset of SBP > 140 mmHg and/or DBP > 90 mmHg in second trimester) are well known. However, elevations in BP that do not exceed the upper threshold of the normal range (SBP < 140 mmHg and DBP < 90 mmHg) may also pose risks to the mother and child. For nonpregnant adults, the risk of cardiovascular disease increases linearly as BP increases, even within the normotensive range (Vasan et al. 2001; Williams et al. 2008). A few studies have examined BP as a continuous measure and found that higher BP, even within the normotensive range, also may impact birth weight and intrauterine growth restriction (Churchill et al. 1997; Fukushima et al. 2012). It is possible that elevated BP, albeit within the clinically normal range, alters uterine and placental perfusion, and impacts fetal growth. Our results suggest that there were greater increases in SBP and PP over pregnancy associated with higher arsenic exposure, leading to greater relative differences at the end of pregnancy. However, the clinical significance of greater increases in BP remains to be explored, and more studies utilizing continuous BP outcome measures are needed to examine the relation between BP elevations within the normal range and health risks.

Pregnancy itself is a cardiovascular stressor. In a rodent study, normal, healthy pregnancies were found to induce long-term alterations in cardiovascular and renal function that were absent in nonparous females (Gallo et al. 2012). Pregnancy-induced hypertension has been associated with increased later life risk of chronic hypertension, endothelial dysfunction, and kidney disease (Henriques et al. 2014; Nisell et al. 1995; Vikse et al. 2008; Wang et al. 2013; Wilson et al. 2003). According to a recent study, women with a history of a hypertensive pregnancy had nearly 60% greater odds of peripheral artery disease compared with those with normotensive histories, even decades after pregnancy (Weissgerber et al. 2013). Additional longitudinal studies are needed to determine whether BP changes during pregnancy, such as those observed in relation to arsenic exposure in our cohort, lead to long-term health consequences for mother and child.

In our study, we found that each 5 μg/L urinary arsenic was associated with an average SBP increase of 0.15 mmHg per month and a 0.78-mmHg (95% CI: 0.05, 1.51; p = 0.035) higher SBP. Although we are unaware of any previous studies of arsenic and BP during pregnancy, recent studies have found that exposure to other environmental contaminants may affect BP during pregnancy with similar magnitudes of effects as observed in our study. Several studies have observed associations between particulate air pollution and increased BP in pregnant women (Lee et al. 2012; van den Hooven et al. 2011), including a prospective study of 431 pregnant women that found third-trimester SBP increased linearly with second-trimester exposure to air particulates (Jedrychowski et al. 2012). The Generation R Study found that a 10-μg/m increase in PM10 exposure was associated with greater increases in SBP over the second and third trimesters: 1.11 (95% CI: 0.43, 1.79) and 2.11 (95% CI: 1.34, 2.89) mmHg, respectively (van den Hooven et al. 2011). A recent U.S. cohort study of air pollution on BP changes over the course of pregnancy found that interquartile increases in PM10 (particulate matter ≤ 10 μm in aerodynamic diameter) and ozone exposure in the first trimester were associated with average SBP increases of 1.9 mmHg (95% CI: 0.84, 2.93) and 1.8 mmHg (95% CI: 1.05, 4.63), respectively, an association that was more pronounced in nonsmoking mothers (Lee et al. 2012). In addition, a cohort study of 1,017 pregnant women in France found an association between midpregnancy blood lead levels and increased risk of pregnancy-induced hypertension in the second and third trimesters (Yazbeck et al. 2009). Although studies of the impacts of environmental toxicants on cardiovascular effects during pregnancy are increasing, more studies are needed to assess the vulnerable times of exposure, as well as the effects of toxicants known to increase cardiovascular disease in nonpregnant adults, including arsenic.

Ingested inorganic arsenic is primarily metabolized via methylation, first to MMA, then to DMA. Arsenic metabolism varies greatly between individuals, and higher MMA proportions are indicative of inefficient methylation (Buchet et al. 1981; Vahter 1999). MMA, thought to be a more toxic metabolite, has been linked to adverse health effects, including cardiovascular effects (Chen et al. 2013; Huang et al. 2009). Because previous work from more highly exposed individuals has indicated that higher PMI may be associated with greater health risks (Chen et al. 2013), one might expect to see stronger effects only in those with high PMI, which could indicate inefficient arsenic metabolism, as opposed to high SMI, which may indicate more efficient methylation and therefore arsenic excretion. However, we observed associations between urinary arsenic and BP both among those with higher PMI or higher SMI, although differences may have occurred by chance. In populations with lower overall levels of exposure, one might predict that the majority of ingested arsenic, once methylated to MMA, would be more easily methylated to DMA. This prediction is consistent with our observations, as well as with those in other U.S. populations, including recent results from the Strong Heart Study, which indicated that higher DMA proportions were linked to cardiovascular disease incidence and mortality, raising the possibility for a role of higher SMI in cardiovascular risk in populations with low arsenic exposure levels (Moon et al. 2013). A low SMI may be a susceptibility factor in more highly exposed populations, such as in Bangladesh. Further, the pregnancy-related health outcomes related to high SMI (i.e., high DMA levels) are less well understood. It is possible that women with altered arsenic metabolism may be more susceptible to arsenic's cardiovascular effects and more likely to experience increases in BP during pregnancy. Interestingly, in late pregnancy, a greater proportion of arsenic is excreted as MMA (Concha et al. 1998; Hopenhayn et al. 2003), possibly representing a detoxification mechanism. Although this pregnancy-related alteration in metabolism is not well understood, it is possible that this mechanism may in part account for the observed association between increased BP in association with both PMI and SMI. Further study of the effect modification by arsenic metabolites is warranted, particularly at the lower levels of arsenic exposure found in U.S. populations.

Urinary arsenic is considered to be a reliable short-term measure of arsenic exposure that appears to remain relatively consistent in adults, even during pregnancy (Ahmed et al. 2011; Gamble et al. 2006). In the present study, we collected urine samples over a narrow gestational time frame, during which concentrations were previously found not to vary (Gilbert-Diamond et al. 2011). To examine the trajectory of BP over pregnancy, we used measurements beginning at 13 weeks gestation; thus, some measurements were taken prior to urine sampling. However, prior studies suggest that total urinary arsenic levels remain relatively constant over pregnancy (Ahmed et al. 2011). However, our single exposure measurement may not be representative of typical exposure levels for all of the women in our study sample, and there may be variability in arsenic exposure levels that we were unable to account for in this study. Further, the study by Gamble et al. (2006) was performed in adults, and urinary arsenic stability may vary between nonpregnant and pregnant adults. Although we did not collect multiple urine samples from participants, we collected maternal toenail samples prior to delivery, which approximately represent the previous 6–9 months of exposure. Among 334 women in our study with both prenatal urinary and toenail arsenic measurements, toenail arsenic was positively correlated with urinary arsenic measurements (r = 0.33, p < 0.001; data not shown). Moreover, use of urine as an arsenic biomarker allows us to account for exposure from other sources, such as diet. Nearly 1 in 8 individuals (12.5%) in this sample had water arsenic levels that exceeded the U.S. EPA maximum contaminant limit of 10 μg/L, which likely represents the primary source of arsenic exposure among these individuals. Further, work from our study area of New Hampshire has found that a variety of foods, including rice, can also significantly increase an individual's arsenic exposure (Cottingham et al. 2013; Gilbert-Diamond et al. 2011).

Our study has some potential limitations. First, we used measurements of BP at prenatal care visits, obtained from medical records. These measurements reflect the types of measurements and patterns that are obtained in routine clinical settings; although standard medical procedures were used, differences in staff and instrumentation may have introduced random variability into our measurements. In addition, BP can fluctuate acutely in relation to anxiety, recent exertion, and caffeine consumption, contributing to measurement error. Although we were not able to account for these factors in our models, we would not expect instrumentation to be related to exposure status and error in the precision of measurement techniques that would likely bias our estimates toward the null. We also were unable to account for dietary factors (i.e., high sodium consumption, nutrient levels) that have the potential to impact BP levels, and due to sample size, we may have been limited in our ability to examine the impact of effect modifiers, such as age or BMI. Our study population of mothers tended to be well-educated and primarily white, which may underrepresent different racial or socioeconomic groups that are at higher risk of gestational hypertension. Nonetheless, internal validity of the study is strengthened by the fact that we have multiple measurements for each woman over the course of pregnancy, detailed medical history, and sociodemographic information from our participants to include in our models. However, some women in our study were missing covariate information. We used multiple imputation methods to impute missing data, and we cannot rule out the possibility that data were not missing completely at random. Further, our choice of mixed models helps to account for random variability. Longitudinal data analysis provides a sensitive tool for characterizing health outcomes that change gradually, such as BP, and repeated measures can be a powerful way to identify small changes that can have a large impact at the population level (Farrington 1991). BP has a strong, continuous positive association with cardiovascular disease (Law et al. 2003; MacMahon et al. 1990; Sagie et al. 1993), and as SBP increases above 115 mmHg, the risk of cardiovascular disease rises continuously (Vasan et al. 2001; Williams et al. 2008). Therefore, the changes observed here have the potential to impact maternal cardiovascular risks (Law et al. 2003).

It is becoming increasingly evident that pregnant women and developing fetuses are particularly vulnerable to environmental insults. Inorganic arsenic consumed in both drinking water and diet may contribute to overall arsenic burden in U.S. pregnant women. Although the adverse cardiovascular effects of arsenic have been investigated in adults, to our knowledge, our study is among the first to examine these impacts during pregnancy. As cardiovascular morbidity and mortality rise worldwide, the potential risk of later-life cardiovascular diseases in mothers and children who are exposed to arsenic during pregnancy makes this a critical area of investigation.

You might also like on "Health & Medical"

Leave a reply